Hydrogenfeul

IAEA-TECDOC-1085
Hydrogen as an energy
carrier and its production
by nuclear power
INTERNATIONAL ATOMIC ENERGY AGENCY /A\

ULr\±
May 1999
30-24
The originating Section of this publication in the IAEA was:
Nuclear Power Technology Development Section
International Atomic Energy Agency
Wagramer Strasse 5
P.O. Box 100
A-1400 Vienna, Austria
The IAEA does not normally maintain stocks of reports in this series.
However, copies of these reports on microfiche or in electronic form can be obtained from
INIS Clearinghouse
International Atomic Energy Agency
Wagramer Strasse 5
P.O. Box 100
A-1400 Vienna, Austria
E-mail: CHOUSE® IAEA.ORG
URL: http://www.iaea.org/programmes/inis/inis.htm
Orders should be accompanied by prepayment of Austrian Schillings 100,-

in the form of a cheque or in the form of IAEA microfiche service coupons
which may be ordered separately from the INIS Clearinghouse.
HYDROGEN AS AN ENERGY CARRIER AND ITS

PRODUCTION BY NUCLEAR POWER
IAEA, VIENNA, 1999
IAEA-TECDOC-1085
ISSN 1011-4289
© IAEA, 1999
Printed by the IAEA in Austria
May 1999
FOREWORD
The impact power generation has on human health and the environment is becoming
an ever increasing concern in the decision-making process when considering the energy
options and power systems required by a country in order to sustain its economic growth and
development. The continuing dependence on conventional fossil fuels and the deterioration
of air quality have spurred the need for clean and efficient power systems resulting from
the development of new technologies.
Hydrogen is a strong emerging candidate with a significant role as a clean, envi-

ronmentally benign and safe-to-handle major energy carrier of the future. Its enhanced
utilization in distributed power generation as well as in propulsion systems for mobile ap-
plications, such as automotive transportation, will help to significantly mitigate the strong
negative effects on the environment particularly prominent in the large population centers
of the world.
It is also nuclear power that will be of utmost importance in the energy supply of
many countries over the next few decades. Nearly 500 nuclear power plants are currently
being operated or are under construction around the world. The development of new,
innovative reactor concepts utilizing passive safety features for process heat and electricity
generation applications are considered by many to play a substantial role in the world's
energy future in helping to reduce greenhouse gas emissions.
This report was developed under a contract with the International Atomic Energy
Agency (IAEA). It documents past activities as well as those currently in progress by many
Member States in the development of hydrogen as an energy carrier and its corresponding
production through the use of nuclear power. It provides an introduction to nuclear
technology as a means of producing hydrogen or other upgraded fuels and to the energy
carrier hydrogen and its main fields of application. Emphasis is placed on high-temperature
reactor technology which can achieve the simultaneous generation of electricity and the
production of high-temperature process heat. This report was compiled and edited by Karl
Verfondern, Forschungszentrum Jiilich (FZJ), Jiilich, Germany, and was reviewed by the
experts named at the end of the report.
The IAEA officers responsible for this publication were L. Brey and G. Woite of the
Division of Nuclear Power.
EDITORIAL NOTE
In preparing this publication for press, staff of the IAEA have made up the pages from the
original manuscripts). The views expressed do not necessarily reflect those of the IAEA, the
governments of the nominating Member States or the nominating organizations.
Throughout the text names of Member States are retained as they were when the text was
compiled.
The use of particular designations of countries or territories does not imply any judgement by
the publisher, the IAEA, as to the legal status of such countries or territories, of their authorities and
institutions or of the delimitation of their boundaries.
The mention of names of specific companies or products (whether or not indicated as
registered) does not imply any intention to infringe proprietary rights, nor should it be construed as
an endorsement or recommendation on the part of the IAEA.
CONTENTS
CHAPTER 1. I N T R O D U C T I O N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . l
1.1. World energy s i t u a t i o n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2. Global environmental i m p a c t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.3. Why nuclear power? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.4. Why hydrogen? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.5. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
References to Chapter 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
CHAPTER 2. NUCLEAR POWER AND ITS POTENTIAL FOR HYDROGEN GENERATION .... 7
2.1. Significance of process heat for the industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2. Nuclear power p l a n t s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2.1. Light-water reactor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.2.2. High-temperature gas-cooled reactor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.3. Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.3.1. Nuclear steam reformer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.3.2. Helium-helium intermediate heat exchanger . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.3.3. Hot gas d u c t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.4. Materials for nuclear power plant and process c o m p o n e n t s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.4.1. Materials testing at the Research Center J u l i c h . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 7
2.4.2. Materials testing at J A E R I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
References to Chapter 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
CHAPTER 3. SAFETY CONSIDERATIONS CONCERNING NUCLEAR POWER PLANTS
ASSOCIATED WITH PROCESS HEAT AND/OR HYDROGEN PRODUCTION... 33
3.1. Safety design concept specific to nuclear heat utilization systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.2. Safety considerations for a nuclear process heat H T G R . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2.1. Activity e n c l o s u r e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2.2. Primary circuit contamination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.2.3. Hydrogen and tritium b e h a v i o r . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.2.4. Disturbance during steam reformer o p e r a t i o n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.2.5. Rupture of helium line in g a s i f i e r . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.2.6. Rupture of process gas f e e d l i n e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.3. Safety and risk assessment for process heat HTGRS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.3.1. A V R - I I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . " . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.3.2. HTR-MODUL for process heat application (170 M W ( t h ) ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.4. Hydrogen behavior in nuclear power plant a c c i d e n t s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.4.1. P h e n o m e n a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.4.2. Hydrogen mitigation s y s t e m s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.4.3. Accidental hydrogen e x p l o s i o n s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.4.4. Containment experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
3.4.5. Containment m o d e l i n g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
3.5. Gas explosion research programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
3.5.1. The German LWR p r o g r a m . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
3.5.2. The German PNP program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
References to Chapter 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ^ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 8
CHAPTER 4. ENERGY SITUATION AND ACTIVITIES ON NUCLEAR POWER

WITH POTENTIAL USE FOR PROCESS HEAT AND HYDROGEN
PRODUCTION IN SOME EXEMPLARY C O U N T R I E S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 3
4.1. C a n a d a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
4.2. C h i n a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 3
4.2.1. Energy situation in C h i n a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
4.2.2. Experimental and theoretical activities in China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.3. G e r m a n y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
4.3.1. Energy situation in Germany . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
4.3.2. Experimental and theoretical activities in Germany . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
4.4. I n d o n e s i a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
4.4.1. Energy situation in I n d o n e s i a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
4.4.2. Experimental and theoretical activities in I n d o n e s i a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
4.5. J a p a n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 9
4.5.1. Energy situation in J a p a n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
4.5.2. Experimental and theoretical activities in Japan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
4.6. The N e t h e r l a n d s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
4.7. Russian Federation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
4.7.1. Energy situation in the Russian F e d e r a t i o n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 8
4.7.2. Experimental and theoretical activities in the Russian Federation . . . . . . . . . . . . . . . . . . . . . . 88
4.8. South A f r i c a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
4.9. U S A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
4.9.1. Energy s i t u a t i o n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
4.9.2. Experimental/theoretical a c t i v i t i e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
4.10. V e n e z u e l a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 3
References to Chapter 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
CHAPTER 5. CONVENTIONAL AND ADVANCED HYDROGEN PRODUCTION METHODS.. 101
5.1. Decomposition and gasification of fossil fuels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
5.1.1. Conversion of natural gas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
5.1.2. Conversion of c o a l . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
5.1.3. Conversion of o i l . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
5.1.4. Conversion of other feedstocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
5.1.5. Recovery from refinery o f f g a s e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 7
5.2. Decomposition of w a t e r . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
5.2.1. E l e c t r o l y s i s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
5.2.2. Thermochemical c y c l e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
5.2.3. Photolysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
5.2.4. Direct water s p l i t t i n g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
5.3. Comparison of hydrogen production methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
5.4. Production of liquid h y d r o g e n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
5.4.1. LH2 production m e t h o d s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
5.4.2. Liquefaction p l a n t s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
5.4.3. Production of slush hydrogen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
References to Chapter 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
CHAPTER 6. STORAGE AND TRANSPORT OF HYDROGEN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
6.1. Storage of h y d r o g e n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
6.1.1. Gaseous hydrogen s t o r a g e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
6.1.2. Liquid hydrogen storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
6.1.3. Hydrogen storage as a metal h y d r i d e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
6.1.4. O t h e r s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
6.1.5. Comparison of storage s y s t e m s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
6.2. Transport and distribution of gaseous and liquid hydrogen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
6.2.1. Pipeline s y s t e m . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
6.2.2. Vehicle transportation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
6.3. Chemical energy transmission systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
6.4. Chemical heat pump systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
CHAPTER 7. APPLICATIONS OF HYDROGEN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
7.1. Hydrogen as a raw material in chemical processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
7.1.1. Ammonia synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
7.1.2. Methanol s y n t h e s i s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
7.1.3. Direct reduction of iron ore . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
7.1.4. Fossil fuel p r o c e s s i n g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
7.1.5. Fischer-Tropsch synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
7.1.6. O t h e r s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
7.2. Hydrogen as a f u e l . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
7.2.1. H e a t i n g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
7.2.2. Fuel cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
7.2.3. Hydrogen powered v e h i c l e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
7.2.4. Stationary gas turbines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
CHAPTER 8. SAFETY RISKS OF A LARGE-SCALE HYDROGEN APPLICATION . . . . . . . . . . . . 191
8.1. Properties of h y d r o g e n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
8.2. Safety measures in handling (cryogenic) hydrogen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
8.2.1. Physiological hazards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
8.2.2. Safety m e a s u r e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
8.3. Effect of hydrogen on materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
8.3.1. Hydrogen e m b r i t t l e m e n t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
8.3.2. Low-temperature i n f l u e n c e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
8.3.3. Materials questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
8.4. Theoretical and experimental investigation of accidental release of h y d r o g e n . . . . . . . . . . . . . . . . . . . 199
8.4.1. LHa pool formation, spreading, and v a p o r i z a t i o n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
8.4.2. Atmospheric dispersion of h y d r o g e n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
8.5. Combustion behavior of h y d r o g e n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
8.5.1. Flame propagation, d e f l a g r a t i o n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 0 9
8.5.2. Flame acceleration and detonation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
8.5.3. Pressure impact and damage analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
8.5.4. Safety d i s t a n c e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2 4
8.6. Experience in handling h y d r o g e n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
8.7. Safety and risk assessment for hydrogen containing systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
8.7.1. General methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
8.7.2. E x a m p l e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
8.8. Environmental issues..........................................................................236
8.8.1. Water vapor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
8.8.2. Nitrogen oxides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
8.8.3. H y d r o g e n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
8.8.4. O t h e r s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
8.8.5. M o d e l i n g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
CHAPTER 9. PAST AND PRESENT ACTIVITIES ON HYDROGEN IN THE WORLD . . . . . . . . . 249
9.1. General overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
9.2. Large-scale hydrogen energy s y s t e m s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
9.2.1. WE-NET p r o j e c t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
9.2.2. Euro-Quebec hydro-hydrogen pilot p r o j e c t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
9.2.3. NHEG p r o j e c t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
9.2.4. PORSHE p r o j e c t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258
9.2.5. HYSOLAR p r o j e c t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 5 9
9.2.6. Solar-Wasserstoff-Bayern p r o j e c t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
9.2.7. PHOEBUS p r o j e c t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 6 1
9.2.8. HYPASSE p r o j e c t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
9.2.9. LH2 a i r p o r t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
9.2.10. CRYOPLANE p r o j e c t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
9.2.11. Biohydrogen production s y s t e m . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
9.3. Local hydrogen s y s t e m s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
9.4. Power generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
9.5. Activities on hydrogen as a fuel in transportation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
9.5.1. A v i a t i o n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
9.5.2. Ground-based vehicles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
CHAPTER 10. TRANSITION TO A FUTURE HYDROGEN ENERGY E C O N O M Y . . . . . . . . . . . . . 287
10.1. Transition from fossil to non-fossil e n e r g y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
10.2. Design of a hydrogen energy e c o n o m y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
10.3. Concept of a nuclear hydrogen e c o n o m y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
10.4. Some interesting assessments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
10.5. Further work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
References to Chapter 10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
CHAPTER 11. FINAL REMARKS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
CHAPTER 12. S U M M A R Y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
APPENDIX A: THERMOCHEMICAL CYCLES FOR HYDROGEN PRODUCTION . . . . . . . . . . . . . 325
A.I. P r i n c i p l e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
A.2. Thermochemical c y c l e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
A.2.1. Oxide s y s t e m . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2 5
A.2.2. Halide family p r o c e s s e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326
A.2.3. Sulfur family processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
A.3. Thermochemical hybrid c y c l e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328
A.3.1. Sulfuric acid bromine hybrid p r o c e s s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
A.3.2. Sulfuric acid hybrid process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
A.3.3. Hydrocarbon hybrid p r o c e s s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 330
A.3.4. Metal-metal hydride hybrid p r o c e s s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331
References to Appendix A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332
APPENDIX B: CHEMICAL ENERGY TRANSMISSION S Y S T E M S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
B.I. Steam reforming system ( E V A / A D A M ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
B.2. COz reforming s y s t e m . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
B.3. Methanol system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
B.4. Methylcyclohexane — toluene s y s t e m . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
B.5. NHs s y s t e m . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339
References to Appendix B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340
APPENDIX C: CHEMICAL HEAT PUMP S Y S T E M S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 4 1
C.I. 2-propanol — acetone system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341
C.2. Cyclohexane — benzene system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341
C.3. Hydride s y s t e m . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 4 1
C.4. Sponge iron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342
A B B R E V I A T I O N S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343
CONTRIBUTORS TO DRAFTING AND R E V I E W . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
Chapter 1
INTRODUCTION
1.1. WORLD ENERGY SITUATION
In 1996, the world's consumption of primary energy amounted to 8380 million TOE1,
which was 3 % more compared with the previous year. Most of the energy, 90 %, originated
from fossil fuels (coal: 27 %, oil: 39.5 %, gas: 23.5 %) and the remainder from nuclear
power (7.4 %) and renewable energies (hydro: 2.6 %) [8]. About 30 % of the primary
energy consumed is used for electricity production. Currently hydro and nuclear are the
only significant non-fossil contributors to electricity production with hydro representing
about 20 % and nuclear about 17 % of global electricity. Other renewables have been able
to penetrate only limited niche markets [5].
As described in a study by the European Commission, global energy supply is

expected to increase by 1.6 % per year. Renewable energies are gaining mainstream
acceptance with a 2.0 % increase per year, nuclear not more than 0.5 % per year. Natural gas
is currently the fastest growing energy source with an annual increase of 2.7 %. The energy
intensity of many products has fallen, but energy consumption is increasing: consumers
have developed new needs, developing countries want to achieve higher economic and
social levels. The energy trade is growing rapidly [7].
The future energy supply mix will depend on the magnitude of demand growth,
changing objectives and social priorities, and the technologies to meet them [5]. The
International Energy Agency (IEA) in Paris does not expect a significant change in the
world's energy mix, because long time periods are required to develop new or improved
energy technologies before they can be commercially deployed. What it does expect is a
strong demand for more energy (1.6 %/yr) due to population increase and rapid economic
growth in some regions of the world. The US Electric Power Research Institute (EPRI)
does not anticipate primary fuel resources to be an economic constraint in the next century
although their absolute and comparative cost might shift with time. A balanced and stable
program without pursuing extreme policies has the best prospects of achieving lowest long-
term social cost, therefore it recommends facing the future with an array of options and
flexible strategies [5].
1.2. GLOBAL ENVIRONMENTAL IMPACT
Most of the world's fuel supply is some sort of hydrogen combined with carbon,
hydrocarbons. Burning hydrocarbons means liberating carbon oxide gases with undesirable
side effects: CO is toxic to life, COi is currently labeled as one of the primary causes of
the greenhouse effect. The other aspect of hydrocarbons is their limited supply on earth.
Although the problem of global warming has been recognized, there is still a
discrepancy between the global trend of CO2 emission and the global target of CO2
reduction. The agreement of 154 countries at the Rio earth summit in 1992 was to pursue
1 TOE (- ton oil equivalent) - 1 428 TCE (- ton coal equivalent) - 41,868 MJ
"sustainable development" by reducing the emission of greenhouse gases to the 1990 level
by the year 2000. Figures from 1996 revealed a new record high with CO2 emissions
increased by 2.8 % compared with 1995. According to a prediction by the PROGNOS
AG in 1991, compared with 1989 global CC>2 emissions will be increased by 67 to 170
% in 2040. Even if the industrial countries were successful in achieving a CO2 reduction
of 80 %, total emissions would still increase by 30 to 100 % due to the enormous energy
demand of the developing countries [4]. For the OECD, there is a predicted 13 % increase
of CC>2 emission between 1990 and 2010 versus the declared goal of a 15 % decrease in the
respective period [6]. CC>2 contributes 40 % to the anthropogenic greenhouse effect [1] with
a long lifetime of up to 100 years in the atmosphere. More than half of the CC>2 emissions
come from energy production. Traffic contributes a great deal to the environmental impact
by using fossil fuels at low efficiency.
A recent analysis by EPRI expects that even with a favorable scenario, total carbon
combustion will rise from the 5.5 gigatons of 1986 to at least 8.6 gigatons in 2060 despite
the introduction of renewables and a relaunch of nuclear development [5]. In contrast, a
reorientation has occurred in some sectors, in particular, in the automotive industry where
low emissions of nitrogen oxides and hydrocarbons are mandatory.
The United Nations Convention on Climate Change has prompted initiatives by

various countries to identify and implement methods for curbing the rise in greenhouse
gas emissions. Studies by the International Panel on Climate Change (IPCC) have shown
that a stabilization of the atmospheric CC«2 would be achieved in the second half of the next
century, if CC>2 emissions are reduced by 60 % within the next few decades. In the short
term, the desired reductions are likely to be achieved through energy savings combined
with the deployment of more advanced, more efficient supply and utilization technologies.
However, it is generally recognized that the longer-term targets will require substantial
increases in the deployment of new and improved technologies.
Three conferences of parties to the United Nations Framework Convention on

Climate Change have been held so far to discuss the global warming issue. The summit in
Toronto in 1988 resulted in the promise of a 20 % CC«2 reduction by 2005; the reality today
is a 16 % increase compared with the 1988 level. During the conference in Rio de Janeiro in
1992, it was decided to go back to the 1990 level by 2000; so far there has been an increase
of 8 % in the OECD countries. At the third world summit in December 1997 in Kyoto,
Japan, the 167 participating countries came to an agreement which, if finally accepted2,
will for the first time be an obligatory agreement according to international law. It includes
the commitment by the 38 industrialized countries to reduce their overall emissions of
the gases, which contribute to the global warming potential, by 5.2 % compared with
1990 and 19953 by the years 2008/2012, respectively. Emission change ranges between
-8 % (EC) and +10 % (!) (Iceland) for the single countries. No goal has been specified
for the developing countries. The introduction of new economic instruments such as the
possiblities of trading emission permits and of taking so-called "phantom" emissions into
2
The draft protocol has legal effect 90 days after it has been ratified by 55 % of the parties of the convention, among them industrial
nations with more than 55 % of all COj emissions [2].
By midth of May 1998, 35 countries have signed the protocol including industrialized countries (among them Japan and Germany) with in
total 39 % of the worldwide COj emissions.
3
The 1990 level is the baseline for the greenhouse gases COj, CR), N2O. The 1995 level is the baseline for the greenhouse gases
sulfur hexafiuoride (S?6), perfluorocarbons (PFCs), hydrofluorocarbons (MFCs).
account meaning that the creation of CC>2 sinks (e.g., reforestation) could be balanced with
emissions, was a matter of controversy and needs further elaboration.
1.3. WHY NUCLEAR POWER?
Limited fossil energy resources and pending environmental issues are likely to in-
crease the significance of non-fossil energy forms - hydro, nuclear, solar, wind, geothermal,
tidal - in the long run. Despite its low growth at present, nuclear power is foreseen by the
World Energy Council as gaining importance in the future representing, besides renewables,
a serious option for decoupling the existing energy demand from CC>2 emissions.
Usually the sites of high energy potentials are not identical with the sites of energy
consumption. This situation offers the following options:
• on-site generation of energy-intensive products

• generation of electricity and transport via grid
• transformation of electricity into chemical energy carriers, e.g., hydrogen, methanol
• direct generation of hydrogen and transport via pipeline.
Nuclear power is considered by many to be the most promising CO2-free energy

technology with long-term fuel supply security. According to a statement by IAEA, in 1996
the nuclear power installed worldwide saved an additional impact on the atmosphere of 2.3
billion tons of CC>2 per year corresponding to 8 % of additional release [3]. Principally
used as base load power plants, nuclear off-peak electricity could be applied for hydrogen
production.
The operation of nuclear power plants in the "Combined Heat and Power" (CHP)
mode, i.e., cogeneration of electricity plus heat including the production of process heat is a
means of significantly raising energy conversion efficiencies. Furthermore the refinement of
fossil fuels, e.g., coal, by nuclear power contributes to the fundamental objectives of energy
policy: (i) energy security, (ii) saving of resources, (iii) protection of the environment. Cost
is a major concern, but nuclear power seems to be competitive in the long term.
1.4. WHY HYDROGEN?
Hydrogen, the most abundant element in the universe, is - as a fuel - clean, powerful,
renewable, and environmentally benign. It can be burnt to water vapor at a very low level
of pollution releasing its chemical energy as heat. It can be stored directly or transported
for later usage. Most of the hydrogen is bound in chemical compounds, i.e., it must be
extracted prior to its use by means of a primary energy resource. The introduction of
hydrogen as a significant contributor to meet the world's energy demand in the future
requires an infrastructure, economy, and last but not least, safety.
An important driving force for development work on hydrogen has always been
developing a further secondary energy carrier, in addition to electricity, for the non-electric
market. Economic solutions for generation, storage, transport, distribution, and application
may help to introduce a new energy concept with hydrogen playing an essential role as
a secondary energy carrier.
The fact is that billions of cubic meters of hydrogen are produced every year, but
unfortunately mostly converted from fossil fuels and only to a small extend used for energy
purposes. In a future low-emission energy economy, the hydrogen energy carrier could play
a pertinent role using its advantages compared with other alternatives. With the energy
carrier hydrogen, all the subtasks of an energy transmission can be performed in a safe and
appropriate way. A major drawback of hydrogen is its hazard potential if inadvertantly
released and mixed with the ambient air to form a flammable gas mixture. Therefore it is
necessary to assess its risks to society.
1.5. OVERVIEW
The report covers all main issues concerning hydrogen and its potential connection
to nuclear power. It presents for either area goals achieved so far and ongoing research
activities. Also secondary areas are described including processes with hydrogen as an
intermediate step.
The first part comprising three chapters deals with nuclear power as the primary
energy source for producing electricity and process heat / steam which could be utilized
for hydrogen production. Chapter 2 treats the design of nuclear power plants for process
heat application and the components required. Safety considerations described in chapter
3 concentrate on the aspects that are peculiar to nuclear process heat plants. International
activities on using nuclear power to be utilized in process heat applications, for example
for hydrogen production in the past, present, and future are listed in chapter 4.
The second part of the report focuses on key technologies as the basic elements of an
economy with hydrogen as an energy carrier. Chapter 5 (plus appendix A) describes the
production methods for hydrogen including those which are still at an early research stage.
Chapter 6 (and appendices B and C) contains the description of methods and components
for the storage, transportation, and distribution of hydrogen. In chapter 7, both major and
minor areas of applications of hydrogen as a feedstock for (petro)chemical processes and,
in particular, as a fuel are presented. The safety aspects which have to be considered
in a hydrogen energy economy including some examples of a safety and risk assessment
are given in chapter 8. This part on hydrogen is terminated by giving a - most probably
incomplete - list of international activities for promoting the use of hydrogen including a
more detailed description of some examples of an autonomous hydrogen energy system or
subsystem, some in operation, others still at the design stage.
The report ends with a rough description of a future large-scale hydrogen energy
economy, in which nuclear power may play a significant role, and the efforts needed to
reach this goal.
REFERENCES TO CHAPTER 1
[1] BEISING, R., HILDEBRAND, M., Emissionen in die Atmosphare und ihre Ein-
flusse auf die globale Klimaentwicklung, Elektrizita'tswirtschaft, 94 (1995) 328-333.
[2] BOLIN, B., The Kyoto Negotiations on Climate Change: A Science Perspective,
Science 279 (1998) 187-188.
[3] ROGNER, H.H., Kyoto Climate Conference: IAEA Statement, Highlights Envi-
ronmental Benefit of Nuclear Power, Press Release PR 97/40, December 8, 1997,
International Atomic Energy Agency, Vienna (1997).
[4] SOCHER, M., RIEKEN, T., TA-Projekt "Risiken bei einem verstarkten Wasser-
stoffeinsatz", TAB Working Report No. 13, Biiro fur Technikfolgen-Abschatzung
des Deutschen Bundestages, Bonn (1992).
[5] STARR, C, SEARL, M.F., ALPERT, S., Energy Sources: A Realistic Outlook,
Science 256 (1992) 981-987.
[6] TENHAGEN, H.J., Treibhausheizung kraftig hochgedreht, Newspaper article in
TAZ, Berlin, May 26, 1997.
[7] WEIRAUCH, W., Natural Gas Fastest Growing Fuel, as World Energy Demand
Increases, Hydrocarbon Processing 75 (1996) November 27.
[8] SCHURMANN, J., Weltweit wachst das Energie-Angebot, Newspaper article in
Handelsblatt, Dusseldorf, July 15, 1997.
NEXT PAGE(S) I
left BLANK I
IXftEb H maUJ DHfliniFJI KmNLiUMul
Chapter 2
NUCLEAR POWER AND ITS POTENTIAL

FOR HYDROGEN GENERATION
The world's primary energy consumption amounts to well over 300,000 peta Joule
and more than two thirds of it are used in the transportation sector or converted to hot water,
steam and space heat. Unlike fossil energy carriers, nuclear power is almost exclusively
used for electricity production. Only a few nuclear power plants are being used for heat
applications. If nuclear power is to utilize its full potential of substituting the energy
carrier oil, it needs to be introduced into the heat market for heating (< 100 °C) or low-
temperature process heat (« 200 °C). This could easily be provided in combination with
electricity production by light-water reactors (LWR). It implies, however, near-customer
sites for the plants or even on-site installation. Even more favorable would be Combined
Heat and Power (CHP) from a high-temperature gas-cooled reactor (HTGR), in particular,
the one-cycle gas turbine variant because of its higher temperature output and its higher
efficiency [4]. Clearly nuclear heat production could play a major and important role in
the non-electric sector. Numerous studies on small and medium-sized reactors have shown
their option for heat application to be technically and economically viable [15].
With respect to a large-scale hydrogen production, nuclear power can play a significant
role if used as a provider of electricity in the electrolysis process or as a provider of high-
temperature heat in fossil fuel conversion. An introductory option would be the use of
cheap surplus electricity. Production of hydrogen as a bulk energy carrier is by a factor
of about 2 too expensive compared with the today's commercial business of natural gas
and oil, however, the trend to include external effects into the energy cost may help to
achieve economic attractiveness [2].
A look at the use of different reactor types demonstrates the dominant role of the
LWR line in commercial electricity production. From 428 nuclear power plants operating
worldwide in 1996 to produce 363 GW(e) or 17 % of the international electricity supply -
plus 62 units under construction to produce 55 more GW(e) - 342 plants are LWRs, 249
PWRs and 93 BWRs, for a total of 311 GW(e) [27]. And the trend to larger power units
was observed. Not more than 5 GW(th) of global nuclear power are presently employed
to supply hot water and steam, mostly in Canada, China, Kazakhstan, Russia, and Ukraine
[25].
Nuclear growth depends on the future public perception of the comparative benefits,
cost, and risks of alternatives. Based on the comparative evaluation of tangible risks to
public health, safety, and environment, nuclear power appears to be a better choice than
coal [39].
2.1. SIGNIFICANCE OF PROCESS HEAT FOR THE INDUSTRY
The structure of energy demand in the industries for the generation of certain process
heat temperature levels is shown in Figs. 2-1 and 2-2 for the examples of Germany and
Japan, respectively [5, 17]. In principle, there is a great demand in the low temperature
Capita! goods
\
Chemistry
0 200 400 600 800 1000 1200 1600

Temperature ——*
Fig. 2-1: Process heat spectrum in the industries as a function of process temperature
in Germany as of 1996, from [5]
350
0 200 400 600 800 1000 1200 1600

Temperature ——*
Fig. 2-2: Process heat spectrum in the industries as a function of process temperature
in Japan as of 1983, from [17]
Table 2-1: Steam vessels in selected industrial branches in Germany as of 1987, from [19]
Accumulated
No. of steam
Branch plant power No. of plant sites
power plants
[tfh]
Paper, pulp 278 9.3 131
Chemical 438 29.1 105
Iron 157 8.9 36
Road Vehicle Construction 117 6.9 21
Sugar 137 5.3 43
Petrochemical 120 7.6 30
Others 323 8.2 139
S 1570 75.3 505
range of 100 - 300 °C to be used in petrochemical and refinery processes or the paper and
textile industries. Other distinct peak demands are given in the high temperature range at
about 900 °C in the chemical and non-ferrous industries and at about 1500 °C in the iron
and steel producing, and glass and ceramics industries. Power plants operated in the CHP
mode generally provide heat below the 500 °C level [5].
For the introduction of nuclear process heat into the energy market, the estimation of
present and future industrial demand for process steam needs to be made first as well
as establishing how nuclear power fits into the energy needs profile [19]. Inevitable
fluctuations in industrial process heat/steam demand have to be expected. Also the consumer
circle for secondary energy is limited for industrial plants compared with public power
plants, thus smaller units would be preferable. On the other hand, nuclear units are
economically operated only at a certain minimum capacity. Units of 500 MW(th) are
thought in [7] to allow economic use of the nuclear option. A more recent study from
1990 [19] takes plants of 2 * 100 MW(th) as a minimum into consideration increasing the
industrial potential of nuclear plants. Table 2-1 summarizes the number of process steam
plants, sites, and steam production capacity in Germany valid for the year 1987 [19].
Temperature requirements range from low level (e.g., hot water, district heating,
desalination) to a high level of up to 1000 °C (e.g., petrochemical industry, water splitting).
Applications at temperatures beyond 1000 °C can be served by nuclear power only in
connection with secondary energy carriers such as electricity, hydrogen, or synthesis gas.
From all industrial branches, the chemical industry is the most pertinent in terms of process
steam representing in Germany about 50 - 70 % of the total potential [19]. Depending on
the nuclear steam generation system, steam can be generated at different levels of pressure
and temperature according to the needs of the process industries. Process steam must be
produced on-site, since its long-distance transportation is not practicable. An estimated
fraction of 30 % of installed steam capacity in the chemical industries could be based
Table 2-2: Steam parameters for different nuclear steam-generating systems, from [10]
Steam parameters
Nuclear system
Pressure [MPa] Temperature [°C] Mass flow [t/h]
SIEMENS PWR 6.5 280 7400
B&WPWR 6.9 312 7280
SIEMENS PHWR 5.6 271 3440
Candu - PHWR 4.7 260 1090
Phenix LMFBR 16.3 510 2100
SIEMENS BWR 5.5 270 1185
THTR-300 18.1 530 918
Fort St. Vrain HTGR 17.3 541 1044
Table 2-3: Survey of Japanese industries suitable for electricity / steam cogeneration, from [42]
Ratio of steam Ratio of

Temperature range
Industry consumption independent power
ra [%] generation [%]
Chemical 300 - 600 18.6 18.6
Paper - Pulp 300 - 600 34.3 49.2
Ceramic - Pottery 800 - 900 5.0 12.6
Iron & Steel > 1000 2.1 16.4
Nonferrous Metal - 0.7 16.8
Table 2-4: Typical heat application plant models for selected industries, from [42]
Steam conditions Steam Power

Production
Industry Temperature Pressure consumption consumption
scale
[°C] [MPa] [tfh] [MW(e)]
250 4 20
400,000 t/yr
Petrochemical 198 1.5 170 60
ethylene
152 0.5 205
750,000 t/yr 188 1.2 10
Paper - Pulp 204
newspaper 134 0.3 235
530 12 22
210,000
Oil refinery 320 2.1 225 55
bbUd
190 0.3 40
10
on nuclear whereas the major fraction will remain to be produced in conventional power
plants [7].
In the field of enhanced oil recovery, high steam pressures are required as could be
provided by gas-cooled reactors; water-cooled reactors would need an additional steam
compression step [25]. Characteristic parameters are given in Table 2-2 for various designs.
For the example of the Japanese industries, process steam temperature range and energy
consumption structure are presented in Tables 2-3 and 2-4.
2.2. NUCLEAR POWER PLANTS
In order to optimally exploit a heat source for power generation, the Carnot cycle
process should work at the largest possible temperature difference. In HTGRs and in fossil-
fueled conventional plants, steam temperatures are in the range of 530 °C. Using the above
systems in the combined heat power (CHP) mode, steam and electricity production in an
HTGR or an conventional plant can be accomplished consecutively by routing the steam
first through a turbine and then to the steam application system (Fig. 2-3).
In contrast, the steam temperature in an LWR is no higher than 320 °C due to the
low coolant outlet temperature allowing steam production at a lower quality level in a
typically parallel arrangement of steam and electricity production (Fig. 2-4). The electricity
output for various process steam outputs at a constant total thermal power of the plant is
shown in Fig. 2-5. Smaller electricity output results in a higher waste heat loss, which on
the other hand is decreasing with increasing steam output. The electricity produced out of
one ton of steam varies between 120 - 270 kWh depending on process steam output for an
HTGR, and between 40-184 kWh for an LWR [7].
For the liquid metal fast breeder reactor (LMFBR), the upper temperature limit of the
steam produced is approximately 500 °C (505 °C in the Russia BN-600 reactor, 495 °C in
the German SNR-300 concept or 483 °C in the Japanese Monju reactor) prior to entry into
the turbine. LMFBRs have not been considered so far for delivering process heat.
In principle, all reactors are able to provide process steam and long-distance heat to
bridge a distance between the locations of heat production and heat demand. Even if LWRs
are not able to provide heat at a temperature level sometimes required by the industry, their
temperature level could be raised by some additional conventional heating. However, only
the HTGR is capable of opening the market for direct high-temperature process heat
applications (Fig. 2-6) [19].
11
320 °C, 2 MPa
Steam generator
530 °C 200 °C, 0.5 MPa
15.2 MPa
120 °C
Feed 21.4 MPa ,,. . Turbine
High-
water p: pressure
pump turbine
28 °c
0.3 MPa 0.004 MPa
Heat consumers
0.2 MPa
Condenser
conventional plant or HTGR
Fig. 2-3: Secondary circuit of industrial power plant with HTGR, from [7]
320 °C, 2 MPa

Steam generator
320 °C 200 °C, 0.5 MPa
2 MPa
120 °C
Feed 21.4 MPa Turbine
water (• '•
pump
28 °C
0.3 MPa 0.004 MPa
Heat consumers
0.2 MPa
Condenser
LWR
Fig. 2-4: Secondary circuit of industrial power plant with LWR, from [7]
12
200- 400-
MW(e) MW(th)
t 150- A 300-
HTGR
<D
ca
LWR
o 1 DO- 5 200-
o <D
"L-
^-»
•4—'
03 HTGR
O
Q) LWR
50- 100-
0 0 i i i i
0 100 200 300 t/h 500 0 100 200 300 t/h 500
Process steam production ——* Process steam production ——*
Fig. 2-5: Electric power output and waste heat as a function of steam output, from [7]
Product Production Reaction Temperature [°C]

Fluid Process 50 500 1000
i i i
Synthesis - Steam reforming 1 - >,' -x |
gas of methane
(H2 + CO)
- Steam gasification
of coal
(fluidized bed)
Hydrogen - Water electrolysis 1 ->, 1

gas
- Steam electrolysis
- Thermochemical II 1 i
splitting of water (HI decomposition) (H2SO4 decomposition)
(IS process)
Methanol - Methanol synthesis I | (Exothermal)
'-,- "v j LWR

Nuclear process heat temperature : . - - '-* VV|LMFBR
7 ; '^-, ; > , - : ;5 "-'-M '<-'•'" -V IHTGR
Fig. 2-6: Reaction temperature ranges for various industrial production processes
dependent on type of nuclear reactor, from [11]
13
2.2.1. Light-Water Reactor
With regard to the heat market, a light-water reactor (LWR) of the size of the German
Biblis A reactor (1200 MW(e)) has the potential to provide 10,000 GJ/h low-temperature
heat in the CHP mode. Since hot water and process steam cannot be transported directly
over long distances, nuclear power can be economically used only hi areas with large
heat consumption density like chemical industrial complexes, or in long-distance heating
systems [4]. With respect to the latter, some of the candidate systems are operated at lower
temperatures and could be used in connection with an LWR, if the catalyst conversion
processes can be properly controlled [14].
The fact that steam generation by LWRs is limited to lower temperature steam («
280 °C) makes the introduction of LWRs into the heat market conceivable for the chemical
industry where large amounts of process steam are required as a heat carrier or as a medium
used for synthesis gas and hydrogen generation.
The procedure of using electricity in the electrolysis of water to produce hydrogen

is not competitive with the methods of extracting the hydrogen from fossil raw materials.
The overall efficiency then is no higher than approx. 35 %. Electrolysis is applied when
cheap electricity is available or high-purity hydrogen is desired [4].
2.2.2. High-Temperature Gas-Cooled Reactor
The high-temperature gas-cooled reactor represents one type of the next generation
of nuclear reactors for safe and reliable operation as well as for efficient and economic
generation of energy. Present HTGR designs are based on about four decades of R&D
including operating experience with six prototype reactors [37].
2.2.2.1. Features of HTGR-Based Chemical Processing
The concept of the high-temperature gas-cooled reactor has important features such as
• production of electricity,
• production of high-temperature heat up to 1000 °C,
• production of high-temperature steam of about 530 °C,
• realization of fundamental safety features, and
• potential for economic attractiveness.
In reference to the CHP mode, the industrial power plant with HTGR could be
named "CHHP", because it provides three energetical connection points: high-temperature
heat, low-temperature heat, and electricity. With respect to electricity production, the
achievement of a 40 - 43 % efficiency is possible when using steam turbines. Gas turbines
reach efficiencies up to 48 %. A combination of both could even reach the 50 % level.
Cogeneration applications are estimated to allow efficiencies in the range of 80 - 90 % [24].
Physical requirements for process heat HTGRs are

• gas outlet temperature of 950 -1100 °C,
which allows for sufficient reaction velocity of chemical process;
14
• gas inlet temperature of 350 - 500 °C,
which is necessary in order to use recuperation heat of product gas;
• lower system pressure,
which raises the efficiency of the chemical processes;
separation of the nuclear system from the chemical system,
which reduces the risk of radioactivity release after an accidental chemical explo-
sion;
• separate components for high-temperature and low-temperature heat consumption,
which allow for different constructions and materials;
• arrangement of primary and secondary circuit in a non-integrated design because
of better accessibility.
Steam applications include the generation of electricity in the steam cycle, the
production of process heat in cogeneration with electricity, and the production of district
heat. Steam on a higher temperature level offers the chance to open up new markets for
process steam applications in the chemical and petrochemical industries or for enhanced
oil recovery. The ratio of electricity to steam production can be adopted over a wide
range according to the needs of the consumer. In heat applications, heat is delivered from
the helium coolant with an outlet temperature at the high end of 750 - 950 °C, for the
conversion of solid and liquid fossil fuels into "clean" gaseous and liquid fuels synthesis
gas, hydrogen, methanol. Chemical energy transmission systems could be most effectively
utilized when integrated with a parallel electric transmission system, taking the high-end
temperature range for driving the heat pipe and the low-end for generating electricity.
Chemical industrial plants could be optimized in terms of product spectrum and

product volume such that the total energy output of an HTGR, whether as heat or steam
or electricity, could be utilized on-site [7].
A direct-cycle reactor system strongly depends on the HTGR fuel quality. The
available high quality TRISO (SiC) fuel4 has not been rigorously tested under gas turbine
conditions, but appears satisfactory under moderate conditions, i.e., reactor outlet tempera-
ture of 850 °C and time-averaged fuel temperature under operating conditions of < 1200
°C. Options to improve fission product retention capability are to enlarge the thickness of
the SiC layer or to replace the SiC coating layer by a ZrC coating layer. TRISO (ZrC)
fuel, however, will require a long and expending development program. ZrC coatings are
under study in Japan and the USA. The effect of a thicker SiC layer has been investigated
within the German HHT project [16, 31].
2.2.2.2. Chemical Processes for Hydrogen Production
A widely applied method for the production of hydrogen is the decomposition and
gasification of fossil fuels. Reaction temperatures are typically in the range > 500 °C,
which could be ideally provided by an HTGR. The non-fossil generation of heat for the
endothermal reforming reaction represents a means of reduction of COi emissions. In
relation to the same amount of CO2 produced, the reforming of natural gas with an external
heat source increases the hydrogen output by 40 - 50 %, in case of coal gasification even by
4
A TRISO coated fuel particle consists of a 0.5 mm diameter UOj kernel surrounded by subsequent layers of buffer, inner pyrocarbon,
silicon carbide (SiC), and outer pyrocarbon. The 35 nm thick SiC layer provides an efficient barrier to all safety relevant radionuclides.
15
0,5x10
900 950 1000 900 950 1000
l°cJ
Fig. 2-7: Dependence of hydrogen and electricity production on helium temperature

at steam generator inlet (T2) with Tp reactor outlet temperature (950 °C), and
TS: reactor intlet temperature (250 °C), from [23]
Table 2-5: Influence of process heat temperature split-up in a 3000 MW(th) nuclear
process heat plant with six loops, from [23]
Temperature split-up [°C] 950...800...250 950...700...250 950...600...250

Steam reformer
Power for process heat [MW(th)] 640 1070 1500
AT [°C] 235 180 137
Heat flux density [kJ/(m2 s K)] 76 64 52
Required heat transfer area [m2] 8,468 16,731 28,667
Power density [MW/m3] 1.4 1.2 1.0
Diameter [m] 3.2 4.4 5.8
Steam generator
Power for electricity [MW(th)] 2360 1930 1500
AT [°C] 148 112 70
Heat flux density [kJ/(m2 s K)] 174 128 81
Required heat transfer area [m2] 13,531 15,089 18,429
Power density [MW/m3] 6.6 4.9 3.1
Diameter [m] 3.5 3.5 3.5
16
Table 2-6: Mass balance for hydrogen production, from [23]
From H2 pro-
Inlet Outlet Mass low- duced in
From Methane
mass mass conver- temp, metha-
scrubber recycled
flow rate flow rate ted separa- nation
[kg/h] [kg/h]
[kg/h] [kg/h] [kg/h] tion step
[kg/h] [kg/h]
H2 406 108,474 122,446 624 121,416 406 114,744
CH4 384,528 135,968 135,968 704 30,528 104,736 48,320
CO 11,620 238,588 42,980 224 31,136 11,620 -
CO2 - 326,876 634,260 634,260 - - -
H2O 1,297,782 884,430 758,682 - - - (20,016)
163,064
E 1,694,336 1,694,336 1,694,336 635,812 183,080 116,762
(183,080)
66 % compared with the conventional processes [35]. Cogenerated electricity is applicable

in H2 production methods based on the decomposition of steam in the high-temperature
electrolysis or in thermochemical cycles. The chemical processes of hydrogen generation
are described in further detail in chapter 5 and appendix A, respectively.
2.2.2.2.1. Reforming Process
The HTGR allows a splitting of the nuclear reactor heat for methane reforming and
for steam generation. According to the selected design, the shares of synthesis gas and
electricity production are variable over a wide range adjusted by the steam generator inlet
temperature, T2. The dependence is shown in Fig. 2-7.
For the industrial processes of ammonia synthesis, hydrocracking, hydrogenation, a

suitable heat split-up would be made at a T2 value in the range of 625 - 670 °C. In the
direct iron ore reduction process, a steam generator inlet temperature, T2, of 810 - 830
°C would be chosen. It is obvious that the heat split-up has a significant influence on the
designs of both steam reformer and steam generator [23]. Some exemplary data are given
in Table 2-5 for different T2 values. A mass balance for hydrogen production based on the
medium T2 value (= 700 °C) is given in Table 2-6.
Unlike conventional fossil-fueled reformer tubes, the helium-heated reforming plant

connected to an HTGR has to meet the much more stringent requirements of a "nuclear"
component in terms of construction, quality assurance, and scheduled re-testing. The
reforming tube placed in the primary circuit has the function of forming a radioactivity
barrier between the primary helium and the process gas. Another major difference is the
manner of heat input, which is convective transport with the helium in the nuclear version.
Synthesis gas is the starting material for numerous chemical products and it can be
processed to hydrogen if desired. It is the intermediate step in the synthesis of methanol
to serve as a motor fuel.
17
2.2.2.2.2. Steam-Coal Gasification
Since heat input for the coal gasification process is a significant cost factor, nuclear
process heat provided by an HTGR was considered as a substitute for conventional firing.
From the perspective of Germany, the most economic hydrogen may be produced from
imported cheap coal in a steam coal gasification process using HTGR heat [2]. Another
important advantage of nuclear coal gasification is its environmental effect since no coal is
fired to provide the reaction heat. Only smaller amounts of pollutants such as SC»2, CC>2,
dust are emitted into the atmosphere. The main problem, however, is the need to transfer
the nuclear heat via two heat exchangers to the coal gasification process.
Fig. 2-8 presents a flow chart of the steam-coal gasification process with the
numerical example based on a 3000 MW(th) HTGR. High-temperature heat of 950 °C
is passed to a secondary helium circuit via a He/He heat exchanger where it enters with a
900 °C temperature a steam gasifier. Hot steam is routed into a coal bed where the coal
is gasified in two steps to give synthesis gas. As a result of the carbon-steam reaction, the
helium is cooled to approx. 810 °C. If desired, a subsequent methanation process could be
added to provide synthetic natural gas (SNG) [9]. Compared with a conventionally fired
coal gasification plant, the nuclear-driven process could increase the output by up to about
60 %, because no coal is spent on direct combustion [4]. As an example, with 680 MW(th)
power from an HTGR plus 3470 t of hard coal, the yield is 2690 Nm3 of SNG [18].
For methanol synthesis, the purified coal gas is compressed to 7.5 MPa. The power
of 680 MW(th) from an HTGR plus 2680 t of hard coal result in 2390 t of methanol [18].
A combined system of coal, steel, and nuclear energy has been discussed for the
production of methanol and raw iron comprising the steps:
(i) partial (50 %) steam gasification of hard coal by means of HTGR process heat,
(ii) Klockner-steel-gas process for iron ore reduction by means of fine coke,
(iii) methanol synthesis from the product gases.
The combination of all three steps appears economically attractive. A balance

assessment assumes an input of 1.4 million t/yr of hard coal and 500 MW(th) delivered
from two HTGR units to result in the production of 1.23 million t/yr of methanol and 0.8
million t/yr of raw iron [1].
2.2.2.2.3. Hydro-Gasification
For the hydro-gasification process (Fig. 2-9), an intermediate circuit is deemed

unnecessary. An alternative to the above described coal gasification process was pursued
by the German Rheinische Braunkohlenwerke AG, Cologne: the nuclear hydriding coal
gasification. Nuclear power is here introduced in a steam reformer for methane splitting.
From the resulting synthesis gas, hydrogen is taken to be the input for the exothermal
gasification reaction.
Coupling hydro-gasification of lignite with a methanol synthesis process is possible

after cooling and purification of the coal gas and mixing it with the reformer gas. The crude
methanol yielded is then separated from the residual gas mixture by condensation. For the
18
4 MPa
j
Helium Helium
Steam
gasitier
__ .
iCoal4l8t/h _
+ |JT - L
Cooler
p ! CH4
H2
CO
950 °C 1 900 °C
^|I Ij-i |C02
| H2O
4 MPa { ' ^^ Degasifier^ I ^p r \ H2S
3000 MW ir\ Ash V i '————— —J 1 ! ^ _
uas
O HTGR ^
\.
He-heat exchanger Coke H
*s
CO2 H purification
4! 300 °C | 4-240 °C I————| ^ I CH4
i^' . . .. ,
5 CO
Steam ^ k |H 2
generator -< F"
H2^—j Methanation
^J—O °'
t
Water output
239-10 3 m 3 /h
Fig. 2-8: Flow chart of the process of steam gasification of coal (hard coal), from [9]
Helium
Lignite
Dryer
2"317~t/h~ ~
Gas Residual coke

3000 MW ir <*•
purifi-
HTGR 286
cation
Steam
reformer
Pre-heater
Gas
Steam ——— purification
gener- Conversion decom-
output
368-10 3 m 3 /h
Fig. 2-9: Flow chart of the process of hydro-gasification of coal (lignite), from [9]
19
example of 680 MW(th) power from an HTGR, the input of 2650 t of lignite would allow
a production of 2530 t of methanol [18]. If synthetic natural gas production is desired, the
coal gas passes a purification and a cryogenic gas separation step to obtain H2, CO, and
CH4 fractions. Some of the CH4 is used for H2 production in the steam reformer to serve
as a feed gas in the gasifier. The remaining methane is the product [18].
The process of hydrogen or methanol production by using a raw iron producing blast
furnace, also for stack gas utilization, can be realized with a process heat HTGR, shown
in [34] for a 250 MW(th) unit (see section 4.3.2.8.)-
A concept for the "non-integrated" employment of an HTGR in the iron and steel
industry has been proposed in [28] as described in section 7.1.3. The nuclear reactor is
sited at a location favorable for the reduction gas generation from which the gas is delivered
via a pipeline to the iron and steel industrial sites. A 3000 MW(th) nuclear power plant
(HTGR) was chosen where 2167 MW(th) are taken for naphta splitting and an operation
time of 8000 h/yr assumed would be able to provide 6.75* 109 Nm3/yr of hydrogen [28].
An HTGR is favorable, if an H2 - CO mixture as a reduction medium is used.
2.2.2.2.4. Biomass Conversion
A process for the conversion of biomass into methanol by means of an HTGR has
been proposed [3]. Wood, represented by the chemical formula C6Hg 64.03.7, is gasified by
superheated steam. Electrolytic hydrogen is added to the product gas and then converted to
methanol or methane. Both heat and electricity are provided by an HTGR. The partial
processes of steam gasification, water splitting and methanol / methane synthesis are
summarized in the overall reaction [3]:
+ 2 H2O + 5.66 H2O -» 6 CH3OH + 2.68 O2

+ 2 H2O + 5.66 H2O -» 6 CH4 + 5.68 O2
Additional requirements to an energy system as to be environmentally benign and

application friendly, the energy carrier of choice would be liquid methanol rather than
gaseous methane due to lower losses and easy and economically competitive storage and
transportation.
2.2.2.2.5. Water Splitting
The high-temperature heat from HTGRs could be used to convert heat energy directly
into chemical energy in the form of the (lower) heat of combustion of hydrogen. There are
three processes of water splitting appropriate for process heat utilization from an HTGR:
• high-temperature electrolysis,
• thermochemical cycle,
• thermochemical-electric hybrid cycle.
There is a great economic potential of nuclear thermochemical cycles, i.e., the

H2 production by means of water and nuclear heat compared with electrolytic H2 from
nuclear electricity. The HTGR is typically considered the high temperature heat source of
20
Battery J Battery A+E
Steam and electricity High-temperature reactor
generation Gas turbine
Electricity Process heat
Battery G H2S04 Battery H

Electrolysis H2SO4 decomposition
SO2
02
H2O
H2O
SO Battery I
02 SO2 / O2 separation
02
02
Fig. 2-10: Diagram of the Westinghouse sulfuric acid hybrid cycle, from [20]
choice, although other heat sources are, of course, also possible. Most of the promising
thermochemical cycles are based on sulfuric acid processes [20] (see section 5.2.2. and
appendix A).
In the Westinghouse sulfuric acid hybrid cycle shown in Fig. 2-10 (see also appendix
A.3.2.), HTGR heat will be used for the H2SO4 decomposition step. Both high-temperature
and electric steps have been experimentally investigated at the Research Center Julich. The
above hybrid cycle has undergone a detailed balance and cost analysis already in a plant
design based on nuclear power [8].
Materials corrosion problems arise by the fact that the system pressure of 4 MPa on
the primary side of the HTGR needs to be also on the secondary side of the process gas.
2.3. COMPONENTS
23.1. Nuclear Steam Reformer
Reformer tube heating with a high-temperature nuclear reactor is performed with

helium, typically at 950 °C, as the heat source. The aim of reaching a heat flux density
similar to that of the conventional method can be achieved by employing a helium-heated
counterflow heat exchanger (see Fig. 2-11). Helium under pressure shows excellent heat
transfer properties. Furthermore, precautions must be taken to minimize the effects of
asymmetry or hot gas streaks in the helium flow as well as a non-uniform process gas
flow. The materials of a helium-heated steam reformer should be selected such that the
21
Closure (to open for
cofolyst removed
Feed Cos Inlet
Product Gos Outlet
Catalyst
625 870 I 950 39/ 40
Fig. 2-11: Principle of a reformer tube for nuclear applications, from [23]
operation is made with a maximum tube wall temperature of 900 °C; for further increase,
research on improved materials outside the range of currently known and tested materials
is needed. A comparison of conventional reformer data with those of a helium heated
reformer is given in Table 2-7 [23].
Experience in construction and operation was gained with the EVA-I and EVA-
11 facilities at the Research Center Jiilich (see section 4.3.2. and appendix B.I.). The
replacement of the catalyst is difficult because of the presence of the internal return pipe
("pigtail"). Using a vacuum cleaner was found to be inefficient. A new hydraulic system
developed at Jiilich reduced the replacement time down to a few minutes [38].
The reference design of the German 60 MW PNP steam reformer consists of approx.
200 single reforming tubes composed to a bundle. Helium at 950 °C enters the reformer
pressure vessel through a short horizontal hot gas duct. With an outlet temperature of 720
°C, the helium is then routed through a short coaxial duct into the steam generator pressure
vessel. After exiting the steam generator at a temperature of 300 °C, the helium flows
back to the steam reformer and through the outer wall to keep it on cold gas temperature
before it is led back to the reactor.
On the secondary side, a steam-methane gas mixture with a ratio of 4:1 enters the
reformer at a temperature of 330 °C and is distributed to all reforming tubes. The feed
22
Table 2-7: Comparison of the data of a conventional with those of a helium-heated
reformer, from [36, 23]
Parameter Conventional Plant Nuclear Plant

Tube length [m] 8- 12 10- 12
Inner diameter [mm] 100 - 150 » 100
Wall thickness [mm] 15-20 w 15
Product gas return pipe outside reformer tube inside reformer tube
Reforming pressure [MPa] 0.1 - 2.5 4
Reforming temperature [°C] 800 - 850 800 - 850
Pressure on heating side [MPa] 0.1 4
Heat transfer Radiation Convection
2
Tube density [tubes/m ] < 1 45
Maximum heating temperature
1400- 1500 950
[°C]
Maximum tube wall temperature
900 - 950 900
[°C]
Maximum pressure difference 0.1 (hot part)
0- 2.5
across tube wall [MPa] 0.4 (cold part)
Maximum temperature difference
30
across tube wall [°C]
Ratio H2O : CH4 2:1 -5:1 2:1 -5:1
2
Mean heat flux [kW/m ] 80 70-80
Heat flux ratio max. over min. 10:1 1.5:1
Product gas flow rate [(Nm3
14 14
H2+CO)/(m2 s)]
Goal of service lifetime [h] 100,000 100,000
gas is heated up by recuperators to 560 °C before it enters the catalyst section reaching a
final temperature of 810 °C while reformed at the same time. The hot product gas is then
cooled in the recuperators and leaves the reformer at a temperature of 480 °C [33].
To overcome the perceived disadvantages of a nuclear steam reformer with its

comparatively low heat transfer and its high system pressure, improvements of the Japanese
design of the HTTR nuclear reformer have been proposed (Fig. 2-12):
(i) increase heat input to process gas by applying a bayonet type of reforming tube,
(ii) increase heat transfer rate by using combined promoters of orifice baffles and
cylindrical thermal radiation pipes, and
(iii) increase conversion rate by optimizing chemical composition of feed gas.
23
Reforming gas 450'C, 4 5Mpa
Improvement of hydrogen Thermal energy
production performance utilization of 78%
• Compact seamless
catalyst tubes
Increasing heal inpul lo
Reformed gas reforming gas
' 600'C
1 Effective utilization of hot reformed gas
Bayonet type of catalyst tubes
2 Increasing helium temperature drop through reformer
Reformer outlet helium temperature 600'C
From helium 23~36MW\
4 8MW in total
From reformed gas 0 -*12MWJ
E Increasing reaction
CD
temperature
Helium side heal transfer enhancement

Orifice baflle+wirenel
K f Maximum reformed gas temperature 800'C
I Maximum heat flux 40kW/m2
Optimization of reforming
gas composition
1 Sleam/Carbon=3 5
2 Excessive methane supply
(resulted in residual methane of 0 3MPa at outlet)
Helium gas OOO'C, 4 1Mpa
Fig. 2-12: Improvement of performance of Japanese helium-heated steam reformer,

from [30]
With these improvements, it is estimated to achieve an H2 production rate of approx.

3800 Nm3/h for a 10 MW heat input [12, 30].
The installation of a steam generator as a downstream cooler of the endothermal

reforming reactor was analyzed to provide stable controllability of the thermodynamic
behavior in the system and to be universally applicable [13].
22.2, Helium-Helium Intermediate Heat Exchanger
An intermediate heat exchanger (IHX) circuit serves as an interface component to

connect the nuclear with the chemical system. It thus
• minimizes contamination of the process heat exchanger and allows it to be placed
outside the reactor containment,
• prevents water and process gas ingress into the nuclear core in case of a tube
rupture, and
• reduces hydrogen permeation from the process gas into the primary helium as well
as tritium permeation in the reverse direction.
24
Collector
Circulator
Fig. 2-13: Principle gas flow in a helium-to-helium intermediate heat exchanger,

from [22]
Table 2-8: Comparison between IHX design data of test modules operated in the
KVK loop and the PNP component [29]
U-tube counter-current compact

Helical type IHX design
IHX design
PNP KVK PNP KVK
Prim. Secon. Prim. Secon. Prim. Secon. Prim. Secon.
Power [MW] 125 10 125 10
Flow rate
36.9 35.6 3.0 2.9 36.9 35.6 2.95 2.85
[kg/s]
Temperature in
950 220 950 220 950 220 950 220
[°C]
Temperature
293 900 293 900 293 900 293 900
out [°C]
No. of tubes 1900 180 1444 117
Tube length
32.7 32.7 17.5 17.5
[m]
25
The IHX is a new type of component in terms of its specific nuclear requirements
and its thermal load. The aim is a service life of 140,000 h at a system pressure of 4 MPa
and temperatures up to 950 °C [29].
Drawbacks of an IHX are its technical problems due to the higher temperature, plus
50 - 70 K, compared with the reformer, its additional electricity demand, 50 MW(e) in a
3000 MW(th) PNP HTGR, which reduces the overall efficiency and last but not least, the
fact that it presents an additional component which may fail.
The most economic medium for the IHX is helium. However, others such as CC>2,
N2, steam, Ne, or Ar could be used also. The design of an IHX is shown in Fig. 2-13.
It is composed of two separate heat exchanging systems for high-end and low-end heat,
here with a heat split-up at 550 °C [22].
For the German PNP project (see section 4.3.2.2.), two concepts of a He - He
intermediate heat exchanger for a heat rating of 125 - 170 MW have been selected: (i) the
helical heat exchanger elaborated by Sulzer/Steinmuller, and (ii) the U-tube compact heat
exchanger elaborated by Balcke-Durr. For both, a 10 MW test plant has been operated in
the KVK loop (see section 4.3.2.6.). Design data of the test modules, as shown in Table
2-8, corresponded to the future nuclear component data with the exception of the helium
flow rate. The test program included fatigue tests (temperature cycles) and creep buckling
tests (cooling in the presence of strong pressure gradients) for the as-manufactured and
the fatigued components, as well as steady-state operation at 40 - 100 % load and non
steady-state tests, e.g., startup / shutdown operation [29].
The helium-to-helium intermediate heat exchanger (IHX) designed and fabricated for
the HTTR in Japan (see section 4.5.2.2.) is a vertical helical coil counterflow type heat
exchanger with 10 MW thermal power. As a primary circuit component, it has to meet the
requirements on structural integrity as a class 1 component during its service life. The tube
bundle consists of 96 heat transfer tubes which have an effective height of 4.87 m, an outer
diameter of 31.8 mm and a wall thickness of 3.5 mm. The tubes are fabricated from the
Ni-base superalloy Hastelloy XR. Radiation plates between the tubes are to increase heat
transfer and restrain flow-induced vibrations. Operating conditions are a temperature of 950
°C and a pressure of 4.1 MPa on the primary side and 905 °C and 4.2 MPa, respectively,
on the secondary side. The construction of the IHX was successfully finished in 1994 and
installed in the HTTR reactor building in 1995 [21].
233. Hot Gas Duct
The hot gas duct is represented by coaxial pipes between core and reformer. The
inner hot gas pipe is insulated on the inside and cooled by counterflow on the outside.
Since the steam generator has to be on the alert in case of a decay heat removal action,
the pipe system has to be safely constructed. In particular, fast pressure transients which
could be destructive should be avoided by means of flow limitation.
For the 3000 MW(th) PNP plant, the inner diameter is 1.3 m, the insulation thickness
is 100 mm (Fig. 2-14). The mass flow rate is 205 kg/s at a speed of 51 m/s. The
26
CORE EXIT I ENTRANCE
CAXIAL DUCTS
It DUCTS -3000H%,-9SO-250*C1
.950*C
1.327m'
50.7mA
Fig. 2-14: Schematic of a hot gas duct between the reactor core and the steam
reformer, from [36]
depressurization rate is limited to 0.1 MPa/s. The component is designed to be removable

as long as no reliable data on its lifetime are available [36].
Experimental work has been conducted at the Research Center Jiilich starting in 1971
to verify thermal efficiency, mechanical performance, applicability of materials and coatings
for the hot gas duct insulation. Facilities used were the ARGUS loop (max. temperature
1000 °C, max. pressure 1 MPa), a high pressure wind tunnel (max. temperature 400 °C,
max. pressure 4 MPa), and the high temperature helium test rig, HHV, (max. temperature
850 °C, max. pressure 5.1 MPa). Also a full-scale test facility, ADI, (max. temperature
950 °C, max. pressure 8 MPa) was operated by HRB for endurance tests as well as the
KVK loop by INTERATOM. Feasibility was demonstrated for the insulation design applied
in the THTR-300, which is composed of 18 layers, each consisting of a 10 mesh gauze
sandwiched between two stainless steel sheets [6].
Protective coatings for the hot gas duct tested were a NiCrAlY basis layer followed by
a ¥203 stabilized ZrC>2 layer. Other components which need to be especially designed are
the isolation valves. Two different kinds have been investigated, axial-type and ball-type
valves. Further tests before application are still necessary [41].
2.4. MATERIALS FOR NUCLEAR POWER PLANT AND PROCESS COMPONENTS
2.4.1. Materials Testing at the Research Center Jiilich
A comprehensive testing program for metallic materials was initiated at the Research
Center Julich within the PNP_and HHT projects in a joint effort with industrial partners
[32]. It is of utmost importance for design and valuation of component behavior during
operation, since temperature limits covered by the existing regulations are exceeded by the
PNP and HHT design limits. Significant components are those with the highest operational
temperatures (> 750 °C), those with a pressure load, and those which represent the boundary
27
of the primary circuit. Considering a 40 years plant lifetime and in between one replacement
of a critical component, the effective operation time should be in the order of 100,000 hours.
Priority tasks have been the selection of high temperature, high strength alloys and
testing of candidate commercial alloys. The material characteristics investigated included
creep behavior, fatigue properties, structural stability, and corrosion resistance [32].
The characterization program for high-temperature alloys included the alloy types:
• INCOLOY 800H
representing Fe-Ni-Cr alloys used for nuclear steam generators,
• HASTELLOY X and NIMONIC 86
representing Ni-Cr-Mo alloys developed for gas turbines,
• INCONEL 617
representing Ni-Cr-Co-Mo alloys for gas turbines,
• NIMONIC 80A,
• ALLOY 713LC
as turbine blade material,
• TZM
on Mo basis with very high creep rupture strengths, however, becomes brittle at
lower temperatures.
Creep rupture strength experiments have been conducted in the ZEMAK I - IV

laboratory facilities in Julich. Fig. 2-15 shows some measurement results. Also fatigue
tests with periodic stress - strain impact have been made to simulate load changes.
Corrosion effects on the metallic surfaces changing the mechanical properties of the
materials have been investigated. Different kinds of impurities in the hot helium have been
observed to cause different corrosion phenomena resulting from the interaction of carbon
dissolved in the alloy with the surface oxide [32].
When exposed to neutron irradiation, heat resistant austenitic alloys tend to embrittle
above 500 °C caused by (n,a) transmutations. The effect can be measured by tensile and
creep testing. Special thermo-mechanical treatment of the alloys was found to successfully
improve their ductibility [32].
In the qualification program for metallic high-temperature materials, lifetimes of steam

reformer and He-He heat exchanger materials have been approved to reach the required
100,000 h. Materials tested in the high temperature range of the KVK facility were the
types 1.4876 (< 800 °C) and 2.4663 (< 1000 °C). Experience gained on the interaction
between metallic and ceramic materials and the helium coolant confirmed the operating
specifications. Furthermore, CFC materials for the hot gas duct are in development. Still
the experience from the PNP and HHT projects has disclosed that the technical solution, in
particular, of materials problems require further research in longer-term projects. It was the
reason that the plan to operate the THTR-300 as a process heat plant was no longer pursued
at that time and rather postponed this plan to be used in a future 900 MW(e) HTGR.
28
1000
N/mm2
100-
"w
1
"5.
10-
£
o
ASME-CC-N 47
for INCOLOY 800 H
Alloys
for HHT/PNP/NFE
500 600 700 800 900 °C 1000

Temperature —>
Fig. 2-15: Creep rapture strength experiments for different alloys under consideration
for HTGRs [32]
2.4.2. Materials Testing at JAERI
The material chosen for the IHX and the steam reformer of the Japanese HTTR
is HASTELLOY XR. Corrosion tests and long-term creep tests in helium atmosphere in
the temperature range 800 - 1000 °C have been conducted reaching 50,000 hours. No
significant degradation in creep properties was observed even after carburization. Creep
rupture was found to be caused by nucleation, growth, and link-up of grain boundary
cavities [26].
Satisfactory behavior of Hastelloy XR material has been found with respect to

weldability, tensile and creep properties through optical microscopy and corresponding
tests in air at room temperature and at high temperatures up to 950 °C, respectively [40].
Also the investigation of the potential use of SiC ceramics for heat exchanger and
vessels, high-temperature properties and oxidation behavior has been started, e.g., on a
29
SiC-coated vessel with potential use for thermal chemical reactions for hydrogen processes.
Other tests conducted at JAERI examine the corrosion exposure in boiling sulfuric acid.
Mechanical bending tests are planned at the Research Center Jiilich.
Further work will be focusing on lifetime prediction methods for gas turbine materials
(superalloys) and experiments on fracture mechanics in superalloys as well as finite element
calculations of multiaxial loaded tubes and validation of multiaxial tests (tension, torsion,
internal pressure).
[I] BARNERT, H., Der Verbund von Kohle, Stahl und Kernenergie, Report Jiil-2085,
Research Center Julich (1986).
[2] BARNERT, H., The Status of the High Temperature Reactor and H2 Related R &
D Work, Hydrogen Production, (2nd IEA Technical Workshop, Julich, FRG, 1991),
STRUCK, B.D. (Ed.), Implementing Agreement for a Programme of Research
and Development on the Production of Hydrogen from Water, Document HUF-
6, Research Center Julich (1991) 113-132.
[3] BARNERT, H., Energy Alcohol from Plant Biomass plus High Temperature Heat,
the CO2-Neutral, Environmentally Benign, and Consumer Friendly Future Alterna-
tive, Report Jul-3089, Research Center Julich (1995).
[4] BECKURTS, K.-H., DffiTRICH, G., Projekt Femwarme Versorgung fur Millionen-
Stadte, Bild der Wissenschaft, 13 (1976) January 64-70.
[5] BIRNBAUM, U., RIENSCHE, E., SUMMING, U., Optimization of Small SOFC
Plants with Flexible Power/Heat Ratios and their Future Application, (5th Int.
Symp., Aachen, 1997), SUMMING, U., et al. (Eds.), Solid Oxide Fuel Cells
V, Volume 97-40, The Electrochemical Society, Inc., Pennington, USA (1997)
112-123.
[6] BROCKERHOFF, P., et al., Status of Design and Testing of Hot Gas Ducts, Nucl.
Eng. Des. 78 (1984) 215-221.
[7] ENGELHARDT, H.R., EinsatzmOglichkeiten und Einsatzpotential von Hochtem-
peraturreaktoren in der Industriegruppe Chemie unter Beriicksichtigung von tech-
nischen, okonomischen und okologischen Aspekten, Report Jul-966-RG, Research
Center Julich (1973).
[8] FARBMAN, G.H., Studies of the Use of Heat from High Temperature Nuclear
Sources for Hydrogen Production Processes, Report NASA-CR-134918, NASA
Lewis Research Center (1976).
[9] FROHLING, W., NEEF, H.J., Synthetisches Naturgas aus Kohle und
Hochtemperatur-Reaktorwarme, KFA Annual Report 1977, Research Center
Julich (1978) 15-24.
[10] GAO, Z., Auslegungsuntersuchung fur den HTR zur ErschlieBung von Schwer-
stollagerstatten, Report Jiil-1977, Research Center Julich (1985).
[II] HADA, K., FUJIMOTO, N., SUDO, Y., Design of Steam Reforming Hydrogen
and Methanol Co-Production System to be Connected to the HTTR, (TCM, Oarai,
1992), High Temperature Applications of Nuclear Energy, Report IAEA-TECDOC-
761, International Atomic Energy Agency, Vienna (1994) 124-134.
30
[12] HAD A, K., Improvements of Reforming Performance of a Nuclear Heated Steam
Reforming Process, Report JAERI-Research-96-054, Japan Atomic Energy Rese-
arch Institute (1996) (in Japanese).
[13] HADA, K., et al., Universally Applicable Design Concept of Stably Controlling an
HTGR Hydrogen Production System, Nippon Genshiryoku Gakkai-Shi 38 (1996)
834-844 (in Japanese).
[14] HANNEMAN, R.E., VAKIL, H., WENTORF, R.H.JR., Closed Loop Chemical
Systems for Energy Transmission, Conversion and Storage, Proc. Intersoc. Energy
Conversion Eng. Conf. (1974) 435-441.
[15] IAEA, Design and Development Status of Small and Medium Reactor Systems
1995, IAEA-TECDOC-881, International Atomic Energy Agency, Vienna (1996).
[16] IAEA, Fuel Performance and Fission Product Behavior in Gas-Cooled Reactors - A
Compilation Produced within the IAEA Coordinated research program on Validation
of Predictive Methods for Fuel and Fission Product Behavior in Gas-Cooled Reac-
tors, IAEA-TECDOC-978, International Atomic Energy Agency, Vienna (1997).
[17] IDE, A., TAKENAKA, Y., MAEDA, S., Utilization of Heat of Modular High
Temperature Gas-cooled Reactors, Proc. ASME-JSME 4th Int. Conf. on Nuclear
Engineering, RAO, A.S., DUFFEY, R.B., ELIAS, D. (Eds.), The American Society
of Mechanical Engineers (1996) 255-261.
[18] JAGER, W., WEISBRODT, I., HORNING, H., Nuclear Process Heat Applications
for the Modular HTR, Nucl. Eng. Des. 78 (1984) 137-145.
[19] JOBSKY, T., Ermittlung von Potentialen zur industriellen ProzeBdampfversorgung
mit nuklearen Anlagen, Report Jiil-2411, Research Center Jiilich (1990).
[20] KNOCHE, K.F., Stand der Arbeiten zur Wasserstoff-Erzeugung mit nuklearer
ProzeBwarme, Chem.-Ing.-Tech. 49 (1977) 238-242.
[21] KOIKEGAMI, H., et al., Design and Fabrication of He-He Intermediate Heat
Exchanger for HTTR, (3rd JAERI Symp., Oarai, 1996), Proc. JAERI-Conf 96-
010, Japan Atomic Energy Research Institute (1996) 305-321.
[22] KUGELER, K., et al., The Pebble-Bed High-Temperature Reactor as a Source of
Nuclear Process Heat, Vol 3: System Considerations on the Nuclear Reactor, Report
Jul-1115-RG, Research Center Jiilich (1974).
Nuclear Process Heat, Vol 4: System Considerations on Nuclear-Heated Steam
Reformers, Report Jul-1116-RG, Research Center Julich (1974).
[24] KUGELER, K., PHLIPPEN, P.-W., The Potential of the Self-Acting Safety Features
of High Temperature Reactors, Kerntechnik 61 (1996) 239-244.
[25] KUPITZ, J., Role of IAEA in Non-Electric Applications of Nuclear Energy, Non-
Electric Applications of Nuclear Energy, (Advisory Group Meeting, Jakarta, Indo-
nesia, 1995), IAEA-TECDOC-923, International Atomic Energy Agency, Vienna
(1997) 119-131.
[26] KURATA, Y., et al., Evaluation of Long-Term Creep Properties of Hastelloy XR
in Simulated High-Temperature Gas-Cooled Reactor Helium, (3rd JAERI Symp.,
Oarai, 1996), Proc. JAERI-Conf 96-010, Japan Atomic Energy Research Institute
(1996) 338-352.
[27] LIEBHOLZ, W.-M., (Ed.), Jahrbuch der Atomwirtschaft 1997, Vol. 28, Verlags-
gruppe Handelsblatt Fachverlag, Dusseldorf (1997).
31
[28] MANTHEY, C, Einsatz von Hochtemperaturreaktoren in der Eisen- und Stah-
lindustrie unter besonderer Berucksichtigung des Einflusses auf die Standort- und
Umweltprobleme dieses Industriezweiges, Report Jul-1180, Research Center Julich
(1975).
[29] MAUS, W., et al., The He / He Heat Exchanger - Design and Semitechnical Testing,
Nucl. Eng. Des. 78 (1984) 195-214.
[30] MIYAMOTO, Y., et al., Overview of HTGR Utilization System Developments
at JAERI, High Temperature Gas Cooled Reactor (HTGR) Development, (Techn.
Comm. Meeting, Johannesburg, South Africa, 1996), IAEA-TECDOC in Prepara-
tion, International Atomic Energy Agency, Vienna (1997).
[31] NABIELEK, H., FUKUDA, K., KAMA, M.J., Fuels Technology, Closed-Cycle
Gas-Turbine Modular High-Temperature Gas-Cooled Reactor, (Int. Workshop,
Cambridge MA, 1991), PENFIELD, S.R., KASTEN, P.R., Massachusetts Institute
of Technology (1991) 4-8 - 4-16.
[32] NICKEL, H., SCHUBERT, F., SCHUSTER, H., Evaluation of Alloys for Advanced
High-Temperature Reactor Systems, Nucl. Eng. Des. 78 (1984) 251-265.
[33] NIEBEN, H.F., et al., Erprobung und Versuchsergebnisse des PNP-Testrohrenspalt-
ofens in der EVA-H-Anlage, Report Jiil-2231, Research Center Julich (1988).
[34] PESCHEL, W., Analyse der Erzeugung von Methanol bzw. Wasserstoff aus
einem modifizierten Hochofengichtgas unter Einsatz von Kernenergie aus dem
Hochtemperaturreaktor, Report Jul-2031, Research Center Julich (1985).
[35] SANDSTEDE, G., Moglichkeiten zur Wasserstoff-Erzeugung mit verminderter
Kohlendioxid-Emission fur zukunftige Energiesysteme, Chem.-Ing.-Tech. 63
(1991) 575-592.
[36] SCHULTEN, R., et al., The Pebble-Bed High-Temperature Reactor as a Source of
Nuclear Process Heat, Vol 1: Conceptual Design, Report Jul-1113-RG, Research
[37] SIMNAD, M.T., GOODJOHN, A.J., KUPJTZ, J., High-Temperature Helium Gas-
Cooled Nuclear Reactors: Past Experience, Current Status and Future Prospects,
Energy 16 (1991) No. 1/2, Special Issue, Pergamon Press, Oxford (1991).
[38] SINGH, Y., et al., The Nuclear Heated Steam Reformer - Design and Semitechnical
Operating Experiences, Nucl. Eng. Des. 78 (1984) 179-194.
[39] STARR, C., SEARL, M.F., ALPERT, S., Energy Sources: A Realistic Outlook,
Science 256 (1992) 981-987.
[40] WATANABE, K., et al., Evaluation of Materials Performance of Hastelloy Alloy
XR for the High Temperature Engineering Test Reactor Components - Weldability
and High Temperature Strength Properties, (3rd JAERI Symp., Oarai, 1996), Proc.
JAERI-Conf 96-010, Japan Atomic Energy Research Institute (1996) 368-382.
[41] WEISBRODT, I.A., Summary Report on Technical Experiences from High-
Temperature Helium Turbomachinery Testing in Germany, (TCM, Beijing, 1995),
Design and Development of Gas Cooled Reactors with Closed Cycle Gas Turbines,
Report IAEA-TECDOC-899, International Atomic Energy Agency, Vienna (1996)
177-248.
[42] YAMADA, M., et al., A Preliminary Study on Co-Generation Systems with the
Modular HTGR for Japanese Conditions, (TCM, Oarai, 1992), High Temperature
Applications of Nuclear Energy, Report IAEA-TECDOC-761, International Atomic
Energy Agency, Vienna (1994) 46-54.
32
Chapter 3
SAFETY CONSIDERATIONS CONCERNING

NUCLEAR POWER PLANTS ASSOCIATED WITH
PROCESS HEAT AND/OR HYDROGEN PRODUCTION
3.1. SAFETY DESIGN CONCEPT SPECIFIC TO NUCLEAR HEAT UTILIZATION

SYSTEMS
Present nuclear power plants must be designed and operated to current safety ob-
jectives and principles in order to achieve a high level of safety. Nuclear power has a
small environmental impact when operating as designed, but faces serious public concerns
about the risk of off-design accidental releases. This has stimulated the current engineering
concentration on reducing the probability of such accidents.
After the severe accidents at the Three-Mile-Island and Chemobyl nuclear power
stations, new designs with improved safety concepts have been proposed focusing on
reducing the probability and consequences of severe accidents, on an improved containment
system, and on more intensive consideration of passive safety characteristics. Different
categories of passive systems have been defined by the International Atomic Energy Agency
in the course of these discussions which appear to be beneficial [17], although no method
has been finally agreed upon which type of system is more appropriate in certain cases.
The criteria for a decision include maintenance, redundancy, diversity, quantification of
reliability or human interaction. The key proposal is that also severe accidents beyond the
design basis will be systematically considered during the design process for future reactors.
Further research and development is required to assure the effectiveness and reliability of
passive safety systems [15].
In many countries of the world, a large number of different R&D and design projects
have been initiated to serve for a wide range of applications. Advanced light-water reactors
(ALWR), also called evolutionary reactors, are to obey the overall design philosophy of
being simple and rugged with high design margin, based on proven technology. The ALWR
safety policy features an integrated design approach to safety based on defense-in-depth
philosophy. This strategy includes the three levels of accident resistance, core damage
prevention, and accident mitigation. Fig. 3-1 lists most of the projects of small and
medium sized reactors and the current status of development as reported in [18]. Pros and
cons as seen at DOE [52] are presented in Fig. 3-2.
A much more far-reaching demand for the introduction of innovative nuclear power
plants in the future is made on a design such that fission product release is made impossible
or, at least, restricted to the plant itself [27]. This means that emergency planning would
no longer be necessary or in more detail:
• no immediate fatalities outside the plant,
• no evacuation of population,
• no contamination outside the plant.
33
Light-Water Reactor Heavy Gas- Liquid
Heat Water Cooled Metal
CHP Electr. Prod. Reactor Reactor Reactor
Prod.
ATS-150 MHTGR Rapid

V* 600
HR-200 ABV SIR ADP
:
Geyser Shakha-92 SBWR; ISIS MHTR MDP
PHWR-
Ruta KLT^B MARS
500
HTR-500 SAFR
K
Triga PIUS BWR-600 GT-MHR 4S
Thermos CAREM VVER PRISM
SES-10 MRX HSBWR BMN-170
Siemens SPWR
Secure-H
Conceptual Design
Basic Design
Detailed Design
ifSS Design under Construction
Fig. 3-1: Small and medium reactor development lines in the world, from [18]
Proponent and Critic Claims

Proponents
• LWRs are cost-competitive with other baseload generation options; ALWRs will be more cost-competitive.
• Nuclear energy is environmentally clean.
• Nuclear fuel cost is low and stable, and this option provides energy security benefits and fuel diversity.
• LWRs have an excellent safety record; ALWRs will be even safer than LWRs.
• LWRs are reliable, ALWRs will be more reliable. At cntical times (e.g., when coal piles freeze, natural gas pipelines reach their capacity, the wind isn't
blowing), nuclear plants continue to generate electricity.
Critics
• Nuclear energy is not cost-competitive.
• It is more important to stop producing spent nuclear fuel. Then we can talk about solutions to proper disposal of that already on hand.
• Fuel costs are low because utilities pay only the marginal cost of enrichment services; the taxpayer paid for the development of this capability and continues
to pay for capability maintenance.
• Nuclear energy is not safe.
• Nuclear energy is not reliable because of major, unplanned repairs (e.g., steam generator replacement).
Fig. 3-2: Pros and cons of an advanced light-water reactor (ALWR), from [52]
34
The demand for nuclear, thermal, chemical, and mechanical stability of a nuclear
reactor needs to ensure the following safety requirements [27]:
• inherent limitation of nuclear power and of fuel temperature,
• passive decay heat removal,
• barrier fuel remains intact, integrity of core against corrosion attack,
• barrier reactor pressure vessel remains intact, integrity of core against mechanical
failure,
• barrier containment remains intact,
• no interdependencies among barriers.
The safety philosophy for future nuclear plants will most probably include the
protection against aircraft crash (penetration and induced vibrations) as well as chemical
explosion waves such that a respective contribution to the risk is minimal.
3.2. SAFETY CONSIDERATIONS FOR A NUCLEAR PROCESS HEAT HTGR
3.2.1. Activity Enclosure
For safety reasons, the system design of an HTGR plant with process heat application
has to assure the restriction of contamination with radionuclides to the primary circuit or
its reduction to the least possible in the case of leakage or permeation. One possibility
is to deploy an intermediate loop as an additional safety barrier to the system. In
such an arrangement, all heat-consuming components are physically separated from the
primary circuit by means of an intermediate heat exchanger. In the case of heat consuming
components to be arranged in the primary circuit, they have to have themselves the quality
of a safety barrier. In case of an accidental release of product gas, the gas mixture escaped
into the reactor building and mixed with air, carries the hazardous potential of a local
explosion.
Safety-related layout of the plant comprises double-walled pipes hi product gas

carrying ducts and deployment of isolation valves. Released process gases should be
routed immediately into the atmosphere to avoid accumulation inside the reactor building.
The prevailing idea currently is to install the heat transferring components of a nuclear
coal gasification plant - gas generator, steam superheater, steam generator - in a secondary
circuit, whereas in a nuclear steam reforming plant, steam reformer and steam generator are
thought to be safely operated within the primary circuit [46]. With respect to the operational
conditions, a compromise has to be found between a high-pressure system with a positive
effect on operational and accidental behavior and a low-pressure system in all circuits to
minimize the load on the components. For the active reactor core, a two-zone fueling should
be chosen to avoid radial temperature gradients to proceed to the heat exchanger [20].
The heat consuming components are, in general, not appropriate to serve as the decay
heat removal system in case of a nuclear accident. Therefore, in the Japanese design of
the HTTR, an auxiliary cooling system is planned to remove the decay heat via forced
convection, as it was also the case in the former German PNP-500 process heat reactor
design. In the most recent German AHTR-500 design, an adequate passive system is being
planned based on natural convection and there is no demand for active safety systems.
35
Thermal loads upon the process heat exchanger do not allow rapid temperature change
rates. Therefore in case of the demand for a decay heat removal system after a fast shut-
down of the reactor, an auxiliary cooling system should be used rather than the main cooling
system. In addition, much of the decay heat would be removed via the core surfaces to
the liner cooling system [10].
In case of a pressure loss accident, the maximum escape rate of the primary coolant
is restricted by flow restrictors. A depressurization will take at least two minutes so that
destructive dynamic forces in the primary system can be excluded. All heat exchanging
components incl. steam reformer are designed to keep their pressure, if there is a pressure
loss in the primary system [10].
A safety feature, in particular for a nuclear power plant on an industrial site, is a

compact arrangement which allows an underground construction. Such an arrangement as
was proposed for the HTGR design PR 500 for process steam production would be safe
against gas explosions and air craft crashes.
3.2.2. Primary Circuit Contamination
The deposition of fission products on primary circuit surfaces and, in particular,

on the reformer tube walls causes difficulties during maintenance and catalyst refilling
procedures, if the activity is intolerably high. Cesium and silver isotopes released during
reactor operation are of major concern. Particularly silver diffuses easily out of the fuel
elements at operating temperature conditions into the coolant and migrates easily into metal
surfaces and is difficult to remove in decontamination operations [32].
In a gas turbine version of the HTiR, half of the radiation dose is estimated to
originate from the silver isotope Ag-llOm. Countermeasures could be either a fuel
temperature reduction which would require a change of the core design, or the use of
advanced fuel with the most efficient barrier against fission product release, the SiC coating
layer, to be substituted by a ZrC layer. The higher retentivity of ZrC against Ag release,
however, needs further confirmation [40].
3.2.2.1. Plateout Research Program
The circulating and plateout activities in the primary coolant circuit are potential
sources of release into the environment in the event of primary coolant leaks or as a
result of the venting of primary coolant in response to overpressuring the primary circuit,
e.g., in response to significant water ingress. The fraction of the circulating activity lost
during such events is essentially the same as the fraction of the primary coolant that is
released, although the radionuclide release can be mitigated by pumpdown through the
helium purification system, if the leak rate is sufficiently slow.
Numerous experimental and theoretical efforts have been made to examine the
plateout distribution of fission products in the primary circuit both under normal operating
and accident conditions. Different in-pile and out-of-pile deposition loops were operated
in Germany, Japan, France, the UK, and the USA to study systematically the ad-/desorption
behavior of fission product on metallic surfaces as a function of temperature and gas
36
flow. The obtained experimental data as well as the measurements from the gas-cooled
reactors AYR, THTR-300, Peach Bottom, Fort St. Vrain were taken to derive plateout
parameters such as desorption energy or penetration coefficient to be used in corresponding
calculation models. The most important computer codes which were tested and more or less
successfully validated, are SPATRA (Germany), PLAIN (Japan), and PADLOC, TRAPIC
(USA) [19].
Within the HHT program, the AYR reactor has been used to conduct several expe-
riments for investigation of atomic or molecular radionuclides in the coolant flow as well
as transport and deposition with graphitic or metallic dust. Cold gas filter tests and dust
experiments were used to measure dust size and concentration as well as vapor-phase and
dust-borne metallic fission product activities in the coolant under normal operating con-
ditions. The in-pile experiments VAMPYR-I and VAMPYR-JJ starting in 1972 were hot
gas sampling tubes to determine diffusion profiles of deposited fission products on various
materials. The helium loop facility LAMINAR in Germany was operated in the turbulent
flow range with Cs-134 or 1-131 to examine their deposition behavior under varying tem-
perature conditions and with desorption periods. Material samples were later inserted in
steamoff/washoff experiments. In the out-of-pile circuit SMOC, cesium transport and depo-
sition under the turbulent flow conditions of a depressurization accident were investigated
by varying in a wide range of gas/wall temperatures and mass flow rates.
In Japan, the Oarai Gas Loop No. 1, OGL-1, installed in the JMTR, was used for
cesium and iodine plateout distribution measurements basically under the normal operating
conditions of the HTTR. Within a joint German/French effort, another plateout experimental
series was conducted in the in-pile loop SAPHIR in the French PEGASE reactor. Also
extensive plateout distribution data were obtained in the UK by examining components
removed from the Dragon HTGR
The US CPL-2 test program was a series of four in-pile loop tests in the French
PEGASE reactor using a small fraction of bare U(>2 kernels as the fission product source.
Plateout data from heat exchanger tube consisting of various types of steels were gathered
under somewhat complex temperature and flow field conditions. Out-of-pile loop data
were obtained in five GA deposition loop tests to study plateout under conditions similar
to sections of an HTGR steam generator or gas turbine. The COMEDIE loop in the French
SILOE MTR composed of an in-pile and an out-of-pile section was capable of providing
integral test data under realistic accident conditions. The plateout section simulated different
metallic components in the primary circuit of an HTGR.
3.2.2.2. Corrosion
Furthermore impurities in the helium coolant, mainly the air constituents, can cause
corrosion effects on the outside reformer tube walls which eventually change its properties.
Measurements of impurity contents in Dragon and AYR revealed a large scattering of the
data. Experimental results obtained within the Dragon project indicate a strong corrosion
of aluminum and titanium, i.e., the formation of Cr-, Mn-, Si-, and Ti-oxide layers, and an
increased corrosion rate in moist helium compared with a dry atmosphere [26].
37
3.23. Hydrogen and Tritium Behavior
Under conditions of nuclear process heat plant operation, hydrogen and tritium
exhibit high mobility at 950 °C causing permeation through the walls of the heat-exchanging
components. Tritium produced in die primary circuit could permeate into the product gas
in the secondary circuit and eventually create a radiation problem to the consumer. In
the opposite direction, hydrogen from the secondary circuit could diffuse into the primary
circuit to cause corrosion of the fuel elements and of the graphite liner in the core with the
formation of methane (which finally could decompose again and deposit carbon in the heat
exchangers). Parameters which influence permeation are, besides temperature and partial
pressure, wall material and thickness as well as surface size and roughness. An upper limit
for the tritium contents in the product gas set by regulations as low as 10 pCi/g, turned
out to be a key problem for the PNP project [47]. In the meantime, the limit has been
raised to 135 pCi/g restricted to refined products of nuclear heat processes from fossil fuels
in domestic applications.
The tritium production in a process heat HTGR is estimated to originate mostly from
activation of Li-6 which is contained in the graphite, and with smaller fractions from fission
and from He-3 activation. The release of tritium into the coolant is based on conservative
assumptions, e.g., 10 % of the fission product tritium from defective coated particles.
Estimated tritium production and release rates for a 170 MW(th) HTR-MODUL plant are
given listed in Table 3-1 [8].
Table 3-1: Tritium production and release rates as estimated for a 170 MW(th)
process heat HTR-MODUL, from [8]
Tritium production Tritium release into coolant

Tritium [103 Bq/s] ([%]) [103 Bq/s] ([%])
source from
Initial phase Equilibrium Initial phase Equilibrium
Fission 898 (14) 1245 (51) 89(4) 126 (12)
Li-6 4721 (76) 846 (34) 1413 (66) 529 (52)
He-3 628 (10) 367 (15) 628 (30) 367 (36)
Possibilities of reducing the permeation streams are the formation of oxide layers (in-
situ or pre-coated) or, as active measures, an improved design of the gas purification system
or even the construction of an intermediate circuit between the primary and secondary circuit
to serve as a sink for both hydrogen and tritium.
• Oxide layers
Oxide layers on the heat exchanging material surfaces were found to be capable
of significantly decreasing the tritium transport. Experimental investigations of
the permeation process in high-temperature alloys were initiated at the Research
Center Julich [49]. Test facilities were "AUWARM", which allowed long-term
(1000 - 3000 h) examination of hydrogen permeation in 16 tube specimens at a
time with temperatures up to 1000 °C and pressures up to 3.2 MPa, "TRTPERM"
operated with an argon-tritium mixture on the primary side for one tube specimen,
38
and "HD-PERM", "SE-PERM", "DEUPERM" for short- and medium-term analy-
ses and pre-selection of materials at low pressures.
Results have shown that in-situ oxidic layers show a large inhibition of permeation
in the medium and high-temperature range > 650 °C. The uncertainty is compa-
ratively large in the lower temperature range. Also measurements from existing
HTGRs (AYR, Peach Bottom) revealed significant uncertainties, meaning that data
are not necessarily transferable to new plants [47].
• Gas Purification System
An effective and powerful gas purification system could be designed by integrating
additional components for hydrogen and tritium gettering from the coolant gas.
Hydride formers such as Ti, Zr, and Ce or a cerium mixed metal were deemed
appropriate. A respective experiment was planned at the AYR reactor to test a Cer
mixed metal gas purification system with a throughput of 8 Nm3/h [8].
• Intermediate Circuit
An additional boundary between primary and secondary circuit would be provided
by an intermediate circuit purified by a sweep gas flow. Adding oxygen or steam
to the sweep gas, a fraction of the tritium will be bound as tritiated water HTO
and no longer be available for permeation plus it would form a protective oxide
layer. An inhibition effect was also found if hydrogen was injected (factor of 2 -
3). This effect, however, needs further understanding [47].
An intermediate circuit was experimentally tested in a facility whose heart was a
double tube consisting of Incoloy 800H. The inner tube simulated the secondary
circuit carrying the product gas. The gap between inner and outer tube represented
the intermediate circuit. The tubes were closed at both ends. Gas inlet/outlet was
given by small pipes into the system. The whole construction was put underneath
a quartz glas recipient which simulated the primary circuit. Argon was used as
a sweep gas. Hydrogen concentrations in the different circuits were measured by
means of gas chromatography, while the tritium concentration in the primary circuit
was measured in ionization chambers. The facility was used to check how partial
pressures of hydrogen and tritium in all circuits were affected by the design of the
intermediate circuit, and to derive permeation constants and activation energies [5].
3.2.4. Disturbance During Steam Reformer Operation
In the hydrogen production system designed for the HTTR, an approximate 4 MW

power share is consumed for the reforming reaction while 6 MW are utilized downstream
for steam production in a steam generator. Analytical studies have been conducted at JAERI
with the transient thermohydraulics computer code IRIS to investigate the steam generator
as a passive heat sink in case of a failure of the heat removal system of the reformer to
ensure its coolability [44]. JAERI has developed a computer model to analyze the transient
behavior of the nuclear reformer system. It is able to treat the chemical reforming reactions
under the influence of gas (methane, feedwater) flow disturbances connected to the transient
thermohydraulics by using the THYDE-HTGR code to account for the reactor side [11].
Tests with the EVA-U plant in Jiilich have shown that its normal operation is possible
even if up to 30 % of the reforming tubes were blocked, meaning that isolation of single
tubes would not disturb the operation, thus increasing the plant flexibility [45].
39
Advanced steam generators with a large steam drum (high reliability, passive cooling
capability) can mitigate steam reformer disturbances. The German once-through type of
steam generator is designed to be integrated inside the reactor pressure vessel. It has been
tested in the EVA-II test facility.
Alternative design of a steam reformer contains a spherical type catalyst which itself
moves downwards while the process gas flows upwards through the catalyst This method
increases the efficiency (smaller residual methane contents). Drawback is a catalyst damage
due to the movement.
3.2.5. Rupture of Helium Line in Gasifier
The accidental potential of a coal gasification plant with a temperature of 950 °C

which has combustible gases in the secondary and/or tertiary loop must be classified as
being higher than that of a conventional nuclear plant. In PNP concepts, an isolation valve
is considered to limit the amount of explosive gas mixture in the containment.
The system pressure in the gasifier is designed to be 2.1 MPa, meaning that in case
of a ruptured helium line which is normally at 2 MPa, no radioactivity could escape from
the primary circuit, but product gas and steam as well as coal particles could enter the
primary helium circuit. Leakage will be detected by moisture sensors, helium lines shut off
from the process gas circuit. The principal consequences of steam entrainment in a steam-
producing HTGR are equivalent to those of a water ingress accident into the core. Coal
dust in the primary circuit provides an additional carrier medium for radioactivity which
could be released into the confinement during a depressurization accident. The escape of
a coal dust - product gas - steam mixture into the confinement would result in a wet dust
which makes an ignition unlikely [25].
3.2.6. Rupture of Process Gas Feedline
3.2.6.1. Release into the Containment
Any process heat plant design implies piping through the containment to connect
the reactor vessel with the chemical plant. The fracture of a pipe could result in the
accumulation of a flammable gas mixture in the containment. Precautions must be taken
to minimize the risk of a fire or gas explosion such as avoidance of explosive gas
ingress, proper detection devices, inerting, sufficient safety distances, appropriate layout of
secondary coolant boundary, explosion-proofed wall, plant isolation valve. For the PNP-
500, the use of two concentric pipes for the process gas carrying lines were recommended.
Alternatives are concrete channels around the gas lines or inerting of the containment [10].
As part of the PNP safety program, Interatom has studied the process gas ingress
into the containment for the scenario of a depressurization plus process gas pipe rupture.
Various containment volumes as well as times for the process gas pipes have been examined
showing that the equilibrium pressure will be in the range of 0.16 - 0.24 MPa at temperatures
40
10 Inerting limit
90
20
30
40
50
Air [Vol %] 50 Helium [Vol %]
60
70
80
AAAAAA
AAAA
90 Flammability
range
10 20 30 40 50 60 70 80 90
Fuel gas (47 Vol % H2 + 10 Vol % CH4 + 43 Vol % CO) [Vol %]
Fig. 3-3: Estimated flammability range of the three-gas mixture fuel gas-helium-air,
from [13]
Safe Distance
C/V:Containment vessel
R/8:Reactor building
I/V:isolation valve
1HX:Intermediate heat exchanger
ECR:Endothermic chemical reactor
FG :Feed gas
Rg. 3-4: Safety design concept in the Japanese HTTR against fire and explosion,
from [33]
41
between 120 and 200 °C. Process gas mass released within an assumed time period of 20
s will be 1100 kg in the primary system winding up to 1700 kg of flammable gases, if all
steam were transformed in the core to H2 and CO. Composition of the gas will then be 47
% H2, 43 % CO, and 10 % CFLt. The fraction that eventually enters the containment is no
higher than 1.3 - 2.4 vol% which, however, does not preclude local flammable mixtures.
For the above composition of the fuel gas and a temperature of 200 °C, the flammable
range within the three-gas mixture fuel gas-helium-air has been estimated as shown in
Fig. 3-3. For a fuel gas fraction of up to 2.4 vol%, no flammable mixture is expected,
if a homogeneous mixture is assumed. Locally flammable mixtures, however, cannot be
excluded during the mixing phase [13].
Under the assumption of a reformer tube rupture creating a leak size of 80 cm2 with
process gas to escape into the containment over 20 s before the stream is automatically
cutoff, an estimated 3 % of the containment volume is occupied by the process gas.
Furthermore, pessimistically assuming that a stoichiometric mixture is formed and explodes,
results in an overpressure of 46 kPa within 3 s which is expected to cause no damage to
the containment [10].
Safety assurance in the design of the HTTR steam reforming system against fire or
explosion of flammable product gas mixtures is the task of the secondary helium piping
system which is designed to limit gas ingress by means of a combination of containment
vessel isolation valves and emergency shutoff valve in the process gas feed line [33].
3.2.6.2. Release into the Environment
According to a guideline on the protection of nuclear power plants against pressure

waves from chemical reactions issued in Germany in 1974 (see section 3.5.1.), the plant
has to be designed such that it withstands the static and dynamic load generated by the
fast deflagration of a large gas cloud. Also minimum safety distances between plant and
explosible gas mixture were fixed to be at least 100 m.
The release of product or feed gas into the environment and the formation of a free
flammable gas cloud could result in heat radiation and blast wave impact from fire /
explosion. For the German PNP design, it was proposed to ensure that the gas be routed
through inerted channels away from the gasifier and flared at a safe spot. This measure,
however, requires additional components in the confinement, e.g., to allow for a separation
of the coal dust from the gas mixture and a cooling of the mixture [25].
The safety concept of the Japanese HTTR for process heat application in the case of
fire and explosion (see Fig. 3-4) includes the appropriate blast wave resistance design of
building structures, gas leak detection device, fire extinguishing system, emergency shutoff
valve, safety distance which could be reduced by fireproof walls. Current JAERI design
regulations do not allow overpressures beyond 10 kPa (Japanese standard is 12 MPa) and
a wall-averaged temperature of the concrete structures beyond 175 °C [33]. Considering
a natural gas vapor cloud explosion after the release of the total tank contents of 168 t
(= 400 m3) of LNG, the above upper limits translate into a safety distance of as much as
1900 m [12]. Presently, a more practicable figure of 470 m is in discussion for the HTTR.
42
An argument for a safety distance reduction could be the deployment of an underground,
non-pressurized LNG storage tank and ventilation system [34].
3.3. SAFETY AND RISK ASSESSMENT FOR PROCESS HEAT HTGRS
The former Institute for Nuclear Safety Research (now: Institute for Safety Research
and Reactor Technology) at the Research Center Julich conducted in the past various studies
on the safety and risk assessment for nuclear power plants or their designs. Among them
were two dedicated to the German process heat HTGR concepts, AVR-n and the 170
MW(th) HTR-MODUL, whose results are presented in the following sections. A brief
description of the general methodology applied is given in section 8.7.1.
33.1. AVR-H
In 1982, the Research Center Julich presented the conceptual design of a 50 MW(th)
nuclear process heat plant with a pebble-bed HTGR, named AVR-n, for which a safety-
related study has been conducted [29]. Its characteristic features are a slim steel pressure
vessel, no separate decay heat removal system, shutdown and control system via reflector
rods, surface cooling system, and a simplified containment. The safety of the reactor is
principally based on passive system features.
The cylindrical core of 4.1 m height and 2.5 m diameter is fueled in an OTTO loading
scheme. The power density is 2.5 MW/m3. The helium coolant inlet / outlet temperatures
are 250 and 985 °C, respectively, at a system pressure of 4 MPa. The reformer consumes
a thermal power of 20 MW. It consists of a bundle of 68 reforming tubes with a length
of 15 m. Inlet / outlet temperatures of the primary helium are 950 and 680 °C, and of the
product gas 825 and 550 °C, respectively. The steam generator consumes the residual 30
MW(th) with helium inlet / outlet temperatures of 670 and 250 °C with 80 % of the power
providing process steam for the reformer.
Standardized protection actions include shutdown of the reactor, of the main cooling
system (which is not required for keeping the reactor within safe limits) and of the reformer
plant. The two (required) shutdown systems consist of in total 18 absorber rods which are
moved in the side reflector. Diversity is given by the employment of different propulsion
systems. In particular, the active cooling of the core by the main cooling system is not
required because fuel temperatures remain within the safety limits. Only for reactor vessel
protection purposes, the surface cooling system in the reactor is necessary which alone is
able to account for heat removal.
Accident sequences investigated were decay heat removal without / with main cooling
system. In the former case, both core and reformer are cutoff from the coolant circulation
and a gradual temperature balance is obtained by heat conduction, radiation and small
internal convection. The activation of the main cooling system delayed by 1 h encounters
a merely changed temperature distribution allowing for a smooth and gentle restart.
Thermodynamic studies have shown that in case of a loss-of-forced-convection

accident, the maximum fuel temperature is 1150 °C, if the reactor remains under pressure,
and no higher than 1250 °C in a depressurized reactor. The predicted maximum temperature
43
for the pressure vessel is RJ 400 °C and for the containment wall « 340 °C inside and
w 100 °C outside. All figures all uncritical. A temperature-induced release of fission
products from the spherical fuel elements can thus be excluded. Besides, both reformer
and steam generator surfaces provide an enormous adsorption potential for fission products
which is utilized to a small extent only. The concrete walls enclosing the reactor vessel
reach maximum temperatures of 300 °C within 10 days, but no damage to the reactor
components is expected.
A depressurization accident with a subsequent water ingress was found to not

represent an intolerable load upon the pressure-keeping containment in terms of peak
pressure and pressure transient. Reactivity transients including false operation of absorber
rods, water ingress, or a decrease of core inlet temperature were also analyzed to cause
no serious damage to the heat exchanging components, unless both reactor shutdown and
helium circulators fail at a time.
In a water ingress accident after pipe rupture, the maximum ingress rate is conserva-
tively assumed to be 5 kg/s with a maximum amount of water of 200 kg in the primary
system. The water causes graphite corrosion of not more than 1 % on the average keeping
fission product release at a relatively low level. The rupture of a process gas line results
in an escape rate of 5 kg/s process gas consisting of water (60 %), hydrogen (6 %) and
CC>2, CO, CH4. Maximum amounts to enter the primary circuit until the reformer is shut
off, is 100 kg of water and 10 kg of hydrogen, low enough to be covered by the water
ingress accident.
3.3.2. HTR-MODUL for Process Heat Application (170 MW(th))
A probabilistic safety analysis has been conducted at the Research Center Jiilich
for the process heat variant of the German modular HTGR with the purpose to identify
differences compared with the electricity generating HTR-MODUL [55]. The process
heat HTR-MODUL consists of a pebble bed core with 360,000 spherical fuel elements
to produce a thermal power of 170 MW. Helium coolant gas inlet / outlet temperatures
are 300 and 950 °C, respectively. The system pressure is 4 MPa. The connection to the
secondary circuit is given by a He / He intermediate heat exchanger.
The safety concept considers two nuclear shutdown systems, a set of six reflector rods
for reactor scram and power control and a "KLAK" system of small absorber balls for cold
and long-term shutdown. Decay heat removal is made via the heat exchanger, an auxiliary
cooling system, and the panel cooling system inside the concrete cavern, or, in case of a
failure of these systems, passively by heat transfer via the surface of the reactor vessel.
The principal barriers against fission product release into the environment are the high
quality TRISO fuel, the reactor pressure vessel, and the reactor building. The calculation
of the fission product release during normal operation of the reactor (Fig. 3-5) which
determines the contamination of the primary circuit and thus the source term in case of
a depressurization or a water ingress accident has again identified silver to be the nuclide
with the largest release fraction.
44
10'
Silver-110m -2.3-10"2
10,-2.
Defective Fuel Particles 2.5-10"4
Strontium-90 2.7-10"8 O
-Irradiation Temperature - 1500 £
A A A A / -1000 «
,VAV Vn VA VA vAVAV vA V Av AV V /V -500 §
10,-8
200 400 600 800 1000 1200
Irradiation time [dj ——•
Fig. 3-5: Accumulated metallic fission product release fraction from a spherical fuel
element in the 170 MW(th) process heat HTR-MODUL, from [55]
Events identified to initiate an accident with the release of radionuclides from the core
are a small or large leak in the primary circuit, the loss of the main heat sink or the panel
cooling system, and the loss of offsite power. Reactivity accidents have been excluded,
because they were deemed insignificant in an HTR-MODUL in terms of fission product
release. The event tree analyses have led to the classification of all accident sequences
into four release categories with frequencies ranging between 1.8*10~5/yr and 5*10~8/yr.
For those scenarios with the largest fission product release expected, deterministic studies
applying conservative input data have been conducted. The severest case, a core heatup in
the depressurized system, is predicted to result in a maximum core temperature of 1525 °C
and a maximum core average of about 1000 °C. From the radiologically relevant nuclides
under such temperature conditions, no cesium or strontium release from the core is expected
due to a high sorptivity of the graphite in the core. Only silver escapes from the primary
circuit in a considerable quantity. The release of iodine is determined by the fraction of
free uranium in the core, which is a question of the fuel quality. Iodine-131 quantities
released are predicted to be 111 GBq in a slow depressurization (small leak) and 46 GBq
in a fast depressurization (large leak).
The higher fuel temperatures under normal operating conditions compared with the
electricity generating variant result in a higher fission product contamination of the primary
circuit outside the active core, a factor of 3 for iodine and a factor of 4 for cesium.
In the case of a depressurization through a large leak, it is the dust-borne cesium and
45
strontium dominating the activity release. Depressurization through a small leak makes
coolant activities enlarged by desorbed plateout activity escape from the primary circuit;
the values, however, are about two orders of magnitude lower.
3.4. HYDROGEN BEHAVIOR IN NUCLEAR POWER PLANT ACCIDENTS
The investigation of explosion hazards associated with the inadvertent existence of

hydrogen in partially or fully confined spaces has been subjected to studies in different
countries to assure the phenomena associated with hydrogen are understood with regard to
reactor safety. Theoretical and experimental investigations have focused on the distribution
of hydrogen-air-steam mixtures and its ignition and combustion. The physical phenomena
and mitigation systems as well as the examples of experiments of the behavior of hydrogen
in nuclear plant accidents are presented in the following sections because of the relevance to
safety for hydrogen production systems. Further discussion on the combustion behavior of
hydrogen and associated research activities for non-nuclear unconfined areas is presented
in section 8.5.
3.4.1. Phenomena
Different categories apply to the hydrogen issue in nuclear reactor containments.

During normal operation, small quantities of HI are generated by radiolysis of the coolant,
44 Ha and 22 Oa molecules per 104 eV of neutron radiation energy. Under accident
conditions, significant amounts of hydrogen could be produced due to radiolysis and
corrosion reactions.
The most important hydrogen source in a core meltdown accident is the in-vessel fuel
element Zircaloy cladding/water reaction. The exothermal nature of the reaction enhances
the H2 production rate. Substantial hydrogen generation begins at temperatures > 1300 K.
At a later stage, additional hydrogen could result at a much slower rate from radiolysis and
aluminum oxidation in the containment. Flooding of an overheated core could terminate the
core melt progression, but could also result in significant further H2 generation and fission
product release. For complete oxidation, H2 generation was computed to range between
450 kg in smaller nuclear power plants and more than 900 kg in larger plants. However,
there is still uncertainty in the precise quantification of hydrogen generation during a core
meltdown event [36]; a rough rule of thumb is the generation of 1 kg of H2 per MW(e)
of the reactor core.
Under accident conditions, steam and hydrogen will be released simultaneously from
the reactor vessel into the containment. Mixing with the containment atmosphere is
by forced convection (pressure differences, jet flow, water spray system) or by natural
convection (density differences). Slow mixing and steam condensation may result in locally
high concentrations of hydrogen; rapid mixing could either prevent the formation of a
flammable mixture by keeping the H2 concentration low enough or result in an explosion
that is global in nature provided large H2 masses are quickly released. Containments with
no recirculation fans to promote forced mixing are more vulnerable to locally high H2
concentrations.
46
The most likely mode of HI combustion in a containment is deflagration. Elevated
temperatures could lead to autoignition of the gas mixture, but there are also other
ignition sources, so that a combustion, supposed the multicomponent gas mixture is in
the flammability range, will be very likely, possibly as a diffusion flame near the release
point after flash-back from the ignition point. Buoyancy plays a major role in the flame
propagation: for upward propagation, 4 vol% Ha are sufficient, whereas at least w 9 vol%
are required for a propagation in downward direction [2]. Intercell flow distribution and
local steam condensation rates have a significant influence on the local gas fractions in each
compartment. A rapid condensation, if possible, may convert an inert atmosphere into a
detonatable mixture, which would threaten the containment integrity and the survival of
safety equipment. Assuming a 75 % cladding/water reaction, the potential of DDT needs
detailed investigations. It is decreased in the presence of diluents (steam, €62) and is
increased by fan coolers, by obstacles in the flow path, or by spray systems which decrease
the steam concentration.
The interaction between evolving pressure waves and internal structures is complex,
because loads on the containment will be variable with position and time. Local deflagrati-
ons or detonations in complex geometries do currently not allow a prognosis and evaluation
of the dynamic load on the containment structure [16].
A radiological problem arises with the hydrogen deflagration in a nuclear contain-

ment In a severe core damage accident, extensive aerosol generation must be expected.
One radiologically relevant compound released as an aerosol in significant amounts is Csl.
Incompletely oxidized "metallic" aerosols such as those formed from iron (Fe) were found
to increase the peak pressure by up to a factor of 3, because the Fe participates as a com-
bustion fuel. Experimental studies have shown that 10 - 15 % of the Csl are decomposed
during hydrogen combustion and liberate gaseous iodine which is less effectively retained
in the containment filters and thus significantly affects the source term. Iodine release was
observed to be independent of the steam contents in the containment atmosphere [42]. Cur-
rent understanding, however, is still poor concerning the effects of hydrogen combustion
on fission product behavior [16].
3.4.2. Hydrogen Mitigation Systems
In case of an accident in a nuclear containment where hydrogen is liberated and

distributed within the building, a variety of methods can be employed for hydrogen
mitigation [16]:
• Mixing devices, e.g. fans or helium injection, can prevent locally high H2 concen-
trations as long as the H2 release is small compared with the containment volume
available, enhancing convective gas exchange with upper regions. Fan mixing could
also serve as a heat removal system. A disadvantageous effect is the generation of
turbulence in the containment atmosphere which reduces the combustion time and
increases the severity of a possible explosion. Drawback of mixing devices is their
dependence on availability of electric power.
• Although not directly dedicated to H2 mitigation, venting systems can act so. It might
be the more favorable system of choice, if in a severe accident, containment integrity
has to be assured for long-term retention of radioactivity and rather allowing short-
47
term release. Venting removes reactants from the building, retards flame acceleration,
and lowers peak pressures. Too small openings, e.g., in the top of ducts, however,
could have the opposite effect due to turbulence generation. The ignition inside
venting devices needs to be further investigated.
• The principal purpose of inerting is to limit the supply of oxygen. Strategies possible
are a pre, post, or partial inerting of buildings. Pre-inerting is being widely applied
in nuclear containments of BWRs as a consequence of the TMI accident. Partial
inerting has the advantage of keeping at least a part of the building accessible by
personnel. Media for inerting used or considered are N2, CC«2, or Halon-1301.
Catalytic, thermal, or flame recombiners allow an immediate and continuous H2
removal. They are characterized by a high reliability and availability and no power
supply is required. Catalytic recombiners also work in non-flammable, steam-inerted
mixtures. Recombination systems can be located inside or outside the containment,
normally some distance away from the possible hydrogen source.
• The deliberate ignition of lean (< 8 vol%) H2-air mixtures in a containment is
considered a possible method of partially consuming hydrogen. However, it always
represents some challenge to building and equipment. It should be done as close
as possible to the H2 source location in order to provide relatively benign pressures
and temperatures. Methods of ignition are by hot surface (in use), by spark, or in a
catalytic system which does not require electric power. Igniters are simple, robust,
and easy to install, but because of the difficulty of anticipating the consequences,
deleterious effects under certain circumstances need to be investigated.
• Containment spray systems generate water droplets or fogs. Their prime purpose
is to cool the equipment and the containment atmosphere when steam is released.
Fogs exhibit a slightly inerting effect, thus reducing the burning rate, whereas larger
droplets of water could even act as an obstacle field which can trigger a flame
acceleration. The consideration of foam has been abandoned, because it could allow
the formation of pockets enclosing high H2 concentrations.
3.43. Accidental Hydrogen Explosions
During the accident in the Three Mile Island nuclear power plant in 1979, an
estimated 150 - 600 kg of hydrogen (« 400 kg as quoted in [36] with more than half
presumed to be caused by reflooding of the melting core) were generated in the primary
system, which were ignited at about 10 h into the accident resulting in a peak overpressure
of 190 kPa, well below the 414 kPa design limit of the containment. The deflagration was
presumed to have been initiated at a 7.3 - 7.9 vol% H2 concentration [2, 16].
Also the Chernobyl accident resulted in a significant metal-water reaction with

hydrogen generation, but no final conclusions could be made so far concerning the H2
behavior and its contribution to the accident [16].
Hydrogen-oxygen offgas5 explosions have occurred in some older boiling water

reactors where no treatment of such gases was given. The explosions took place in
various components from the condenser air injector downstream to the vent stack and
5
Offgases released into the primary coolant during normal operation consist of air leaked into the turbine condenser, activated gases,
fission gases, and Hj and Oj from radiolytic decomposition of steam. A typical mixture could contain roughly a 6 - 8 % share of hydrogen.
48
partially caused radioactivity release into the plant. Among other ignition sources were
mechanical sparks from closing valves or electrostatic discharges. Equipment damage,
however, was minimal in most of the 25 cases reported [43]. Modern offgas treatment
equipment tries to eliminate explosive mixtures as soon as possible by dilution with steam
and passing them through a catalytic recombiner and activity filters before routing to the
vent stack [53]. The offgas explosions are part of the incidents given under "Intentional
operation" in Table 8-4 in section 8.6 on hydrogen accident statistics.
3.4.4. Containment Experiments
Hydrogen-air-steam combustion experiments have been conducted to simulate local

detonations in confined structures and to validate the respective numerical detonation codes.
3.4.4.1. Experimental Activities in Germany
A hazardous potential was thought to exist in the AVR HTGR, Jiilich, with a
hydrogen-carbon monoxide-helium-air mixture, which could evolve in a water ingress
accident scenario with an estimated 8 % H2 + CO concentration resulting from the corrosive
reaction between steam and hot fuel element graphite. Ignition experiments in a combustion
chamber were conducted in 1971 to examine the lower combustibility limit in respective
gas mixtures. The tests have shown that a mixture with 22 % H2 + CO was not flammable
at temperatures up to 120 °C [54].
In Germany, the Battelle Institute has employed the Battelle model containment
(BMC) which is a 1/100 down-scaled model of a typical PWR containment. Between
2SOOO
Fig. 3-6: Hydrogen concentration transients in a six-compartment structure of the

Battelle model containment with inverse thermal stratification (the numbers are
H2 sensor positions as shown in the diagram at the top), from [28]
49
1977 and 1982, a six-compartment geometry with a total free volume of 640 m3 was
used to measure the concentration distribution of hydrogen released at rates of 0.2 and
1.0 Nm3/h, respectively. Under isothermal conditions, a homogeneous distribution of the
hydrogen was found. For the condition of an inverse thermal stratification (top: hot, bottom:
cold), higher H2 concentrations in the bottom compartment due to the larger density were
measured until the concentration difference reached 2.2 vol% and a convective flow through
all compartments eventually started (Fig. 3-6). In a combustion test series, hydrogen-air
mixtures with 9 % H2 were ignited in 2-, 3-, or 4-chamber structures [28]. Jet ignition was
observed to create highly transient deflagrative combustion at H2 concentrations around
10 %.
Another test series in the BMC was dedicated to deliberate ignition measures using
a 5-compartment structure with 209 m3 volume. Spark igniters caused deflagrative com-
bustion at H2 concentrations of 5.2 - 5.6 vol% and steam fractions up to around 50 vol%.
Higher steam contents suppressed the deflagration. Catalytic recombiners require a starting
threshold value which was 3.4 vol% H2 for the device used, but are then able to remove
hydrogen down to a remaining concentration of 0.3 - 0.4 vol% [22].
At the end of 1987, the investigation of hydrogen distribution on a real-size scale was
started in the containment of the decommissioned German HeiBdampfreaktor (HDR) (hot
steam reactor) multi-compartment facility in Karlstein with dimensions of 60 m height, 20
m diameter and a free gas volume of 11,300 m3. HDR consists of 72 subcompartments
connected by about 300 flow paths with different and complicated geometry. A series of
seven H2 distribution experiments was conducted in 1989. In the test T31.5 considered as
an international standard problem for benchmark calculations, 321 of a water vapor mixture
have been fed into one compartment within 60 s, another 2 t after 30 min and eventually 160
kg of a 15 % H2 plus 85 % He light gas mixture after 36 min. Pressures, temperatures, and
light gas concentrations were measured at various positions. The distribution tests have
exhibited a stratification in terms of both temperature and H2 concentration with higher
values above the inlet location. There is only a slow exchange leading to equilibrium.
Conclusions drawn from the distribution tests were positive in that significant amounts of
energy were found to be withdrawn by the inner containment structures, but were also
negative in that not anticipated distribution inhomogeneities were observed which do not
allow a prognosis of the consequences of an H2 explosion based on the assumption of a
homogeneous mixture [23].
The HDR combustion tests were restricted to a two-chamber section with 530 m3
volume filled with a mixture of H2, steam, and air. Flow obstacles were removed on purpose
to allow a comparison with earlier test series. In these tests, the influence of different
ignition positions on the combustion has been investigated. A variety of pressure-time
histories was observed depending on the location of ignition, of compartment geometry,
and of initial H2 concentration. Jet ignition of a H2-25 vol% steam-air mixture resulted
in the largest pressure loads peaking at 130 kPa for 10 vol% H2 and 185 kPa for 12
vol% H2 [24, 39]. Steam concentrations greater than 35 % strongly reduce the flame
front propagation, whereas concentrations between 55 and 60 % completely suppress an
ignition [23]. A problem in these tests was the undocumented hydrogen stratification prior
to ignition.
50
100
c
o>
CD
O
20 40 60 80 100
% Steam
Fig. 3-7: Ternary diagram for hydrogen-steam-air mixtures at 100 °C deduced
from the HDR experiments El 1.2 simulating a small leak at a high location
in the containment, El 1.4 simulating a small leak at a low location, and El 1.5
simulating a large leak at a low location, from [3]
Flammability data have been derived for various ternary mixtures including H2, CO,
air, H2O, N2, He, and CO2 at different temperature levels. A general correlation for a
ternary system between room and autoignition temperature is given in [36]. The ternary
diagram in Fig. 3-7 shows the deflagration and detonation ranges on a large-reactor typical
scale for H2-steam-air mixtures depending on the mixture composition.
3.4.4.2. Experimental Activities in the USA
A part of the US Nuclear Regulatory Commission's (NRC) severe accident research

program was dedicated to hydrogen issues in LWR containment designs under core melt-
down conditions. The analysis included the in-vessel and ex-vessel hydrogen generation
as well as its mixing and distribution in the containment.
The Hanford Engineering Development Laboratories (HEDL) have used a 20 m

high, 7.6 m diameter and « 300 m3 volume annulus-shaped one-compartment test facility
simulating the lower part of a PWR ice-condenser containment6. Hydrogen was replaced
by helium released at 150 °C as a high-speed jet into the compartment which was at a
80 - 90°C temperature level. Maximum concentration differences of about 3 vol% during
source release and up to 7.5 % only after switching off the source and subsequent local
6
A PWR ice-condenser containment has a smaller volume and a lower design pressure compared with others, because ice is used as a
heat sink to condense the steam deliberated in an accident.
51
AICC: 3O VOL * STEAM •
70 OEO C. 1 ATM
o AICC: « VO*. * STEAM

a
^. 30 oea c. t ATM
X
«
E
ai
cc
«
co
UJ
CC 2
a.
OS VOL % STEAM. To- 3O1 2 OEQ C
O30 VOL * STEAM. To- 7 O ± 6 DEO C
A40 VOL % STEAM. T o - 7 4 DEQ C
i i i I i i i I i i I
e a to 12
HYDROGEN CONCENTRATION. VOL %
Fig. 3-8: Experimental and calculational pressure ratios during combustion tests with
quiescent premixed H2 mixtures in the NTS model containment, from [7]
condensation effects were observed [28]. The dominant mixing process was by the high
velocity gas jet.
The EPRI experimental (and analytical) program comprised the study of H2 combu-
stion and control in a hydrogen-air-steam-water spray environment focusing on deliberate
ignition methods. In several projects, investigations were done in test facilities of various
scales ranging between 0.002 and 2100 m3 of volume. Large-scale demonstration tests
at the Nevada Test Site (NTS) have been conducted in a 2100 m3 inner volume one-
compartment spherical dewar representing a 1/4 scale model of a nuclear containment.
Goals were the investigation of H2 combustion in a premixed atmosphere or with a con-
tinuous injection flow, the study of equipment operability and thermal response, and the
evaluation of the performance of mitigation methods. Results for the quiescent, premi-
xed ignition tests with different H2-air-steam mixtures are shown in Fig. 3-8 indicating a
continuous increase in pressure ratio with increasing H2 concentration. The combustion
tends to be more incomplete towards lean mixtures. In the continuous injection tests, a
stratified atmosphere was built up at low injection rates, whereas uniform concentrations
were established immediately when water sprays were applied. Rapid mixing in the vessel
was also achieved at high injection rates of 1 kg/min HI and 9.1 kg/min steam [7, 50,
51]. Combustion of lean mixtures ( 4 - 8 vol%) was generally observed to be incomplete
as well as in the case of steam addition. Turbulence was found to drastically reduce time
of combustion and increase the degree of completeness of combustion [36].
In cooperation and with funding by the Japanese NUPEC, the US NRC is investigating
at the Sandia National Laboratory the containment building safety in a series of pressure
52
tests with nitrogen up to the failure limit of a 1/10 scaled vessel of 5.8 m height. The
failure occurred at a pressure of 4.66 MPa, the prediction was 4.50 MPa, the design basis
value is 0.77 MPa. A second test containment will be a 1/4 scaled model with the final
pressurization-to-failure test at the end of 1999 [31].
The Russian combustion tests in the large-scale channel RUT are described in more
detail in section 8.5.2.2.
3.4.5. Containment Modeling
Many codes have been used to calculate the transport, convective mixing, and
combustion of hydrogen in containments, some of which are shortly described in the
following. Some more CFD computer codes are being briefly described in the sections
8.4.2. on atmospheric dispersion and 8.5.1. and 8.5.2. on combustion processes.
A multi-zone computer code WAVCO (Wasserstoff-Verteilung im Containment =

Hydrogen Distribution in the Containment) has been developed by the Siemens company
for predicting thermodynamics and hydrogen distribution during severe LWR accidents.
The model solves an equation system consisting of separate mass and energy balances
for both atmosphere (superheated or saturated steam and up to nine noncondensibles) and
sump of each zone. Also mass balances for each component are used to determine the
actual gas distribution. Validation was made against above mentioned hydrogen/helium
distribution experiments of Battelle and HEDL with good agreement between calculated
and measured data [38].
The Sandia code HECTR (Hydrogen Event: Containment Transient Response) is a

lumped parameter7 analysis code for modeling the containment atmosphere under accident
conditions involving release, transport, and also combustion of hydrogen [37]. It can handle
saturated and superheated conditions and it covers both short-term transients and long-term
convection up to several hundred days.
The Sandia code CONTAIN is a lumped parameter code with mechanistical models
for simulating the physical and chemical conditions in the nuclear containment to predict
hydrogen and steam concentration distribution as well as the consumption of H2 by
respective combustion. Assuming a core meltdown accident and no vessel breach, i.e.,
no corrosion/concrete interaction, the code has predicted a thermally stratified containment
atmosphere with relatively low temperatures in the central and lower regions which would
permit steam condensation. Concerning H2 deflagration, CONTAIN predicts respective
burns, if sprays are used for steam removal [56].
The development of the computer code RALOC at the German GRS for the simu-
lation of the distribution of hydrogen, air, and steam in a LWR containment after severe
accidents has started in 1974. RALOC is based on a lumped parameter approach with a
differential equation system which describes the composition of the containment atmos-
phere, temperature, and the transport phenomena of convection and diffusion. Validation
7
The lumped parameter or lumped volume technique uses control volumes connected by flow junctions. Gas exchange through the
junctions is by convection and diffusion and is influenced by a certain flow resistance. In each zone, thermodynamic equilibrium and
saturated steam-water conditions are assumed. The approach allows a modeling of gas distribution in highly complex multicompartmental
structures. It does, however, not consider dynamic effects such as jets and local turbulence [16].
53
has been made against HDR distribution tests [21]. Due to continuous improvement, it is
being considered now the most sophisticated lumped parameter code [16].
In cooperation with the Kurchatov Institute in Russia, the Research Center Karlsruhe
(FZK) is developing the 3D computer code COM3D to simulate fast turbulent combustion
in severe LWR accidents. A k-e submodel is used to describe turbulence. The chemical
kinetics submodel is adapted to hydrogen-air-steam mixtures. The reaction rate is calculated
by using an eddy dissipation model which depends on the turbulent time scale and an
empirical constant. The code validation was made against combustion experiments in
the 12 m FZK. tube and the Russian RUT facility (see section 8.5.2.). Agreement between
calculational and measured results was optimal for the empirical constant being in a narrow
range even for experiments with a different scale, geometry, and gas mixtures [4]. As a
second step in the combustion modeling, a so-called PDF approach8 is currently under
development.
The k-e model BASSIM (Battelle Stromungssimulator) of the German Battelle Inge-
nieurtechnik is a CFD model applicable to multi-dimensional transient compressible fluid
flow problems considering heat and mass transport and chemical reactions in body-fitted
grids. It has been employed to investigate flame development and pressure rise in hydrogen
deflagration as measured in BMC and HDR tests. BASSIM could successfully reproduce
the observed thermal stratification in the tests.
The computer code GASFLOW has been developed at the Los Alamos National
Laboratory to simulate the distribution of a hydrogen-steam-air mixture in a multicompart-
ment structure of a nuclear power plant containment, combustion processes after ignition,
and the respective dynamic quasi steady state load due to H2 combustion. Extensive model
development at the Research Center Karlsruhe has brought the code into full containment
applications. GASFLOW solves the Navier-Stokes equations in a 3D Eulerian grid and
allows the description of processes on different time scales. The k-e model is one option
of turbulence modeling. A unique feature of the code is the simulation of countermeasures
such as the activation of igniters and/or recombinators or the feeding of inciting gases. The
code has been validated against numerous HDR and BMC experiments [39].
The general purpose computational fluid dynamics code CFX is a commercially

available software of AEA Technology, Harwell, UK, to simulate a wide range of fluid flow
and heat transfer processes based on the Navier-Stokes equations in complex 3D geometries.
CFX comprises a selection of advanced turbulence models including the k-e and Reynolds
stress model plus a selection of numerical solution methods to accurately and efficiently
describe subsonic or supersonic flows. The code is able to handle in body-fitted, multi-
block grids any number of fluids and separate phases: liquids, solids, gases or chemical
species. It includes a particle transport model according to the Lagrangian scheme as well as
submodels for combustion, radiation, and chemical kinetics [1]. Combustion phenomena in
confined areas including slow/fast deflagration, with weak or strong ignition, and transition
to detonation, are currently been investigated within a CEC program on reactor safety using
the CFX code. The R&D studies comprise the validation of state-of-the-art predictive
8
The employment of the "Probabilistic Density Function", PDF, is a statistical method for the determination of the mean combustion
rate in a computational gnd cell
54
methods for the description of combustion processes and mitigation measures (deliberate
ignition) by a comparison with respective experimental work [9].
3.5. GAS EXPLOSION RESEARCH PROGRAMS
Gas explosion research programs have been initiated in various countries either one
focusing on specific aspects of the offsite explosion and the consequences of flammable gas
clouds, in particular, the impact of the pressure wave on the nuclear containment. Germany
has started safety programs related to both light-water reactors and high-temperature gas-
cooled reactors, the latter in connection with process heat applications. The USA have
conducted an LNG investigation program to ensure the safety of LNG transportation and
terminals. Both France and the European Community have conducted gas cloud explosion
programs dedicated to light-water reactors. Norway has made special off-shore safety
studies related to fire and explosion on off-shore drilling platforms.
3.5.1. The German LWR Program
Between 1979 and 1983, a research program on gas explosions has been conducted
as a part of the program "Research on the Safety of Light-Water Reactors" funded by the
German federal government [48]. The principal areas investigated were:
(1) Conceivable mechanisms and configurations necessary for the evolution of
detonation-like explosions,
(2) Relationship between the characteristic features of incoming blast waves and the
response of building structures,
(3) Strength of blast waves assuming the detonation of a realistic gas cloud
with the overall goal to develop the methodology for a safety valuation of accidents
with the explosion of transport gas nearby nuclear power stations.
A BMI9 guideline from 1974, "Protection of Nuclear Power Plants Against Pressure
Waves from Chemical Explosions by Design of the Nuclear Power Plant with Respect to
Their Stability and Induced Waves as well as by Safety Distances", has defined a pressure-
time function (Fig. 3-9) according to which a nuclear power plant has to be designed as a
protection against extend explosions and a safety distance given by the general equation
as introduced in section 8.5.4. with a k-factor of 8:
R = 8 * M1/3
where R is the safety distance in m (minimum 100 m) and M the explosible mass
in kg underlying the assumption of a homogeneously mixed, stoichiometric gas cloud of
hemispherical shape and with ignition in its center. For multiply unsaturated hydrocarbons,
M is the maximum mass released as well as for compressed gases, for pressurized liquid
hydrocarbons, it is 50 % (or k •» 6.3 m/kg173) and for cryogenics or hydrocarbons under
standard conditions, it is 10 % (or k -= 3.7 m/kg1/3).
For a nuclear coal gasification plant, the above safety distance relation is probably not
practicable. Therefore it has to be guaranteed by respective studies, that the load upon the
nuclear reactor building in case of an outdoors explosion is covered by the design limits.
BMI is the German Federal Ministry of the Interior
55
Load function
\
\
\
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 s 1.0
Time after pressure increase
Fig. 3-9: Pressure-time function according to the BMI guideline from 1974, from [14]
The ratio of flame front velocity to speed of sound influencing flame acceleration in
the gas cloud was theoretically shown to be at least 0.45 to cause a pressure wave which
is still too weak to cause a detonation. The initiation of a detonation is only possible, if
the ignition power density is high enough given by a explosive. Deflagrative ignition, e.g.,
flame, spark, hot surface, do not develop a detonation in an unconfined (real) gas cloud
[35]. The required safety distance between an LH2 (or LNG) tank and a nuclear reactor
building according to the above definition would then be represented by curve #5 of Fig.
8-14 in section 8.5.4.
Results of the research program can be summarized as follows:

(a) The release of liquefied gases (no hydrogen) results in the formation of a flat
(heavy) gas cloud. A reliable prediction of cloud size, gas concentrations, and
flammable portion is not possible. There are indications that the BMI guideline
provides conservative results.
(b) Only a strong ignition source has a significant influence on the flame spreading.
The state of turbulence prior to and during the explosion is a decisive parameter for
flame speed and overpressure. No fast deflagration with overpressures > 0.01 MPa
are expected after explosions in open terrain; an exception may be the catastrophic
failure of a pressure vessel.
Partial confinement and/or obstruction, e.g., a lane between buildings, is necessary
to cause overpressures > 0.01 MPa with significant impact on building structu-
res. The blast wave is weakened in obstruction-free areas, but presumably less
pronounced the larger the scale.
(c) Prerequisite for DDT is the generation of very high flame speeds. A characteristic
parameter is the detonation cell size with a higher reactivity of the fuel the smaller
its cell size. The direct initiation of a detonation is only feasible with strong igniters
and, thus, appears to be unlikely for accidents with mixed gas clouds.
56
(d) The blast wave of a deflagration may exhibit a steep pressure rise, multiple peaks
or an underpressure phase creating a complex load function enhanced by reflection
or pressure wave convergence.
(e) In comparison with a linear pressure slope, a stronger impact on containment and
and components is expected for load functions characterized by multiple peaks or
detonation or underpressure. The absolute impact, however, is in the range of
elasticity and, thus, in accordance with the design.
Existing safety margins of modern nuclear containments guarantee a safe enclosure.
3.5.2. The German PNP Program
Delivery of process heat implies the coexistence of nuclear reactor and chemical plant,
e.g., coal gasification, hydrogen production, ADAM-EVA transportation system. The main
goals of the PNP gas explosion program were, in general, to improve understanding of the
complex process of chemical explosions and its effect on the environment, and in particular,
to demonstrate that a power station for nuclear coal gasification or process steam production
is safely designed against explosions from outside, if the BMI guideline is applied. This
is deemed true for explosions of transported hydrocarbons and minimum distance rules
applied. However, some experimental studies on gas explosions under conditions typical
for process gases inside the plant have revealed that under unfavorable circumstances,
overpressures can exceed the respective maximum figures from the guideline. Experimental
results were often complex and difficult to interpret. The experimental data base for the
phase of flame acceleration is sufficient for a qualitative understanding of the phenomena
and sound statements on flame speed and pressure within the frame of a safety analysis
for an industrial complex. The transition to detonation itself, however, is strongly and
decisively dependent on the turbulence structure which can only insufficiently be simulated
in numerical models. It allows a derivation of specific DDT boundary conditions only in
very concrete accident scenarios.
The PNP safety program [41] focused on the formation and explosion of clouds
of the process gases Hk, CO, air and other representative combustible gases such as
acetylene, ethylene, propane, methane, and the damage caused by the blast wave to simple
structural elements (window pane, brick wall) after fast deflagration and detonation. Areas
of investigations were
(i) flame front velocity in given scenarios and respective mechanisms of flame acce-
leration,
(ii) criteria for the transition to detonation, and
(iii) criteria for a passing on of a detonation.
Tasks within the study were dealing with deflagration damage analysis, mechanisms
of combustion, regulations and limits comprising subtasks about
• identification of PNP specific accident scenarios which result in the release of
flammable gas mixtures and assessment of corresponding gas masses to be expec-
ted,
• material-specific aspects such as burning velocity, flammability limits, flame sprea-
ding,
57
• the effects of a chemical explosion on building structures in relation to different
degrees of partial confinement,
• evaluation and analysis of process gas cloud explosion accidents and experiments,
• derivation of the relation between a deflagrative pressure wave and characteristic
destruction lines, upper limits of destruction, "design pressure wave",
• deduction of a load function by a damage analysis of the accident in Beek10,
• experimental investigation of deflagrative combustion and its enhancement in the
presence of initial turbulence in a channel with open top,
• Balloon explosion tests, and
• development of a computer code for the spherical spreading of a flame according
to the piston model.
Experimental studies within the PNP gas cloud program were conducted and eva-
luated by the Fraunhofer Institute for Chemical Technology using flammable mixtures of
hydrogen and other fuels with air in different shapes (sphere, hemisphere, tube), confined or
unconfined. They are described in more detail in section 8.5.2. Experimental results from
other test programs in the USA (explosion series in "FLAME") or Norway (jet ignition
with transition to detonation in acetylene-air mixtures) were also taken into account. All
tests were close to the go - no go boundary between deflagration and detonation.
An overall statement concerning the deflagration - detonation transition (DDT) was

made within the frame of the PNP gas cloud program saying that the mechanisms and
flame acceleration are qualitatively understood, however, they cannot be described on a
quantitative basis, a statement that is, in principle, still valid today [6, 30, 41], although
computational efforts are progressing tremendously.
[1] AEA TECHNOLOGY, CFX 4.1: User Guide, AEA Technology (1995).
[2] BERMAN, M., CUMMINGS, J.C., Hydrogen Behavior in Light-Water Reactors,
Nuclear Safety 25 (1984) 53-74.
[3] BREITUNG, W., REDLINGER, R., A Model for Structural Response to Hydrogen
Combustion Loads in Severe Accidents, Nucl. Tech. Ill (1995) 420-425.
[4] BREITUNG, W., KOTCHOURKO, A., Numerische Simulation von turbulenten
Wasserstoff-Verbrennungen bei schweren Kemreaktorunfallen, FZK-Nachrichten 28
(1996) 175-191.
[5] BURGHOLZ, L.M., Das Zwischenkreislaufexperiment - Theorie und Praxis, Re-
port Jul-2230, Research Center Julich (1988).
[6] CHAN, C.K., TENNANKORE, K.N., A State-of-the-Art Report on Flame Acce-
leration and Transition to Detonation in Hydrogen / Air / Diluent Mixtures, Draft
Report, NEA Committee on the Safety of Nuclear Installations (CSNI) (1991).
[7] EDESKUTY, F.J., HAUGH, J.J., THOMPSON, R.T., Safety Aspects of Large-
Scale Combustion of Hydrogen, (6th World Hydrogen Energy Conf., Vienna, Aus-
tria, 1986), VEZIROGLU, T.N., et al., Hydrogen Energy Progress VI, International
Association for Hydrogen Energy (1986) 147-158.
10
In 1975, a violent vapor cloud explosion occurred in Beek, The Netherlands. The accident in a naphta cracker installation was caused
by a leakage of the product gases hydrogen, ethylene, and other hydrocarbons. A vapor cloud of an estimated 5500 kg of gas drifted into
the plant where it eventually ignited. The accident claimed several fatalities; damage was found up to a distance of 4.5 km.
58
[8] EICHLER, R., Reinigung inerter Gaskreisläufe nuklearer Energieerzeugungsanla-
gen von Tritium und Wasserstoff - Auslegung eines Gasreinigungssystems, Report
Jül-2008, Research Center Jülich (1985).
[9] FINESCHI, F., Defense in Depth against the Hydrogen Risk - A European Research
Programme, Proc. 34th Annual Conf. of the Canadian Nuclear Association (1994).
[10] FRÖHLING, W., et al., Safety Concept and Operational Criteria of a Nuclear
Process Heat Plant, Nucl. Eng. Des. 78 (1984) 167-177.
[11] FUJIMOTO, N., et al., Safety Analysis and Considerations for HTTR Steam
Reforming Hydrogen / Methanol Co-Production System, (TCM, Oarai, 1992),
High Temperature Applications of Nuclear Energy, Report IAEA-TECDOC-761,
International Atomic Energy Agency, Vienna (1994) 86-91.
[12] FUMIZAWA, M., Safety Concept of Heat Application Systems in JAERI - Fire and
Explosion Accidents, Presentation at the Research Center Jülich, March 11 (1997).
[13] GHT/HRB, Prozeßgasaustritt und zündfähige Gemische, PNP Quarterly Report
H/79, PNP Project (1979).
[14] GESELLSCHAFT FÜR REAKTORSICHERHEIT, Deutsche Risikostudie Kern-
kraftwerke, Fachband 4: Einwirkungen von außen (einschließlich anlagenintemer
Brände), Verlag TÜV Rheinland, Köln (1980).
[15] KICKEN, E.F., Passive Safety Systems, a Possibility of Enhancing Reactor Safety,
Kerntechnik 61 (1996) 207-209.
[16] IAEA, Hydrogen in Water-Cooled Nuclear Power Reactors, International Atomic
Energy Agency and Commission of the European Communities, Vienna (1990).
[17] IAEA, Development of Safety Principles for the Design of Future Nuclear Power
Plants, IAEA-TECDOC-801, International Atomic Energy Agency, Vienna (1995).
[19] IAEA, Fuel Performance and Fission Product Behavior in Gas-Cooled Reactors - A
Compilation Produced within the IAEA Coordinated research program on Validation
of Predictive Methods for Fuel and Fission Product Behavior in Gas-Cooled Reac-
tors, IAEA-TECDOC-978, International Atomic Energy Agency, Vienna (1997).
[20] JÄGER, W., WEISBRODT, L, HORNING, H., Nuclear Process Heat Applications
for the Modular HTR, Nucl. Eng. Des. 78 (1984) 137-145.
[21] JAHN, H., HÜTTERMANN, B., SCHWINGES, B., Analytical Methods for the
Prediction of Hydrogen Distributions in Reactor Containments, Kerntechnik 54
(1989) 153-158.
[22] KANZLEITER, T., Modellcontainment-Versuche zum Wasserstoffabbau bei ausle-
gungsüberschreitenden Ereignissen, (Proc. Jahrestagung Kerntechnik '92, Karls-
ruhe, 1992), Inforum GmbH, Bonn (1992) 207-210.
[23] KARWAT, H., Hydrogen Mitigation in Steel Shell Containments of Pressurized
Water Reactors, Kerntechnik 59 (1994) 171-177.
[24] KATZENMEIER, G., MÜLLER-DIETSCHE, W., Großversuche am ehemaligen
Kernkraftwerk HDR, Atomwirtschaft 36 (1991) March 134-137.
[25] KÖNIG, S., BARNERT, H., SINGH, J., Prinzip-Auslegung und sicherheitstechni-
sche Untersuchung der Wasserdampf-Kohle-Vergasung von Braunkohle mit HTR-
Wärme, Internal Report KFA-ISR-IB-5/91, Research Center Jülich (1991).
59
Reformers, Report Jul-1116-RG, Research Center Jiilich (1974).
[27] KUGELER, K., PHLIPPEN, P.-W., The Potential of the Self-Acting Safety Features
of High Temperature Reactors, Kerntechnik 61 (1996) 239-244.
[28] LANGER, G., STOCK, M., Experimental Investigations on Hydrogen Behavior in
Reactor Containments, Kerntechnik 53 (1988) 39-46.
[29] MERTENS, J. (Ed.), Nukleare ProzeBwarmeanlage AVR-U, Sicherheitstechnische
Untersuchungen, Report Jul-Spez-199, Research Center Jiilich (1983).
[30] MOEN, I.O., Transition to Detonation in Fuel-Air Explosive Clouds, J. Hazardous
Materials 33 (1993) 159-192.
[31] MURPHY, B., Japanese Industry Turns to Sandia to Test Nuclear Reactor Contain-
ment Building Safety, Sandia LabNews, December 20, 1996.
[32] NABIELEK, H., FUKUDA, K., KANIA, M.J., Fuels Technology, Closed-Cycle
Gas-Turbine Modular High-Temperature Gas-Cooled Reactor, (Int. Workshop,
Cambridge MA, 1991), PENFIELD, S.R., KASTEN, P.R., Massachusetts Institute
of Technology (1991) 4-8 - 4-16.
[33] NISHIHARA, T., et al., Safety Considerations and Countermeasures Against Fire
and Explosion at an HTGR-Hydrogen Production System (Proposal of Safety
Design Concept), (3rd JAERI Symp., Oarai, 1996), Proc. JAERI-Conf 96-010,
Japan Atomic Energy Research Institute (1996) 264-271.
[34] NISHIHARA, T., HADA, K., SHIOZAWA, S., Proposal of Safety Design Metho-
dologies for an HTGR-Hydrogen Production System (Mainly on Countermeasures
against Fire and Explosion), Report JAERI-Research-97-022, Japan Atomic Energy
Research Institute (1997) (in Japanese).
[35] PFORTNER, H., Zundverhalten von Erdgas/Luft-Gemischen in freien Wolken, gwf-
gas/erdgas 120 (1979) 19-24.
[36] PLYS, M.G., Hydrogen Production and Combustion in Severe Reactor Accidents:
An Integral Assessment Perspective, Nucl. Tech. 101 (1993) 400-410.
[37] PONG, L.T., Assessment of the Combustion Model in the HECTR Code, Re-
port NUREG/CR-5590 and SAND90-7080, U.S. Nuclear Regulatory Commission,
Washington (1990).
[38] PREUSSER, G., The Multi-Compartment Code WAVCO, Kerntechnik 53 (1988)
47-52.
[39] ROYL, P., et al., Dreidimensionale Simulationen von Wasserstoffverteilung und
-verbrennung im auBeren Sicherheitsbehalter eines Druckwasserreaktors, FZK-
Nachrichten 28 (1996) 192-208.
[40] SAWA, K., et al., A Study of Silver Behavior in Gas-Turbine High Temperature
Gas-Cooled Reactor, (TCM, Beijing, 1995), Design and Development of Gas
Cooled Reactors with Closed Cycle Gas Turbines, Report IAEA-TECDOC-899,
[41] SCHILDKNECHT, M., STOCK, M., Statusbericht zum Kenntnisstand des
tlbergangs Deflagration-Detonation unter besonderer Berucksichtigung der Ziel-
setzung des PNP-Gaswolkenprogramms, Final Report BF-R-66.459-2, Battelle
Institute, Frankfurt (1987).
60
[42] SCHMTTT, R.E., et al., Experimentelle Untersuchung fiber das Verhalten von
Csl bei H2-Verbrennung, (Proc. Jahrestagung Kerntechnik '94, Stuttgart, 1994),
Inforum GmbH, Bonn (1994) 179-182.
[43] SCHODEL, J.P., Hydrogen - A Safety Risk?, Hydrogen as an Energy Vector
Its Production, Use and Transportation, (CEC Sem., Brussels, 1978), Report EUR
6085, Commission of the European Communities (1978) 567-581.
[44] SHIBATA, T., et al., Availability of Steam Generator Against Thermal Disturbance
of Hydrogen Production System Coupled to HTGR, (3rd JAERI Symp., Oarai,
1996), Proc. JAERI-Conf 96-010, Japan Atomic Energy Research Institute (1996)
289-293.
[45] SINGH, Y., et al., The Nuclear Heated Steam Reformer - Design and Semitechnical
Operating Experiences, Nucl. Eng. Des. 78 (1984) 179-194.
[46] SINGH, J., HADA, K., JAERI/KFA Cooperation on Nuclear Heat Utilization
System Design and Safety, Internal Report KFA-ISR-IB-16/94, Research Center
Julich (1994).
[47] STEINWARZ, W., et al., Distribution of Tritium in a Nuclear Process Heat Plant
with HTR, Nucl. Eng. Des. 78 (1984) 267-272.
[48] STOCK, M., GEIGER, W., Teilforschungsprogramm Gasexplosionen: Zusammen-
fassende Darstellung und Auswertung, Final Report BIeV-R-64.181-4, Battelle In-
stitute, Frankfurt (1984).
[49] STOVER, D., et al., Zusammenfassende Darstellung von Experimenten und Er-
gebnissen zur Permeationsproblematik an In Situ oxidierten Hochtemperaturlegie-
rungen, in: SCHAFER, J. (Ed.), Permeation der Wasserstoffisotope durch metalli-
sche Werkstoffe bei hohen Temperaturen, Report Jul-1747, Research Center Julich
(1981) 1-21.
[50] TAMANINI, F., URAL, E.A., CHAFFEE, J.L., Hydrogen Combustion Experiments
in a 1/4-Scale Model of a Nuclear Power Plant Containment, (22nd Int. Symp. on
Combustion, Seattle, 1988) The Combustion Institute (1988) 1715-1722.
[51 ] THOMPSON, L., EPRI Research on Hydrogen Combustion and Control for Nuclear
Reactor Safety, (4th World Hydrogen Energy Conf., Pasadena, FRG, 1982), Hy-
drogen Energy Progress IV (1982) 1675-1684.
[52] US DEPARTMENT OF ENERGY, Selected Technology Profiles, in: INTER-
NATIONAL ENERGY AGENCY, Comparing Energy Technologies, OECD/IEA
(1996) 237-296.
[53] VERNA, B.J., Off-Gas System Explosions - Part 1, Nuclear News 19 (1976) No.
14 53-54, - Part 2, Nuclear News 20 (1977) No. 1 43-44, - Part 3, Nuclear News
20 (1977) No. 4 41-42.
[54] WISCHNEWSKI, R., Untersuchungen zur Wassergasbildung bei Storfallen an
HTR-Reaktoren am Beispiel einer geplanten HeiBgastemperaturerhohung auf 950
°C am AVR-Reaktor, AVR Report, Arbeitsgemeinschaft Versuchs-Reaktor (1974).
[55] WOLTERS, J. (Ed.), Probabilistic Safety Analysis and Assessment on Possible
Urban Siting of the Modular HTR for Process Heat Application, Internal Report
KFA-ISR-IB-3/90, Research Center Julich (1990).
[56] YANG, J.W., MUSICKI, Z., NIMNUAL, S., Hydrogen Combustion, Control, and
Value-Impact Analysis for PWR Dry Containments, Report NUREG/CR-5662, U.S.
Nuclear Regulatory Commission, Washington (1991).
NEXT PAGE(S)
left BLANK
61
Chapter 4
ENERGY SITUATION AND ACTIVITIES ON NUCLEAR

POWER WITH POTENTIAL USE FOR PROCESS HEAT AND
HYDROGEN PRODUCTION IN SOME EXEMPLARY COUNTRIES
4.1. CANADA
Canada has an enormous potential of domestic natural energy resources, in particular

natural gas and water. The electricity production in 1993 was approx. 610 TWh with
62 % from hydro power, 23 % from fossil fuels, and 15 % from nuclear power. Since
1969, Canada is a net energy exporting country. In most parts of the country, low-cost
hydro-electricity is available and technologies are being developed to benefit from this
privileged situation. Renewable energies (others than large-scale hydro) accounted for 6
% of the total Canadian primary energy demand in 1994. Canada is the world leader in
long-distance electric power transmission. It is also ranking first with 18,000 IcWh in the
per capita electricity consumption. CC«2 emissions on a per capita basis are 16.1 t. The
goal set at the Kyoto summit is a 6 % reduction relative to the 1990 level compared with
a 9 % increase as was observed in the period 1990-95 [16].
On a per capita basis, the largest producer and consumer of hydrogen in the world
is Canada producing more than 2 million t/yr. Primary sources for hydrogen are methane,
hydrogen sulfide, and water. Hydrogen plays an important role in the Province of Alberta
where 55 % of the production and 63 % of the consumption of all hydrogen in Canada
takes place. Alberta's research activities are concenu-ating on hydrogen production, for
example from the widely available H2S.
4.2. CHINA
4.2.1. Energy Situation in China
Two decades of enormous economic growth in the People's Republic of China at a rate
of 9.4 % per year (1978 - 1994) have led to a tremendous need for energy and electricity,
and the demand for more is increasing at a rapid pace, in particular, in the industrially
developing regions of East and South China [92, 91]. The power plant capacity reached
about 200 GW at the end of 1994, out of which three quarters were provided by coal causing
a serious transportation problem. The remainder is provided by hydro power and approx.
1 % by nuclear power. Today China has the third largest energy system in the world and
is ranking 4th in the electricity production [91]. Still, at least 10 % of China's population
has no access to electricity and, due to the large population, the electricity consumption
per capita is no higher than 674 kWh/yr compared with, e.g., the world's average of 2200
or to Germany's of 6700 kWh/yr (data of 1992).
63
In terms of global anthropogenic CO2 emission, China with 2373 Mt in 1990 is
ranking 2nd in the world already, and no obligation of reduction has been put on the
country during the Kyoto summit. It is expected to take the lead by 2010 with more than
5000 Mt/yr of emitted CO2.
Chinese primary energy consumption is dominated by coal. Most part of the large
domestic resources of coal are situated in North China currently hauled at a rate of 1.2
billion t/yr and utilized for electricity production and as a raw material in the chemical and
steel industry. Transportation and distribution of the coal around the country causes an
enormous logistic problem resulting in plans to erect large on-site coal power plants and
rather transport the electricity. Coal burning is made with almost no preceding cleaning and
produces 85 % of the total CC>2 emissions which undoubtedly turns into an international
concern. Oil and natural gas is found mainly in the north and consumed as a raw material
in the chemical industry and not for electricity production. Oil production will amount
in 1997 to 160 million t, still China has turned into an oil importer after years as an oil
exporter. Hydro power capacity was 44.7 GW in 1993 compared with 380 GW which are
deemed to be economically available. At present, the largest hydro power plant is under
construction at the Yangtse river with a capacity of 18.2 GW corresponding to about a
tenth of China's present total electric power. In 1992 - 1994, the first nuclear power plants
with a 300 MW(e) and 2 * 900 MW(e), respectively, have started commercial operation
to provide a nuclear share of 1.3 % (beginning of 1997) in the electricity generation.
Renewable energies are currently insignificant. Still, surprisingly China has become one of
the world largest users of renewables (if including hydro), with firewood to be the largest
resource, to be highly important for energy provision in rural and remote areas (wind,
solar, geothermal, tidal) [3, 92]. A recently developed strategy, called the "China Sunlight
Program", running up to 2010, is expected to further promote renewable energies [3], e.g.,
1 GW by wind power.
The increasing demand for hydrogen in China is basically met by hydrocarbon steam
reforming process. Presently there are 23 hydrogen production units in operation with a
total capacity of 582,600 Nm3/h [27].
In the future, China's power capacity development is estimated to reach 310 GW(e)
in the year 2000 and 800 GW(e) in 2020. Because of the large distances between resources
and sites of consumption, nuclear power is expected to gain more importance. The next
two nuclear stations with 2 * 935 MW(e) and 2 * 575 MW(e), respectively, are planned to
go into operation in 2001 and 2002. Nuclear power in the future is considered for various
applications [79]:
• as baseload electricity where coal is short,
• for district heating in areas where most coal is consumed to serve as a clean
energy source, or
• in the process heat industries (e.g., HTGR) to provide steam in the petrochemical
industry and for coal gasification and liquefaction.
64
4.2.2. Experimental and Theoretical Activities in China
4.2.2.1. Nuclear Reactor Concepts
A 10 MW(th) HTGR test module, HTR-10 [89], is under construction since 1995 by
the Institute of Nuclear Energy Technology (INET) of Tsinghua University in Beijing. It is
a pebble bed reactor with 27,000 spherical fuel elements and a coolant outlet temperature
of 700 °C (later 950 °C). Its operation is expected to start end of 1999.
The HTR-10 is designed as a module for multipurpose research. In a first phase,

the HTR-10 is planned to be operated in connection with a steam turbine which works
in a cogeneration mode to provide electricity and heat. In a second phase, the coolant
outlet temperature will be raised from 700 to 900 °C to be operated in a gas turbine cycle
[79]. A long-term test program for nuclear process heat applications has been proposed
in different phases [89]:
a. feasibility study
b. simulation experiments on a laboratory scale for coal gasification by means of hot
helium to acquire technology, experience, and to optimize operational parameters
c. nuclear coal gasification test in a pilot facility where the HTR-10 is to be operated
at 950 °C and to demonstrate IHX operation (see Fig. 4-1)
He 950°C He 800 C 3.0MPa
C.)SW»
IHX: Intermediate Heal Exchanger, SC: Steam Generator. SR: Steam Reformer
ffS: Water Seperotor. CH: Mcthonizolion Device RC: Recuperator PH: Prehentcr
Fig. 4-1: HTR-10 circuit with process heat applications, from [89]
65
A 5 MW water-cooled nuclear heating reactor (NHR-5) is being operated by
INET since 1985 for demonstrating nuclear heating (in winter), cooling (in summer) and
desalination. A follow-up plant, NHR-200, with 200 MW has been designed; a respective
construction permit has been released in 1997 [81].
R&D efforts on innovative nuclear power reactor designs were conducted in inter-
national cooperation. The Chinese design of an AC-600 reactor which has been accepted
by the International Atomic Energy Agency (IAEA) as a progressive concept is currently
investigated by the Nuclear Power Institute of China [92].
4.2.2.2. Exploitation of Very Heavy Oil Fiejds
A concept was proposed exploiting very heavy oil fields in the north-east of China
by the use of process steam from an HTGR [22]. This oil with an estimated total amount of
90 million tons is buried in a depth of 1500 - 1700 m. By means of injection of 370 - 390
°C steam, the very heavy oil resources are heated up for easier recovery. The consumption
of steam is approx. 3 - 5 t per ton of oil delivered. The energy input is estimated to be
30 - 50 % of that contained in the produced oil. Estimated oil production rate is 1.5 - 2
million t/yr. Design data for the oil production by steam injection are for the steam-soak
process a pressure of 16 MPa and a temperature of 370 °C, and for the steam-drive process
a pressure of 10 MPa and a temperature of 350 °C.
Two pebble bed HTGR designs for 1000 MW(th) (5 units of HTR-OTTO-200 or 3
units of HTR-MEDUL-334) were shown to be able to provide steam with the required
parameters to open up the very heavy oil resources [22].
A 4 x 200 MW(th) modular HTGR plant has been proposed to meet increased demand
for process steam and electricity in the Chinese petroleum industry. Coal conversion
processes by means of HTGR are projected to mitigate future shortage of oil supply [80].
4.3. FEDERAL REPUBLIC OF GERMANY
4.3.1. Energy Situation in Germany
Primary energy consumption in Germany totaled 340 million TOE in 199511 with
significant contributions from oil (40.3 %), gas (20.0 %), hard coal (14.5 %), lignite (12.2
%), nuclear (10.1 %) and hydro (1.7 %), corresponding to an energy consumption per capita
of 4.169 TOE. The consumption of primary energy has been stagnant since the 1980s due
to energy saving and industrial restructuring and is expected to remain on this level until
2010, while gross internal production as well as electricity consumption have still been
growing. Growth of the latter, however, was observed to be getting slower [84]. The share
of renewable energies will increase from 2 % in 1989 to 3.5 % in 2010 with hydro as the
most significant contribution [77]. The total installed electric power is presently 122 GW.
The electricity generation in 1997 was 450* 109 kWh with 52 % provided by coal, 35.5 %
by nuclear, 6.4 % by natural gas, 5.1 % by hydro and renewables, and 1 % by oil. The
11
Figures from 1996 are a total primary energy consumption of 350 million TOE with contributions from oil (39.5 %), gas (21.6 '
hard coal (13.9 %), lignite (11.5 %), and nuclear (12.0 %).
66
current opinion is that beyond the 20 units presently operated, no additional nuclear power
plant will be constructed due to lack of consensus among the major political parties.
The industrial energy market in Germany has moved from coal and oil to more natural
gas and electricity. Also the specific energy consumption in the industrial production has
decreased from 1500 to about 850 kWh/1000 DM Gross Value Added (GVA) between 1970
and 1994. However, the energy demand structure did merely change. A fraction of 65 %
of the industrially used energy is consumed in the process heat production [15].
The CO2 emission in 1995 was 894 million t/yr of CO2 which is by 12 % less than
the 1990 level, however, with an increase of 2 % in 1996 compared with the year before.
Traffic causes 17 % of the total CC>2 emissions and more than 50 % of the total NOX
emissions with 85 % of all traffic emissions originating from road traffic (figures from
1987) [32] and a further increase is predicted. The need for CC>2 reducing technologies as
demanded on the Toronto conference in 198812 is in contrast to the realization of the inner
European market with the consequence of a strong competition on the energy sector that
might turn out unfavorable for the renewables [77]. According to the 1997 earth summit in
Kyoto, the European Community (which is treated as a whole "bubble") is obliged to reduce
greenhouse gas emissions by 8 % (observed in 1990-95 for the EC: -2 %) with Germany
- still sticking to its ambitious goal - taking over a 25 % portion within the bubble.
In 1994, the Enquete Commission "Protection of the Earth Atmosphere" has recom-
mended to achieve an energy mix in Germany by the year 2050 with a 25 % share only
from fossil fuels and another 25 % share from renewable energies and 50 % from nuclear
power or imported energy.
Hydrogen production in Germany in 1988 was 19.1*109 Nm3 [70] with the main
sources crude oil (46 %), followed by coal (30 %) and natural gas (19 %). Hydrogen
consumption is by 33 % non-energetic, by 19 % in the petrochemical industry, and by
48 % direct energetic as a fuel gas (coke gas) [65]. As much as 8*109 Nm3 hydrogen
per year are estimated to be generated in chemical processes as an undesired byproduct
and mostly flared without any energy gain. As a large-scale energy carrier, hydrogen in
Germany is presumed to be not competitive until 2010, even under a given demand of a
30 % CC>2 reduction. Only in a prognosis for the year 2040 assuming the need for a 60
% CO2 reduction, the establishment of a hydrogen energy economy will be competitive,
even necessary [77]13.
43.2. Experimental and Theoretical Activities in Germany
4.3.2.1. Reactor Designs for Process Heat Applications
In Germany, the potential of reducing CC>2 output through different hydrogen sub-
stitutions is not very high because of the specific energy demand structure. Due to its
12
In Toronto, the recommendation given was to reduce worldwide CO? emission by 20 % until 2005 and by at least 50 % until 2050.
13
In this study, the assumption on CO2 emission is a parameter in an optimization system to determine the structure of energy carriers to
be utilized and the technology options to be selected. This optimization system determines the economic potential of hydrogen in Germany.
Fundamental prerequisite is the assumption that hydrogen be produced by electrolysis from solar or hydro power. Concerning the origin of
the hydrogen, several strategies were examined: (i) "diversification", allowing limited solar electricity and LH2 imports, (ii) "EC autarchy",
limiting H2 import to < 1800 PJ and only from EC countries, and (iii) "electricity", including domestic photovoltaics and no limitation of
solar electricity import.
67
high production cost, only hydro power or off-peak nuclear power are considered to be an
economically comparable alternative primary energy source for hydrogen production. A
breakthrough for the non-fossil hydrogen economy is not expected until the year 2020 [40].
At present, nuclear power in Germany does merely contribute to the heat market. So
far it is just one plant, PWR Stade, to deliver 40 MW(th) to an industrial company in the
vicinity of the plant. Saturated steam of 190 °C at a pressure of 1.05 MPa is provided at
a rate of 60 t/h for condensation of a saline solution. Total delivery since 1983 winded up
to 5.4 million tons of steam equivalent to 690 GWh by the end of 1996.
In a project of the German chemical company BASF, the installation of a nuclear

power plant for cogeneration of electricity and process steam was planned. It was designed
in two units for the generation of 1200 MW electricity and 3000 t/h of steam. Technical
problems, e.g. storage of steam, and safety issues in connection with the central location
within the chemical complex have finally stopped this project.
A concept of an evolutionary reactor is pursued with the joint French / German

"European Pressurized Water Reactor", EPR, a 1525 MW(e) plant with evolutionary steam
generating system and innovative double-walled containment [20]. A three years basic
design phase as a prerequisite for the beginning of the licensing procedure was finished
in 1997. The characteristic feature is a core catcher to restrict a possible core melt to
the power plant itself. The joint effort by Germany and France, however, finds in both
countries a situation where no further base load is required. The EPR, confirmed as a future
standard in France, is projected to substitute decommissioned nuclear plants.
The 46 MW(th) AYR reactor in Jiilich as a first-of-its-kind pebble bed HTGR has
been operated for 21 years with a high percentage of availability demonstrating that nuclear
power can be converted into high-temperature heat. For more than 10 years, the reactor was
operated with an average outlet temperature of 950 °C. The operation allowed successful
testing of a large number of high quality HTGR fuel elements [88].
A design for the reconstruction of the AVR into a process heat plant, AVR-II, has
confirmed the technical feasibility. The installation of a segmental baffle type reformer was
planned characterized by the existence of openings between baffles and reforming tubes
to allow for a partition of the helium coolant into horizontal and vertical flows to reduce
radial temperature gradients.
The prototype pebble bed reactor THTR-300 [50] has demonstrated the expected
large-scale applicability of HTGR technology [10] and the relatively low level of radio-
activity in the HTGR plant. The measured generator power of 303 MW(e) translates into
an efficiency of 39.7 %. The early shutdown after 423 efpd was not related to the re-
actor concept neither to any safety concerns, but primarily associated with technical and
economic problems.
The present German HTGR reference design is the HTR-MODUL containing a large
number of safety features with the possibility of process heat applications [76]. A two-
vessel side-by-side design was selected for safety reasons and for better accessibility in
terms of maintenance and inspection. The active core consists of 360,000 fuel balls in a
9.43 m tall and 3 m diameter cylindrical pebble bed to produce 200 MW(th) power in the
68
electricity-generating version and 170 MW(th), respectively, in the process heat version.
The results of a safety study for the latter are described in further detail in section 3.3.2. An
application for a site-independent advanced concept license for a two-module facility was
filed in 1987, but later stopped due to loss of interest by the potential user. A respective
TUV audit was practically completed and positive.
The HTR-100 with 258 MW(th) was designed as a reactor for special applications
including process steam generation with possible cogeneration of electricity. It applies the
AYR principle of an upstream coolant flow with the steam generator arranged above the
top reflector. Gas outlet temperature is 740 °C. As a twin plant in the cogeneration mode,
it can provide 170 - 500 t of process steam per hour (1.6 MPa, 270 °C) plus a gross
electricity output of 175 - 100 MW [66].
An industrial power plant with HTGR, Ruhr 100, was proposed for upgrading of hard
coal. It was designed to produce synthetic natural gas by means of a Lurgi pressurized
gasifier with the option of cogenerating methanol and, by applying the Mobil oil process,
gaseous hydrocarbon fuels.
The "gas-plus-steam-turbine-cycle" concept for an HTGR, HTR-GST, with 200

MW(th) has been considered to maximize thermal efficiency. Assuming a gas turbine
inlet temperature of 1050 °C, the calculated net efficiency is 54.5 % [12].
The AHTR 500 is a further development of the HTR-MODUL design with 500
MW(th). The helium coolant is heated up from 330 to 950 °C. The system pressure is 2
MPa. The new feature of this reactor design is a central graphite column to provide an
additional heat sink. It contains a passive decay heat removal system on the basis of natural
convection which runs also during normal operation. No intermediate circuit is foreseen
for the connection with a coal gasification system [41],
The concept of a 900 MW(e) HTR (thermal power 2250 MW) has been developed to
operate in an indirect cycle for electricity generation with the extraction of process steam to
be used for coal gasification employing an improved Lurgi process. The plant is to provide
110 kg/s of process steam at 400 °C and 10 MPa.
4.3.2.2. "Prototype Plant Nuclear Process Heat" (PNP) Project
The 'Trototype Plant Nuclear Process Heat" (PNP) project was founded in 1972
(and terminated in 1992) with the goals of
• developing HTGRs for high gas outlet temperatures of 950 °C as a source of
process heat for coal gasification,
• developing and testing components for heat transfer to the process plant,
• developing and testing processes and experimental facilities for steam-coal gasifi-
cation and hydro-gasification.
A first concept for high-temperature gas-cooled reactor for process steam production
was proposed in [68]. The PR-500 was designed for 500 MW(th) to produce 523 t/h of
steam at 265 °C and 2 MPa plus a net electricity of 55 MW(e). The helium coolant was
69
heated up from 265 to 865 °C. Three loops either one with heat exchanger and blower were
arranged around the pebble bed reactor outside the prestressed concrete pressure vessel.
The PNP-3000 reactor plant was designed for the generation of process heat for a
steam reformer (1071 MW) and of electricity with 540 °C / 19.5 MPa turbine steam (1929
MW). The helium coolant is heated up from 250 to 950 °C at a primary system pressure
of 4 MPa. Nuclear power density is 5 MW/m3. Four (or more) loops are connected to
the reactor either one with circulator, steam generator plus two steam reformers. Eight
steam reforming furnaces are to be operated at 825 °C with a methane conversion of about
65 % [69].
Technical feasibility of a nuclear process heat reactor for the refinement of coal has
been established, main components developed, and its basic licensing capability confirmed.
The economic competitiveness compared with non-nuclear alternatives could be achieveable
by using improvement potential, however, nowadays looking at low oil prizes realistically
not existing. Studies on the technical feasibility and economic competitiveness of the
processes for nuclear coal refinement were completed in 1987 [11].
4.3.2.3. Nuclear Long-Distance Energy Transportation
The steam reforming process was experimentally investigated in the "Nuclear Long-
Distance Energy" (NFE) project. Two different bundles of steam reformer tubes were
tested: (i) the "orifice baffle" design by Lurgi, and (ii) the "counter current concentric"
design by Steinmuller. Convective helium heating was verified for both designs [28]. The
reaction rate for bundle (ii) was found to be dependent on pressure and active velocity
in the empty tube, but independent of temperature indicating a heat-controlled reaction.
Postoperation inspections confirmed the integrity of all reformer tubes.
The complete EVA-II / ADAM-IT system was operated for a total time of 10,150
hours including steady state conditions at both full and partial load, transient procedures,
and also special test situations such as tube blockage [28] (see also section B.I.).
The nuclear variant of using CHsOH as an energy carrier taking the HTGR as a
primary source for the required process heat is schematically shown in Fig. 4-2. A
congenial system is the methanol-carbon monoxide system.
4.3.2.4. EVA-n Pilot Plant
Within the PNP project, a test reformer tube bundle, EVA-II, with a power of 5 MW
has been investigated to prove design and engineering and to demonstrate its feasibility
as being used in nuclear process heat applications such as hydro-gasification of coal or
long-distance energy transportation (see Fig. 4-3). EVA-n was successfully operated over
5800 h within the "Nuclear Long-Distance Transportation" project NFE in connection with
the methanation plant ADAM-II (see also section B.I.) and then modified to account for
PNP steam reforming conditions.
The nuclear heat source was again simulated by an electric heater. The new EVA-n
reformer constructed by the Steinmuller company has used the concentric tube design. It
70
CH4
——i Gas-
Steam- CH3OH
H2O reformer synthesis
fication
Auto-
thermal C02
CH3OH
Gas-
Steam- CH3OH
H2O reformer synthesis
fication
t
1
UU2
-N
( HTGFA
I heat )
_^ 50
50%% COs-emission reduced
no C02-net emission
Fig. 4-2: Methanol as an energy carrier with different CC>2 emissions, from [Research
Center Julich, ISR]
consisted of a bundle of 18 reforming tubes. All other features remained essentially the
same. The test program was conducted in two series with the operating conditions as given
in Table 4-1.
In December 1986, the facility was shut down and a postexamination program was
started. The tests have confirmed the expected thermodynamic and chemical processing
behavior and also the predictive character of respective calculation models. After a total
of 5200 h operation time, the reforming tubes were optically still in excellent shape. The
procedures of catalyst changing need further improvement [53].
71
w V
T «c 4O 40
P bar 41.4 3&S
rh kg/s aeia 1^34
CH, nLVaL a95i Q123
H, • acos aesi
CO H — QO96
CO, « acno O.O98
| «=Ef
I boBng«^v I
br^ni
W~T1
Fig. 4-3: Simplified flow chart of test facility EVA-II, from [53]
Table 4-1: Operating conditions in EVA-II test program, from [53]
Series A Series B
Helium mass flow [kg/s] 2.574 3.850
Feed gas mass flow [kg/s] 1.908 2.776
Helium inlet temperature [°C] 950 950
Feed gas inlet temperature [°C] 700 700
Feed gas temperature at catalyst entry [°C] 572 548
Product gas temperature at catalyst exit [° C] 810 791
Product gas outlet temperature [°C] 457 458
Thermal power [MW] 3.4 5.0
4.3.2.5. "High-Temperature Reactor with Helium Turbine of Large Power" (HHT) Project
Another subject of research parallel to PNP was the project of a closed-cycle "High-
Temperature Reactor with Helium Turbine of Large Power (HHT)" initiated in 1968,
started in 1972, and terminated in 1981 [83]. Main goals in this project, of course, were
72
to achieve a higher efficiency and to reduce the number of components. But it was also
considered to present a broad variety of utilizations, in particular, in the lower process
temperature range. The heart of the HHT concept was a 2700 MW(th) high-temperature
reactor to which three parallel loops were connected, either one containing a 350 MW(e)
helium gas turbine, a recuperator, and a cooler. The helium coolant was heated up from
460 to 868 °C. All components were arranged in a prestressed concrete pressure vessel.
Maintenance capability of turbine and hot gas ducts mainly depend on Cs-137 contamination
which increases steadily during its long lifetime and produces a high 7-dose via its daughter
nuclide Ba-137m [83].
Research activities in connection with HHT were

• a materials research program starting at BBC and at the Research Center Julich to
investigate the lifetime of the turbine, hot gas duct, and insulation materials in a
helium atmosphere,
• the development and testing of the hot gas duct and insulation in the project ARGAS
at the Research Center Julich, and
• the investigation of fission product deposition behavior in the primary helium circuit
(VAMPYR, SAPHIR).
The R&D program included two experimental setups both demonstrating helium as
a closed-loop Brayton cycle working fluid [44]:
(i) a 50 MW cogeneration power plant for electricity and district heating, EVO, in
Oberhausen, FRG, by employing a helium turbine. Operation time was 24,000 h
with 11,500 h at 750 °C temperature,
(ii) the "High-Hemperature Helium Test Plant", HHV, at the Research Center Julich
for the development of helium turbomachinery and components. Helium is routed
through the test section at a rate, of 200 kg/s with a maximum temperature of 1000
°C. Operation was 1100 h with 325 h at a greater than 850 °C temperature.
4.3.2.6. Component Experimental Loop (KVK)
The 10 MW Component Experimental Loop (KVK) ("Komponenten-Versuchs-

Kreislauf') was a high-temperature helium circuit operated by Siemens/Interatom in Ber-
gisch Gladbach. It consisted of a primary and a secondary helium loop and a water steam
system. Its flexible design, however, also allowed for a single loop operation. Various
components were tested under the operating conditions of coal gasification.
Two concepts of a He - He intermediate heat exchanger for a heat rating of 125 - 170
MW have been selected. For both, a 10 MW test plant has been operated in the KVK loop
verifying the operation of reformers with convective helium. A 10 MW decay heat removal
system cooler, hot gas ducts including insulation and liner, hot gas valves, and a steam
generator were other components of the KVK loop. Furthermore, a helium purification
system was operated in a bypass of the main system. Starting in 1982, the KVK facility
was operated for 18,400 h with approx. 7000 h above 900 °C [28]. Hot gas duct with
internal insulation was operated at temperatures up to 950 °C. The KVK experimental loop
has demonstrated reliability and availability even of newly developed components.
73
4.3.2.7. Coal Gasification
From 1976 to 1984 within the frame of the PNP project, the former German company
Bergbau-Forschung, now: Deutsche Montan Technologic (DMT), has established and
successfully tested catalytic and non-catalytic steam-coal gasification in a 1.2 MW semi-
technical scale experimental facility, WKV, developed for hard coal, in the city of Essen.
Heat was provided by helium electrically heated up to 950 °C via a steam generator. The
plant was in hot operation for approx. 23,000 h with more than 13,000 h under gasification
conditions. The maximum capacity was 230 kg/h of coal. The total amount of hard coal
gasified was 1600 t [38].
The hydro-gasification process as an alternative to the above described steam-

coal gasification process was pursued by the German Rheinische Braunkohlenwerke AG,
Cologne. Its principle was successfully verified in a semi-technical test facility, HKV, in
Wesseling, which operated for about 27,000 h with more than 12,000 h under gasification
conditions from 1976 to 1982. The throughput was 320 kg/h of lignite corresponding to
a power of 1.5 MW. The total amount of coal gasified was 1800 t. The largest non-stop
operation was 748 h. Carbon gasification rates up to 82 % were achieved with methane
contents in the dry raw gas up to 48 % [62]. Since 1983 until 1985, a followup pilot plant
was operated with a throughput of 9.6 t/h of lignite corresponding to a power of 50 MW
and an SNG production of about 8000 Nm3/h. The temperature was between 850 and 950
°C, pressure was between 6 and 12 MPa. The inside diameter of the gasifier was 1 m.
The translation into a large-scale gasification plant is still in the design phase. Focus is
laid upon the non-catalytic gasification.
Higher coolant outlet temperatures allow for a smaller reaction volume and a higher
coal throughput, respectively. The intermediate heat exchanger is required because of
the difficulties expected when handling coal and ash within the reactor containment.
Furthermore coal gasifiers have a shorter lifetime than steam reformers which means that
the former should be outside the nuclear plant for shutoff and repair purposes [69].
4.3.2.8. Hydrogen / Methanol from Raw Iron Furnace
In Fig. 4-4, a schematic of an iron ore reduction plant is shown using hydrogen
generated by a nuclear reforming process. The hydrogen is heated up to 900 °C and then
routed into the blast furnace to act as a reduction gas [45].
A proposal was made in [56] to deploy a stack gas processing plant for either
• the generation of hydrogen
in a homogeneous water gas reaction and removal of Na and residual CO by
converting all CO with steam, or for
• the generation of methanol
by applying the proper ratios of H2, CO and CO2 as well as removal of NI- The
HI to CO ratio is in the range of 2.4 - 4.5, the CO2 fraction is 5 % for optimal
synthesis at the expense of a reduced iron production.
A balance of such a plant with the combined production of raw iron and methanol /
hydrogen is summarized in Table 4-2. Compared with the reference case 1 using a rich ore
74
Ore pellets
Oil "I 67 % Fe
Natural gas > Conversion
Coal J
V\feter splitting
Condenser
850 - 900 °C
H2 /VTA H2. H2o Reduction
He
1 temperature
Com-
950 °C pressor
Furnace
HTGR H2 long-distance
pipeline (20 °C) - H2
3 Cooling
1070 Mm zone
250 °C 750 'C H2/t SI
370 Mm3 H2/t SI Sponge iron (SI)

R = 95 %; 80 °C
Fig. 4-4: Direct iron ore reduction plant, from [45]
Table 4-2: Balance of direct iron ore reduction plant, from [56]
Hydrogen Methanol
Case 1 Case 2 Case3 Case 3
(rich ore) (lean ore) (O2/H2O wind) (O2/H2O wind)
HTGR [MW(th)] 250 250 250 250
Coke [t/h] 70 74 72 72
Stack gas [mol%]
CO 37 40 65 65
C02 12 12 8 8
H2O 6 7 16 16
H2 2 3 10 10
N2 43 38 1 1
Hydrogen of 99.9
71,160 81,030 104,980 -
mol% [Nm3/h]
Methanol [t/h] - - - 59
Raw iron [t/h] 81 87 73 73
Net electricity
28 23 34 5
output [MW(e)]
Thermal efficiency
49.0 50.8 57.6 55.6
[%]
75
JH,_________________________
Methanol Methanal Methanal
synthesis synthesis electrolysis
2H2+CO 1/2O2
o
Fig. 4-5: Schematic of a nuclear plant design for the "methane-methanol-methanal"

process with the main material streams, from [9]
and an oxygen/steam enriched furnace wind, case 2 treats an alternative for lean ore which
demands more energy for the reduction step. In case 3, a pure oxygen/steam furnace wind
is used which has the advantage of an almost nitrogen-free stack gas whose treatment is
easier and less expensive. The process energy required is considered here to be provided
by a 250 MW(th) HTGR. The thermal efficiency for the methanol production could even
be increased from 55.6 to 70.9 %, if the hydrogen needed for the synthesis is provided
from outside without using the internally converted hydrogen [56].
4.3.2.9. Hydrogen Production
The basic design of a nuclear plant for the hydrocarbon-hybrid or the "methane-
methanol-methanal" process is schematically shown in Fig. 4-5 [9]. It is based on an
HTGR to cogenerate low-temperature energy to be used in the steam generator and high-
temperature energy to be used in the steam reformer. The latter is taken for methane splitting
as the first step of the hybrid cycle to catalytically produce synthetic gas, a mixture of HI
and CO. A certain fraction of the hydrogen contents is separated by a two-fold application
of the shift reaction. The remaining synthesis gas is then consumed in the process of
methanol production. In step (3), the methanol is split into methanal (= formaldehyd) and
hydrogen. The heat required for this reaction is taken either from the steam generator or
from the previous exothermal reaction. Both products methanal and hydrogen are fed into
the electrolyzer to regain the methane which is recycled and to obtain the desired byproduct
oxygen. According to a first design, 2.6 MW thermal power from an HTGR are required
to generate 300,000 Nm3/h of hydrogen [9].
76
Research work has been conducted at the Research Center Julich on the Westing-
house sulfuric acid hybrid cycle. The heat consuming step of sulfuric acid splitting was
successfully realized in bench-scale experiments under HTGR conditions at 4 MPa and
with 950 °C heat from a furnace.
4.4. INDONESIA
4.4.1. Energy Situation in Indonesia
Indonesia is an oil and gas producing country with primary energy shares of 60 %
and 33 %, respectively. Coal although abundantly available consists of low-quality lignite
only; its share is about 6 %. The Indonesian domestic energy consumption rate is presently
increasing by 15 % per year. In the 1970s, several giant natural gas fields have been
discovered making natural gas a major export commodity. In 1994, the LNG production
amounted to about 25 million t with 90 % of it exported to Japan, the residue to South
Korea and Taiwan. The latest find, the off-shore Natuna natural gas field, is estimated
to have an energy content of 1.5 TWy which would allow a production of 38 million t/yr
of LNG over 20 years. Strong competition among international petrochemical companies
is going on around this so-called "35 Billion plus" US $ project (with 10 Billion US $
needed for carbon dioxide extraction and re-injection). The Natuna gas feature of a very
large content of CC>2 of 71 % is hoped to become not a burden but rather an economic
challenge to introduce new technologies by means of gas processing to be the incentive
for a rapid economic growth in South East Asia [2]. Proven gas reserves in Asia stand at
some 10,650 billion Nm3 with a share of 7.2 % of the world's total reserve. Hydro power
is currently 1.7 GW restricted to Java island.
4.4.2. Experimental and Theoretical Activities in Indonesia
4.4.2.1. Exploitation of the Natuna Gas Field by Means of HTGR
A feasibility study has been initiated by the IAEA to investigate the application of
the HTGR for cogeneration purposes of COz conversion, desalination, and H2 production
in terms of economy and technology [17].
The prerequisite of minimal emission of CC>2 into the atmosphere has suggested the
nuclear option in exploiting the Natuna gas field. Six alternatives have been identified as
being feasible and economically competitive, distinguished by the desired product methanol
or methane (or hydrogen) and by the COz consumption [14]:
(a) steam and COz reforming and HZ from electrolysis for methanol synthesis with 15
% consumption of the Natuna CO2,
(b) steam and CC>2 reforming and HZ from electrolysis for methanol synthesis with 25
% consumption of the Natuna CC>2 (see Fig. 4-6 and Table 4-3),
(c) steam and COz reforming and HZ from electrolysis for methanol synthesis with
100 % consumption of the Natuna CO2,
(d) COz reforming and HZ from electrolysis for methane and methanol production with
100 % consumption of the Natuna COz,
77
(e) Natuna gas and H2 from electrolysis for methane production with 100 % consump-
tion of the Natuna CO2,
(f) separation of methane from the Natuna gas and H2 from electrolysis plus Natuna
CO2 for methanol synthesis with 100 % consumption of the Natuna CO2-
Natuna
—————————————————————————————————«>^ OO
f^f^ ®
(D gas Oonaratnr -
2
CH4,
C02 ©
(RI) ®© (RS)
CH4 \~S H2, \^^
^%^» -*j- ^"^ •>•" ^^V^*^,^^^1;^. ^ * V^Vx ^^^1^^^^?Tii|^'^-S?^''1^**- j^«
c
(^ j i_j /~\ '——-* | Steam -.; °2 ? Methanol '^ C§/ CH3OH
. . . » ' reforming >: J * * synthesis § *
/ H2O
-3- ^* i-* î ' S^
(R2)
"" " ^S£t* "* - ^ Jî -^
© H2
C^
vV O
v2/ F |ppfrol\/î<5
H2O
IS^?^S'?
Fig. 4-6: Use of Natuna natural gas for methanol synthesis, variant (b) (see text),
sources and sinks of species involved, from [14]
Table 4-3: Use of Natuna natural gas for methanol synthesis, variant (b) (see text),
from [14]
Species flow n, An [kmol/s]

1 2 3 Rl 4 5 R2 6 7 R3 8
CH4 +4.5 4.5 - ,*•-. ,
,s!s.-^>>~. Hi
.„>!
-4.5
" I- J
fti
CO2 + 12 3 3 3 ft 3 -3 9
-r -.ra<ft
H2O +9
-^
4.5 -4.5 9IB:. 4.5 -4.5 v , f: -•"• +3 3
H2 if*-:-:, -I +13.5
^, ,.,<•-! --* 13.5 ?<,^/ii t;^, +4.5 ;
4.5 18 -18 V/i
j-
CO V
,' V
' "^^-.'^
+4.5 4.5 1 !'/-'« - 4.5 -4.5 >' 41
02 !
*^^ **
r ^
+2.25 2.25 2.25
M (D - -- +7.5 7.5
(1) M = Methanol = CH3OH
78
Another process is the production of formic acid, CH2O2, which would result in a
complete conversion of the Natuna gas:
3 CH4 + 9 C02 + 6 H20 -» 12 CH2O2
Formic acid is a product of the chemical industry and could again attention as a
future fuel for fuel cells [13].
An HTGR is proposed to be operated in cogeneration mode to provide the reaction

heat for the reforming processes as well as electricity for water decomposition by electro-
lysis. Resulting products are the marketable fuels methanol as a motor fuel and, of lower
value, methane as a heating fuel. Taking account of the huge size of the Natuna gas field,
the thermal power of about 20 modular HTGRs with 200 MW(th) each is required [14].
A working group has been established by the National Atomic Energy Agency in
Indonesia to concentrate on coal conversion to synfuel as a substitute for depleting oil
or to synthetic natural gas or to synthesis gas, CO2 conversion to methanol, hydrogen
production, and enhanced oil recovery [1]. In connection with coal liquefaction, an HTGR
could provide process heat for dewatering of the coal, thermal treatment of the slurry, and
electricity production; this plan, however, needs further elaboration [29].
4.4.2.2. Oil Recovery by Means of HTGR
The thermally enhanced oil recovery process by steam flooding is a tertiary recovery
method. A four-unit HTGR power plant with 200 MW(th) each for cogeneration of steam
(530 °C, 15 MPa) and electricity (25 - 30 MW(e)) has been proposed to exploit the Duri
oil field which currently represents the largest steam flood project in the world. Peak
production has reached 300,000 bbl/d which requires 4,000 - 5,000 t/h of steam consuming
20 % of the oil produced. However, the present oil price development does not allow the
economic employment of HTGRs for this type of oil recovery [57].
4.4.2.3. Hydrogen Production
Two hydrogen production plants are presently being operated in Indonesia used in oil
refinery, ammonia synthesis, and methanol production. The capacities are 1.8*109 Nm3
and 2.1*109 Nm3, respectively. The H2 is generated from steam reforming of natural gas.
It is projected to replace fossil fuels in industrial processes, in the transportation sector,
and in households.
4.5. JAPAN
4.5.1. Energy Situation in Japan
The significant growth of Japan's economy in the 1960s was based on cheap petroleum
imports, which, however, ceased with the first oil crisis. The oil portion in primary energy
consumption was lowered from 77 % in 1973 to 56 % in 1985, while shares of the
alternatives nuclear and natural gas, also coal, were rising. It has become and is expected
79
to remain the country with the most imports of LNG covering 70 % of the worldwide
traded LNG. The Japanese energy consumption in 1993 of 507 million TOE (= 2.12*1013
MJ) was covered by oil (53 %), gas (17 %), coal (16 %), nuclear (11 %), and hydro (4 %)
[61]. A portion of 60 % of the energy consumption in Japan is non-electric. About one
third of the CC>2 emissions are from fossil-fueled power plants. The committment made
on the Kyoto conference in 1997 for Japan is a reduction of the emissions by 6 % relative
to 1990 compared with an observed increase of 8 % by 1995 [16]. According to MTTI,
however, even the minor goal of greenhouse gas reduction to the 1990 level increases
Japan's nuclear dependence demanding 20 more units of the 1300 MW(e) class by 2010.
Japan was the first industrial country to sign the Kyoto protocol.
The largest share in Japan's electricity production of 907*109 kWh (1993) is provided
by nuclear power with 34 % from 47 units; others are LNG (29 %), oil (18 %), hydro (11
%), and coal (8 %) [4, 61]. In mid 1997, Japan has operated 53 nuclear power stations;
also Japan has the two largest nuclear sites of the world, Fukushima with 10 BWR units
and a total of 9096 MW(e), and Kashiwazaki-Kariwa with 7 BWR units and a total of
8212 MW(e). Because of shortage of land, Japan has a preference for large nuclear power
units with a strong and diverse program supported by the big industrial companies. The
different designs for future reactors currently being pursued are considered promising and
are expected to form Japan's next generation of reactors [33]. Reactors planned and under
construction are expected to raise the nuclear share to 42 % by 2010 [46]. At present,
however, the Japanese nuclear policy is being re-examined after the Monju accident which
has aggravated the already serious difficulties of getting public acceptance [46].
Japan depends heavily on imported energy (84 % of the primary energy) and is thus
highly energy-vulnerable. Therefore intensive research efforts are being conducted on new
energy carriers. Fuel cell technology is considered one of the most important new energy
technologies. The most likely chance for entry of SOFC based power generation plants
is seen for Japan because of its very high gas and electricity price level combined with
stringent emission requirements [8]. The WE-NET project focusing on hydrogen to follow
LNG is one example, research on gas hydrates14 is another one.
4.5.2. Experimental and Theoretical Activities in Japan
4.5.2.1. Process Heat from Light-Water Reactors
For the purpose of a stable, long-term energy security, promoting industries, and
development of local areas, Japan is currently investigating the potential of light-water
reactor process heat applications. The chemical industries in Japan occupy a share of about
one sixth of the electricity which is mostly consumed in the form of heat. Designs have
been presented to utilize steam from 1100 MW(e) class PWRs or BWRs.
The LWR is separated from the chemical plant by an intermediate circuit to minimize
the risk of a radioactivity escape into the process gas circuit. A further advantage will be
the accumulation of steam generated by cheap off-peak electricity. The basic technology
14
Gas hydrate is a mixture of methane embedded in ice crystals at -23 °C buried at the ground of the oceans. One m3 of gas hydrate
transforms under normal conditions into 164 m3 of methane plus 0.8 m3 of water.
80
is available. The demonstration of process heat applications is considered by using a small
or medium-sized LWR [19].
The final goal of heat utilization of LWRs is described in a Japanese study [19]
bypresenting the concept of a new city for approx. 50,000 inhabitants with the shape of half
a circle and situated at the coast line. The nuclear energy supply center is located in the
(circle) center to provide heat and electricity to belts of industries, greens, residences,
arable land in a concentric arrangement. The energy center consists of four nuclear power
units, two PWRs, two BWRs, with 1000 MW(e) each.
4.5.2.2. H igh-Temperature Reactors
Utilization of heat from the HTGR has been studied during the 1970s and 1980s in
USA and Germany. In Japan, the experimental multi-purpose VHTR was designed by
JAERI and a nuclear steel-making plant was studied supported by AIST. But the project
was premature at that time and discontinued in 1980. Still the HTGR was and is considered
a promising solution to meet the increased demand for clean energy on the basis of the
energy carrier hydrogen to be used in industry and transportation. The Japanese market is
foreseen to hold for 80 GW thermal power provided by HTGRs [34].
__Starting in 1991, the 30 MW(th) High-Temperature Engineering Test Reactor,

HTTR, is being constructed in Oarai [58]. The main components are the reactor pressure
vessel containing a prismatic core, a main cooling system with a 10 MW He-He
intermediate heat exchanger (IHX), plus a 20 MW pressurized water cooler (PWC), and an
auxiliary cooling system as well as a reactor vessel cooling system. In the first operation
phase, the helium-cooled reactor will have an outlet temperature of 850 C. HTTR is
planned to be critical in 1998.
With the introduction of the second core, the HTTR will be the first-of-its-kind
nuclear reactor to be connected via a He-He heat exchanger to a high-temperature process
heat utilization (natural gas steam reforming) system with an outlet temperature of 950 ^C.
The IHX is a counter-current and helically wound tube type shell-and-tube heat exchanger
designed to provide helium of about 900 C at 4.1 MPa. Due to the high pressure and the
low temperature of 800 C in the reformer, the CH4 conversion rate is expected to be no
higher than 65 %. Start of its operation in the HTTR is planned with the installation of the
second core expected for 2002 [44]. Promoted by the International Atomic Energy Agency
(IAEA), a Coordinated Research Program with international cooperation has been started in
1994 on the design and evaluation of heat utilization systems with the goal to identify the
most promising technologies to be demonstrated at the HTTR.
The FAPIG (First Atomic Power Industry Group) representing 22 (now 18) Japanese
industrial companies from different branches, organized a working group to study how to
use heat from HTGRs to mitigate environmental and resource problems, to stabilize supply,
and to promote effective use of energy. Three types of heat utilization plants based on
HTGRs have been modelled, for each of which the system outline, basic plant layout, social
effects, and technical issues were studied [35, 85]:
(i) a cogeneration plant for a typical chemical plant with two modular HTGRs of 450
MW(th) with a helium outlet temperature of 700 C, each to provide 640 t/h of
81
steam of 200 °C plus 210 MW of electricity with an estimated overall efficiency
of 74 %,
(ii) a coal gasification ammonia production plant with four modular HTGRs plus IHX
of 170 MW(th) each and a coolant outlet temperature of 950 °C to produce 1700
t/d of ammonia, coal gasification rate is 2200 t/d in a fluidized bed, and
(iii) a hydrogen production plant with four 200 MW(th) modular HTGRs with a coolant
outlet temperature up to 1050 °C and applying both the iodine-sulfur process and
the high-temperature steam electrolysis process to produce 37,090 and 110,000,
respectively, Nm3/h of hydrogen.
The concept of an HTGR plant with a closed-cycle gas turbine is also being pursued
by FAPIG and a proposal has been made for the construction of a demonstration test facility.
4.5.2.3. Hydrogen and Methanol Cogeneration by Steam Reforming
Top candidate to be tested by J AERI in connection with the HTTR is steam reforming
of natural gas for hydrogen and hydrogen/methanol coproduction including carbon
dioxide recycling [26]. The thermal energy of 10 MW is planned to be transferred to
a secondary helium circuit via the IHX. The total power needed to provide the reforming
process heat is 4.8 MW with 3.6 MW transferred from the helium gas to the steam
reformer and 1.2 MW transferred from the outlet to the inlet process gas [82]. There
are differences compared with a conventional fossil-fired system, in particular, a lower
maximum temperature and a higher process gas pressure, so that a lower hydrogen
production is expected. However, an optimization of the operation conditions should reach
compatibility with the conventional system (see Table 4-4). The heat utilization efficiency
is expected to be 78 % being close to the figure of 80 - 85 % of a conventional reforming
system. The goal is the production of 3800 Nm3/h of hydrogen or, in the cogeneration
mode, 1390 Nm3/h of hydrogen plus 1930 kg/h of methanol, while the input is 950 kg/h
of natural gas plus 10 MW of process heat. A schematic of the coproduction process is
shown in Fig. 4-7.
Prior to the installation of the steam reforming system in the HTTR, the operation
of an out-of-pile plant downscaled to 1/20 - 1/30 of the size projected for the HTTR, is
planned to investigate system characteristics, to verify operation and control technologies,
and to confirm safety and performance under simulated HTTR operational conditions [49].
Pilot plant designing has started in 1995; its construction has begun and is expected to be
completed in 1998.
4.5.2.4. Nuclear Coal Gasification
Another nuclear heat application system that offers reduction potential is the conver-
sion of coal by gasification to other synthetic fuels. The gasification process selected is
dependent on what type of fuel is desired. Hydro-gasification is appropriate for H2 or CH4
production, whereas partial oxidation and steam gasification are better suited for methanol
production. The latter process has been proposed for a nuclear coal reforming system with
methane reformer and steam gasifier based on a 450 MW(th) HTGR. With an input of 93
t/h of water and 34.5 t/h of coal and 27,490 Nm3/h of methane, the methanol production
rate would be 101 t/h plus 68 MW(e) electricity [31].
82
Table 4-4: Comparison of operation conditions and performance of conventional with
nuclear-heated steam reformers, from [26]
Steam reformer
Helium-heated
Fossil-fired Helium-heated
(Improved design)
1 -3
Process gas 4.5
(depending on final > 4
pressure [MPa] (at inlet)
products)
Maximum process
gas temperature 850 - 950 <750 800
[°C]
Maximum heat flux
to catalyst zone 50-80 10-20 40
[kW/m2]
Efficiency [%] 80-85 « 50 78
CC>2 emission [t/h]
3 0 0
(based on 10 MW)
I Methanol Methanol
•* Synthesis
Reactor 1930kg/h
Reactor
(30 MW) Hydrogen
1390Nm3/h
LNG
950kg/h
Feedwater
2010kg/h
Heat Load Controller

Secondary Helium
Fig. 4-7: Schematic of the HTTR steam reforming hydrogen / methanol coproduction
system, from [25]
83
4 6 w t % H2SO<
SECTION 1 SECTION 1
Fig. 4-8: Schematic of JAERI bench-scale apparatus for the "iodine-sulfur process"
thermochemical cycle for hydrogen production (LL = liquid-liquid, VL - vapor-
liquid), from [73]
The hydrogen production step of methane reforming could also be substituted by

high-temperature electrolysis. The calculated balance here is 48 t/h of sea water plus 27
t/h of coal plus 15 t/h of air as input to yield 42 t/h of methanol. Excess hydrogen could
be stored in tanks and retrieved for methanol synthesis, supposed the nuclear power needs
to be used in peak periods [31].
4.5.2.5. Nuclear Hydrogen Production by JTiermochemical Cycles
A bench-scale test facility for hydrogen production using the thermochemical iodine-
sulfur (IS) process has been established at JAERI to verify the hydrogen production,
to study the conditions for the reactions, and to gain experience for a large-scale plant
[74]. The three reactions (see appendix A.2.3.) are performed in separate sections of the
apparatus, the Bunsen reaction and the sulfuric acid decomposition at the same time to
avoid SC>2 storage (see Fig. 4-8). The process requires temperatures of 800 to 900 °C. Its
feasibility was successfully demonstrated in a glass, quartz, and teflon lab-scale apparatus.
In the course of six cycles completed, the total amounts of H2 and 62 produced were 16.4
1 and 9.9 1, respectively. The thermal efficiency achieved, however, was much smaller than
the theoretical one of 47 - 50 % [44]. In late 1997, the continuous operation of the IS
process cycle as a closed loop over 48 h resulted in the production of 44.8 1 of Ha [75].
84
Parallelly a simulation has been started to formulate a flow sheet for the cycle
operation. The IS process and its characteristics have been verified. Agreement has been
achieved between experimental data and the simulation model. A further improvement of
the process is needed to acquire a higher thermal efficiency and to select better structural
materials which are compatible on an industrial scale. Further demonstration steps are
the employment of a metallic reactor for hydrogen production at a rate of 1 Nm3/h and a
scale-up process to a 100 Nm3/h rate [44]. The IS process is one of the candidates to be
connected to the HTTR following the steam reforming system.
The 4-step UT-3 thermochemical cycle with bromine-calcium-iron developed at

Tokyo University is considered in Japan to be superior to many other thermochemical
cycles. It has been successfully transferred into a bench-scale continuous model plant
MASCOT. The system (Fig. 4-9) consists of four reactor furnaces containing the solid
reactants CaBr2, CaO, Fe2Os, and FeBr2, respectively, which are manufactured as spherical
pellets. Only gases are passed through the reactors which eases material flow control.
Eleven cycles have been completed with a yield of 0.2 - 0.3 1 of hydrogen per cycle [59,
90].
A modified UT-3 cycle has been developed where the reactions are carried out
adiabatically. A process evaluation promises a higher thermochemical efficiency compared
with the non-adiabatic version. The combination of the modified process with an HTGR
for hydrogen and electricity cogeneration is one aspect of the current research efforts. The
concept of a plant has been developed for the production of 30,000 Nm3/h. The hydrogen
production rate varies with the thermal efficiency and the nuclear heat input (see Fig. 4-10)
[59].
The UT-3 cycle has also been studied based on solar as the thermal energy source.
Its flow sheet needs to suit the intermittent characteristics of the heat source as day/night
operation. Kinetic studies have shown that it takes about one hour for a complete cycle.
Hydrogen generation rate is designed to be 2000 Nm3/h. Surplus solar energy that is not
immediately used could be stored and retrieved at night to allow a continuously running
process. The calculated thermal efficiency of the solar cycle is 49.5 % [60].
4.5.2.6. Hydrogen Production by High-Temperature Electrolysis of Steam
Another candidate for nuclear process heat application is hydrogen production by

means of high-temperature electrolysis of steam. It has already been tested by JAERI in
a bench-scale facility. With the experience gained, design data on the process characteristics
are to be derived. Experiments conducted in a serial arrangement of 12 tubular-type cells
at a temperature of 850 °C resulted in a maximum yield of 4 Nl/h of hydrogen, which was
increased to 7.6 Nl/h for a temperature of 950 °C. The maximum efficiency was with 80 %
rather low. Tests have started with a planar-type cell. Hydrogen production densities were
shown to agree well with the theoretical values, however, on a still very low efficiency
level [54].
Next steps are the cell optimization, use of new construction materials, and completion
of the system with ancillary equipment. The construction of a prototype system is planned
with a capacity of 10 and later 100 Nm3/h [30].
85
Hydrolysis
analyzer
\
HBr.HsO H2O
1. Electric furnace 3. Plunger pump
2. Evaporator 4. Condenser
Fig. 4-9: UT-3 cycle system MASKOT, from [90]
240 60 ,_
220
15 25 30
3
PRODUCTION H2 [ NnrrVh x 10" ]
Fig. 4-10: Nuclear heat required and process thermal efficiency as a function of
hydrogen production rate, from [59]
A 350 MW(th) HTGR is planned to be connected to 38 high-temperature steam

electrolysis units, either one in a cylindrical vessel of 5 m height and 4 m diameter. The
electric power required will be 139 MW resulting in a total hydrogen production of 39,000
Nm3/h [48].
86
Thermal output
300'C
>100<J8C
I
jr Low Temperature
™ •* 'Pressure Vessel
High Temperature
Pressure Vessel
100-300«C
Thermal Input
Compressor
~soo°c
Thermal Input
•I 9^- Hydrogen ges flow
————> Helium gas flow

HX; Heal exchanger
Fig. 4-11: High-temperature chemical heat pump test system at JAERI, from [36]
4.5.2.7. HTTR High-Temperature Chemical Heat Pump System
A high-temperature chemical heat pump system has been designed at JAERI to heat
up waste heat of about 500 °C to above 1000 °C by using Ti/Cr metal hydride intermetallic
compounds (Fig. 4-11). This method would improve the total energy efficiency of the
hydrogen production system combined with the process heat provision by the HTTR. The
system consists of a low-temperature and a high-temperature vessel with the metal hydride
material. The exothermal process of hydrogen absorption by the metal is able to heat up
helium to approx. 1000 °C as was demonstrated in a test system [36].
4.6. THE NETHERLANDS
Within the so-called WHITE ("Widely Applicable High Temperature Reactor")

program, the design of a small-sized HTGR for combined heat and power generation,
INCOGEN, is being developed. The reactor of choice is a 40 MW(th) direct-cycle pebble-
bed reactor with a so-called peu-a-peu15 loading scheme [67].
15
The peu-a-peu loading scheme is characterized by a slow, but steady fueling of the core resulting in just a small amount of surplus
reactivity. One single loading cycle could encompass the whole reactor lifetime. For the INCOGEN concept, a 10-years cycle is considered.
87
4.7. RUSSIAN FEDERATION
4.7.1. Energy Situation in Russia
The energy economy in Russia is the largest provider of foreign exchange in inter-
national trading and usually considered the backbone of the Russian economy. It is also
strongly influenced by the present economic crisis in the country. The region of the former
USSR was the only in the world where the energy consumption was decreasing in 1996,
minus 2.5 %, compared with the year before due to a lower demand by the domestic in-
dustries of the CIS16 countries. Also the consumption per capita was reduced by a third
in the past five years and has reached European level, approx. 3 TOE or 125 GJ [7]. The
shares of electric capacity are 9 % nuclear, 22 % hydro, and 69 % fossil fuels [55]. The
consequence has been a 30 % reduction of the CO2 emissions compared with the 1990
level. According to the Kyoto agreement from 1997, Russia's greenhouse gas emissions
have to stabilize at the 1990 level, thus seeming predestined to trade emission permits to
the high-polluting countries.
The present economy crisis in Russia is thought of being a threat to energy supply
security in the immediate future. The ongoing restructuring of the energy economy has
led to a significant fraction of idle nuclear and hydro power capacity [55]. Current nuclear
power in Russia allows the production of 20 GW electricity with a significant fraction
dedicated for export to the other CIS countries or to the Scandinavian countries. However,
within the next 10 years, about 60 % of the reactors are expected to reach the end of
their lifetime. The Russian nuclear program is largely based on 1000 MW(e) units, but the
500 - 600 MW range is well represented in the development program. The nuclear option
seems to have much better economics compared with conventional sources for application
in remote areas, e.g. heating. The Russian Federation has also designs for HTGRs in the
range of 50 - 400 MW(e). The VGM reactor with 200 MW(e) is intended for process heat
application with helium outlet temperatures up to 900 °C [33].
4.7.2. Experimental and Theoretical Activities in Russia
Due to great experience in the development and construction of gas turbines, designs
of HTGRs are being considered using a direct helium cycle, a cycle with intermediate heat
exchanger, and the combined gas-steam cycle. The MVGR-GT - 200 MW(th) is a pebble
bed HTGR in a closed-cycle gas turbine system. Helium coolant temperatures are 400 °C
at the inlet and 850 - 950 °C at the outlet. The HTGR-10 MW(th) is thought of as being
appropriate for decentralized power supply. Its design includes an intermediate loop and
a combined gas-steam cycle.
The GT-MHR - 600 MW(th) concept of a direct cycle HTGR is a joint development
by the US General Atomics and Russia's Minatom starting in 1995 and since 1997 supported
by the French Framatome and Fuji Electric, Japan. It is planned to be built as modules
of 250 - 285 MW(e) each with a prismatic core for commercial electricity generation and,
in particular, for the burning of weapon-grade plutonium. A prototype and a plutonium
16
CIS - Community of Independent States is the successor of the USSR.
88
fuel production facility is intended to be constructed at the Siberian Chemical Collective
Combine in Seversk [39].
The conceptual design of a 200 MW modular high-temperature gas-cooled reactor,

VGM, has been developed and is intended to be used for process heat applications in
various industrial branches including oil refineries, petrochemical plants, and the production
of hydrogen, methanol, and synthetic fuels. To assure a high level of safety, the system is
designed to have a helium-to-helium IHX and a helium-to-silicon oil process heat exchanger
(PHX) [23]. The single-cycle variant, VGMP, with 215 MW(th) is completely orientated
towards production of process heat. A design including three units has been made for
connection to a standard oil-refinery plant [23],
4.8. SOUTH AFRICA
In 1996, the installed electricity capacity in South Africa was 38.5 GW with 90 %
provided by coal-fired power plants. The only non-coal power station for base load is
nuclear (two units of 965 MW(e) each). The fact of a continuously growing demand for
more electricity in South Africa has raised serious consideration on installing new power
plants. An alternative contestant to a conventional coal plant which must be erected near the
coal mining, is a modular high-temperature direct-cycle reactor which could be erected in
the electricity consuming coastal area and which is considered to increase diversity. The
pre-conceptual design of a pebble bed HTGR with 900 C gas outlet and 558 C gas inlet
temperature and a peu-a-peu loading scheme for pure electricity production is currently
investigated. Reference size will be 100 MW(e) or 222 MW(th). The cost target fixed at
< 1000 US $/kW is probably achievable only under series construction conditions (>10
modules) [21, 52]. The option of a gasification system for upgrading the enormous
domestic coal reserves should be given a thought.
The company Sasol operates a synthetic fuel production facility consisting of 48 coal
gasification (Lurgi) units. The product gas is generated at a rate of 2.1 million Nm3/h with
a 38 % hydrogen share and is used for gasoline and diesel production by means of Fischer-
Tropsch synthesis [78].
4.9. USA
4.9.1. Energy Situation
In 1994, electricity production in the United States winded up to 3*10 12 kWh from a
mix of energy sources with coal (55 %), nuclear (20 %), natural gas (11 %), hydro (9 %),
oil (3 %) as the most significant energy sources. Renewable energies supply 2 % of the
national electricity and 8 % of the energy demand, respectively, with 90 % of the current
solar, wind, and geothermal generation residing in California [6]. The United States are
ranking first in the annual energy consumption of more than 6 TOE per capita in 1996.
With 20 %, the USA are also ranking first in the world's COz emissions. Greenhouse gas
emissions have increased in 1996 by 3.4 % along with the energy consumption (plus 3.2 %)
compared with the year before and are 8.3 % above the 1990 level. The USA's obligation
from the Kyoto summit in 1997 is a greenhouse gas emission reduction of 7 % below the
1990 level.
89
A fraction of 45 % of all industrial energy consumption in the US is used at
temperatures below 300 C, a temperature compatible with LWRs. The huge hydrocarbon
reserves that exist in the USA in the form of coal and natural gas could be easily used for
effective alternate transportation fuel production.
Despite the fact that in the past 25 years no further nuclear power plant beyond the
existing 107 units (1997) in the USA was licensed and no new plant is planned (for
economic reasons), the US Department of Energy (DOE) has recently proposed to increase
the budget for nuclear R&D efforts by 44 %. The transition to smaller nuclear power units
is expected to reduce the financial risk of the US utilities and to provide a better adaptation
to the demand for further electric power [37]. The current strategy aims at the prolongation
for the operational license of the plants.
Renewable energies have been demonstrated to be technologically successful,

however, they have to compete against the conventional energy carriers whose prizes are
artificially kept on a low level. Photovoltaics (PV) is currently encountering a strongly
increasing market for off-grid electric power needs in the USA and also internationally.
Research continues to improve system performance and reduce system cost. Small PV
systems for remote applications are commercialized. Large-scale systems involving units
with up to 8 MW are currently in the demonstration phase; units of 100 MW are expected
before the year 2000 [87].
4.9.2. Experimental / Theoretical Activities
The 115 MW(th) Peach Bottom HTGR operated successfully for seven years until it
was shut down for decommissioning in 1974. It served as an excellent test bed for fuel
development. An extensive R&D program was conducted with the goal of generating real-
time integral data to validate HTGR design methods.
The Fort St. Vrain (FSV) HTGR was designed to produce 842 MW(th) and 330
MW(e). Its operation between 1979 and 1989 provided invaluable fuel performance,
fission product release and plentiful data that have been used for further validation of
respective calculational tools. Fort St. Vrain represents now the world's first successful
decommissioning effort for a long-term operating commercial nuclear reactor.
Within the US HTGR program established by the Gas-Cooled Reactor Associates

(GCRA), a reference conceptual design of an 1170 MW(th) HTGR plant was developed
based on a prismatic core for steam / electricity cogeneration, which is readily adaptable to
various applications, Expected performance parameters for the reference design are coolant
inlet / outlet temperatures of 318 and 686 °C at a system pressure of 7.2 MPa, a net electric
output of 120 MW, and cogenerated process steam of 356 °C at 4.5 MPa with the amount
of 1225 t/h. Multipurpose applications proposed for this design include [43]
• a "Sensible Energy Storage" (SES) system based on a molten salt (sodium nitrate,
potassium nitrate) circuit for thermal energy storage and transport,
• a gas turbine cogeneration system utilizing the higher hot gas temperature of > 850
°C yielding a higher efficiency, and
• a process heat system employing the chemical reforming process with synthetic
fuel or chemical feedstock as the principal output.
90
°2
SEPARATOR
SEPARATOR
(a) PROCESS DIAGRAM (b) MHR APPLICATION
Fig. 4-12: Schematic of hydrogen production by the IS thermochemical cycle using

MHR process heat, from [71]
Feeds Wastes Products
MethanoT>
Electricity^-
I Water makeupN
Fig. 4-13: Schematic of methanol processing based on a process heat MHTGR, from
[63]
91
The MHTGR-GT is considered a second generation meltdown-proof nuclear power
plant combining the experience of both gas-cooled reactor technology and industrial gas-
turbine technology. The core consists of 84 columns of graphitic fuel elements in an annular
arrangement. Helium inlet / outlet temperatures are 590 and 844 °C, and the reactor power
is 450 MW(th) with a projected efficiency of 46 % if optimized for electricity production.
A characteristic feature of the closed-loop Brayton cycle is its potential, besides electricity
generation, for efficiently utilizing waste heat, e.g., for desalination, district heating, or
steam generation.
The use of an MHTGR was studied for methanol synthesis from coal without CC>2
production. Besides coal and steam, the process requires a supplemental hydrocarbon feed
or ideally hydrogen, also methane could be used, plus a non-combustion source of high-
temperature heat which is ideally an HTGR. Based on coal with 60 % carbon content and
a conversion rate of 80 %, a balance of the process was predicted such that with an input
of 162 t/h of coal and 69.4 t/h of methane and an electric power of 300 MW yields 347
t/h of methanol [64].
A follow-up design is given in the GT-MHR (Modular Helium Reactor) (see also
section 4.7.2.) with a higher power output of 600 MW(th). A standard plant is planned
consisting of four of those units. Helium inlet/outlet temperatures are 485 and 850 °C,
respectively. The cycle efficiency is predicted to be 47 % [51]. Follow-on evaluations
which need to be done include the study of transient response of plant components to
normal and off-normal events, impact of turbine contamination, and confirmation of plant
efficiency [47].
The process steam/cogeneration variant of the MHR ("PS/C-MHR") has been studied
at General Atomics for applications in heavy oil recovery, coal conversion, and in steel and
aluminum processes. The process heat variant ("PH-MHR") provides the potential for fuels
production including methane, synthetic gasoline, hydrogen, or methanol. The processes
favored are hydro gasification of coal and methane reforming [72]. Also the employment of
the iodine-sulfur (IS) thermochemical cycle for hydrogen production originally developed
by General Atomics is conceived being applicable using MHR process heat; a respective
diagram is depicted in Fig. 4-12 [71].
The generation of electrolytic hydrogen is achieved by selecting an MHR system

in the electricity generation mode [72]. The two major safety aspects, higher operating
temperatures and tritium permeation to the product gas, were evaluated and shown not to
be a serious concern for the nuclear process heat plant [63].
A schematic of methanol processing on the basis of an MHTGR-PH is shown in

Fig. 4-13, in this case with CH* as supplemental feed. The performance of the process
components is depending on the operating conditions. Thermal efficiencies are expected
in case studies to be up to 61 % [63].
The Advanced Pressurized Reactor AP-600 (Westinghouse) final design certification

by NRC is expected for 1998. A larger unit of an advanced PWR, the System 80+, with
1300 MW(e) is being developed by ABB/Construction Eng. [86]. These designs with
enhanced passive safety features are pursued in the USA to provide additional electricity
and replace retired plants.
92
Also the boiling water reactor is the basis for an innovative design with passive safety
installations, sized approx. 1300 MW(e) for the "evolutionary" type designed by General
Electric in cooperation with Hitachi/Toshiba and approx. 600 MW(e) for the passively safe
reactor. Of the former type, two units have commenced commercial operation in Japan
since 1996 [86].
4.10. VENEZUELA
Venezuela is one of the world's largest energy suppliers, in particular, of extra-

heavy crude oils and bitumens. Continuous steam injection could raise the oil recovery
rates. Studies have been conducted in Venezuela to apply a high-temperature gas-cooled
reactor to the chemical processes of extracting and upgrading the domestic heavy crude oil
resources, of the production of synthetic fuel, and of the gasification of extra heavy oil and
the so-called Orimulsion (mixture of bitumens and water) fuel [18].
[1] AHIMSA, D., Opening Remarks, Non-Electric Applications of Nuclear Energy,

(Advisory Group Meeting, Jakarta, Indonesia, 1995), IAEA-TECDOC-923, Inter-
national Atomic Energy Agency, Vienna (1997) 13-14.
[2] ANON, RI Calls on SE Asia to Give Added Value to Abundant Natural Gas
Supply, World Wide Web, http.7/www.indocon.com/itolnews/past/sept/fri27/hl.htm,
The Indonesia Times (1996).
[3] ANON, People's Republic of China, World Wide Web, http://solstice.crest.org/
sustainable/gem/gemdocs/ecre/asia-pac/china, Internet Information Service of the
Center for Renewable Energy and Sustainable Technology (Solstice) (1997).
[4] ANON, Japan, World Wide Web, http://solstice.crest.org/sustainable/gem/gemdocs/
ecre/asia-pac/japan, Internet Information Service of the Center for Renewable
Energy and Sustainable Technology (Solstice) (1997).
[5] ANON, Chinese Taipei, World Wide Web, http://solstice.crest.org/sustainable/gem/
gemdocs/ecre/asia-pac/taipei, Internet Information Service of the Center for Rene-
wable Energy and Sustainable Technology (Solstice) (1997).
[6] ANON, United States, World Wide Web, http://solstice.crest.org/sustainable/gem/
gemdocs/ecre/asia-pac/usa, Internet Information Service of the Center for Renewa-
ble Energy and Sustainable Technology (Solstice) (1997).
[7] ANON, Der Weltenergieverbrauch steigt und steigt - Foratom: Kernenergie tragt
zur Nachhaltigkeit bei, SVA-Bulletin, No. 14, Schweizerische Vereinigung fiir
Atomenergie, Bern (1997).
[8] BAKER, J.N., EVANS, R.J.D., Initial Studies on the Market Potential for SPFC
and SOFC Systems (Summary Report), Report No. 3719, ETSU F/03/00080/REP,
EA Technology (1996).
[9] BARNERT, H., Anmerkungen zur thermochemischen Produktion von Wasserstoff
aus Wasser mittels Hochtemperaturreaktor-Warme, Report Jill-1660, Research Cen-
ter Jiilich (1980).
93
[10] BARNERT, H., The Status of the High Temperature Reactor and H2 Related R &
D Work, Hydrogen Production, (2nd ffiA Technical Workshop, Jiilich, FRG, 1991),
6, Research Center Jiilich (1991) 113-132.
[11] BARNERT, H., Progress and Safety Aspects in Process heat Utilization from
Nuklear Systems, (3rd JAERI Sem. HTGR Technologies, Tokai, 1994), Proc.
[12] BARNERT, H., KUGELER, K., HTR Plus Modern Turbine Technology for Higher
Efficiencies, (TCM, Beijing, 1995), Design and Development of Gas Cooled Re-
actors with Closed Cycle Gas Turbines, Report IAEA-TECDOC-899, International
Atomic Energy Agency, Vienna (1996) 67-82.
[13] BARNERT, H., Remarks to and New Proposals for the Application of Very CCV
Rich Natural Gas from the Natuna Gas Field in Indonesia, Non-Electric Applications
of Nuclear Energy, (Advisory Group Meeting, Jakarta, Indonesia, 1995), IAEA-
TECDOC-923, International Atomic Energy Agency, Vienna (1997) 337-342.
[14] BARNERT, H., Results of Cost Estimations for the Exploitation of the Natuna Gas
Field in Indonesia, Using the HTR, High Temperature Gas Cooled Reactor (HTGR)
Development, (Techn. Comm. Meeting, Johannesburg, South Africa, 1996), IAEA-
TECDOC in Preparation, International Atomic Energy Agency, Vienna (1997).
[15] BIRNBAUM, U., RIENSCHE, E., STIMMING, U., Optimization of Small SOFC
Plants with Flexible Power/Heat Ratios and their Future Application, (5th Int.
Symp., Aachen, 1997), STIMMING, U., et al. (Eds.), Solid Oxide Fuel Cells
V, Volume 97-40, The Electrochemical Society, Inc., Pennington, USA (1997)
112-123.
[16] BOLIN, B., The Kyoto Negotiations on Climate Change: A Science Perspective,
Science 279 (1998) 187-188.
[17] BREY, L., Report of Travel, Note TA96101034OA, International Atomic Energy
Agency (1996).
[18] CARVAJAL-OSORIO, H., Prospective High-Temperature Nuclear Power Applica-
tions in Venezuela for Heavy Oil Exploitation, (TCM, Oarai, 1992), High Tempe-
rature Applications of Nuclear Energy, Report IAEA-TECDOC-761, International
Atomic Energy Agency, Vienna (1994) 60-66.
[19] COMMITTEE ON USES OF REACTOR HEAT, Heat Application of Light-Water
Reactors and High-Temperature Gas-Cooled Reactors, Report, Japan Atomic Indu-
strial Forum (1981).
[20] FISCHER, U., BOUTEBLLE, F., The EPR Nuclear Island, atw 42 (1997) 757-760.
[21] FOX, M., MULDER, E., SA Pebble Bed Modular Reactor Study in Perspective -
(PBMR-SA Study), Integrator of Systems Technology (1996).
[22] GAO, Z., Auslegungsuntersuchung fur den HTR zur ErschlieBung von Schwerst-
ollagerstatten, Report Jul-1977, Research Center Julich (1985).
[23] GOLOVKO, V.F., et al., State of HTGR Development in Russia, (TCM, Bei-
jing, 1995), Design and Development of Gas Cooled Reactors with Closed Cycle
Gas Turbines, Report IAEA-TECDOC-899, International Atomic Energy Agency,
Vienna ^996) 31-46.
94
[24] GREVEN, R.G., Hydrogen Production Projects of the Ontario Ministry of
Energy, Hydrogen Production, (2nd TEA Technical Workshop, Julich, FRG, 1991),
and Development on the Production of Hydrogen from Water, Document HUF-6,
Research Center Julich (1991) 11-18.
[25] HADA, K., FUJIMOTO, N., SUDO, Y., Design of Steam Reforming Hydrogen
and Methanol Co-Production System to be Connected to the HTTR, (TCM, Oarai,
1992), High Temperature Applications of Nuclear Energy, Report IAEA-TECDOC-
[26] HADA, K., et al., Design of a Steam Reforming System to be Connected to the
HTTR, (3rd JAERI Symp., Oarai, 1996), Proc. JAERI-Conf 96-010, Japan Atomic
Energy Research Institute (1996) 229-240.
[27] HAO, S., Hydrocarbon Steam Reforming Process, Feedstock and Catalysts for Hy-
drogen Production in China, (llth World Hydrogen Energy Conf., Stuttgart, FRG,
1996), VEZIROGLU, T.N., et al., Hydrogen Energy Progress XI, International
[28] HARTH, R., JANSING, W., TEUBNER, H., Experience Gained from the EVA JJ
and KVK Operation, Nucl. Eng. Des. 121 (1990) 173-182.
[29] HARTINIATI, A.S., et al., Prospect of Coal Liquefaction in Indonesia, Non-Electric
Applications of Nuclear Energy, (Advisory Group Meeting, Jakarta, Indonesia,
1995), IAEA-TECDOC-923, International Atomic Energy Agency, Vienna (1997)
37-50.
[30] HINO, R., MIYAMOTO, Y., Hydrogen Production by High-Temperature Elec-
trolysis of Steam, (TCM, Oarai, 1992), High Temperature Applications of Nuclear
Energy, Report IAEA-TECDOC-761, International Atomic Energy Agency, Vienna
(1994) 119-124.
[31] HISHIDA, M., et al., Nuclear Energy Conversion Systems for Arresting Global
Wanning, Advanced Energy Conversion Systems and Related Technologies, (Int.
Symp., Nagoya, 1995).
[32] HOHLEIN, B., Neue Energietrager fur den Verkehr, Monographien des Forschungs-
zentrums Julich, Vol 5, Research Center Julich (1991).
[34] IDE, A., TASAKA, K., Outlook on Heat Utilization and H2 Production Potential
of HTR in Japan, (TCM, Oarai, 1992), High Temperature Applications of Nuclear
Energy, Report IAEA-TECDOC-761, International Atomic Energy Agency, Vienna
(1994) 55-59.
[35] IDE, A., TAKENAKA, Y., MAEDA, S., Utilization of Heat of Modular High
Temperature Gas-cooled Reactors, Proc. ASME-JSME 4th Int. Conf. on Nuclear
Engineering, The American Society of Mechanical Engineers (1996) 255-261.
[36] ISHIYAMA, S., UGACHI, H., ETO, M., The Performance of High Temperature
Chemical Heat Pump System with Metal Hydride Reaction, Hydrogen and Clean
Energy (Int. Symp., Tokyo, 1995), NEDO (1995) 337-340.
[37] JOBSKY, T., Ermittlung von Potentialen zur industriellen ProzeBdampfversorgung
mit nuklearen Anlagen, Report Jul-2411, Research Center Julich (1990).
[38] KIRCHHOFF, R., et al., Operation of a Semi-Technical Pilot Plant for Nuclear
Aided Steam Gasification of Coal, Nucl. Eng. Des. 78 (1984) 233-239.
95
[39] KIRYUSHIN, A.I., et al., Project of the GT-MHR High-Temperature Helium
Reactor with Gas Turbine, Nucl. Eng. Des. 173 (1997) 119-129.
[40] KOMOROWSKI, K., MALINOWSKI, P., Hydrogen Production and Utilization
Programme Sponsored by the German Federal Ministry for Research and Techno-
logy (BMFT), Hydrogen Production, (2nd IEA Technical Workshop, Julich, FRG,
1991), STRUCK, B.D. (Ed.), Implementing Agreement for a Programme of Re-
search and Development on the Production of Hydrogen from Water, Document
HUF-6, Research Center Julich (1991) 19-26.
[41] KONIG, S., BARNERT, H., SINGH, J., Prinzip-Auslegung und sicherheitstechni-
Warme, Internal Report KFA-ISR-IB-5/91, Research Center Julich (1991).
[42] KUPITZ, J., Role of IAEA in Non-Electric Applications of Nuclear Energy, Non-
Electric Applications of Nuclear Energy, (Advisory Group Meeting, Jakarta, Indo-
nesia, 1995), IAEA-TECDOC-923, International Atomic Energy Agency, Vienna
(1997) 119-131.
[43] LAWYER, G.B., DAVIS, D.E., The High Temperature Gas-Cooled Reactor: A
Flexible Energy Option, Alternative Energy Sources, (3rd Int. Conf., Miami Beach,
1980), VEZIROGLU, T.N., (Ed.), Alternative Energy Sources m, Vol. 5: Nuclear
Energy / Synthetic Fuels, Hemisphere Publishing Corp., Washington (1983) 3-21.
[44] LEWKOWICZ, I., CLEVELAND, J., Design and Evaluation of Heat Utilization
Systems for the HTTR through International Cooperation, (3rd JAERI Symp., Oarai,
193-212.
[45] MANTHEY, C., Einsatz von Hochtemperaturreaktoren in der Eisen- und Stah-
lindustrie unter besonderer Beriicksichtigung des Einflusses auf die Standort- und
Umweltprobleme dieses Industriezweiges, Report Jiil-1180, Research Center Julich
(1975).
[46] MASTERS, R.F., Japan: Profile of Japan's Nuclear Power Programme, Nucl. Eng.
Int. 42 (1997) 14-15.
[47] MEARS, L.D., PENFIELD, S.R.JR., Technical and Economical Prospects for the
Gas-Turbine MHTGR, (TCM, Oarai, 1992), High Temperature Applications of
Nuclear Energy, Report IAEA-TECDOC-761, International Atomic Energy Agency,
Vienna (1994) 33-41.
[48] MTYAMOTO, Y., et al., Methanol Synthesis from Recycled Carbon Dioxide and
Hydrogen from High Temperature Steam Electrolysis with the Nuclear Heat of an
HTGR, (TCM, Oarai, 1992), High Temperature Applications of Nuclear Energy,
79-85.
[49] MIYAMOTO, Y., et al., Overview of HTGR Utilization System Developments
at JAERI, High Temperature Gas Cooled Reactor (HTGR) Development, (Techn.
Comm. Meeting, Johannesburg, South Africa, 1996), IAEA-TECDOC in Prepara-
tion, International Atomic Energy Agency, Vienna (1997).
[50] MULLER, H.W., SCHONING, J., The Development and Characteristics of the
German High Temperature Reactor, BNES Journal 12 (1973) 95-105.
[51] NEYLAN, A.J., SIMON, W.A., Status of the GT-MHR, (3rd JAERI Symp., Oarai,
89-96.
96
[52] NICHOLLS, R., Utility Requirements for HTGRs, High Temperature Gas Cooled
Reactor (HTGR) Development, (Techn. Comm. Meeting, Johannesburg, South Af-
rica, 1996), IAEA-TECDOC in Preparation, International Atomic Energy Agency,
Vienna (1997).
[53] NIE8EN, H.F., et al., Erprobung und Versuchsergebnisse des PNP-
Testrohrenspaltofens in der EVA-II-Anlage, Report Jul-2231, Research Center
Julich (1988).
[54] ONUKI, K., et al,, Studies on Water Splitting Hydrogen Production, (3rd JAERI
Symp., Oarai, 1996), Proc. JAERI-Conf 96-010, Japan Atomic Energy Research
Institute (1996) 462-474.
[55] OPITZ, P., Zur Transformation der russischen Energiewirtschaft, in: HAKE, J.-F.,
et al. (Eds.), 3. Ferienkurs Energieforschung, Vol. 3200007, Research Center
Julich (1997) 403-412.
einem modifizierten Hochofengichtgas unter Einsatz von Kemenergie aus dem
Hochtemperaturreaktor, Report Jul-2031, Research Center Julich (1985).
[57] RAHMAN, A., et al., Thermal Enhanced Oil Recovery in Indonesia, Prospect of
HTGR Application, Non-Electric Applications of Nuclear Energy, (Advisory Group
Meeting, Jakarta, Indonesia, 1995), IAEA-TECDOC-923, International Atomic
[58] SATTO, S., Present Status of the HTTR Project at JAERI, (TCM, Oarai, 1992),
High Temperature Applications of Nuclear Energy, Report IAEA-TECDOC-761,
[59] SAKURAI, M., et al., Nuclear Hydrogen Production by Adiabatic UT-3 Ther-
mochemical Process, (llth World Hydrogen Energy Conf., Stuttgart, FRG, 1996),
VEZIROGLU, T.N., et al., Hydrogen Energy Progress XI, International Association
for Hydrogen Energy (1996) 837-842.
[60] SAKURAI, M., et al., Solar UT-3 Thermochemical Cycle for Hydrogen Production,
Solar Energy 57 (1996) 51-58.
[61] SATO, O., Future Perspective of Nuclear Energy Utilization and Expected Role of
HTGR - JAERI's Energy Systems Analysis Research, (3rd JAERI Symp., Oarai,
229-240.
[62] SCHARF, H.-L, SCHRADER, L., TEGGERS, H., Results from the Operation of
a Semitechnical Test Plant for Brown Coal Hydrogasification, Nucl. Eng. Des. 78
(1984) 223-231.
[63] SCHLEICHER, R.W., ENGLER, D., LABAR, M.P., Methanol from Coal without
CO2 Production via the Modular High Temperature Gas-Cooled Reactor, Proc. 54th
American Power Conf. (1992) 490-495.
[64] SCHLEICHER, R.W., LEWIS, A.C.JR., Modular High Temperature Gas-Cooled
Reactor Heat Source for Coal Conversion, (TCM, Oarai, 1992), High Temperature
Applications of Nuclear Energy, Report IAEA-TECDOC-761, International Atomic
[65] SCHOLZ, W.H., Verfahren zur groBtechnischen Erzeugung von Wasserstoff und
ihre Umweltproblematik, Linde Berichte aus Technik und Wissenschaft 67 (1992)
13-21.
97
[66] SCHONING, J., Small and Medium-Sized High-Temperature Reactots for Genera-
tion of Electricity, Process Steam and District Heat, Kerntechnik 52 (1988) 101-106.
[67] SCHRAM, R.P.C., CORDFUNKE, E.H.P., VAN HEEK, A.I., High-Temperature
Reactor Developments in The Netherlands, (3rd JAERI Symp., Oarai, 1996), Proc.
[68] SCHULTEN, R., et al., Industriekernkraftwerk mit Hochtemperaturreaktor PR 500 -
"OTTO-Prinzip" - zur Erzeugung von ProzeBdampf, Report Jul-941-RG, Research
[69] SCHULTEN, R., et al., The Pebble-Bed High-Temperature Reactor as a Source of
Nuclear Process Heat, Vol 1: Conceptual Design, Report Jul-1113-RG, Research
[70] SEMKE, S., Energieforschung in Deutschland, in: HAKE, J.-F., et al. (Eds.), 2.
Ferienkurs Energieforschung, Konferenzen des Forschungszentrums Julich, Vol 20,
Part H, Research Center Julich (1996) 919-940.
[71] SHENOY, A., Hydrogen Production Using Modular Helium Reactor, (30th Inter-
society Energy Conversion Eng. Conf., Orlando, 1995).
[72] SHENOY, A., Modular Helium Reactor for Non-Electric Applications, Non-Electric
Applications of Nuclear Energy, (Advisory Group Meeting, Jakarta, Indonesia,
1995), IAEA-TECDOC-923, International Atomic Energy Agency, Vienna (1997)
253-277.
[73] SHIMIZU, S., ONUKI, K., NAKAJIMA, H., Bench-Scale Studies of the Iodine-
Sulfur Process, Hydrogen Production, (2nd IEA Technical Workshop, Julich, FRG,
[74] SHIMIZU, S., NAKAJIMA, H., ONUKI, K., A Progress Report on Bench Scale
Studies of the Iodine-Sulfur Process for Thermochemical Hydrogen Production,
(TCM, Oarai, 1992), High Temperature Applications of Nuclear Energy, Report
IAEA-TECDOC-761, International Atomic Energy Agency, Vienna (1994) 114-
119.
[75] SHIMIZU, S., et al., Closed-Cycle Continuous Hydrogen Production Test by
Thermochemical IS Process, Proc. on Chemical Engineering, 24 (1998) No. 2
352-355 (in Japanese).
[76] SIEMENS/INTERATOM, Hochtemperaturreaktor-Modul-Kraftwerksanlage, Safety
Report, Siemens/Interatom (1988).
stoffeinsatz", TAB Working Report No. 13, Euro fur Technikfolgen-Abschatzung
[78] STREICHER, R., REJMERT, R., GroBtechnische Wasserstofferzeugung - Stand der
technik und Entwicklungstendenzen, Kernenergie und Wasserstoff - ein Energie-
system mit Zukunft?, (Top. Meeting, Travemiinde, 1988), Deutsches Atomforum
e.V., Bonn (1988).
[79] SUN, Y., et al., Present Status of the HTR-10 Project, (3rd JAERI Symp., Oarai,
39-49.
98
[80] SUN, Y., et al., The HTR-10 Test Reactor Project and Potential Use of HTGR for
Non-Electric Application in China, Non-Electric Applications of Nuclear Energy,
(Advisory Group Meeting, Jakarta, Indonesia, 1995), IAEA-TECDOC-923, Inter-
national Atomic Energy Agency, Vienna (1997) 211-222.
[81] SUN, Y., Institute for Nuclear Energy Technology, Tsinghua University, Beijing,
Personal Communication (1997).
[82] TANAKA, T., et al., Construction of the HTTR and its Testing Program for
Advanced HTGR Development, (TCM, Beijing, 1995), Design and Development
of Gas Cooled Reactors with Closed Cycle Gas Turbines, Report LAEA-TECDOC-
[83] TEUCHERT, E., et al., OTTO-Kugelhaufenreaktor fur eine 1000 MWe-
Heliumturbinenanlage, Report Jul-1070-RG, Research'Center Jiilich (1974).
[84] THEENHAUS, R., HAKE, J.-F., Energie: Grundlage fur zukiinftige Enrwicklun-
gen, in: HAKE, J.-F., et al. (Eds.), 2. Ferienkurs Energieforschung, Konferenzen
des Forschungszentrums Julich, Vol 20, Part I, Research Center Julich (1996) 1-12.
[85] TONOGOUCHI, M., et al., Plant Concept of Heat Utilization of High Temperature
Gas-Cooled Reactors, (3rd JAERI Symp., Oarai, 1996), Proc. JAERI-Conf 96-010,
Japan Atomic Energy Research Institute (1996) 429-434.
[86] UIC, Advanced Reactors, Nuclear Issues Briefing Paper 16, World Wide Web,
http://www.uic.com.au/nipl6.htm, Uranium Information Centre, Melbourne (1997).
NATIONAL ENERGY AGENCY, Comparing Energy Technologies, OECD/IEA
(1996) 237-296.
[88] VDI, AVR - Experimental High-Temperature Reactor: 21 Years of Successful
Operation for a Future Energy Technology, Association of German Engineers,
VDI-Verlag, Dusseldorf (1990).
[89] WANG, D., et al., High Temperature Process Heat Application of Nuclear Energy
in China, (TCM, Oarai, 1992), High Temperature Applications of Nuclear Energy,
42-46.
[90] YOSHIDA, K., Hydrogen Production by UT-3 Thermochemical Water Decomposi-
tion Cycle, Hydrogen and Clean Energy (Int. Symp., Tokyo, 1995), NEDO (1995)
39-46.
[91] ZHA, K., Energie-Entwicklungspolitik in China unter besonderer BerQcksichtigung
der Elektrizitatswirtschaft, Elektrizitatswirtschaft 94 (1995) 1170-1179.
[92] ZHANG, W., Entwicklung der Kemenergie im asiatischen Raum, speziell in China,
in: HAKE, J.-F., et al. (Eds.), 2. Ferienkurs Energieforschung, Konferenzen des
Forschungszentrums Julich, Vol 20, Part I, Research Center Julich (1996) 893-917.
NEXT PAGE(S)
left BLANK
99
Chapter 5
CONVENTIONAL AND ADVANCED

HYDROGEN PRODUCTION METHODS
The most obvious pathways for the production of hydrogen from different primary
energy sources are shown in Fig. 5-1. The efficiency of a conversion method is strongly
dependent on the level of temperature at which the respective chemical reactions take place.
The world hydrogen production today is 500* 109 Nm3 which corresponds to 2 % of
the world's primary energy demand. Most part of it, 97 %, is produced from fossil fuels.
5.1. DECOMPOSITION AND GASIFICATION OF FOSSIL FUELS
5.1.1. Conversion of Natural Gas
5.1.1.1. Steam Reforming
The most widely used method of hydrogen generation is steam reforming of na-
tural gas, which offers many advantages: Natural gas is available in large amounts; its
commercial production in 1992 winded up to 2.1*1012 m3. It is clean and can relatively
easily be converted to hydrogen.
Primary energy sources Secondary energy sources Chemical energy source

Nuclear energy
- partial oxidation—.
Coal gasification
COK6
. refining .
Petroleum heavy hydrocarbons synthesis gas
light hydrocarbons
purification, or
Natural gas refinery gases
ammonia
fermentation. biogas l_ — reforming . synthesis/cracking,
Biomass or methanol
1 metabolism.
synthesis/cracking
or reforming
[ Waste materialsj • thermoelectric cycles.
Geothermal water electrolysis.
HYDROGEN
Solar ——— photovoltaic
i electric power
direct turbine
Hydro power generator sets
Wind power
Fig. 5-1: Sources and production of hydrogen, from [21]
101
During the steam reforming process, natural gas is mixed with steam to react in a
wide range of high temperatures, typically > 500 °C up to 950 °C, in the presence of a
nickel catalyst. The steam reforming reactions are:
CH4 + H2O <—> CO + 3 H2 - 206 kJ/mol

CH4 + 2 H2O <—> CO2 + 4 H2 - 165 kJ/mol
or for light hydrocarbons in general:
CnHm + n H2O <——> n CO + (1/2 m + n) H2
At a stoichiometric ratio, the conversion rate to the right-hand side is maximal for
high temperatures and relatively low pressures. However, excess steam of 300 % away
from the stoichiometric mixture is generally injected to move the equilibrium towards more
CO2 in the homogeneous water-gas shift reaction
CO + H2O <——> CO2 + H2 + 41 kJ/mol
in order to achieve a higher H2 yield and to preclude carbon deposition due to the
Boudouard reaction which is also catalyzed by nickel:
2 CO <—» CO2 + C + 172 kJ/mol
The yield of the reforming process is a mixture of H2, CO, CO2, plus residual steam
and still unreformed methane, the so-called reformer gas. The equilibrium composition of
the gases depends on the process parameters temperature and pressure which are chosen
according to the desired subsequent synthesis gas applications. The example shown in Fig.
5-2 is for a 4 MPa system pressure and a water to methane ratio of 2. The mixture of
H2 and CO left after purification is what is called synthesis gas which is an important
feedstock in the chemical industry.
If the reforming reaction is carried out at additional presence of a selected adsorbent

for CO2, the reaction temperature is significantly lowered down to 300 - 500 °C while
achieving the same conversion of methane to hydrogen plus the purity of the hydrogen
is higher. This new process, sorption-enhanced reaction (SER), reduces the traditional
downstream hydrogen purification step and could make small plants more economic [1].
5.1.1.2. CO? Reforming
If the steam is completely or partly replaced by carbon dioxide, the result is another
H2 to CO ratio, shifted towards a more CO-rich synthesis gas. The catalytic reforming of
CH4 with CO2, first proposed in 1928 by Fischer and Tropsch, involves an environmental
benefit, because two greenhouse gases are combined here resulting in a product gas which
102
0
300 600 900 °C 1200
Temperature —*
Fig. 5-2: Equilibrium composition of gases involved in the methane reforming

process as a function of temperature, from [16]
might be more favorable for certain applications, e.g., the synthesis of valuable oxygenated
chemicals [68].
CH4 + CO2 2 CO + 2 H2 - 247 kJ/mol
Side reactions are the reverse water gas shift reaction, the Boudouard reaction and
methane cracking, where the former two are suppressed at temperatures > 820 °C and the
latter two form carbon in the temperature range 550 - 700 °C [68].
However, currently no industrial technology for the CO2 reforming of methane is

established. The reasons for this are the rapid deactivation of conventional catalysts if used
without the presence of steam, and the relatively high soot formation [68].
5.1.1.3. Partial Oxidation
Partial oxidation is able to convert methane and other hydrocarbons heavier than
naphta, catalyzed or non-catalyzed, where steam reforming is not applicable:
CH4 + 1/2 02 CO + 2 H2 + 71 kJ/mol
103
or for hydrocarbons in general:
CnHm + 1/2 n O2 «—» n CO + 1/2 m H2
The catalytic process works at about 600 °C using methane or naphta feedstock.
The non-catalytic process works at temperatures between 1100 and 1500 °C and in the
pressure range 2-6 MPa using hydrocarbons from methane to heavy oil and coal. Pure
oxygen from a preceding air decomposition step is taken because of the difficulty to
clean the hydrogen from the nitrogen. Immediate feeding of water prevents undesired
reverse reactions (quenching). Despite its lower efficiency of approx. 50 % compared with
steam reforming, partial oxidation provides, on the other hand, exothermicity and a greater
selectivity for synthesis gas production as well as advantages for certain applications such
as compactness, rapid start-up or load change, and lower overall cost [21, 68].
If steam is added to the fuel and the oxidant, it is possible to heat balance the
exothermal partial oxidation reaction with the endothermal reforming reaction. The reaction
is then said to be auto thermal, meaning that no external heat source is required what
raises the efficiency.
A simulation model is currently applied to investigate the partial oxidation of methane

in a fluidized bed of catalyst particles with some tenths of a millimeter diameter, rather than
in a solid bed. This procedure is expected to provide a much better heat exchange. Total
cost reduction is predicted to be about 20 %, which, however, still needs verification [72].
5.1.1.4. Reduction of Metal Oxides
In the temperature range 900 - 1100 °C, gallium (also tungsten) oxide is reduced to
the metal by use of methane:
3 CH4 + Ga2O3 «—> 2 Ga + 3 CO + 6 H2
with the gallium oxide recycled in an oxygen-rich atmosphere. The problem in this
reaction is that the activity of the metal decreases over repeated oxidation / reduction
cycles [21].
5.1.1.5. Thermal Cracking
An alternative to the steam reforming of methane is single-step thermocatalytic

decomposition or "thermal cracking". Methane and other hydrocarbons decompose at
temperatures between 700 and 980 °C and, if no air is present, form hydrogen and carbon:
CH4 <—— C + 2 H2 - 75 kJ/mol
This process has been practiced over many years for the production of carbon black
for rubber tire vulcanization and for the printing industry. Despite the advantage that no
104
carbon dioxides are formed, it is often considered an unwanted reaction because of the
formation of solid carbon [21].
Cracking was studied for various catalysts in both continuous flow and pulse catalytic
reactors. Variations in the operational parameters were found to produce CH4 - H2 mixtures
with a wide range of hydrogen concentrations, from 30 to 98 vol% [39].
For the thermal cracking, a tandem furnace can be operated where a methane-air
flame is used to heatup firebrick to w 1400 °C. The air is then turned off and the methane
decomposes on the hot brick until the temperature drops below 800 °C. The H2 enriched
effluent gas stream is taken to heatup the second furnace. A continuous process is thought of
as being possible also, but has not yet been practiced on a commercial scale [55]. A small-
scale catalytic propane cracker has been realized at the Duisburg University, Germany (see
section 9.4.).
The Kvaerner process is the non-catalized thermal decomposition of hydrocarbons

with no emission of CO2. Pyrolysis is done by means of an electrically heated plasma
burner. Products are high-purity hydrogen plus carbon black, also an important industrial
product. The specific energy demand is with 1.01 kWh/Nm3 of hydrogen about a quarter of
that of an advanced electrolyzer. A pilot plant based on the Kvaerner process is currently
operated with a capacity of 8 million Nm3 of H2 and 2000 t of soot [60]. The construction
of a larger facility in Canada is planned. It will be fueled by natural gas and optionally
oil providing carbon black in the amount of 20,000 (later 90,000) t/yr and hydrogen in the
amount of 50 (later 225) million Nm3 [8].
High-temperature, non-catalytic reforming of hydrocarbons using plasma technology

is investigated in the USA at the MIT Plasma Fusion Center. Plasma reforming is flexible
in fuel because it consumes any hydrocarbon including heavy fractions. The fast kinetic
rates can greatly reduce volume and weight of the reformer. An experimental facility was
constructed and initial tests have been performed [18].
5.1.1.6. Methanol Dissociation
Methanol is endothermally cracked over a catalyst to produce synthesis gas plus traces
of methanol, ether, and methane. A two-stage membrane separation system extracts the
hydrogen from a CO-rich fuel that fires the cracker (see Fig. 5-3).
CH3OH -+ CO + 2 H2 - 91 kJ/mol
The catalytic decomposition of methanol by steam reforming as taking place in

methanol fuel cells at temperatures > 250 °C is described by the reaction:
CH3OH -i- H2O -» CO2 + 3 H2 - 50 kJ/mol
A water to methanol ratio of greater than 1 plus considering that also the methanol
cracking process (see first reaction equation) occurs, the only significant products are H2,
CO2, and CO.
105
EXHAUST
j j H2, CO, etc ^ 90+ %JH MEMBRANE

SEPARATORS
CRACKER WITH BURNER,

-^TTTT—— r^rr-.——...................
HEAT EXCHANGER
AND CATALYST
UJ UJ
HIGH CO
?
co 9S+%H 2 OUT
PRESSURE V*
^f
PUMP
r
LIQUID'
°
O 1kBACK PRESSURE
METHANOL IN
II .^ ...... . .
CO-RICH CRACKER FUEL
REGULATORS
A'R
Fig. 5-3- Thermal cracking of methanol with a membrane separation system, from
[34]
CAJR8ON D1OX3OE
CARBON DIOXIDE PSA
Fig. 5-4: Process flow sheet of hydrogen production from steam reforming of natural
gas, from [BOC Gases]
106
5.1.1.7. Steam Reforming Process Technology
In the conventional steam reforming process (Fig. 5-4), the feed gas passes a
desulfurization step to protect the catalyst against poisoning and is then routed to the
steam reforming tubes which are externally heated mainly by heat radiation. The typical
reforming temperature is 750 - 850 °C and the reforming pressure is 2 - 3 MPa depending
on the application of the product gas. In the conventional process, the required energy is
supplied by the combustion of light hydrocarbons. Combustion gases at a temperature of
1500 °C flow downwards through the steam reformer in parallel with the feed gas flow for
the reforming process. Large amounts of heat are transferred in the upper section of the
tube, but heat transfer decreases by about one order of magnitude from top to bottom. The
average heat flux density is approx. 80 kW/m2.
A typical reforming fraction in practice is 80 %. The CO in the gas mixture is

converted after cooling in the shift reactor to CO2- The gas mixture is then purified by
passing special adsorber beds to obtain pure (> 99.99 %) hydrogen [38].
Steam reforming is a mature and well-established technology practiced on an indu-

strial scale. Today's reformer plants for hydrogen production vary in size, from typically
around 100 Nm3/h capacity up to the largest currently existing with a maximum H2 pro-
duction rate of 140,000 Nm3/h. Large-scale reformers could be used for H2 supply to fuel
cell applications at the 100 MW scale. Recent reformer improvement developments have
focused on high efficiency, good load changing capability, and low cost through the use
of cheaper materials and production methods [21].
A dominant problem for all synthesis gas processes is the metal-dusting corrosion
(MDC) phenomenon. Further improvement of the reforming technology requires a mini-
mization of MDC. Aging of the alloys and defects in the metal oxide layer as well as the
gas atmosphere might explain this corrosion effect. A possibility for MDC prevention is
the displacement of CO by a purge gas [28].
Membrane reformers take advantage of a useful feature of hydrogen, namely its

ability to selectively permeate through Pd or Pd alloy membranes. As has been shown on
a laboratory scale, the hydrogen is relatively clean and its continuous removal increases
the methane conversion level [21].
Although small-scale steam reformers are commercially available today, relatively

few small units have been constructed. Small size and suitability on a small scale makes
them beneficial for integration into a fuel cell power plant Fuel cell manufacturers are
developing advanced miniaturized designs which are more compact for mobile applications.
A 50 kW multi-fuel partial oxidation processor coupled to an SPFC stack has been

presented in October 1997 by Arthur D. Little Inc., USA, and test-operated in the meantime
for more than 3000 h. The fuels that can be applied are gasoline, methanol, and ethanol; in
a later stage, also diesel, oil, methane and propane processing will be possible [27]. In the
UK, a bench-scale steam reformer system processing gasoline and diesel was successfully
demonstrated for over 50 hours each. The H2 concentration in the reformate was typically
107
- Gas inlet
Insulation
A
Gas outlet
Window v Absorber
Fig. 5-5: Schematic of a directly absorbing receiver reactor, from [60]
76 %, the CO concentration less than 0.5 %. Both fuel processor developments are to be
employed in fuel cell vehicles [7].
High temperature fuel cells (MCFC and SOFC) allow for the reforming process to
take place inside the fuel cell stack which lowers the requirement for cell cooling and
reduces cost due to absence of the external reformer vessel. A future SOFC application
could also be the production of hydrogen (and electricity) by internal reforming of natural
gas where more H2 as a component of the synthesis gas is produced than can be converted
electrochemically into electricity; the heat losses from the fuel cell operation would be used
as the endothermal heat source for the reforming step [65].
A solar methane reforming process has been experimentally investigated in a joint

German/Spanish project ASTERIX (Advanced Steam Reforming in Heat Exchange). Pri-
mary energy is solar at a temperature of 1000 °C. The convectively heated tube bundle
reactor has a power of 80 - 170 kW. ASTERIX is designed for a product gas output of
115 Nm3/h [60].
Other solar methane reforming projects are CAESAR (Catalytically enhanced Solar
Absorption Receiver) in Germany with COi reforming and SCR (Solar Chemical Reactor)
in Israel. The latter is a directly absorbing receiver reactor (Fig. 5-5) with a power of 200
- 300 kW. The absorber structure allows an almost homogeneous distribution of solar flux
and thus temperature. Conversion rates of 80 % have been achieved. Advantages are high
heating rate and energy density, also the possibility to combine the external solar reformer
gas to run a gas turbine, thus saving fuel up to 26 % [60].
108
5.1.2. Conversion of Coal
The most abundant fossil fuel on earth is coal. Its conversion to liquid and gaseous
hydrocarbons has been commercially employed worldwide. In the first half of this century,
it was the principal method of hydrogen production in the coke furnace process. The gas
production from coal besides the anaerobic combustion process is realized by means of
a gasification medium which reacts with the coal at temperatures > 800 °C. All organic
constituents of the coal will be converted with long enough residence times. The gasification
medium used is either steam ("steam-coal gasification") or hydrogen ("hydro-gasification").
If air or oxygen is injected into the gasifier, part of the coal is burnt directly leading to
an autothermal reaction.
5.1.2.1. Steam-Coal Gasification
During steam-coal gasification, two consecutive chemical processes take place [30,
50]. The first process, the pyrolysis reaction
CHxOy <—» (1-y) C + y CO + x/2 H2

CHxOy <—> (l-y-x/8) C + y CO + x/4 H2 + x/8 CH4
is a very rapid expelling of all volatile constituents of the coal. In the subsequent,
much slower step, steam of 600 - 1000 °C is needed to convert the residual organic solids
in the heterogeneous water gas reaction:
C + H2O *—» CO + H2 - 163 kJ/mol
with further increase of the H2 yield in the shift reaction. The thermodynamic
equilibrium composition of the product gas mixture is a function of temperature and
pressure. As can be drawn from Fig. 5-6, the production of hydrogen and carbon monoxide
is optimal at high temperatures and low pressures. The heat of the hot gas is recuperated
for use in the production of high-pressure steam; the heat withdrawal must be quick to
avoid reverse chemical reactions.
Despite its small hydrogen contents, coal gasification with oxygen and steam is used
to generate about 18 % of the world hydrogen production. Also the production of methane
or methanol or gasoline from the coal conversion step by modifying downstream treatment
is conventional chemical engineering. Pulverized coal is rapidly partially oxidized by
oxygen and steam in a fluidized bed at about atmospheric pressure. 30 - 40 % of the coal
are transformed into CO2 to supply splitting energy of water. The reaction rate strongly
increases with temperature. Temperatures up to 2000 °C and pressures up to 3 MPa are
typically chosen.
Pertinent criteria for applicability and economy of steam-coal gasification are the
characteristics of the coal to be gasified. The (geologically) older the coal, the smaller is
its reactivity and the higher is the temperature required [30]. Main disadvantages of coal
gasification are the handling of solid material streams, which is generally more difficult
109
01 MPa 4MPa
100 100-
0
600 700 800 900 1000 °C 1200 600 700 800 900 1000 °C 1200
Temperature ——>• Temperature ——*
Fig. 5-6: Equilibrium composition of the product gas after steam-coal gasification as
a function of temperature and (left) 0.1 MPa, (right) 4 MPa, from [30]
and more expensive than gaseous or liquid streams, and the large amounts of CC>2, SC>2,
and ash produced during the process.
Different types of autothermal coal gasification processes used on a large scale are
existent as shown in Fig. 5-7, Lurgi, Winkler, Koppers-Totzek, and their modifications,
respectively. Their characteristic features are summarized in Table 5-1.
The TEXACO gasification process operates at high pressures (« 5.5 MPa) and high
steam contents. The synthesis gas typically contains 34 % H2 and 48 % CO. Hydrogen
of > 97 % purity is yielded at a pressure of 4 MPa to reduce energy consumption during
H2 pressurization [55].
At present, coal gasification is primarily used for ammonia synthesis in the fertilizer
industry and for synthesis gas production to be employed in the hydrocarbon and methanol
synthesis with large-scale production facilities, in particular, in the developing countries.
Hydrogen production is of minor importance.
The option of integrating the steam-coal gasification process in a combined heat

and power facility with a gas turbine step preceding the water/steam process, is currently
considered the cleanest and most efficient (34 —»• 38 %) coal-fueled technique so far, and
the combined-cycle portion could be switched to different fuels, e.g., natural gas or oil.
The Integrated Gasification Combined Cycle (IGCC) technology with its intermediate
stage of synthesis gas allows the removal of most carbon components before combustion.
The separated COi stream is of high purity and therefore suited for disposal. Adopting
improvements in the turbine system, the overall efficiency is anticipated to be raised to 52
% [64]. Respective facilities have been realized so far in Germany and the United States
110
Table 5-1: Characteristic features of the steam-coal gasification processes after Lurgi,
Winkler, Koppers-Totzek, from [30]
Lurgi Winkler Koppers-Totzek

Reactor Solid bed Fluidized bed Flue stream
Grain size [mm] 10-30 1 - 10 <0.1
Steam over oygen
9 -5 2.5- 1 0.5 - 0.02
ratio
Movement of fuel countercurrent flow vortex cocurrent flow cocurrent flow
Residence time of
60-90 15 -60 <0.02
fuel [min]
Requirements to must not bake high reactivity, melting point of ash
fuel or decay must not decay < 1450 °C
Maximum gas
outlet temperature 370 - 600 800 - 950 1400 - 1600
PC]
Pressure [MPa] 2 -3 atmospheric atmospheric
Composition of
product gas [vol%]
CO + H2 62 84 60 + 29
CH4 12 2 0.1
tar, oil, phenols,
By-products none none
gasoline, waste water
Winkler Koppers-Totzek
Product gas
Vapor
Product
gas
Vapor —*•
+ Oxygen Vapor —*•
+ Oxygen
Solid bed Fluidized bed Flue stream
Fig. 5-7: Gas generator types for steam-coal gasification according to Lurgi (left),
Winkler (middle), and Koppers-Totzek (right), from [66]
111
[30]. Some 4000 MW(e) of these IGCC power systems are planned for installation in the
USA, Europe, and Asia [54]. The world's largest IGCC plant with 300 MW(e) is almost
completed in Spain with an expected efficiency of 45 %.
A 300 MW IGCC power plant with electricity and methanol cogeneration has been
proposed where 88 % of the carbon introduced by the coal can be extracted as CC>2
with an acceptable amount of energy and investment cost which is partially utilized in
combination with H2 from an external carbon-free source for methanol production. The
overall balance for a 354 MW power plant (234 MW for the gas turbine and 120 MW for
the steam generator) is an input of 2300 t of coal, 780 t of hydrogen, and 5500 t of CC>2
(as intermediate product) per day to achieve a daily output of 3800 t of methanol and a
total net power output of 310 MW [44].
At a very early R&D level is the use of coal gasification integrated in a high-
temperature fuel cell as a stationary power system (see section 7.2.2.3.).
5.1.2.2. Hydro-Gasification
In the hydro-gasification process, coal is converted to synthetic natural gas in a

fluidized bed at temperatures of around 800 °C in an exothermal reaction:
C + 2 H2 -» CH4 + 86 kJ/mol
Parallel reactions are the steam reforming and the water-gas shift reactions to produce
synthesis gas. The feed gas hydrogen could be taken either from a partial conversion of
methane from the product gas in a steam reformer or from a conversion of the residual char
with oxygen and steam in a high temperature Winkler process. A high gasification degree
can be reached already with relatively short residence times (9 - 80 min) [50]. In order to
obtain a high conversion rate of coal, the CHU fraction in the product gas of the reforming
step (which will then be the feed gas for the hydro-gasification) should be no higher than
5 %, which requires a low-temperature separation step. Both reactions are carried out at
system pressures of 4 MPa. The amount of 1.5 Nm3 of hydrogen is required to produce
1 Nm3 of synthetic natural gas [32].
After cleaning and separation, the methane is either utilized as a final product or
fed into the reformer while the product hydrogen is available after the reforming process
in the synthesis gas, subtracted by the portion that is required for the gasification step.
The advantage of hydro-gasification compared with steam-coal gasification is its 200 K
lower pre-heating temperature which reduces potential corrosive attack. A major drawback,
however, is the relatively large amount of residual coke of up to 40 % [13]. Standardized
reformer equipment for power industries is available.
5.1.2.3. Steam-Iron Process
The steam-iron process is another old hydrogen generation process. Although based
on coal, it is actually a cycle process where hydrogen is generated from the decomposition
of steam by reacting with iron oxide. The cycle, however, is not completely closed, since
112
coal is consumed and CO2 is emitted. The synthesis gas produced in the coal gasification
step with steam is then reacted with iron oxide to generate reduced forms of iron oxides:
Fe3O4 + H2 -» 3 FeO + H2O

Fe3O4 + CO -» 3 FeO + CO2
FeO + H2 -* Fe + H2O
FeO + CO -» Fe + CO2
In the following step, the reduced species are re-oxidized with water to form the
original oxides plus a hydrogen enriched gas:
Fe + H2O -» FeO + H2
3 FeO + H2O -+ Fe3O4 + H2 815 - 870 °C
Reduction and oxidation step in two separate reactors allow for a continuous hydrogen
production. A major drawback is the small conversion rate of 60 % of the synthesis gas in
the reduction step. The effluent from the steam-iron reactor contains 37 % H2 plus 61 %
steam and 96 % H2, respectively, if the steam is condensed. The remaining heating value
plus sensible heat at 825 °C, however, can be used to cogenerate electricity. With a plant
capacity of 110,000 Nm3/h of H2, the byproduct electric power is 158 MW [55].
5.1.2.4. Gasification in a Molten-Iron Process
Temperatures of around 1400 °C in the molten-iron process, MIP, are capable of

cracking almost any chemical compound, in particular that of coal. Product gases are
CO and H2. Other constituents are remaining as slag in the iron bath. Process heat is
provided by the exothermal partial oxidation of C to CO. The MIP process is still in the
development stage.
5.1.2.5. Coal Cracking Process
The HYDROCARB coal cracking is an advanced process where natural gas is

synthesized from the hydrogenation of the carbon containing raw material. Coal is routed
to a thermal cracker where it is decomposed to carbon black as a clean fuel and hydrogen
as a byproduct fuel. HYDROCARB represents the possibility of a CO2-free fossil-based
hydrogen economy. The reaction in the hydropyrolyzer operating on bituminous coal is:
CHo.gOo.o8So.oi6No.oi5 + -» C + 0.32 H2 + 0.08 H2O + 0.008 N2 +

ash + CaCO3 ash + 0.016 CaS
The HYDROCARB process is expected to become highly competitive with all other
coal-based processes [55].
113
5.1.2.6. Byproduct from Coke Production
Coke furnace gas was formerly the major source for hydrogen until replaced by
more efficient petroleum-based processes and by coal gasification. However, it is still
an important and convenient method to obtain hydrogen on places where economically
attractive.
CHo.gOo.2 -» 0.8 C + 0.2 CO + 0.4 H2
Modem coke furnaces produce about 350 m3 of coke furnace gas out of 1 ton of coal.
Half of it is used to run the furnace at 750 - 850 °C and atmospheric pressure, the other
half is processed for hydrogen extraction by means of either low temperature condensation
or by pressure swing adsorption.
From the worldwide coke production from hard coal, an estimated (theoretical) 60
billion Nm3 of hydrogen could be recovered [14].
5.13. Conversion of Oil
5.1.3.1. Partial Oxidation of Heavy Oils
The partial oxidation of heavy oil, in particular the remainder from crude oil
distillation, is one of the most wide spread methods of hydrogen production. It involves
the conversion of steam, oxygen, and hydrocarbons to hydrogen and carbon oxides with
or without catalyst The catalytic process appropriate for light hydrocarbons occurs at
about 590 °C. The non-catalytic process includes also heavy oils and coal and works at
temperatures of 1150 - 1315 °C and pressures of 1 - 10 MPa. Technological processes of
oil gasification developed by Shell and Texaco operate at 8 MPa. The following reactions
take place in a mixture of oil (residual oil: n=l, m=1.3; coal: n=l, m=0.8), oxygen and
steam [55]:
CnHm + n/2 O2 -+ n CO + m/2 H2

CnHm + n H2O -» n CO + (n+m)/2 H2
CO + H2O *—> CO2 + H2
The partial oxidation delivers the energy for the steam splitting. The resulting
synthesis gas with a typical composition of 46 % H2, 46 % CO, 6 % CO2, 1 % CH4,
1 % N2 + Ar at 6 MPa can be either used for methanol synthesis or is further catalytically
converted to increase the H2 yield. Purification is done in several steps. Maximum unit
capacity is in the order of 80,000 Nm3/h of H2 [55].
5.1.3.2. Hydrocracking of Heavy Oil
Hydrocracking plants are used to convert heavy oils into lighter oil and gasoline
(naphta) fractions. This process requires an input of hydrogen and heat where the
former could be provided by naphta splitting. Different properties of raw materials
114
imply the selection of different process conditions. Hydrocracking typically involves
feedstock preparation by hydro-treating to remove metals, sulfur, nitrogen, and oxygen.
The conversion process of the heavy feedstock is made in several hydrocracking stages
which are high-pressure (10 MPa) catalytic processes in the temperature range of 320 -
400 °C. The mass balance is:
1 kg heavy oil + 0.494 Nm3 H2 -* 4.22 Nm3 H2 + 0.041 kg H2S

+ 34.4 MJ heat +0.104 kg residues + 10.7 MJ heat
resulting in a net hydrogen production of 3.7 Nm3 H2 per kg heavy oil [35].
Partial oxidation of heavy oil is occasionally practiced on a large scale (> 55,000
3
Nm /h of H2). Heavy oil is reacted with pure oxygen in a less than stoichiometric ratio.
The product gas of CO and H2 at temperatures of 1200 - 1350 °C is cooled and then
subjected to the shift reaction to obtain pure hydrogen [37].
5.1.4. Conversion of Other Feedstocks
Thermal processes can also be applied to feedstocks or fuels others than natural gas
and coal for hydrogen production. These include municipal solid waste and biomass
such as peat, wood, gases, and agricultural residues. Its gasification leads to a low calory
gas which is less convenient for H2 production and better suited for heat or electricity
production. Biomass acquires its H-atoms from water (with the release of O). In order
to produce methanol with a higher content of H-atoms per C-atom, the input of energy is
required. The amount of hydrogen produced during the conversion step depends on type
of fuel/feedstock, the availability of steam and oxygen, and the temperature of the reaction.
Besides the opportunity of a renewable clean energy supply, this process is CO2-neutral.
It is attractive for decentralized application.
0.5 H2O + 0.3 O2 «—> CO2 + 0.9 H2 + 92 kJ/mol
The autothermal gasification of biomass in a fluidized bed results in a synthesis

gas with about 30 % H2, 30 % CO2, 30 % CO, and 5 - 10 % CH4 plus some higher
hydrocarbons. For methanol synthesis, either H2 needs to be added or CO converted in
the shift reaction to CO2 plus H2 with separation of the CO2 portion. The latter method is
usually applied. A catalytically supported gasification of biomass (wood) has been reported
to work in a fluidized bed at 650 °C. Experiments on a laboratory scale have shown, that
80 - 90 % of the feedstock were converted to synthesis gas consisting by 85 vol% of CO
and H2 [9]. Commercial plants for wood or waste biomass conversion are available already
applying autothermal processes and using air instead of oxygen [49]. Other feedstocks like
com pulp, glucose or sugar cane juices were also found in experiments to produce hydrogen
[48]. Under specific physical and chemical conditions, the H2 or CH4 concentration in the
tail gas is sufficiently high for integration of a hydrogen or methane powered fuel cell plant
into the gasification process.
A first-of-its-kind European power plant is planned in Bad Brflckenau, Germany, to
start its operation in 1999. The conversion of 2000 - 3000 t/yr of biomass to hydrogen-rich
115
gas will allow an output of 600 kW of electricity from H2 consuming fuel cells plus 800
kW of heat from combustion of the residual gas in an engine. The overall efficiency is
expected to range between 50 and 60 % [6].
Methanol generation from biomass gasification is the heart of the Swiss project
BIOMETH to be later extended by the introduction of electrolytic hydrogen for a complete
conversion of the carbon oxides [56]. Gasification of biomass, where the heat for the
endothermal reaction is externally provided, results in high heat value and high hydrogen
contents of the product gas. It has been realized on a laboratory scale with a capacity
of 2 kg biomass throughput per hour [31]. The Paul Scherrer Institute (PSI) has recently
realized a demonstration plant with a capacity of 30,000 tons of dry wood to produce about
7500 t of methanol plus electricity and process heat [5].
In the USA, the erection of a pilot plant for methanol production applying the
HYDROCARB process to biomass feedstock is planned. The designed capacity is 360
1/d of methanol to be used as a transportation fuel. The HYDROCARB appears to be
superior to other biomass gasification processes [2].
Sewage waste usually generates in an anaerobic process methane as the major gas.
However, its treatment with charged electrodes results in a hydrogen production. The
process has been verified in two different types of lab-scale chemical reactors which allowed
a continuous H2 generation, if the sewage sludge was continuously supplied. The electric
energy required is much smaller than in water electrolysis [48]. A first-of-its-kind fuel
cell plant in Yokohama, Japan, has started its operation fueled by methane which is being
derived from processed sewerage.
Fast pyrolysis is a process with no need of oxygen that converts biomass into
a mixture of low-molecular weight oxygenates to be subsequently cracked and steam
reformed. A hydrogen yield of 80 % of the theoretical maximum has been obtained in
experimental studies. The marketing of the byproducts seems possible. Focus of research is
being laid upon the catalytic reforming step to minimize carbon formation on the catalyst
[19, 67].
Hydrogen sulfide, H2S, is sometimes contained in natural gas with a fraction up to

25 % or is a byproduct of various petrochemical processes and is usually considered a
waste gas. The widely used treatment of H2S according to the Claus process only allows
for sulfur production plus it leaves waste in the form of SO and polluted water. Therefore
H2S is projected to gain economic importance if decomposed in a waste-free process to
hydrogen and sulfur. Achievable hydrogen from this resource is estimated to amount to 1
million tons per year [12]. The endothermal reaction
H2S <——> H2 + 1/2 S2 - 92 kJ/mol
leads to an equilibrium conversion of 24 % when taking place at a temperature of 1100

°C. A yield of 35 % has been achieved in tests at 1200 °C. A calculation model has been
used to optimize the design of a tubular heat exchanger type reactor. The "Zero Emissions
Sulfur Process" catalytically decomposes H2S in the presence of CO2 at 900 - 1000 °C.
The products are sulfur and CO-H2 process gas [62]. Ceramic tubes will be required due
116
to the corrosive nature of H2S [20]. In Russian tests, an experimental electric arc reactor
has been used to decompose H2S at a temperature of about 2000 K. Different technologies
that are being tested include thermal, thermochemical, electrochemical, photochemical, and
plasmochemical methods.
Hydrogen production by decomposition of H2S has been realized in Orenburg,

Russia, in an experimental small-scale industrial plant of 1 MW power. Separating the
CO2/H2S gas mixture from the natural gas exploited in a nearby gas field, the reforming
reaction is:
2 H2S + CO2 H2 + CO + H2O + 2 S
with low energy cost. Heat is provided by a plasma chemical unit containing two
500 kW plasmarrons. The plasma chemical technology allows the production of H2 as a
byproduct of the natural gas exploitation plus avoiding the emission of the greenhouse gas
CO2 into the atmosphere [10].
5.1.5. Recovery from Refinery Offgases
A significant component of refinery offgases is hydrogen which could be extracted

for further use. Some examples of typical hydrogen concentrations in offgases are listed
in Table 5-2. Presently the offgases in the petrochemical industries are usually flared
because their utilization is either considered not economic or the infrastructure for their
utilization is not existent. These byproduct gases, however, represent an attractive potential
as a hydrogen source, in particular those from petrochemical reforming processes or olefin
production. The British company BOC operates a cryogenic hydrogen and argon recovery
unit for a large ammonia plant in Tulsa, USA.
The central recovery unit could be suitably designed also for hydrocarbon byproducts
extraction [42]. Recovery methods for hydrogen are pressure swing adsorption, cryogenics,
membrane, or a combination. Newly discovered polymer membranes which are stable up to
400 °C could help separate significant amounts of hydrogen to reduce energy consumption
in the refineries and/or provide further revenues from selling the hydrogen [4].
Table 5-2: Typical hydrogen concentrations in various offgases, from [42]
Offgas source Hydrogen contents [vol%]

Naphta reforming 65 -90
Hydroprocessing
High pressure 75 -90
Low pressure 50 -75
Fluid catalytic cracking 10 -20
Toluene hydrodealkylation 50 -60
Ethylene synthesis 70 -90
Methanol synthesis 70 -90
117
5.2. DECOMPOSITION OF WATER
5.2.1. Electrolysis
About 4 % of the world hydrogen production is made by electrolysis.
5.2.1.1. Principle
Water electrolysis is basically applied, if cheap electricity is available or high-purity

hydrogen is required. The electrolytic decomposition of water with the cell reaction
H2O -* H2 + 1/2 O2 - 242 kJ/mol

requires electric energy which is dependent on the reaction enthalpy (or gross heat
of combustion), the reaction entropy, and the temperature. The ideal (reversible) voltage
of decomposition is 1.229 V. But heat is also essential for the operation of an electrolysis
cell. If this energy is to be supplied in electric form, the theoretical potential needs to
be increased by another 0.252 V. Irreversible processes in the reaction mechanism at the
electrodes and heat production imply the real decomposition voltage to be always higher.
The excess voltage should be minimized to allow for a high electrolyzer efficiency. Typical
cell voltages are about 1.85 to 2.05 V which translates into efficiencies of 72 - 80 %17 [43].
The effective electricity consumption is approx. 4.5 kWh/Nm3 at standard conditions. The
electrolysis of distilled water takes place without any side reactions resulting in high-purity
gases.
The principle of water splitting is described in Fig. 5-8. Two water molecules are
reduced at the cathode to one molecule of hydrogen and two hydroxyl ions. The latter
migrate under the influence of the electric field through a porous diaphragm to the anode
where they are discharged to 1/2 molecule of oxygen and one molecule of water. Both
oxygen and hydrogen escape from the electrodes as a gas.
The minimum feed water demand of an electrolyzer is 0.8 1/Nm3 of hydrogen, in
practice, about 1 1/Nm3 is required. Sea water must be desalinated first which consumes an
energy of 40 - 100 kWh/m3 of water or in the order of 1 % of the H2 heat of combustion.
5.2.1.2. Alkaline Water Electrolysis
Conventional electrolytic cells use aqueous solutions of KOH or NaOH or NaCl or

use solid polymer matrices as the electrolyte. In industrial plants, the alkaline medium is
preferred, because corrosion is more easily controlled and cheaper materials can be used
than in acidic electrolysis technology. The alkaline water electrolysis using a 25 % potash
lye as the electrolyte consumes about 4 kWh/Nm3 including energy losses and related
energy demands for ancillary equipment. It is a mature technology since decades.
17
The efficiency is the ratio of the energy content of the produced hydrogen to electric input. The given figure of 60 - 70 %, however,
is dropped to 20 - 25 %, if the output is related to the input of primary energy.
118
= * 4S v
02
Anode Cathode
1/2 O 2
2 OH 2e 2 Hj O ••• 2e
H20 2 OH
Electrolyte Anode Diaphragm Cathode
Fig. 5-8: Principle of water splitting process in an alkaline electrolyzer
Salt Removal of other salts

• •
"»... r
Generation of salt solution
23 % NaCI 27 % NaCI
solution Air solution
Electrolysis cells
+ Graphite anode NaCI
solution
Amalgam
decomposer
Water —» Hydrogen
|H2
fchlonne CI2 ]
i^_ _ .. i
(O CO
Fig. 5-9: Electrolytic chlorine (and hydrogen) production using the mercury process
119
The chloralkali water electrolysis is the only large-scale technological method to
be commercialized, where the H2 is actually a byproduct of the chlorine production and
mostly used as the thermal energy source and substitute of natural gas. A solution of salt
in water is electrolytically decomposed into hydrogen and soda lye (cathode) and chlorine
(anode) as shown in Fig. 5-9 for the mercury process:
2 NaCl + 2 H2O -> 2 NaOH + H2 + C12 - electricity
During the annual production of 35 million t of chlorine worldwide, approx. 20

billion Nm3 of hydrogen are being generated, which is more than the H2 production by
water electrolysis. Modem large-scale plants allow a production of 300,000 t/yr of chlorine
with the byproduct hydrogen. In many cases, the hydrogen is vented to the atmosphere if
not needed on site. A 1.6 MW fuel cell (PAFC) power plant is planned in Eastern Europe
to be fueled by waste hydrogen from the chloralkali process. It is said, however, that
industrial chlorine production does not have a sufficiently high electrolytic efficiency and
productivity to form the basis of a large-scale hydrogen production.
Direct processes with the generation of hydrogen as a byproduct are the decomposition
of HC1 or HE
5.2.1.3. High-Pressure Electrolysis
Hydrogen and oxygen production under pressure, in the range between 0.6 and
20 MPa, has the advantages of smaller-sized gas separators, a reduced voltage loss
increasing the current density reducing the electricity consumption by 20 % (Zdansky-Lonza
process). For subsequent high-pressure storage of the produced hydrogen, the expensive
first piston compression is saved. The main disadvantage is the more cost-intensive design
of pressurized systems [66].
5.2.1.4. Solid Polymer Electrolyte Water Electrolysis
Solid polymer electrolyte water electrolysis (SPEWE) is considered to be a promi-

sing method because it can operate at high current densities due to the extreme volume
reduction compared with the KOH electrolyte, and low cell alloys. It is based on the use
of a proton conducting membrane of sulfonic acid type, Nafion, with a tenth of a milli-
meter thickness, which acts both as an acid and as a separating wall. In these membrane
electrolyzers, only clean water is used. SPEWE has the advantages of cell compactness,
simplicity in design and operation, lack of corrosion problem. A considerable handicap,
however, is represented by the high cost of the membrane production and the need of noble
metal primary electrodes. More than 40 polymeric materials known for their thermal and
chemical resistance were examined, some of which showed good stability, some were even
stable up to 300 °C. A schematic of the SPEWE module is shown in Fig. 5-10.
Efforts are made to test engineering polymer materials for intermediate temperatures
200 - 400 °C which are in the range of speeding-up kinetics at still reduced corrosive effects.
120
I 2 10 9 7 6 5 8 7 6 5 3 2 I
1: Flange for compression 6: Membrane-electrodes composite

2: Insulator plate 7: Anodic support collector
3: Cathode side bus bar 8: Bipolar collector
4: Cathodic end plate 9: Anodic end plate
5: Cathodic support collector 10: Anode side bus bar
Fig. 5-10: Schematic of a solid polymer electrolyte water electrolysis module

consisting of 5 cells with a total electrode area of 500 cm2, production capacity
is 2 Nm3/h, from [59]
5.2.1.5. High-Temperature Electrolysis
The process of electrolysis of steam operated at higher temperatures (800 - 1000 °C)
offers several advantages:
i. the thermodynamic electric energy required could be reduced, and
ii. the activation barriers at the electrolyte surfaces would be easier to overcome
resulting in an improvement of the efficiency.
Very low electric energy consumption can be obtained if high-temperature heat and
low-temperature heat or steam provided by HTGR or solar energy are fed to the system. The
high-temperature water vapor electrolysis process could be realized with the development
of metal oxide diaphragms as the solid, oxygen ionic conducting electrolyte. A water vapor
electrolyzer is typically operated at 900 °C and consumes about 3 kWh of electricity for
1 Nm3 of H2 produced.
Zirconia-based ceramic membranes have been selected, which show a high conduc-
tivity for oxygen ions at operating temperatures. The electricity demand at 1000 °C is
30 % lower compared with that at room temperature. The most widely used solid oxide
electrolyte is a calcium and yttrium stabilized zirconium oxide. _It is formed as small tubes
of about 2 cm diameter through which the hot steam is routed (see Fig. 5-11). The steam
is split at the inside surface, the oxygen ions migrate through the ceramic, the hydrogen is
enriched in the steam and is afterwards extracted. The principle of the high-temperature
121
1: anode 4: conducting material
2: solid electrolyte 5: cathode
3: electrically insulating material
Fig. 5-11: Schematic of a high-temperature electrolysis tube (Hot Elly), from [70]
electrolysis of steam corresponds to the reverse operation of a solid oxide fuel cell. Some
variations with tubular or brick-like cell forms have been developed and tested. Main goal
of research is the low-cost production of efficient, reliable, and durable electrolysis cells.
5.2.1.6. Electrolyzer Technology
The technological development of electrolyzers started with a mono polar cell consi-
sting of a cathode part and an anode part separated by a diaphragm. In multi-cell systems,
bipolar plates are used carrying the cathode material for one cell and on its backside the
anode material for the neighbor cell. The functions of the bipolar plate are the continuous
supply of the membrane electrode with H2 on one side and with O2 or air on the other
side and the regulation of the water balance by providing moisture for the membrane on
the H2 side and remove the product water on the O2 side.
The development of electrolyzers is affected in particular by material problems. Cells

need to be tight, electrically insulated, and resistant against corrosive attack. The reaction
velocity is dependent on the electrodes' materials. The replacement of normal electrodes
by catalyst electrodes reduces the specific energy consumption significantly. On the other
hand, it is rising after some years of operation due to aging of the porous electrodes.
The operation of an electrolysis plant is simple and can be interrupted according to the
fluctuating energy demand [66].
122
Table 5-3: Present electrolyzer developments
Energy
Cell Current
Electrolyzer Anode (A) Operating Cell Voltage consumption
efficiency density Remarks
type Cathode (C) conditions [V] [kWh/Nm3
[%] [kA/m2]
H2]
Commercial large-scale units
70 - 90 °C
proven technology, simple,
ambient 77-80 1.84-2.25 1.3 -2.5 4.3 - 4.9
low efficiency, corrosive
pressure
electrolyte
80 °C Norsk Hydro
ambient 83 1.8 3 4.3 > 400 kW
pressure capacity 100-400 Nm3/h
Lurgi
90 °C
79 1.7 4 4.2 25 MW
Alkaline 3.2 MPa
(A): Ni capacity 5100 Nm3/h
(25-30 %
(C): Ni, steel 30 - 85 °C
KOH) 69 3.5 5.0 Hysolar 350 kW
0.3 - 0.9 MPa
100 °C Research Center JUlich
87 1.7 5 4.0
0.11 MPa Hysolar 10 kW
Hydrogen Systems
110 °C
90 1.7 5 3.9 < 100 kW
0.5 MPa
commercial 5-10 kW
80 °C
Electrolyzer Corp.
ambient 80 1.9 2.5 4.4
the only monopolar
pressure
to
OJ
Table 5-3: Present electrolyzer developments (continued)
Energy
Cell Current
[%] [kA/m2]
H2]
Lab scale, prototype units,
90 - 145 °C has commercialization
80-90 1.5-3.0 2-20 3.8 - 4.3
<4MPa potential, severe corrosion at
higher temp.
Advanced
alkaline (A): Ni Research Center Julich
120 °C 1.5 25 3.8
(25-40 % (C): Ni several 10 kW
KOH) < 150 °C
82 1.8 10 4.1 1 00 kW prototype GHW
>3MPa
150-250°C,
70-80
3 - 10 MPa
Inorganic
lab scale, prototype units,
Membrane
(A): spinel oxide 90 - 120 °C non-noble metal cat.,
alkaline 82-91 1.6- 1.9 2- 10 3.6 - 4.0
(14 . 15 % (C): NiS < 4 MPa membrane performance
needs to be demonstrated
NaOH)
100 kW
80 - 150 °C compact, non-corrosive
(A): Ti 85 -90 1.4-2.0 2.5 - 20
Solid polymer <4MPa electrolyte, noble metal cat,
(C): Carbon with expensive membrane
(Nafion)
Pt
100- 130°C GIRI, Osaka,
88 1.7 10
<4MPa production rate 1 .3 Nm3/h
Table 5-3: Present electrolyzer developments (continued)
Energy
Cell Current
[%] [kA/m2]
H2]
Intermediate
temperature
(A): Ni simple double cell
Molten 300 - 600 °C 82-91 1.3 - 1.4 >2
(C): Mb configuration
Carbonate
(KOH/NaOH)
very small lab scale units,
High- (A): Ni-NiO 800 - 1000 °C non-corrosive electrolyte,
90- 100 0.95 - 1.3 3- 10 3.5
temperature (C): Ni <3MPa severe materials and
(solid Y2O3 fabrication problem
stab. ZrO2) tubular
95-115 3.3 2.6 - 3.2
10 kW Hot Elly
to
Electrolyzers are being developed in various countries. Although very promising,
they are still at an early stage of technology. An electrolysis plant can inherently operate
over a wide range of capacities which is attractive for load leveling. Conventional alkaline
water electrolysis is limited to small and medium-sized plants (0.5 - 5 MW corresponding
to about 100 - 1000 Nm3/h) unless cheap electricity is available. There are only a few
large-scale facilities which have H2 production capacities in the order of 33,000 Nm3/h
(approx. 150 MW), e.g. in Assuan, Egypt, with cheap hydro power used for the ammonia
synthesis. The typical size is in the range of 100 - 1000 Nm3/h corresponding to 0.5 - 5
MW. Some efforts are listed in Table 5-3.
One of the world leaders in electrolysis for Ha production is the Norwegian company
Norsk Hydro, which built the first electrolyzer in 1927 used in the ammonia production.
Since then, the company has supplied a total of 461 electrolyzer units of different sizes
(between 0.5 and 565 Nm3/h of Ha) with a total capacity of approx. 30,000 Nm3/h of
hydrogen gas. The German Gesellschaft fur Hochleistungselektrolyseure zur Wasserstof-
ferzeugung mbH (GHW) has developed a 100 kW high-pressure, high-efficiency advanced
alkaline pilot electrolyzer for both continuous and intermittent operation and operated it
for more than 5000 h. Due to higher efficiencies, advanced alkaline electrolyzers can be
smaller in size. Intermittent and part-load operation is possible. The positive experience
allows a scale-up into the MW range [52].
Membrane electrolysis technology is well established and appropriate for smaller

scale facilities. One application is its use as an oxygen generator in submarines, where the
hydrogen is considered only a byproduct. Drawback is the high-cost membrane production
[14].
In the high-temperature range, Westinghouse is leading the development of the thin

layer tube concept with pilot plants up to 25 kW. Other companies pursuing the planar
concept since the end 1980s, have reached the kW scale [23]. Test operation of high-
temperature electrolysis cells basically confirm the predictions of achievable efficiencies.
The concept of a tubular solid oxide hot water vapor electrolysis (Hot Elly) has
been investigated in Germany within the frame of a technology program between 1975 and
1987. It was terminated with the successful demonstration of a 2 kW lab-scale pilot plant
[23]. The tubular concept was replaced by the planar concept, after respective ceramics
production technologies were sufficiently developed.
The high-temperature technology is able to create so-called regenerative cells which

can work either in the electrolyzer mode or in the fuel cell (SOFC) mode (see section
7.2.2.). Hot Elly efficiency is expected to rise up to 111 % in the future including the
utilization of the process heat in the combined heat power operation mode [53].
A lower cost system is achieved if electricity and steam is generated from a coal-
fired power plant and then routed to a high-temperature electrolysis plant for steam
decomposition with an overall thermal efficiency of 35 - 38 % (based on 35 % efficiency
of electricity production from coal) [55].
An advanced hybrid system combines high temperature electrolysis with coal gasifi-
cation. The latter produces synthesis gas plus steam at a high temperature which is used
126
for the electrolysis step. In return, the electrolytic oxygen is utilized in the gasification
step. In the combined process, two thirds of the H2 are produced in the gasifier and one
third in the electrolyzer [55].
There are currently no comprehensive guidelines or codes for the design and instal-
lation of small and medium-sized electrolyzers. Considering future operation mainly by
private users, international guidelines for quality criteria, safety requirements, licensing
procedures, and testing methods should be implemented [57].
5.2.2. Thermochemical Cycles
Since a direct thermolysis of water, which requires temperatures of > 2500 °C, is not
practicable under normal circumstances, the splitting process is subdivided into different
partial reactions, each running on a lower temperature level. The thermal energy required
is heat of 800 to 900 °C. Some 2000 - 3000 potential thermochemical cycles have been
tested and checked in terms of appropriate reaction temperatures and velocities and in
the economic respect. Thermochemical cycles are Carnot cycle-limited meaning that high
temperatures could improve the conversion efficiency. Major problems arise due to the
large material flows, the introduction of impurities, and the potential creation of toxic
and environmentally unacceptable species. The main goal is to achieve satisfactory overall
energy efficiencies which are currently not higher than around 40 % and, in connection with
separation steps and severe materials and equipment design difficulties, do not allow large-
scale thermochemical process applications so far [43]. Studies are focusing on reaction
kinetics, thermodynamics, reactant separation, material stability, flow sheeting, and cost
analysis.
Thermochemical multistage cycles are usually classified in families according to the

chemicals involved. Cycles can consist of up to 8 equations, mainly 3 - 6, with up to 5
elements others than Ha and 02 involved [26].
The thermochemical hybrid process is a combined cycle process with both ther-
mochemical and electrolytic reactions of water splitting. The hybrid process offers the
possibility of running low-temperature reactions on electricity. The expectations for rea-
lization of hybrid processes are similar to those for purely thermochemical processes [13].
Various hybrid processes are energetically possible, but not always feasible. Important
criteria are the minimum voltage for the electrolysis step, realizability, efficiency.
The most advanced and promising thermo(electro)chemical processes are described

in more detail in appendix A.
5.23. Photolysis
In a photolytic process, light energy is directly absorbed to either convert the photon
energy into electricity or store part of it as chemical energy in the water splitting reaction.
This process is of great interest because solar energy can provide an unlimited energy
source for hydrogen production.
127
5.2.3.1. Photoelectrochemical Systems
In photoelectrochemical (PEC) systems, inorganic materials are being used for direct
conversion of (sun) light into hydrogen at a semiconductor-electrolyte contact [17]. The
PEC direct water splitting by solar energy was first reported in 1972. The process is
based on the band gap absorption of photons by which electrons are elevated from the
valence band to the conducting band creating electron-hole pairs. Following this charge
separation, the excitation energy is utilized for the splitting of water molecules at the n-type
semiconducting anode surface. The band gap must be greater than the water decomposition
potential (> 1.23 eV) to enable the necessary charge transfer. Constraints are given by the
high voltage required for dissociation and the corrosiveness of the aqueous electrolytes. For
some reason, only SrTiOs was found so far to be really able to directly split water [36].
Solid state devices with appropriate catalytic surfaces seem to widen working conditions
[47]. Catalytic photo-decomposition of water is also possible using n-type semiconducting
powders [73]. The theoretical limit for the solar conversion efficiency of PEC water
splitting is 29 % [46].
In Japan, a photoelectrochemical system for outdoor operation, i.e., solar, has been
constructed. Photoanodes were Ti plates with a 1 fim thick TiO2 surface layer with a total
area of 0.17 m2, cathodes were of Pt, and the electrolyte was KC1. The result was an
output of 1.1 1 of hydrogen per day, the conversion efficiency was about 0.4 % which is
considered relatively high compared with the small absorption wavelength range (3 % of
the entire solar spectrum) by TiC>2 [25]. In a more recent effort using a multi-band gap
PEC cell structure, an efficiency of 12.4 % has been achieved [3].
In another approach, semiconductor materials and devices which are stable in aqueous
environment are taken to apply sunlight for direct water splitting. The conversion efficiency,
however, is still insignificant, since only a very small portion of the solar spectrum is used.
Photolytic response, however, could be shifted into the visible region using chomophores
called sensitizers [73]. Research is focusing on the development of transparent and highly
active catalysts and techniques of surface modification to increase lifetime and efficiency
of PEC devices.
5.2.3.2. Photocatalvtic Systems
Photocatalytic production of hydrogen can be realized from water in the presence

of various types of catalysts. One system is the sulfide which is often found as the major
constituent of natural gas:
H2S -+ H2 + S - 0.21 eV/mol
Another one is the use of alcohols with dispersed or supported catalysts [29]:
(Rl) (R2) - CHOH -^ (Rl) (R2) - CO + H2
In Japan, the carbonate method is being studied where the addition of carbonate was
found to lead to a much more efficient decomposition of water.
128
5.2.3.3. Photobiological Systems
In photobiological systems, hydrogen is produced as a result of the way certain

biological organisms use sunlight in an energy storing fuel production reaction. In principle,
two ways are distinguished:
a) release of hydrogen from photolytical decomposition of water
b) extraction of hydrogen stored in biomass which serves as nutrition for the micro-
organisms
Photosynthesis reactions are:
H2O + light -» H2 + 1/2 O2

H2 + CO2 + light —> plant tissue
CH3COOH + 2 H2O + light -» 4 H2 + 2 CO2
Some photosynthetic algae and bacteria, in particular the anaerobic ones, absorb
energy from sunlight and use it for water splitting in the process of generating plant tissue or
creating heavier hydrocarbons or oxygen for assimilation. The hydrogen production seems
to be a precursor step from the anaerobic microorganism prior to fixing the hydrogen
with carbon from its CO2 uptake. Competent light-driven H2 producers are cultured
cyanobacteria and purple bacteria. Efficiencies under ideal conditions approach approx.
10 % [17].
The direct decomposition of water by bacteria has only a very low hydrogen yield and
hydrogen extraction is difficult. The organic acids required as a substrate may be gained
by organic waste conversion. Investigations have been made concerning the development
of photoelectrolysis cells and the search for appropriate catalysts.
Among the biological options, systems of intact cells of photosynthetic bacteria are
the most advanced. They evolve hydrogen from thermally gasified biomass (synthesis gas)
using several different enzymatic mechanisms with near-term commercial potential for
biological H2 production from biomass. Research is now concentrating on maximization
of bioactivity and durability [69].
The commercialization of the photosynthetic processes requires a drastic increase of

the conversion efficiency as well as an improvement and upgrading of the microorganisms.
The latter could be realized by genetic engineering and the development of mass cultivation
techniques [58]. Other problems are the rapid separation of the produced oxygen and
hydrogen to avoid explosion hazards and the removal of decayed microorganisms.
German scientists have discovered a suitable organism for H2 production preferably in

sunny climates. Within the "Erg-Chebbi project", a photobioreactor, a so-called heliomite,
has been developed which consists of a cone-shaped wooden construction (Fig. 5-12, left).
The substrate with the bacteria is slowly led through glass tubes of 2.8 cm diameter which
are radiated by solar (or artificial) light. The product gas contained 90 % H2 and 6 % N2.
The highest production rate measured was 2.5 1 of hydrogen per hour per m2 of radiated
reactor surface. Some 100 of such heliomites were deployed across a 60 by 60 m2 area
129
Light
Purple bacteria
Fig. 5-12: Schematics of photobiological hydrogen production arrangements

a heliomite farm (left) [45] and a plate loop reactor (right) [63]
reaching a yield of 285 1 of hydrogen each. A 10 MW heliomite farm would require some
10,000 of such mini generators. A further development was the combination with green
algae to form an artificial symbiosis, where the green algae react with the bacterial CO2 to
produce C>2 which in reverse serves as nutriment for the bacteria [45].
A plate loop or sandwich reactor for photobiological processes using purple non-
sulfur bacteria has been designed at the RWTH Aachen (Fig. 5-12, right). The reactor is
separated by a central plate on which the bacteria are growing. Solar insulation triggers
a natural convection of the nutrition suspension to flush around the bacteria. In outdoor
experiments, a hydrogen production rate of 2 l/(h m2) was achieved [63].
A two-stage process has been proposed [15] to use green algae cultivated in large open
ponds to produce a high carbohydrate algal biomass which is then fed into a photobioreactor
to produce hydrogen in a light-driven reaction. Such a fermentation device could be an
arrangement of closed glas tubes of about 2.5 m in length and 3 - 5 cm in diameter.
5.2.4. Direct Water Splitting
The direct splitting of water requires a high energy input. In the plasma arc process,
water is heated up to 5000 °C by an electric field resulting in the cracking products H,
H2, O, O2, OH, HO2, and H2O. A fraction of 50 vol% of H and H2 is possible. To avoid
recombination, a rapid quenching of the plasma gases, e.g., by a cryogenic liquid, must be
made. The process consumes a lot of energy and thus is very expensive [55].
130
5.3. COMPARISON OF HYDROGEN PRODUCTION METHODS
If hydrogen production processes are compared, the interdependence of efficiency,

capital investment and value of the byproduct has to be taken into account. From the
reaction equations, it can be derived that an essential part of the hydrogen is gained from
water, e.g., 50 % in the case of steam reforming of natural gas plus CO conversion. On
the other hand, all carbon in the raw materials is finally converted to CO2 and released
into the atmosphere; the least unfavorable process is steam reforming with a CO2 to H2
ratio of 0.25, the worst in coal gasification with a respective ratio of 1. The use of heavier
raw materials is connected with a toss of efficiency since a larger mass of C-carriers and
water need to be heated up to reaction temperatures [51].
Reforming of natural gas and partial oxidation of heavy oils are presently the least
expensive and most frequently applied methods of hydrogen production (see Table 5-4).
Methane cracking is considered in [55] to have even lower net H2 production cost than
reforming because of its very low capital investment which compensates the not very high
efficiency. There is a high degression of cost if moving towards upscaling of the plant in
contrast to electrolysis which is composed of many single cells. Electrolysis is the most
expensive process unless cheap electricity is available [38],
The production of hydrogen (or methanol) from coal gasification and using a CO2-free
primary energy source (e.g., nuclear) is not an ideal alternative in terms of the necessary
reduction of CO2 emissions in the future, because the reduction is no higher than 20 - 30
% compared with the coal feed. In this respect, gasification of CO2 -neutral biomass would
be the better alternative [11], although also non-CO2 gases with a global warming potential
are emitted with its combustion.
As can be seen from the reaction equations, the splitting of water requires different
amounts of energy dependent on how much of the carbon contained in the (fossil) raw
materials is converted into CO2- The energy released during this conversion is consumed
in the splitting process and does not need to be introduced from outside. The theoretical
energies required for the production of 1 Nm3 of hydrogen are listed in Table 5-4.
Electrolytic hydrogen has a small, but stable market. It is preferred if there is a

fluctuating demand, a need for small amounts or the requirement for high-purity hydrogen
[41]. It strongly competes with the direct use of electricity. From all electricity generating
concepts, the most economic are hydro and wind compared with photovoltaics as the most
expensive [71]. Cost from renewable sources, however, is projected to drop significantly
in the next decade.
Hydrogen production in thermochemical (hybrid) cycles has been demonstrated to be

feasible with an overall efficiency that can be higher than 35 %, however, is inherently less
efficient than conventional processes. Technological improvements are still possible. A
strong cost reduction, however, is not expected. Economic competitiveness appears to be
likely, if nuclear heat (HTGR) could be used. Using solar primary energy promises high
efficiency and no major technical problems although still connected to many uncertainties
[26].
131
Table 5-4: Characteristics of splitting processes for hydrogen production (H2 heat of
combustion: 3.54 kWh/Nm3), from [various sources, e.g. 26, 38, 51]
Energy required Costs

Fraction
[kWh/Nm3 of H2] relative
Splitting Efficiency of pro-
Status to Cftt
process in [%] duction
in theory steam
practice [%P
reform.(1)
Methane steam
0.78 2-2.5 mature 70- 80 1
reforming
48
Methane
mature 54 (0.9)
cracking
Partial oxidation 1.8
0.94 4.9 mature 70
of heavy oil (1-6)
30
Naphta
mature
reforming
Coal gasification 1.4-2.6
1.01 8.6 mature 60
(TEXACO) (2.3)
Partial oxidation
mature 55 18
of coal
HYDROCARB R&D (0-9)
Steam-iron R&D 46 (1-9)
Chloralkali
mature byproduct
electrolysis
Water R&D- 5- 10
3.54 4.9 27(3)
electrolysis mature (3.0)
High- 4
temperature R&D 48 (2-2)
electrolysis
Thermochemical early
35-45 6
cycles R&D
Biomass early
2-2.4
conversion R&D
-
early
Photolysis < 10
R&D
(1) as estimated by DOE, 1993 [26], in parentheses estimation by Steinberg, 1987

[55]
(2) data from 1988 [51]
(3) electricity from fossils with 40 % efficiency
132
For solar hydrogen technologies, the efficiency is the key to a practical economic
system. Systems with efficiencies less than 10 % will hardly have a chance to survive.
Photoelectrochemical and photobiological devices have not exceeded yet the stage of
laboratory outdoor demonstration and still need to show their potential fo economic H2
generation [17]. Besides, renewable energy systems require much more land than fossil fuel
systems providing the same amount of end-use energy. Land use and water requirements
may be, in particular, a serious issue in the case of biomass energy systems [33].
Since coal is abundantly available in the world and thus should about maintain its
cost level in the future, R&D efforts should concentrate on coal-based processes. The
advanced processes of methane cracking and coal cracking (HYDROCARB) have a huge
economic potential [55].
5.4. PRODUCTION OF LIQUID HYDROGEN
In contrast to the gases in air, the production and liquefaction of hydrogen is conducted
in two separate energy-intensive steps. During the cooling process, condensation heat
and heat of the ortho-para conversion are removed. The minimum work required for
the liquefaction of hydrogen (at ortho-para equilibrium) is 3.92 kWh/kg. A realistic value,
however, for the whole process of H2 cooling from ambient to boiling point and condensing
is 10 - 13 kWh/kg.
If an electric power of 500 MW is to be generated from hydrogen, the corresponding

capacity of LH2 production required is about 600 t/d.
5.4.1. LH2 Production Methods
The Linde-Hampson process uses a thermodynamic process. Isothermal compres-

sion and subsequent cooling along an isobar is done in a heat exchanger. Joule-Thomson
expansion connected with an irreversible change in entropy is used as the refrigeration pro-
cedure. Despite its simplicity and reliability, this method is now less attractive compared
with new ones where cooling is primarily carried out in reversible processes (expander)
and where less energy is required.
A commonly applied method in large-scale liquefaction plants is the Claude process.

The necessary refrigeration is provided in four principal steps leading to the liquefaction
of hydrogen (Fig. 5-13):
I. compression of hydrogen gas, removal of compression heat,
n. pre-cooling with liquid nitrogen (80 K),
ffl. expanding and thus cooling of a part of the hydrogen in an expander resulting in
a further pre-cooling of the residual hydrogen (80 - 30 K),
IV. expanding of the residual hydrogen in a Joule-Thomson valve until liquefaction
(30 - 20 K).
The first expansion process would be sufficient for liquefaction, however, Joule-
Thomson expansion is applied for the final step (from approx. 26 K down) to avoid
two-phase flow in the expander. Variations of the dual-pressure Claude process as a large-
133
Compression
Gaseous H2
Heat exchanger
Pre-cooling
II with LN2
-196°C
Expansion
Heat exchanger
EE23 - 253 °C
Fig. 5-13: Schematic of the Claude process for hydrogen liquefaction, from [Messer
Griesheim]
scale liquefaction procedure has proven to be the optimal solution [43]. Maximum impurity
level in the hydrogen gas is in the order of 1 ppm.
Magnetic refrigeration takes advantage of the entropy difference and the adiabatic
temperature change upon application or removal of magnetic fields in the working material.
It uses isentropic demagnetization of a ferromagnetic material as the cooling process. A
respective plant has neither been built nor utilized; the method is still on a research stage.
But it looks promising, because it offers a compact cooling system with an extremely
long lifetime, lower capital investment since no expanders or compressors are required, an
estimated liquefaction work of 7.3 kWh/kg, and it promises higher efficiencies [24].
5.42. Liquefaction Plants
A considerable amount of LH2 was required starting with the US Apollo space
program leading to the design and construction of large-scale plants. The liquefaction
plants in the world are summarized in Table 5-5 presently revealing a total capacity of 265
tons of LH2 per day, but not more than 19 t/d in Europe. Two major plants in the USA,
West Palm Beach, FL, with a 31 t/d capacity and Sacramento, CA, with a 54 t/d capacity,
ceased operation in the meantime.
Based on current experience, the construction of plants with a capacity of up to 250

t/d as an upper economic limit is expected to be feasible. The goal in the Japanese WE-NET
134
Table 5-5: Hydrogen liquefaction plants in the world
Capacity
Place Operator [metric Operation since
tons/day]
America
Montreal, Canada Air Liquid Canada Inc. 10 1986
Montreal, Canada BOC, USA 14
Sarnia, Canada Air Products, USA 30 1962
Kourou,
L'Air Liquide, France 5 1990
French Guiana
Mclntosh, USA Praxair 24 1995
New Orleans, USA Air Products, USA 68 1977
Niagara Falls, USA Praxair 2* 18 1981 / 1989
Ontario, USA Praxair 20 1962
Pace, USA Air Products, USA 30 1994
Sacramento, USA Air Products, USA 6 1986
Europe
Lille, France L'Air Liquide, France 10 1987
Ingolstadt, Germany Linde, Germany 4.4 1991
Rozenburg, 1987
Air Products, USA 5
The Netherlands
Asia
Amagasaki, Japan Iwatani Gas, Japan 1.2 1978
Pacific Hydrogen Co,
Ooita, Japan 1.4 1986
Japan
project is a plant capacity of 600 t/d. At present, the compression step limits the liquefaction
capacity to 50 - 60 t/d, which appears to^>e a convenient module size. An assumed value
for the liquefaction work of 10.8 kWh/kg representing approx. 30 % of the net heat of
combustion of hydrogen is considered conservative, improvement of components could
lead to a figure of 9.8 kWh/kg. A combined system of advanced pressurized electrolysis
and liquefaction promises an improvement in efficiency.
5.4.3. Production of Slush Hydrogen
Most slush hydrogen has been produced so far by means of the freeze-thaw method
[22]. In this technique the gaseous phase above the liquid hydrogen is pumped to pull a
vacuum until the liquid is cooled down below the freezing point (13.8 K) and a film of
solid hydrogen forms on the surface. The system pressure is then stopped and the solid
135
phase allowed to settle into the liquid. After decreasing the pressure again, the formation
of a new solid layer on the surface is initiated.
The time of the freezing period can be extended by spraying triple point liquid
hydrogen onto the surface, the spray / freeze method, which could be a step towards a
continuous process with no further need of a thawing period [40].
Some other production methods apply liquid helium for cooling. In the Auger
method, a hollow cylinder is placed in liquid hydrogen. An annular interior of the cylinder
is then cooled with liquid helium. Hydrogen freezes on the surface of the heat exchanger
and is scraped off by means of a rotating auger. This method provides a stable, continuous,
and contamination-free slush production. Another way is to have the cold helium bubbling
through the liquid hydrogen.
A Small Scale Slush Hydrogen Facility has been constructed by NASA to study
ways of optimizing the SLHj production process and provide a test bed for advanced
instrumentation. The tank with a total volume of 0.76 m3 is vacuum jacketed and wrapped
in several inches multilayer insulation plus a liquid nitrogen shield in the upper tank part. A
1 kW heater is used to simulate a heat leak or a warming up. The tank also offers the option
to install a liquid helium slush auger for research on this method of production. In a 17
m long and 0.05 m diameter vacuum jacketed transfer line, studies on flow characteristics
and instrumentation can be conducted [40].
A smaller slush hydrogen production system has been constructed in Japan with a
volume of 60 liters. Both the "freeze-thaw" and the "auger" production method have been
investigated revealing the creation of finer fractions with a higher fluidity with the latter
method [61].
[1] ANAND, M., Sorption-Enhanced Reaction (SER) Process for Production of Hy-
drogen, World Wide Web, http://www.eren.doe.gov/hydrogen/sorption.htm, US De-
partment of Energy (1997).
[2] ANON, Pilot Plant Demonstration of Methanol Production Using the
HYDROCARB Process with Biomass Feedstock, World Wide Web,
http://es.inel.gov/new/funding/serdp/fy93en7.html, EnviroSense (1996).
[3] ANON, 10 % Efficient Photoelectrochemical Cell (PEC), World Wide Web,
http://www.nrel.gov/basic_sciences/chemsci.html, National Renewable Energy La-
boratory, Golden (1998), and KHASELEV, O., TURNER, J.A., A Monolit-
hic Photovoltaic-Photoelectrochemical Device for Hydrogen Production via Water
Splitting, Science 280 (1998) 425-427.
[4] ANON, Membran hilft bei der Trennung von Gasen - Gewinnung von Wasserstoff
wird zum lohnenden Geschaft, Newspaper article in Blick durch die Wirtschaft,
Frankfurt, September 23, 1997,.
[5] ANON, Reaktor gewinnt Brennstoffe aus Biomasse - Erzeugung von Methanol aus
pflanzlichen Abfallen, insbesondere aus Holz, Newspaper article in Blick durch die
Wirtschaft, Frankfurt, December 1, 1997.
136
[6] ANON, Wasserstoff aus Biomasse, Wasserstoff-Spiegel No. 1/98, Deutscher
Wasserstoff-Verband (1998).
[7] ANON, Gasoline-Type Fuels for Fuel Cell Vehicles, Fuel Cells UK Newsletter,
ETSU, Harwell (1998) Issue 6
[8] ANON, Erste Anlage nach neuem Verfahren zur Wasserstoffgewinnung aus fossilen
Brennstoffen wird in Canada gebaut, Brennstoffzelle (1998) Issue l, Deutsches
Brennstoffzellenforum e.V.
[9] ARAUZO, J., et a!., Catalytic Pyrogasification of Biomass - Evaluation of Modified
Nickel Catalysts, Ind. and Eng. Chem. Res. 36 (1997) 67-75.
[ 10] B AG AUTDINOV, A.Z., et al., Clean Method of Hydrogen Production - Plasmache-
mical Dissociation of Hydrogen Sulfide, Hydrogen and Clean Energy (Int. Symp.,
Tokyo, 1995), NEDO (1995) 269-272.
[11] BARNERT, H., Energy Alcohol from Plant Biomass plus High Temperature Heat,
the Œ>2-Neutral, Environmentally Benign, and Consumer Friendly Future Alterna-
tive, Report Jul-3089, Research Center Jülich (1995).
[12] BECKMANN, G., Wasserstoff- ein Energiemittel?, Nachr. Chem. Tech. Lab. 39
(1991) 503-508.
[13] BECKURTS, K.-H., DIETRICH, G., Projekt Femwärme Versorgung für Millionen-
Städte, Bild der Wissenschaft 13 (1976) January 64-70.
[14] BEHRENS, D. (Ed.), Wasserstofftechnologie - Perspektiven für Forschung und
Entwicklung, DECHEMA, Frankfurt/M (1986).
[15] BENEMANN, J.R., Feasibility Analysis of Photobiological Hydrogen Production,
(10th World Hydrogen Energy Conf., Cocoa Beach, USA, 1994), BLOCK, D.L.,
VEZIROGLU, T.N., Hydrogen Energy Progress X, International Association for
Hydrogen Energy (1994) 931-940.
[16] BOLTENDAHL, U., HARTH, R., Wärmetransport auf kaltem Wege, Bild der
Wissenschaft 17 (1980) April 44-55.
[17] BOUTON, J.R., Solar Photoproduction of Hydrogen, IEA Technical Report
IEA/H2/TR-96, Internal Energy Agency (1996) and Solar Energy 57 (1996) 37-50.
[18] BROMBERG, L., Hydrogen Production from Plasma Reforming, World Wide
Web, http://www.eren.doe.gov/hydrogen/hyplsma3.htm, US Department of Energy
(1997).
[19] CHORNET, E., et al., Biomass to Hydrogen via Fast Pyrolysis and Catalytic
Steam Reforming, (llth World Hydrogen Energy Conf., Stuttgart, FRG, 1996),
[20] COX, B.G., PRÜDEN, B.B., JEJE, A.A., Ceramic Reactor/Exchanger for H2S
Dissociation, (Proc. 7th Canadian Hydrogen Workshop, 1995, Quebec City),
MEHTA, S.K., BÖSE, T.K. (Ed.), Canadian Hydrogen Association (1995) 329-352.
[21] DICKS, A.L., Hydrogen Generation from Natural Gas for the Fuel Cell Systems
of Tomorrow, J. Power Sources 61 (1996) 113-124.
[22] EDESKUTY, F.J., STEWART, W.F., Safety in the Handling of Cryogenic Fluids,
The International Cryogenics Monograph Series, Plenum Press, New York (1996).
[23] ERDLE, E., Hochtemperaturelektrolyse von Wasserdampf HOT ELLY, Wasserstoff
als Energieträger, (Status Seminar, Würzburg, 1995), Projektträger Biologie, Ener-
gie, Ökologie, Research Center Jülich (1995) 1-10.
137
[24] FOLDEAKI, M., CHAHINE, R., BOSE, T.K., Material Selection for Magnetic
Refrigerator Design, (Proc. 7th Canadian Hydrogen Workshop, 1995, Quebec City),
MEHTA, S.K., BOSE, T.K. (Ed.), Canadian Hydrogen Association (1995) 439-453.
[25] FUJISHIMA, A., Photocatalytic Reactions - Water Splitting and Environmental
Applications, Hydrogen and Clean Energy (Int. Symp., Tokyo, 1995), NEDO
(1995) 23-30.
[26] FUNK, J.E., Hydrogen Production, Tutorial during the 1 Oth World Hydrogen Energy
Conf., Cocoa Beach, USA, June 19, 1994.
[27] HOFFMANN, P., (Ed.), Hydrogen & Fuel Cells Letter 12 (1997) November.
[28] HOHMANN, F.W., Improve Steam Reformer Performance, Hydrocarbon Proces-
sing 75 (1996) March 71-74.
[29] KALYONASUNDARAM, K., GRATZEL, M. (Eds.), Photosensitization and Photo-
catalysis Using Inorganic and Organometallic Compounds, Kluwer Academic Press
(1993).
[30] KONIG, S., BARNERT, H., SINGH, J., Prinzip-Auslegung und sicherheitstechni-
Warme, Internal Report KFA-ISR-IB-5/91, Research Center Julich (1991).
[31] KUBIAK, H, PAPAMICHALIS, A., VAN HEEK, K.H., Production of Hydrogen
by Allothermal Gasification of Biomass, (llth World Hydrogen Energy Conf.,
Stuttgart, FRG, 1996), VEZIROGLU, T.N., et al., Hydrogen Energy Progress XI,
International Association for Hydrogen Energy (1996) 621-629.
Nuclear Process Heal, Vol 4: System Considerations on Nuclear-Heated Steam
[33] LIPMAN, T.E., DELUCCffl, M.A., Hydrogen-Fuelled Vehicles, Int. J. of Vehicle
Design 17 (1996) 562-589.
[34] LYNCH, F.E., Hydrogen Storage, Hydrogen and Clean Energy (Int. Symp., Tokyo,
1995), NEDO (1995) 99-106.
[35] MANTHEY, C, Einsatz von Hochtemperaturreaktoren in der Eisen- und Stah-
lindustrie unter besonderer Beriicksichtigung des Einflusses auf die Standort- und
Umweltprobleme dieses Industriezweiges, Report Jul-1180, Research Center Julich
(1975).
[36] MEISSNER, D., Photoelectrochemical Solar Energy Conversion, in: Solar Techno-
logy, Ullmann's Encyclopedia of Industrial Chemistry, VCH Verlagsgesellschaft,
Weinheim (1993) 400-406.
[37] MOORE, R.B., RAMAN, V., Hydrogen Infrastructure for Fuel Cell Transportation,
(11th World Hydrogen Energy Conf., Stuttgart, FRG, 1996), VEZIROGLU, T.N., et
al., Hydrogen Energy Progress XI, International Association for Hydrogen Energy
(1996) 133-142.
[38] MULLER, R., TANGER, U., Solarer Wasserstoff, VGB Kraftwerkstechnik 72
(1992) 197-203.
[39] MURADOV, N., Production of Hydrogen by Thermocatalytic Cracking of Na-
tural Gas, World Wide Web, http://www.eren.doe.gov/hydrogen/crackin.htm, US
Department of Energy (1997).
[40] NASA, Small Scale Slush Hydrogen Test Facility, World Wide Web,
http://lerc.nasa.gov/WWW/slush, NASA (1997).
138
[41] OPPERMANN, M., STREICHER, R., Fortschrittliche Wasserelektrolyse, Wasser-
stoff als Energietrager, (Status Seminar, Wiirzburg, 1995), Projekttrager Biologic,
Energie, Okologie, Research Center Julich (1995) 11-20.
[42] PACALOWSKA, B., WHYSALL, M., NARASIMHAN, M.V., Improve Hydrogen
Recovery from Refinery Offgases, Hydrocarbon Processing 75 (1996) November
55-59.
[43] PESCHKA, W., Liquid Hydrogen: Fuel of the Future, Springer-Verlag Wien New
York (1992).
[44] PRUSCHEK, R., et ah, Minimum CO2 Emissions from IGCC Power Plants by
Methanol Production with HI, (11th World Hydrogen Energy Conf., Stuttgart, FRG,
1996), VEZIROGLU, T.N., et ah, Hydrogen Energy Progress XI, International
[45] RECHENBERG, L, Hydrogen Production by Means of an Artificial Bacterial Algal
Symbiosis, Report on Experiments in the Sahara, (llth World Hydrogen Energy
Conf., Stuttgart, FRG, 1996), VEZIROGLU, T.N., et ah, Hydrogen Energy Progress
XI, International Association for Hydrogen Energy (1996) 2427-2435.
[46] REINEKE, R., Wirkungsgrade elektrochemischer Zellen, in: Winsel, A. (Ed.),
Elektrochemie in Energie- und Umwelttechnik, Dechema-Monographien 124 (1991)
623-627.
[47] ROCHELEAU, R., Photoelectrochemical Production of Hydrogen, World Wide
Web, http://www.eren.doe.gov/hydrogen/photelch.htm, US Department of Energy
(1997).
[48] ROYCHOWDHURY, S., COX, D., Hydrogen Production Potential From Sewage
Waste, Proc. New Energy Systems & Conversion, (3rd Int. Conf., Kazan, 1997)
59-62.
[49] SANDSTEDE, G., Moglichkeiten zur Wasserstoff-Erzeugung mit verminderter
Kohlendioxid-Emission fur zukiinftige Energiesysteme, Chem.-Ing.-Tech. 63
(1991) 575-592.
[50] SCHARF, H.-L, SCHRADER, L., TEGGERS, H., Results from the Operation of
a Semitechnical Test Plant for Brown Coal Hydrogasification, Nuch Eng. Des. 78
(1984) 223-231.
[51] SCHOLZ, W.H., Verfahren zur groBtechnischen Erzeugung von Wasserstoff und
ihre Umweltproblematik, Linde Berichte aus Technik und Wissenschaft 67 (1992)
13-21.
[52] SCHUG, C.A., Operational Characteristics of High-Pressure, High-Efficiency
Water-Hydrogen-Electrolysis, (11th World Hydrogen Energy Conf., Stuttgart, FRG,
1996), VEZIROGLU, T.N., et ah, Hydrogen Energy Progress XI, International As-
sociation for Hydrogen Energy (1996) 569-578.
stoff einsatz", TAB Working Report No. 13, Biiro fur Technikfolgen-Abschatzung
[54] STARR, C, SEARL, M.F., ALPERT, S., Energy Sources: A Realistic Outlook,
Science 256 (1992) 981-987.
[55] STEINBERG, M., CHENG, H.C., Modern and Prospective Technologies for Hy-
drogen Production from Fossil Fuels, Int. J. Hydrogen Energy 14 (1989) 797-820.
139
[56] STUCKI, S., SCHUCAN, T., Speicherung und Transport von Wasserstoff in Form
organischer Verbindungen, VDI Berichte No. 1129, VDI-Verlag, Dusseldorf (1994)
175-194.
[57] SZAMER, R., Safety Requirements for Small and Medium Sized Electrolysis
Plants, (11th World Hydrogen Energy Conf., Stuttgart, FRG, 1996), VEZIROGLU,
T.N., et al., Hydrogen Energy Progress XI, International Association for Hydrogen
Energy (1996) 2213-2219.
[58] TAKASAKI, K., KOHNO, M., HATTORI, T., Development of Environmentally
Friendly Technology for the Production of Hydrogen, (11th World Hydrogen Energy
Conf., Stuttgart, FRG, 1996), VEZIROGLU, T.N., et al., Hydrogen Energy Progress
XI, International Association for Hydrogen Energy (1996) 61-70.
[59] TAKENAKA, H., R&D on Solid Polymer Electrolyte Water Electrolysis in Japan,
Hydrogen and Clean Energy (Int. Symp., Tokyo, 1995), NEDO (1995) 165-170.
[60] TAMME, R., LEFDAL, P.M., Neue ressourcenschonende H2-Herstellungsverfahren
- Kohlenwasserstoffspaltung und Reforming Prozesse, VDI Berichte No. 1201,
VDI-Verlag, Dusseldorf (1995) 13-26.
[61] TAMURA, H., et al., Experiment on Slush Hydrogen Production with Auger
Method, Hydrogen and Clean Energy (Int. Symp., Tokyo, 1995), NEDO (1995)
219-222.
[62] TOWLER, G.P., LYNN, S., Sulfur Recovery with Reduced Emissions, Low Capital
Investment and Hydrogen Co-Production, Chem. Engng. Commun. 155 (1996)
113-143.
[63] TRAMM-WERNER, S., et al., Photobiological Hydrogen Production Using a New
Plate Loop Reactor, (llth World Hydrogen Energy Conf., Stuttgart, FRG, 1996),
[64] VALENTI, M., Bringing Coal into the 21st Century, Mechanical Engineering 117
(1995) 68-74.
[65] VOLLMAR, H.E., DRENCKHAHN, W., Stationary PEMFCs and SOFCs in the
City of Tomorrow, European Fuel Cell News 5 (1998) January 17-18.
[66] WALDE, H., KROPP, B., Wasserstoff als Energietrager
[67] WANG, D., et al., The Fast Pyrolysis/Reforming Route for the Production of H2
from Biomass and Wastes, (Proc. 7th Canadian Hydrogen Workshop, 1995, Quebec
City), MEHTA, S.K., BOSE, T.K. (Ed.), Canadian Hydrogen Association (1995)
353-364.
[68] WANG, S., LU, G.Q., Carbon Dioxide Reforming of Methane to Produce Synthesis
Gas over Metal-Supported Catalysts: State of the Art, Energy & Fuels 10 (1996)
896-904.
[69] WEAVER, P., Microbes and Bioreactors for Photobiological Hydrogen, World Wide
Web, http://www.eren.doe.gov/hydrogen/micrbes.htm, US Department of Energy
(1997).
[70] WEBER, R., Wasserstoff - Wie aus Ideen Chancen werden, Informationszentrale
der Elektrizitatswirtschaft (IZE), Frankfurt (1991).
[71] WURSTER, R., Hydrogen Research and Development Activities in Europe, Auf
dem Weg zur Wasserstoffenergie - Wie kommen wir weiter?, (BAM-Seminar,
Berlin, 1995), Federal Institute for Materials Research and Testing, Berlin (1997)
11-24.
140
[72] WURZEL, T., Ruhr-Universitat Bochum, Dissertation in preparation and newspaper
article in Handelsblatt, Diisseldorf, June 3, 1997.
[73] YILDIZ, A., Electrochemical and Photoelectrochemical Hydrogen Production, Uti-
lization of Hydrogen and Future Aspects, (Conf., Akcay, Turkey, 1994), YURUM,
Y. (ed.), NATO Advanced Study Institute on Hydrogen Energy System, Kluwer
Academic Publishers (1995) 45-52.
[74] YURtJM, Y., Hydrogen Production Methods, Utilization of Hydrogen and Future
Aspects, (Conf., Akcay, Turkey, 1994), YURtM, Y. (ed.), NATO Advanced Study
Institute on Hydrogen Energy System, Kluwer Academic Publishers (1995) 15-30.
NEXT PAGE(S)
left BLANK
141
Chapter 6
STORAGE AND TRANSPORT OF HYDROGEN
The commercial provision of energy worldwide has turned into a very complex
structure, not limited by national boundaries. A system for grid-bound energy carriers
(electricity, gas) has evolved which represents 35 % of the final energy market. A major
portion of 60 % is characterized as a storageable commodity. Storage represents the
temporal decoupling of generation and consumption. Most countries are dependent on
imports or exports of energy. Primary energies are transported sometimes over several
thousands of km, whereas most of the secondary energies are transported over the shorter
distance range of 50 - 500 km. The delivery to the consumer is usually made in a distribution
system with distances < 50 km. The consumer expects a supply with energy which suits his
spatial and temporal needs ("supply security"). This situation translates then into a question
of storability of energy. Unlike primary energies, secondary energy carriers exhibit high
transportation losses, comparatively poor storability, and higher transportation cost (see
Table 6-1) [39].
Table 6-1: Energy carrier transportation, from [64]
Energy Transportation properties Typical

Energy carrier contents Storage distance
[kWh/kg](1) Speed Losses [km]
capability
Primary energy carriers
Oil 9- 11 slow / fast small good > 103
Natural gas 9- 11 fast high medium >10 2
Hard coal 8 slow small good > 102
Lignite 2-3 slow small good 101 - 102
Secondary energy carriers
Electricity <600kV very fast high none 10 - 104
Gaseous hydrogen 33 fast high medium > 102
Liquid hydrogen 33 slow small poor > 103
Steam <300°C fast high none 1 - 10
Hot water 300 °C (AT) medium high poor 10-1 - 10
Solid biomass 3 -5 slow small good > 10
Liquid biomass 8- 10 slow / fast small good > 103
Gaseous biomass 4 - 7 MJ/m3 fast high medium > 102
(1) unless otherwise indicated
143
6.1. STORAGE OF HYDROGEN
The storage of hydrogen can increase the cost of hydrogen by 30 - 300 %, which
means that there must be some benefit in order to give economic viability [61]. Possibilities
of storage of hydrogen include physical storage via compression or liquefaction, chemical
storage in hydrogen carriers (e.g., metal or liquid hydrides), and gas-on-solid adsorption.
6.1.1. Gaseous Hydrogen Storage
6.1.1.1. Pressure Vessel
Gaseous hydrogen at ambient temperature can be compressed in high pressure tanks

which are easy to handle. Pressures range from < 1 MPa to around 30 MPa. The high
pressure is the dangerous aspect for this type of storage tanks. They must be periodically
tested and inspected to ensure their safety. Pressure vessels are economically operated
only on the basis of short refilling cycles. Pressure vessel materials are low carbon steels
or special alloys or high-strength non-metal composites. The Al/carbon composite is the
lightest and strongest, but also the most expensive. Advanced designs use composite
pressure vessels with a metallic liner and attainable pressures of 30 MPa. Cooling the
compressed gas to cryogenic temperatures ("cryo-hydrogen"), e.g., by using liquid nitrogen,
results in an increase of the storage density by up to a factor of 3.5 [33]. High-pressure
tube storage, however, is confined to relatively small gas volumes.
Typical pressure vessels, e.g., as provided by the German gas company Messer
Griesheim, are 50 1 steel bottles containing 8.9 Nm3 at a pressure of 20 MPa. For a
large-scale consumption (« 250 Nm3), bundles of 28 bottles are used interconnected by
high-pressure piping. Compressed natural gas (CNG) storage systems in cars have an
experience of more than 3 million vehicle years of use with an excellent safety record [21].
The data base for H2-fueled vehicles is very limited thus far.
6.1.1.2. Mass Storage
Natural gas pipeline grids are usually subdivided into a high-pressure grid (> 1.6
MPa), a low-pressure grid, and a third one for special industrial companies only. For gas
storage, the high-pressure grid is used. The pipeline diameter is about 0.8 m and system
pressure around 7 MPa meaning that the pipeline system itself has already an enormous
storage capability.
Seasonal fluctuations of gas supply on interregional basis require a storage capacity

in the order of some 108 to 109 Nm3 which can be economically realized only in no-vessel
underground mass storage used during peak consumption periods: (i) porous storage and
(ii) cavern storage. Geological formations with the porous structure (« 1 pm) of permeable
layers underneath a gas-tight blanket, which can be either an aquifer or a depleted natural
gas resource or a specifically formed cavern in a salt mine. Caverns are usually constructed
and operated in salt formations in which large open void spaces are introduced by a process
where water is pumped through a bore hole into a salt layer and exits as a brine. A cavern
volume is in the order of 300,000 m3. The storage capacity is dependent on void volume
and the allowable pressure which is determined by the strength of the surrounding body.
144
A minimum pressure is required to serve as a support of the void which is independent of
the gas stored and is usually in the range of 15 - 30 % of the total capacity [57]. Maximum
pressures can be up to 4 MPa. Storage rates are in the order of 75,000 Nm3/h, release
rates are in the order of 320,000 Nm3/h. Gas losses are in the order of 1 - 3 % per year.
Typical ratios of the capacities of a 100 km high-pressure pipeline section to cavern storage
to porous storage are 70 to 1300 to 20,000.
The operation of underground natural gas storage sites started in the USA in the
1950s. Today in continental Europe, there are 80 underground storage sites for natural gas
with a total capacity of 40*109 Nm3. In Germany, the first cavern for natural gas started
its operation in 1966. In 1993, 11 caverns, aquifers, and porous storage sites were operated
in Germany for short-term peak consumption with a total capacity of 9*109 Nm3 [17, 57].
A large-scale hydrogen economy would need a similar large-scale underground storage

method which, however, is expected to exhibit higher specific storage cost and somewhat
higher gas losses. The basic feasibility of HI underground storage has been demonstrated.
Experience exists with an aquifer at Beynes, France, operated with no problems from 1957
to 1974 where town gas containing 50 % hydrogen was stored. The capacity was 330* 106
Nm3. Since 1971, the city of Kiel, FRG, uses a salt cavern in 1300 m depth as a storage
site for town gas with 65 % hydrogen contents at pressures of 8 - 16 MPa. In Teeside,
UK, the British company ICI operates three salt caverns in 366 m depth for hydrogen gas
storage with a system pressure of 5 MPa [61]; storage capacity is of the order of 2 million
Nm3 each. Various other porous underground storage sites were in use already during the
coke gas era [15].
6.1.2. Liquid Hydrogen Storage
Hydrogen is in its liquid state (LH2) at a temperature as low as 20 K. It requires about

30 % of its energy for cooling and liquefaction. LH2 must be stored in a superinsulated
tank, called "dewar" or "cryostat", and even then, depending on the insulation quality and
the surface-to-volume ratio, a certain fraction is an unavoidable boiloff to keep the rest
cold. Safety is another concern. Energy can potentially be recovered when the cryogen is
vaporized. The theoretical maximum is 10 % of the liquefaction energy. There are many
opportunities of recovery; its realization is part of downstream engineering consideration
[61].
6.1.2.1. Basic Design of a Cryogenic Tank
A storage tank to hold a cryogenic liquid requires an effective insulation since

temperature difference is the driving force for heat to enter the tank. It usually means a
second vessel around the fluid container [13]. A supporting structure and interconnections
between the two vessels have to assure that the inner shell can freely undergo contraction /
expansion, also for numerous temperature cycles (refills) that may occur during its lifetime.
An even more complicated structural design is necessary for large vessels. Typical materials
used for a cryogenic tank are carbon steel for the outer shell and stainless steel or aluminum
for the inner vessel. Tubing from the inner vessel to the outside is generally made from
stainless steel with vacuum-sealed transition joints on the cold side [51].
145
A multilayer insulation consisting of a reflective foil fixed on the outside of the inner
vessel is usually chosen to minimize the transport of radiation heat With increasing number
of radiation shields, however, additional heat transfer is introduced due to conduction via
physical contact An optimal number is between about 60 and 100 layers [51]. The whole
remainder of the intermediate space, in larger vessels about 1 m in thickness, serves as a
vacuum jacket to avoid heat transport by convection or residual gas conduction. A typical
pressure value in this space is < 0.013 Pa. Appropriate getter materials help maintain the
vacuum over longer times.
An alternative which is used for large capacity tanks, is to fill the annular space
with perlite powder of reflective particles. Tiny spaces in between avoid heat convection
resulting in a less demanding vacuum of < 1.3 Pa. On the other hand, a shifting of
the perlite powder during contraction of the inner vessel, a phenomenon called "perlite
compaction" needs to be prevented in order to avoid breakage of the supporting structure
[13]. A further decrease in heat conduction compared with the powder can be achieved
by using hollow glass microspheres of boron silicate which have diameters in the range of
15 - 150 fim and a very low weight [51].
In large containers, the boiloff rate as a result of heat leakage can be significantly
reduced by cooling the insulation with cold venting gas. For this purpose, metallic shields
are installed with the insulation. In smaller containers, also liquid nitrogen cooling is
being successfully used for boiloff reduction, but they are not as handy as the vapor-cooled
container because of their weight and periodic refilling [51]. Multilayer super-insulation
tanks exist at sizes of 300 m3 with around 100 radiation shields. Storage and transport
containers for LH2 have been developed practically for all sizes. The autonomy of the
system, i.e., the time with no loss of the product, is typically 30 days.
Further improvement is foreseen with the introduction of high-temperature supracon-

ducting materials which allow the contact-free placement of the inner vessel in the outer
con tamer ("magnetic levitation") [33].
6.1.2.2. Passenger Car Tanks
The schematic of an LH^ (or LNG) tank for a passenger car is shown in Fig. 6-1.
The tank materials typically used are austenitic steel and cold-tough aluminum alloys. A
minimum wall thickness of 2 mm is prescribed in Germany. The tanks are checked upon
fabrication in leakage and pressure tests [50]. Present cylindrical double-walled storage
tanks have a capacity of up to 150 1. The 30 mm thick insulation consists of 70 layers of
aluminum foil and glass fibers allowing a boiloff loss of not more than 1.5 %/d. The total
weight is 50 - 60 kg. Maximum operation pressure is 0.6 MPa.
An extensive experimental study as a part of the Euro-Quebec project has been

conducted by BMW to investigate the safety of LH2 storage tanks for passenger cars
[50]. In particular, vibration and acceleration testing, air flooding of the vacuum, vessel
deformation and perforation, and fire testing have been made to check the stability of the
supporting structure and the performance in conceivable traffic accidents (see Fig. 6-2).
Burst tests revealed a safety factor for the LH2 tank walls of more than 14. The results
have been used for design improvement. Preparation of the tanks with a safety burst
146
solenoid shut off solenoid control valve superinsulation
safety
vent line
vacuum
valve/
and burst disc
mechanically
optimized
suspension
liquid level gauge

for pressure regulation
Fig. 6-1: Schematic of LH2 tank for passenger cars, from [Messer Griesheim]
f Safety tests with liquid hydrogen storage vessels
——— t I
Acceleration Vibration Flooding of vacuum
Deformation Perforation Firebrand
Fig. 6-2: BMW safety testing on LH2 (or LNG) tanks for passenger cars, from [42]
device, a circular notch, prevented the catastrophic failure and rather dampened flow and
pressure relief, but its reliability over the whole lifetime has yet to be proven. Fire tests
meeting IAEA requirements have exposed the tank to a temperature of 900 °C. Safety
valves opened after 4 minutes. The LH2 was completely vaporized after 10 more minutes
before the inner aluminum tank started to melt while the outer tank hardly exhibited any
changes. The tanks have demonstrated good-natured behavior during depressurization via
the safety valve. Neither test resulted in LH2 pool formation.
147
6.1.2.3. UJ2 Model Container for Maritime Transportation
Within the frame of the Euro-Quebec project various technical concepts for the
transportation and distribution of liquid hydrogen have been developed. A 61 m3 model
container has been constructed in a joint effort between the Federal Institute of Materials
Research and Testing (BAM), Berlin, responsible for infrastructure and tank operation, and
the Germanischer Lloyd (GL), Hamburg, taking care of the scientific program including the
evaluation of the test results. Its purpose is to demonstrate the safe storage and transport
of large quantities of LH2 [69]. It serves as a model simulating a future tank for maritime
transport and storage. The cylindrical double-walled tank has a diameter of approx. 5 m
and a length of 9 m with a designed system pressure of 0.5 MPa. The insulation between
inner and outer tank consists of 30 layers of aluminum foil at the inside plus another 0.5
m vacuum space. For instrumentation, 67 temperature and pressure sensors are deployed
to examine the thermal behavior of the tank.
Tests were conducted, the first with an LH2 tank of such a size, to investigate the
thermodynamic behavior of the liquid hydrogen in the tank. One important goal was to
examine the pressure rise in the closed tank. Three test series were conducted [52, 69]:
• Open vaporization at atmospheric pressure to determine the energy input into the
tank from the boiloff quantities and provide a homogeneous temperature distribution
for subsequent tests
• Closure of tank and measurements of pressure increase rates simulating conditions
of a maritime transport
• Intentional pressure increase by means of a vaporizer device to investigate the
energy input by hydrogen preheated in a heat exchanger
The boiloff rate was measured to be approx. 1.3 %/d (about 0.7 m3/d) for the filled
open tank, while it was decreasing to 0.6 %/d at filling levels below 40 %. The decrease of
the boiloff rate is attributed to the stronger heat uptake of the steel at higher temperatures
which occur when in contact with the gas phase. In the self-pressurization tests, mass
transfer from liquid to gas was found to have a dominant influence. The tests with the
model tank on the BAM test site near Berlin have been terminated in the meantime, the
tank is now being offered to third parties for further experimentation.
The fluiddynamic processes in a tank are very complex and are currently subject
of investigation. A calculation model has been developed to enable the simulation of all
conceivable states of the tank. The heat transfer from outside into the inner tank was
measured to be 200 W whereas a value of 97 W was predicted [69].
6.1.2.4. Large-Scale Stationary Storage Tanks
For a large-scale liquid hydrogen storage, tanks of spherical shape are used to
minimize boiloff losses. Two spherical dewars with a capacity of 190 m3 of LH2 plus
10 % ullage each were components at the Nuclear Rocket Development Station, Nevada,
during the ROVER project starting in 1955 (see section 9.5.1.). The insulation consisted
of a vacuum jacket and a 0.9 m perlite powder layer allowing an estimated boiloff rate of
148
0.3 %/d [11]. Later a 1900 m3 LHa dewar was in use with an evacuated perlite insulation
and a boiloff rate down to 0.08 %/d.
The largest LH2 tank constructed so far is the NASA 3407 m3 vacuum perlite-
insulated spherical storage tank at the Kennedy Space Center, Florida, used in the US
space shuttle program. The outer sphere is made from carbon steel with an inside diameter
of 21.34 m and the inner sphere is made from austenitic stainless steel with an inside
diameter of 18.75 m; the ullage is about 10 %. The tank has a boiloff rate of 0.03 %
or approx. 800 1 per day [51]. For comparison purposes: the LH2 storage tank within
the External Tank (47 m height, 8.4 m diameter) of the US Space Shuttle has a volume
of almost 1600 m3.
The largest stationary LH2 storage in Japan is at the Tanegashima space center in
two spherical double-walled tanks with an outer diameter of 12.6 m and a capacity of 540
m3 each.
The Baikal- 1 nuclear rocket test facility in Semipalatinsk, Kazakhstan, is reported

to have constructed in the 1970s three large underground tanks for LH2 storage with an
estimated 18 m in diameter. In 1982, one of the tanks developed a leak and became
unusable [53],
6.1.2.5. Future Storage Tank Designs
A concept of a composite LH2 container has been developed in Canada where

the outer shell is fabricated from fiber reinforced plastics. The tank volume is 60 m3.
Such vessels of which a first set has been manufactured already, are to be used for the
transportation of smaller amounts of LH2 onboard conventional container ships, railway
carriers, or as a storage medium near power or refueling stations [34, 35].
The LH2 storage container considered for the Euro-Quebec barge carrier is designed
as a cylindrical pressure vessel type with a volume similar to that of the NASA tank.
Geometric data are a length of 22 m outside and 18.7 m inside, a diameter of 14.7 m
outside and 14.0 m inside. The total weight is 1000 tons including 213 tons of LH2- To
keep the vessel at cryogenic temperatures, 5 % of the LH2 contents is left in the vessel
on its way back to Canada. The tank must guarantee a stand-alone period of more than
25 days. The containers are designed to serve both as a transportation vessel and as an
onshore storage vessel in order to avoid the undesired losses of LH2 if it were to be pumped
from one into another container. This concept eases at the same time the operation of the
ships [18, 35].
Improved concepts of insulations have been developed for future large-size

storage tanks, e.g. a spherical tank with inner / outer diameters of 36 m and 40 m,
respectively, which means a volume of about 24,400 m3 or six times the NASA tank
volume. The characteristic feature of the new insulation design is the partition of the
interspace volume of approx. 9000 m3 into many (some 104) small insulation boxes which
can be arranged as a combination of vacuum and foam layers on the inner tank wall. The
insulation character is maintained even if a single box has lost its vacuum. Target boiloff
is 0.1 % per day. Experimental and theoretical investigations have been conducted for
149
design optimization, in particular, the design of the supporting structure which is subjected
to dynamic loads during maritime transportation. The new insulation concept was found
to exhibit a thermal conductivity which is on the average one order of magnitude lower
than that of conventional foam insulations [40].
Tank specifications for a large-scale LH2 storage tank as projected within the WE-
NET project, are a storage capacity of 50,000 m3 with a target boiloff rate of 0.1 % per
day. Various insulation concepts are currently being under investigation [32].
Also within the WE-NET project, a metal hydride/LH2 hybrid storage system is
proposed, where a hydrogen storage alloy serves as a buffer for a liquid hydrogen storage
for the sake of ensuring a stable supply plus it absorbs flash hydrogen and boiloff losses
[47].
6.13. Hydrogen Storage as a Metal Hydride
Metal hydride (MeH) is an intermetallic compound that traps hydrogen at moderate

pressures. Large amounts of hydrogen can be stored in an exothermal diffusional process.
Me + n/2 H2 -> MeHn + heat
The gas/solid reaction can take place at any temperature in the wide range between
-80 and +600 °C. Storage temperature is at ambient, storage pressure between 3 and 6 MPa.
Heating is required to restore the hydrogen. Some drawbacks are its cost and its heavy
weight, and its requirement for very pure hydrogen. Metal hydride advantages include
convenience, compactness, stable storage, and its intrinsic safety. Its specific storage
capacity of 700 Nl of H2 per 1 of system volume is higher than that of liquid hydrogen. In
order to obtain alloys suitable for different applications, the properties of hydrides can be
controlled by changing the composition and microscopic structure or by treating the surface,
e.g., substituting or adding other metal elements. Examples of appropriate materials are
FeTiH, LaNiHs, MgH2, or MgNiKU. Container designs vary from simple low-pressure
vessels to high-pressure, high-temperature heat exchangers. After repeated use, the metal
alloy disintegrates into very fine powder which is toxic. Current research activities are
concentrating on improving the performance both of the alloy, e.g., increase of capacity,
lifetime, reactivity, and of the container, e.g., decrease of volume and weight, improvement
of durability.
A new class of hydrogen absorbing alloys are nanocrystalline metal hydrides. These
are polycrystals with a typical grain size < 20 nm. Their advantages are to maintain their
integrity upon cycling and to show enhanced absorption / desorption kinetics. Examples
of materials which can be synthesized easily by mechanical alloying, are FeTi and Mg2Ti.
A major drawback is the reduced maximum capacity [58].
Metal hydrides can serve as a purification system for hydrogen. Charging a hydride
container with impure hydrogen allows only the hydrogen to reversibly react with the alloy
while the impurities are concentrated in the void spaces. Venting a small fraction of the
hydrogen removes the impurities and makes a flow of ultrahigh purity hydrogen available
[49].
150
Ni-metal hydride batteries are a highly promising approach to a small and medium-
scale energy storage for portable appliances. The first of its kind were designed for space
applications in the 1980s. The basic electrochemical net reaction is
H2 + 2 NiOOH <—> 2 Ni(OH)2
The Ni-hydrogen battery has the advantages of high capacity, long lifetime, high
energy and power density, recyclability and thus environmental acceptability. Typical
energy densities are 50 - 70 Wh/kg [37]. The NiMeH battery is rechargeable up to
1200 times. Disadvantages are a high self-discharging, long charging times, its sensibility
towards temperature and over-charging as well as complex and expensive manufacture [5].
A storage device that couples a high-temperature hydride to a phase change material

has been tested by Arthur D. Little Inc. This system uses the heat released during
hydrogenation to melt the material. In the reverse process, the released solidification
energy is taken to liberate the hydrogen from the hydride compound. The material used
is a Mg powder coated with Ni by means of chemical vapor deposition. The system was
successfully tested as an essentially isothermal and isobaric process [29].
6.1.4. Others
6.1.4.1. Glass Microspheres
New high-strength glass microspheres with 25 - 500 fim in diameter and with 1
fj,m wall thickness are loaded in a high-pressure bath of hydrogen (« 60 MPa) at higher
temperatures where the tiny bubbles are highly permeable to hydrogen. After cooling the
hydrogen is trapped. Once stored, the pressure in the carrying tank can be low. They are
inert, resistant to contamination and only require moderate heat to release the hydrogen.
Experiments are being done with crushing the spheres to increase the hydrogen release
rate. The most advanced glass microspheres exhibit a burst pressure of 1000 MPa [38].
Engineered microspheres could make storage and transport of large amounts of hydrogen
much more efficient and, if transported by rail, even could economically compete with
liquid hydrogen [55].
6.1.4.2. Zeolites
Zeolite is an aluminum-silicate with a 3D structure which contains pores and voids.

The storage principle of zeolites is that guest molecules (= hydrogen) are forced into the
hollow spaces of the host material at high temperatures and high pressures. After cooling
and relaxation, the hydrogen is encapsuled in intracristalline voids. It is recovered in a
heatup process [66]. Tests have shown that no permanent storage was possible unless
cooled down to temperatures < 120 K. Then the storage capacity was found to be 45 cm3
151
hydrogen per gram dry zeolite (corresponding to 0.4 wt%). Experiments at the Taiyuan
University of Technology in China have reached 1.9 wt% [66]. Zeolites can be effectively
used for purification of hydrogen gas.
6.1.4.3. Gas-on-solid Adsorption
Activated Carbon is prepared from special biological or mineralogical carbonaceous

materials by chemical activation, low-temperature (400 - 500 °C) carbonization to remove
volatile orgam'cs, and high-temperature (800 - 1000 °C) oxidation to develop porosity and
surface areas of 300 - 2500 m2/g [3]. The effect of physisorption of the hydrogen is most
pronounced at cryogenic temperatures. Adsorption properties are dependent on the pore
size distribution. The principal capacity is originating in particular from the micropores
(< 2 nm). The mesopores' contribution is not more than 5 - 10 %, whereas that of the
macropores (> 50 run) is insignificant. A volumetric density of 50 g/1 of H2 has been
verified at 77 K and 3.5 MPa [7]. One of the best currently available carbons contained
in a carbon fiber-wrapped pressure vessel at 87 K and 6 MPa would store hydrogen at 4.2
wt% and 16.8 kg/m3 [3]. Latent heats of sorption must be removed during charging and
supplied while discharging the container. Equilibrium times are less than 5 minutes.
A recently discovered option of hydrogen storage is adsorption on a material known

as carbon nanotubules. These tiny tube-like configurations have a very large surface per
unit weight, minimal macroporosity, uniform and tailored microporosity, and high thermal
conductivity [3]. The amount of hydrogen stored was observed to be 10 wt% on the single-
wall nanotubes (SWNT) weight basis. SWNTs are found, e.g., in arc-generated soots.
Acting adsorption forces could lead to effective H2 storage under ambient temperature
conditions. Research is concentrating on SWNT production at a high yield [10].
The carbon sorption system is claimed to offer higher gravimetric and volumetric
density of pressurized hydrogen with minimal cost as well as a high degree of dormancy and
safety. Efficiency, however, is reduced when introducing impurities. The need of cooling
system and vessel working under high pressures and low temperatures may complicate the
whole system [45].
A new storage system consisting of graphite layers was reported in early 1997 by
US scientists to store 30 1 of molecular hydrogen per gram. This corresponds to about 75
% hydrogen storage per weight, which could represent a breakthrough compared with the
figures of 4 % valid for nanotubes. The material is a densily packed structure of graphite
fibers of 5 - 100 mm length and 5 - 100 nm diameter. Storage pressure is about 4-5
MPa. Filling time is between 4 and 24 hours. This tremendous storage capability is still
unexplained; it would give a hydrogen-fueled car a range of as much as 8000 km (!) [26].
Attempts will be made in near future to verify these results with masses of 200 mg to 1 kg.
6.1.4.4. Liquid Carrier Storage
Hydrogen can be reversibly stored in a chemical compound in a liquid state with,

for example, methanol, toluene (methylcyclohexane), ammonia as the carrier from which
152
it must be extracted (dehydrogenation) prior to utilization unless the products are used
directly (see section 6.3).
6.1.5. Comparison of Storage Systems
Table 6-2 summarizes pros and cons associated with the various types of hydrogen
storage. The Bentley plot in Fig. 6-3 gives some perspective on volumetric versus
gravimetric energy densities of different hydrogen storage systems (which include vessel,
heat exchangers, insulation etc.) compared with conventional liquid fuels and future storage
designs [28]. The DOE goal for vehicular hydrogen storage systems is 6.5 wt% and 65
kg/m3. If the above mentioned recently discovered graphite material with 75 % hydrogen
storage per weight and its technological feasibility could be verified, it would mean a
tremendous qualitative step forward.
200
New graphite fibers •
1 Standard ISO 40 ft
Gardner Cryogenics Diesel
2 L'Air liquide
100- 3 EQHHPP Tank design
4 WE-NET Tank design Gasoline
X
50-
^——^x. rcryo-compressed (80
O)
Hydride \ •• Carbon
Carbc nanotubes
oo
systems ) * Long-term battery
c
O
T3
.0
20-
v!!^4- adsorption I I W Advanced
compressed
"C
"53 (150 K)
Advanced
E compressed
10- • Mid-term battery (ambient
temperature)
i
0.5 2 5 1 0 2 0 50 100
Gravimetric density (% HJ2) ——>
Fig. 6-3: Gravimetric and volumetric hydrogen (and other fuels) storage systems,
from [28, 33, 71]
153
Table 6-2: Advantages and disadvantages of hydrogen storage technologies, from [3]
storage type Advantages Disadvantages

Compressed gas Mature technology Bulky
High gravimetric density compression costs
dormancy safety
Cryogenic Liquid High volumetric and Liquefaction costs
gravimetric density Boiloff loss
Purity Poor performance in small systems
Metal hydride High volumetric density Poor gravimetric density
Safety High dissociation temperature
Impurities
Gas-on-solid adsorption Fair volumetric and Compression and cooling costs
gravimetric density Complexity
Potentially low costs
Chemical storage High gravimetric density Toxicity
Unlimited storage Impurities
Rehydrogenation reaction
Conversion limitation
6.2. TRANSPORT AND DISTRIBUTION OF GASEOUS AND LIQUID HYDROGEN
Larger distances between the locations of heat production (power plant) and heat de-
mand (consumer) require tanks and methods of transportation of heat/energy to distribution
systems.
6.2.1. Pipeline System
6.2.1.1. Gas Pipeline Grid
In most countries, gas pipeline systems serve both transportation and distribution
purposes. An exception is the Russian Federation whose natural gas pipeline is a pure
long-distance transportation. An important component of a gas grid are the compressor
stations to maintain the necessary system pressure. Their energy consumption corresponds
to about 0.2 - 0.5 % of the amount of gas transported per compressor station which amounts
to the order of 10 % for the long-distance gas transportation [39]. Design specifications of
a future natural gas pipeline are a diameter of 1.6 m, the deployment of an inner coating, a
system pressure of 12 MPa, and an annual transport capacity of 54* 109 Nm3. Compressor
consumption will be down to 5 % on a length of 6000 km [57].
The net heat of combustion of hydrogen per unit volume is by a factor of 3.3
lower than that of methane. The transport of the same energy quantity thus requires
a correspondingly increased compression for H2 propulsion, in reality a factor of 3.3 -
3.8. Differences between natural gas and hydrogen are given by different energy transport
154
capacities (Wobbe index) and the different compressibility behavior as a function of
pressure resulting in higher friction losses for hydrogen. Furthermore the H2 compression
necessitates multi-step compressors whereas natural gas compression is usually designed
for one step. A simplification is given in that no pre-heating is required because of
the positive Thomson-Joule effect liberating heat upon hydrogen depressurization [15].
Economic operation implies the transport of very large quantities.
The transport of gaseous hydrogen in a pipeline system is a technology, which has

long been applied. More than 750 km of hydrogen pipelines (as of 1990) have been put in
place worldwide [43], an estimation of 700 miles is given in [46]. The German company
Huls AG has started as early as 1938 to operate a H2 distribution grid in the Rhine/Ruhr
region. This pipeline system, since 1993 being operated by the British BOC, with a total
length of 250 km and connecting 14 H2 producing and consuming companies, is the largest
in Europe. The hydrogen is provided by steam reforming and as a byproduct from the
chlorine production by the chlorine-alkali electrolysis process. The pipe diameter varies
between 0.17 and 0.27 m and the throughput is 250* 106 m3/yr at a system pressure of
2.2 MPa. Since 1969, the US company Air Products is operating a 232 km hydrogen
pipeline in Houston, USA, with a pipe diameter of 0.15 - 0.2 m diameter at a pressure
of 6 MPa. The company operates two more Kb grids in New Orleans, USA, with 96
km and in Rotterdam, The Netherlands, with 48 km. L'Air Liquide in France operates
since 1966 a H2 pipeline network which has reached today an extension of about 600 km
connecting North France, Belgium, and The Netherlands. Another one in South Africa
is of 80 km length [43]. Circumferential stresses of < 56 N/mm2 and the relatively low
degree of purity of the piped hydrogen are reasons for the good performance of the existing
H2 pipelines so far. Long-distance pipeline transportation of hydrogen gas has not reached
an intercontinental dimension up to now.
The use of the existing natural gas pipeline grid for the transportation of pure hydrogen
would require tremendous modification efforts. But the technical problems of increased
propulsion or material degradation appear to be solvable [15]. Safety requirements for H2
pipeline transportation are more stringent compared with natural gas. The question whether
the existing natural gas pipeline grid can be used for hydrogen, cannot finally be answered
since the presently used steel, StE480.7TM, has not yet been proven to guarantee safe
operation under hydrogen transportation conditions. Another disadvantage is that hydrogen
is more prone to leaking than natural gas; grid losses are expected to be 3 % for hydrogen
compared with 2 % for natural gas. If hydrogen, however, is added to the natural gas up
to a fraction of 5 vol%, the existing pipe infrastructure can be used without changes [39].
Minimal changes are required for blends with up to 20 %.
The project of a "Trans-Asian Gas Pipeline Network" has been proposed connecting
natural gas fields with consumer markets of 19 supplying, consuming, and transit countries
with a total length of 46,200 km. It is small compared with the existing NG pipeline grids
in the European Community with 800,000 km (1990) or in the USA with 440,000 km. The
Asian pipeline is designed to transport natural gas in first stage, but is apt to be converted
to hydrogen in the future [25].
Summarizing, it can be stated that a large-scale hydrogen transport and distribution

is technically feasible taking account of the peculiarities of hydrogen (physical, material-
155
specific) compared with natural gas. Transportation cost is expected to be in the same order
as natural gas for regional distribution and to be higher for a large-scale transportation [15].
In the long term for large geographically concentrated demands, pipeline distribution might
ultimately yield the lowest delivered fuel cost [48].
6.2.1.2. Pipeline Grid for Cryogenic Liquids
An alternative to the gas transport is liquid natural gas (LNG) transportation. A

much higher density is gained at the expense of additional losses due to liquefaction, ship
transport, and boiloff, estimated to be in the range of 12 - 20 % [39]. Relay bases as
distribution systems are being operated for LH2, LNG, and LPG. These satellites bases are
usually smaller than the primary bases.
Pipeline transportation of liquid hydrogen is realized on a small scale and short

range. Stainless steel is usually taken for the inner line with low heat conduction spacers
as a support in the vacuum jacket. The Kennedy Space Center in Florida uses an LH2
and LOX pipeline of 500 m length with an inner pipe diameter of 0.15 m. Flow rates
achieved are up to 250 m3 LHa per minute and 100 m3 LOX per minute, respectively [12].
Transfer is realized by applying pressure, no pumps. Major concerns besides heat leakage
is the mechanical stress imposed on the inner line due to contraction / expansion, pressure
oscillations upon cooldown, or two-phase flow.
In the German city of Jena, a 5 km liquid helium pipeline is operated to provide

consumers with 25 m3/yr of LHe which is at a temperature of 4 K.
6.2.2. Vehicle Transportation
The type of storage system to be chosen largely depends on size, driving cycle, and
range of the vehicle. A low mass is required, if acceleration plays a significant role, e.g., in
aviation, while it is of less importance for trains or ships. Also tank geometry for onboard
Hj storage needs to be compatible with the vehicle structure aiming at a minimum of
surface-to-volume ratio. The lifetime of the storage system should suit the vehicle lifetime.
6.2.2.1. Road Transportation
Loads of LHi tank trucks are in the range of 30 to 70 m3 with boiloff rates of 0.3 -
0.5 % per day. Due to restrictions in the tank size, the upper limit of the diameter is around
2.5 m. Cylindrical vessels are used with surface to volume ratios of only 10 % more than
that of a sphere [59]. Messer Griesheim trucks have a transport capacity of 30,800 m3 (at
0.1 MPa, 15 °C) of hydrogen, corresponding to 42 m3 of LH2- Road transport capacities
of Gardner Cryogenics are ranging from 5.7 up to 64 m3 of liquid hydrogen with normal
boiloff losses less than 0.2 % per day.
In 1994 hydrogen transportation on European roads amounted to 311.2 million Mm3,

about 40 % of which were transported in Germany translating into 50,000 deliveries per
year or about 200 per working day [62]. The number of liquid hydrogen deliveries is
lower, estimations are approx. 2300 LH2 deliveries per year to customer sites. In North
156
America, the number of LHa truck deliveries is estimated to be 15,000 per year providing
approx. 180 (metric) tons of LH2 to customers [8, 63].
Substituting conventional gasoline fuel by liquid hydrogen would increase the number
of tank truck transports for distribution by a factor of three.
Gaseous hydrogen transportation is generally realized in 20 MPa pressure vessels.

Common practice is the use of trailers loaded with either gas bottle bundles with a capacity
of 4000 Nm3 or a set of 9 large bottle vessels with a total capacity of 4300 Nm3. A new
development of a trailer is now in operation holding 200 single bottles with a transport
capacity of 6000 m3 (at 0.1 MPa, 15 °C) of hydrogen and a filling pressure limited to 20
MPa. In Germany, several hundred hydrogen trailers are being operated [62]. In North
America, the amount of hydrogen gas transported by truck is estimated to be around 10
% of the liquid volume [8].
6.2.2.2. Ship Transportation
LNG maritime transportation is a since long applied technology on a large-scale

base. In 1992, 76 LNG tank ships were operated with an average capacity of 100,000
m3 of LNG [35]. Commercial transportation of liquid hydrogen via pipeline does not
appear realistic [68]. A tank ship transport of LH^ is reasonable especially if transported
over long distances. The comparatively low volumetric energy density of LH2, however,
requires very large tank ships and a much higher fraction of the hydrogen transported is
needed for the transport itself and is estimated to be in the range of 23 - 38 % of the energy
transported compared with about 5 % for an LNG tank^ship [68]. In terms of safety, LH2
ship designs may principally capitalize on the broad long-term experience with LNG and
LPG maritime transportation.
Liquid hydrogen barges for fuel supply in the French and US space programs.
Barges with 900 m3 of LH2 storage containers were used during the NASA Apollo project.
The European Ariane project was supplied with LH2 by overseas transportation from New
Orleans to Kourou, French Guiana, in 20 m3 storage vessels with either LN2 or vapor cooled
multilayer insulation [51]. Maritime LH2 transports to Kourou have been discontinued with
beginning of the on-site liquefaction plant operation in 1990 [6].
The barge carrier considered within the Euro-Quebec project to transport LH2 to
Europe is designed as a dock ship with a total length of 180 m and a width of 29 m
carrying 5 barges to contain a total of 15,000 m3 of LH2- Its remarkable safety features
are a double-walled hull, a distance of 6.5 m between side shell and barge, and a distance
of 11 m between ship bottom and vessel. The ship is powered by 10 MW diesel engines
and reaches a speed of w 18 kt18 (which is about 33 km/h). The LH2 tank consists of
two concentric vessels with a total length of 22 m and 18.7 m, respectively, for the outer
and inner vessel and with an outer/inner diameter of 14.7 m and 14 m, respectively. The
net volume of the tank is 3047 m3 (corresponding to 213 tons of LH with an ullage of
15 %. The cargo is stored at an initial pressure of 0.125 MPa; the upper design limit for
the tank is 0.5 MPa [18, 35].
1 kt - 1 knot - unit of speed of one nautical mile an hour - 1.852 km/h.
157
SWATH ship Dock ship
Fig. 6-4: Liquid hydrogen tank ship design of the SWATH ship and the dock ship,
from [35]
Primary energy (hydro) Etot =100
Etr= 104 Efficiencies
10
Hydro conversion
= 09
Electrolysis
Liquefacton
137
£kq = 0 25
48
Fuel Available energy

con-
sump-
tion Exergetical efficiency
= 042
EO-, Ei,q - energy portion of transportation, production, liquefaction

w
ei» whq = work requirement for electrolysis, liquefaction
CH2 - chemical exergy, Aeuq = thermal exergy
Fig. 6-5: Exergy flow diagram for SWATH ship transportation of liquid hydrogen
based on Etot corresponding to 310 MJ/kg H2, modified from [35]
158
Two follow-on LH2 tank ship designs, the SWATH (Small Waterplane Area Twin
Hull) ship and the dock ship, have been developed by the German companies Howaldts-
werke Deutsche Werft, Noell-LGA Gastechnik, and Germanischer Lloyd. Both ships have
a load capacity of 125,000 m3 corresponding to a weight of 8,150 tons (see Fig. 6-4). The
ships are hydrogen-fueled for which liquid hydrogen as well as the unavoidable boiloff los-
ses (« 0.1 % per day) are used. The SWATH ship is superior to the dock ship in terms of
lower weight and lower energy consumption per sea mile. The SWATH ship with a length
of more than 300 m carries four spherical LH2 tanks. As the propulsion system, a gas
turbine with steam injection of 41 MW has been proposed. This system allows to regain
a part of the technical work that was introduced into the liquefaction process ("Enhanced
Cryogen Exergy Recovery System", ECERS). Waste gas energy is utilized for steam gene-
ration. The efficiency is estimated to be 51 %. Although this type of propulsion system is
known in its single components, it has not yet been realized. However, the technical feasi-
bility was proven [68]. A flow diagram of the exergy (= energy that can be converted with
no restriction into another form of energy) for a maritime LH2 transport with the SWATH
ship is shown in Fig. 6-5. The overall exergetic efficiency is estimated to be 42 % [35].
Studies on a future hydrogen sea transportation system in Japan is based on its broad
experience on large-scale LNG transportation with the first tanker ship starting in 1959.
Within the Japanese WE-NET project, two preliminary concepts of a cryogenic tank
ship have been proposed based on the assumption of fueling a 1000 MW(e) power plant
which consumes 1200 t/d of LH2- Furthermore assuming two ships to travel 6000 nautical
miles within 10 days, the load capacity will be 200,000 m3 of LH2- The two concepts
are distinguished by different tank structure designs. One is the ISI SPB self-supporting
prismatic tank system, the other one the MOSS self-supporting spherical tank system with
capacities of 100,000 and 50,000 m3, respectively. The boiloff losses of estimated 0.2 - 0.4
% per day are designed to be utilized for the.H2 propulsion system of the twin hull ship,
whose power is calculated in first approximation to be provided by two 30 MW engines
to allow a speed of 20 - 25 kt (« 37 - 47 km/h) [1].
The Japanese Ship Research Institute in the Ministry of Transport has developed
within the PORSHE project (see section 9.2.4) a concept of a 125,000 m3 LH2 tank ship.
An internal reheat hydrogen gas turbine has been designed as an engine for an ultra-high
speed marine vehicle. The engine must have a power of > 300 MW to transport a total
weight of 5000 t at a speed level of 100 kt (w 185 km/h) [24].
6.2.2.3. Railway Transportation
Transportation of LH2 in rail cars has started in the 1960s by the Linde company
using a 107 m3 tank with a multiple layer insulation. The measured boiloff rate was 0.2
%/d. The US company Praxair operates a fleet with 16 hydrogen railcars. The hydrogen is
saturated at a working overpressure of the tank of 55 kPa with a pressure control system
to open the relief valve at an overpressure of 117 kPa. The amounts of LH2 transported in
railcars in recent years were approx. 70 tons or about 10 railcar loads per month [8].
159
6.2.2.4. Aircraft Transportation
The transportation of liquid hydrogen as a payload onboard an aircraft is being

considered an alternative delivery scheme to the ship transport including the option of
a ciyo-fueling of the aircraft itself. The advantages of air transportation is the light
weight of LH2, a reduced boiloff loss, the short duration of an aircraft roundtrip, the direct
accessibility of the production site, and the reduced buffer storage required. Drawback is
the large number of deliveries in a large-scale supply connected with an extensive ground
infrastructure [41] and last, but not least, the lack of any practical experience.
In phase I of the Euro-Quebec project, air transportation of LH2 across the Atlantic
was considered a possible option. The full payload potential of an Airbus A3 00-600 would
cover the entire daily production of 42.7 t of LH2- The overall energy yield, however, was
estimated to be down to 35 % compared with a 10 points higher yield for LH2 transportation
by ship. The fact that the energy demand of transportation in an aircraft was as high as
the payload, was deemed unacceptable [9].
Within the WE-NET project the aircraft transportation capacity based on 5000 km
distance was investigated in a study to correspond to a payload of 454 tons and a speed
of 400 kt (« 740 km/h) to allow two airplanes to operate a 1000 MW power plant. For a
payload of 1500 t and a speed of 300 kt (« 560 km/h), one plane would be sufficient [47].
6.3. CHEMICAL ENERGY TRANSMISSION SYSTEMS
The key components of a chemical energy transmission system (CETS) or a chemical

heat pipe are a primary energy source (fossil, renewable, nuclear) to provide heat to an
input catalyst reactor. In this reactor, a mixture of appropriate materials is chosen for the
desired endothermal reaction, i.e., to store energy in the newly created products. After
transportation of the product gas mixture at ambient temperatures, the reverse exothermal
chemical reaction in an output catalyst converter helps to extract the stored heat for
consumption as high-grade or low-grade heat or in a turbine. In a closed cycle, product
materials are then returned to the heat source. Unlike liquid hydrogen, chemical compounds
with hydrogen provide an economic method of a seasonal storage of energy [60].
The reactions must be reversible, rapid, and as complete as possible to avoid unwanted
secondary products. They should have a high efficiency and a high reaction enthalpy in
order to minimize the masses to be transported, plant and pipe sizes as well as operating cost.
They must be controlable to allow intermittent storage/retrieval operation. Furthermore,
educts and products should be inexpensive, non-corrosive, non-toxic, and easy and safe to
handle [19]. CETS have to compete against hydrogen pipeline systems or the electricity
grid.
Several candidates for a closed-loop reversible chemical energy system with varying
operating temperature ranges and storage capabilities have been detected and are listed
in Table 6-3 [20]. Certain limitations of the long-distance heat transport are given by
its restricted storage capabilities, by the limited temperature ranges and by the complex
distribution systems required [2].
160
Table 6-3: Organic systems with hydrogen as the energy carrier as GETS candidates,
from [20]
Temperature Reaction
Closed-Loop CETS range enthalpy
[kj/mol]
CO + 3H 2 «— > CH4 + H2O 700- 1200 250
2CO + 2H 2 <— » CH4 + CO2 700- 1200 247
C2H2 + H20 «— > CH3CHO 134
C2H4 + H2 <—— > C2H6 138
C6H6 + 3H 2 «—» C6H12 500- 750 207
C7H8 + 3H 2 <— » C7H14 450- 700 215
C7H8 + 4H 2 <— > C 7 Hj6 252
Ci0H8 + 5H 2 «—»- C10H18 450- 700 359
Some of the important CETS are

• steam reforming system with natural gas
theoretically and experimentally investigated and demonstrated at the Research
Center Julich, known as the EVA/ADAM system to prove the ability of the HTGR
as a primary energy provider,
• CO2 reforming system with natural gas
studied as a closed cycle in Israel based on solar primary energy,
• methanol system
as CO2-neutral cycle, if atmospheric CO2 is used as the carbon source,
• methylcyclohexane (MCH) - toluene cycle
as transportation or seasonal storage method considered as one option in the Euro-
Quebec project and the WE-NET project,
« ammonia system
investigated in Australia as a closed cycle system based on solar primary energy.
The above CETS systems are described in more detail in appendix B.
Japan has made a comparison of the transmission efficiencies (= ratio of output

energy to input energy) for various hydrogenation - dehydrogenation systems. Based on
the assumptions of 4000 MW hydro power, electrolysis with a 90 % efficiency, and a
maritime transportation distance of 5000 km, transmission efficiencies have been derived as
listed in Table 6-4. The table, however, does not reflect production cost or environmental
aspects [70].
Within the conceptual design of the WE-NET project, a comparison was made
between the methanol system and the liquid hydrogen system. The energy balance over the
whole path from electricity over the steps electrolysis, methanol synthesis, transportation,
reforming to hydrogen, combustion turbine power output reveals an overall efficiency of
161
Table 6-4: Transmission efficiencies for various types of vectorization, from [16, 70]
Energy
Energy Storage Working Transmission
needed for
Vector density capacity temperature efficiency
storage
[kWh/kg(1)] [wt%] [K] [%]
[%«]
Liquid
33.3 100 20 25 -33 70
hydrogen
Long
distance 33.3 100 >273 88-92
pipeline
Methanol /
Methyl - 2.9 69
Formate(2)
Methanol /
5.6 12.0 >273 28 68
CO2
Ammonia 5.14 17.6 >240 22-28 68
Methanol /
5.6 12.5 >273 66
CO
Cyclohexane 2.03 6.2 >573 30 60
(1) in relation to net heat of combustion

(2) in relation to gross heat of combustion
about 25 % compared with 38 % in the LHa system. On the other hand, LH2 facility cost
is expected to be higher because of the need for heat insulation. Nevertheless, the LH2
system was deemed superior to the methanol system [47].
6.4. CHEMICAL HEAT PUMP SYSTEMS
Chemical heat pump systems are important, if the utilization of low-quality heats
(« 80 °C), e.g., solar or geothermal or nuclear (LWR) heat, waste heat in factories, is
considered for heat storage and temperature upgrading to improve the total efficiency of
the system. Among the many chemical cycles that have been considered to store and
transport heat energy in the form of chemical reaction enthalpy, the coupled processes of
hydrogenation / dehydrogenation can be adapted here.
A chemical heat pump system consists in principle of two reactors H and L where
the following processes proceed:
H: A(s) + C(g) AC(s) + QH

L: B(s) + C(g) BC(s) + QL
162
Reactor L contains the compound BC where heat is added to cause an endothermal
reaction liberating C as a gas. C is routed to the reactor H which contains A to undergo
an exothermal reaction where high-temperature heat becomes available.
The reactions are of either liquid-gas (wet) or solid-gas (dry) type. Dry processes
show high reaction rates enabling compactness, but a poor heat and mass transfer behavior
compared with wet processes [19]. The examples of the 2-propanol - acetone system [44,
56], the cyclohexane-benzene system [31], the hydride system [30], and the sponge iron
system [54] are given in more detail in appendix C.
[I] ABE, A., et al., Studies of the Large-Scale Sea Transportation of Liquid Hydrogen,
Int. J. Hydrogen Energy 23 (1998) 115-121.
[2] BECKURTS, K.-H., DIETRICH, G., Proj'ekt Fernwarme Versorgung fur Millionen-
Stadte, Bild der Wissenschaft, 13 (1976) January 64-70.
[3] BEKKEDAHL, T.A., HEBEN, M.J., Advanced Materials for Hydrogen Storage:
Carbon Nanotubules, (DOE/NREL Hydrogen Program Review, Livermore, 1994).
[4] BOLTENDAHL, U., HARTH, R., Warmetransport auf kaltem Wege, Bild der
[5] BRUDERER, H., Die Krux mit dem Akku, Technische Rundschau, January 17,
1997.
[6] CAMUS, J.-M., L'Air Liquide, Sassenage, Personal Communication (1998).
[7] CHAHINE, R., BOSE, T.K., Characterization and Optimization of Adsorbents for
Hydrogen Storage, (llth World Hydrogen Energy Conf., Stuttgart, FRG, 1996),
VEZUROGLU, T.N., et al., Hydrogen Energy Progress XI, International Association
[8] DANEELI, E., Praxair, Danbury, Personal Communication (1997).
[9] DECHEMA, A Study for the Generation, Inter-Continental Transport, and Use
of Hydrogen as a Source of Clean Energy on the Basis of Large-Scale and
Cheap Hydro-Electricity, Final Report on Contract No. EN3S-0024-D(B), Deutsche
Gesellschaft fur Chemisches Apparatewesen, Frankfurt (1987).
[10] DILLON, A.C., JONES, K.M., HEBEN, M.J., Carbon Nanotube Materials for
Hydrogen Storage, 1996 US-DOE Hydrogen Program Review (Conf., Miami,
1996), Proc. Vol 2, US-Department of Energy (1996) 747-763.
[II] EDESKUTY, F.J., Nuclear Propulsion, in: VANCE, R. W. (Ed.), Cryogenic
Technology, John Wiley & Sons Inc., New York (1963).
[12] EDESKUTY, F.J., WILLIAMSON, K.D., Storage and Handling of Cryogens, Adv.
Cryo. Eng. 17 (1972) 56-68.
[13] EDESKUTY, F.J., STEWART, W.F., Safety in the Handling of Cryogenic Fluids,
[14] EPSTEIN, M., Recent Test Results of CO2 Reforming of Methane in a Closed
Loop, (3rd JAERI Symp., Oarai, 1996), Proc. JAERI-Conf 96-010, Japan Atomic
Energy Research Institute (1996) 228.
[15] FASOLD, H.-G., Wasserstoffgas - ein potentieller EnergietrSger des 21. Jahrhun-
derts? Mdglichkeiten von Transport, Verteilung und Speicherung, Gas Erdgas 129
(1988) 281-291.
163
[16] FTNNSTROM, B., Hydrogen - A Study of Efficiencies, in: IEA, Programme of
Research and Development on the Production of Hydrogen from Water, Annual
Progress Report 1989, International Energy Agency (1990).
[17] PURER, G., Untertagespeicherung von Gas: Energie nach Bedarf, Spektrum der
Wissenschaft 7 (1984) September 26-40.
[18] GIACOMAZZI, G., GRETZ, J., Euro-Quebec Hydro-Hydrogen Project (EQHHPP):
A Challenge to Cryogenic Technology, Cryogenics 33 (1993) 767-771.
[19] HAHNE, E., Heat Storage Media, in: Solar Technology, Ullmann's Encyclopedia
of Industrial Chemistry, VCH Verlagsgesellschaft, Weinheim (1993) 406-418.
[20] HANNEMAN, R.E., VAKIL, H., WENTORF, R.H.Jr., Closed Loop Chemical
[21] HANSEL, J.G., MATERN, G.W., MILLER, R.N., Safety Considerations in the
Design of Hydrogen-Powered Vehicles, Int. J. Hydrogen Energy 18 (1993) 783-
790.
[22] HARTH, R., JANSING, W., TEUBNER, H., Experience Gained from the EVA E
[23] HEIMANN, G., Eine riesige Thermoskanne voll Wasserstoff, Newspaper article in
Der Tagesspiegel, Berlin, August 29, 1996.
[24] HIRAOKA, K., et al., Research and Development of Internal Reheat Hydrogen
Gas Turbine at Ship Research Institute, Hydrogen and Clean Energy (Int. Symp.,
Tokyo, 1995), NEDO (1995) 345-348.
[25] HIRATA, M., A Proposal of Trans-Asian Gas Pipeline Network - Hydrogen
Transportation in 21 Century, Hydrogen and Clean Energy (Int. Symp., Tokyo,
1995), NEDO (1995) 149-156.
[26] HOFFMANN, P., (Ed.), Hydrogen & Fuel Cells Letter 12 (1997) February.
[27] HOHLEIN, B., et al., Methane from Synthesis Gas and Operation of High-
Temperature Methanation, Nucl. Eng. Des. 78 (1984) 241-250.
[28] HYNEK, S., MOORE, R., Liquefied Hydrogen Storage Aboard Automobiles?,
(Proc. 7th Canadian Hydrogen Workshop, 1995, Quebec City), MEHTA, S.K.,
BOSE, T.K. (Ed.), Canadian Hydrogen Association (1995) 231-240.
[29] HYNEK, S.J., FULLER, W.D., Stationary Hydrogen Storage Using a Phase Change
Material, (llth World Hydrogen Energy Conf., Stuttgart, FRG, 1996), VEZTRO-
GLU, T.N., et al., Hydrogen Energy Progress XI, International Association for
[30] ISHIYAMA, S., UGACffl, H., ETO, M., Performance of High Temperature Chemi-
cal Heat Pump with Metal Hydride Reaction, (3rd JAERI Symp., Oarai, 1996), Proc.
[31] JTOH, N., Liquid Chemical Carriers in Hydrogen Storage and Transport, Hydrogen
and Clean Energy (Int. Symp., Tokyo, 1995), NEDO (1995) 279-282.
[32] IWATA, A., KAMIYA, S., KAWAGOE, E., Technical Development for Large
Storage of Liquid Hydrogen in WE-NET, (llth World Hydrogen Energy Conf.,
[33] KESTEN, M., MEYER, G., Offene und geloste Probleme bei Wasserstofftransport
und -speicherung, Auf dem Weg zur Wasserstoffenergie - Wie kommen wir weiter?,
164
(BAM-Seminar, Berlin, 1995), Federal Institute for Materials Research and Testing,
Berlin (1997) 47-55.
[34] KLUYSKENS, D., Hydro Quebec, Montreal, Personal Communication (1997).
[35] PETERSEN, U., WURSIG, G., KRAPP, R., Design and Safety Considerations for
Large-Scale Sea-Bome Hydrogen Transport, Int. J. Hydrogen Energy 19 (1994)
597-604.
[36] LOVEGROVE, K.M., High Pressure Ammonia Dissociation Experiments for Solar
Energy Transport and Storage, Int. J. Energy Research 20 (1996) 965-978.
[37] LUND, P., Improved Possibilities in Energy Storage Through Hydrogen Techno-
logy, (llth World Hydrogen Energy Conf., Stuttgart, FRG, 1996), VEZIROGLU,
Energy (1996) 981-992.
[38] LYNCH, F.E., Hydrogen Storage, Hydrogen and Clean Energy (Int. Symp., Tokyo,
1995), NEDO (1995) 99-106.
[39] MARKEWTTZ, P., Leitungsgebundene Energietrager, in: HAKE, J.-F., et al.
(Eds.), 2. Ferienkurs Energieforschung, Konferenzen des Forschungszentrums
Julich, Vol 20, Part I, Research Center Julich (1996) 257- 289.
[40] MATTHIES, R., HORN, R., Novel Insulation System for Large Scale Liquid
Hydrogen Tanks, (llth World Hydrogen Energy Conf., Stuttgart, FRG, 1996),
[41] MERATLA, Z., Large Scale Liquid Hydrogen Transport by Air, (llth World Hy-
drogen Energy Conf., Stuttgart, FRG, 1996), VEZIROGLU, T.N., et al., Hydrogen
Energy Progress XI, International Association for Hydrogen Energy (1996) 1347-
1353.
[42] MICHEL, F., et al., Onboard Equipment for Liquid Hydrogen Vehicles, (llth
World Hydrogen Energy Conf., Stuttgart, FRG, 1996), VEZIROGLU, T.N., et
(1996) 1063-1077.
[43] MOHTTPOUR, M., PIERCE, C.L., GRAHAM, P., Design Basis Developed for H2
Pipeline, Oil & Gas Journal, May 28 (1990) 83-94.
[44] MURADOV, N., T-RAISSI, A., DOI, T., Dehydrogenation of 2-Propanol Using
Heteropolyacids for Solar-Driven Chemical Heat Pump Application, (llth World
Hydrogen Energy Conf., Stuttgart, FRG, 1996), VEZIROGLU, T.N., et al., Hy-
drogen Energy Progress XI, International Association for Hydrogen Energy (1996)
2679-2685.
[45] NAKATSUGAWA, I., LUO, A., Recent Process of Hydrogen Storage and Trans-
portation Technologies in North America, Hydrogen and Clean Energy (Int. Symp.,
Tokyo, 1995), NEDO (1995) 275-278.
[46] NATIONAL HYDROGEN ASSOCIATION, Handling Hydrogen Safely, World
Wide Web, http://www.ttcorp.com/nha/hhs_2.htm, National Hydrogen Association
(1997).
[47] NEDO, International Clean Energy Network Using Hydrogen Conversion (WE-
NET), 1995 Annual Summary Report on Results NEDO-WE-NET-95, NEDO,
Tokyo (1996).
[48] OGDEN, J., Hydrogen Energy Systems Studies, World Wide Web,
http://www.eren.doe.gov/hydrogen/hydengy.htm, US Department of Energy (1997).
165
[49] OSUMI, Y., Present Status and Prospects od Development of Hydrogen Absorbing
Alloys, Hydrogen and Clean Energy (Int. Symp., Tokyo, 1995), NEDO (1995)
131-138.
[50] PEHR, K., Experimental Examinations on the Worst Case Behavior of LH2/LNG
Tanks for Passenger Cars, (llth World Hydrogen Energy Conf., Stuttgart, FRG,
York (1992).
[52] PETERSEN, U., WURSIG, G., WOHREN, N., Investigation into the Operating
Behaviour of 61 M3 Liquid Hydrogen Tank, to be presented at the Hypothesis n,
August 18-22, 1997, Crimstad, Norway (1997).
[53] PIKE, J., The Death-Beam Gap, World Wide Web,
http://www.fas.org/spp/eprint/keegan.htm, Federation of American Scien-
tists, Washington (1992).
[54] PLZAK, V., et al., Investigations of the Kinetics and Catalysis of the Iron/Steam-
Magnetite/Hydrogen-System in Low Temperature Regime with Regard of its Use to
Produce and Store Hydrogen, (11th World Hydrogen Energy Conf., Stuttgart, FRG,
[55] RAMBACH, G., Hydrogen Storage in Engineered Microspheres, World Wide Web,
http://www.eren.doe.gov/hydrogen/hydstor.htm, US Department of Energy (1997).
[56] SATTO, Y., Chemical Heat Pump and Hydrogen Transportation, Hydrogen and
Clean Energy (Int. Symp., Tokyo, 1995), NEDO (1995) 123-130.
[57] SCHMIDT, T., GOTZEN, P., Unterirdische Speicherung von Erdgas in Salzkaver-
nen und Porenspeichem, VDI Berichte No. 1129, VDI-Verlag, Dusseldorf (1994)
101-120.
[58] SCHULZ, R., BOILY, S., Nanocrystalline Metal Hydrides, (Proc. 7th Canadian
Hydrogen Workshop, 1995, Quebec City), MEHTA, S.K., BOSE, T.K. (Ed.),
Canadian Hydrogen Association (1995) 405-411.
[59] SHERIF, S.A., ZEYTINOGLU, N., VEZIROGLU, T.N., Liquid Hydrogen: Po-
tential, Problems, and a Proposed Research Program, Int. J. Hydrogen Energy 22
(1997) 683-688.
175-194.
[61] TAYLOR, J.B., Technical and Economic Assessment of Methods for the Storage
of Large Quantities of Hydrogen, Int. J. Hydrogen Energy 11 (1986) 5-22.
[62] TRILL, R., Sicherheit bei Transport, Lagerung und Nutzung von gasformigem und
flussigem Wasserstoff, VDI Berichte No. 1201, VDI-Verlag, Dusseldorf (1995)
245-249.
[63] TRILL, R., Linde AG, Hollriegelskreuth, Personal Communication (1997).
[64] VO6, A., WffiSE, A., KALTSCHMITT, M., Technische, okonomische und
okologische Aspekte eines "GLOBAL LINK", VDI Berichte No. 1129, VDI-
Verlag, Dusseldorf , M) 195-213.
166
896-904.
[66] WEITKAMP, J., et al., Wasserstoff-Speicherung in Zeoliten, Wasserstoff als Ener-
gietrager, (Coll. of Sonderforschungsbereich 270, Stuttgart, 1994) (1994) 287-300.
[67] WtJRSIG, G.-M., Shipping Liquid Hydrogen, MER (1991) December 10-14.
[68] WtJRSIG, G.-M., Beitrag zur Auslegung von mit Wasserstoff betriebenen Haupt-
antriebsanlagen fur Flussigwasserstoff-Tankschiffe, Ph.D. Thesis, Verlag Mainz,
Wissenschaftsverlag Aachen (1996).
[69] WtJRSIG, G., SCHMIDTCHEN, U., Flussig-Wasserstoff sicher transportieren
- Versuche zum Verhalten von tiefkaltem Flussiggas an einem 61 m3-FlQssig-
Wasserstoff-Tank, Tt 38 (1997) No. 4 26-30.
[70] WURSTER, R., Randbedingungen und Systemaspekte fur Transport und Speiche-
rung von Wasserstoff, VDI Berichte No. 1129, VDI-Verlag, Dusseldorf (1994)
141-158.
[71] WURSTER, R., Transport von Energie in Form von Wasserstoff uber groBe Ent-
fernungen, Wasserstoff als Energietrager, (Status Seminar, Wurzburg, 1995), Pro-
jekttrager Biologic, Energie, Okologie, Research Center Julich (1995) 129-140.
[72] ZAIDMAN, B., WIENER, H., SASSON, Y., Formate Salts as Chemical Carriers in
Hydrogen Storage and Transportation, Int. J. Hydrogen Energy 11 (1986) 341-347.
NEXT PAQE(S)
left BLANK
167
Chapter 7
APPLICATIONS OF HYDROGEN
Hydrogen has many and versatile uses as a secondary energy carrier and a secon-
dary raw material: With 48 % consumption, it plays a significant role in non-energetic
applications as a chemical raw material and intermediate product for industrial and pe-
trochemical processes. Furthermore, it is largely used in energetic applications indirectly,
e.g., in the production of clean synthetic fuels (20 %) or directly as a fuel for producing
process heat in the chemical industries (32 %) [45]. From the annual world production of
approx. 500* 109 Nm3 of hydrogen, about 70 % are consumed in the chemical industries.
It possesses, however, a huge potential to generate mechanical energy, heat, or electricity
for a future large-scale use.
7.1. HYDROGEN AS A RAW MATERIAL IN CHEMICAL PROCESSES
The chemical hydrogen economy started already at the turn to the 20th century when
coke furnaces generated process gas and town gas with an up to 60 % hydrogen component.
Since then a whole variety of non-energic and indirectly energetic applications has been
developed for hydrogen as a part of the synthesis gas such as
• ammonia synthesis
• methanol synthesis
• direct reduction of iron ore
• fossil fuel processing (hydro cracking)
• Fischer-Tropsch synthesis
• methanation in long-distance energy transportation (see section 4.3.2.3. and ap-
pendix B)
• hydro-gasification (see section 5.1.2.2.)
7.1.1. Ammonia Synthesis
More than half of the hydrogen used in the chemical industries or 40 % of the world
production, approx. 200* 109 Nm3, is consumed in ammonia synthesis
N2 + 3 H2 -* 2 NH3 + 46 kJ/mol
Feed gas for the steam reformer is methane or gasoline. High dilution with steam is
chosen to keep the methane contents on a low level. Adding air in a secondary reformer
leads to partial oxidation of the residual methane and of the CO. After separation of the
CO2, the product gas is a mixture of nitrogen and hydrogen whose ratio (desired 1:3) is
adjusted by the operating conditions. The system pressure is about 5 MPa, the synthesis
temperature is 400 °C [27].
The production of one ton of ammonia requires about 2000 Nm3 hydrogen as well
as approx. 800 kWh for compression of the synthesis gas and provision of the nitrogen
(by air liquefaction).
169
Modern industrial production of ammonia in the so-called Haber-Bosch process allows
a daily output of 1000 - 2000 t of liquid NH3 corresponding to a hydrogen consumption
of 80,000 - 160,000 Nm3/h. The use of high gasification pressures makes the subsequent
compression step superfluous. Ammonia is worldwide used as a fertilizer with facilities
for storage, safe handling, transportation, and distribution being available. It could also be
used as a fuel for automobile transportation to replace CC>2 producing fuels.
7.1.2. Methanol Synthesis
The world production of methanol is currently estimated to be about 27 million

(metric) tons per year (1995). The methanol synthesis consumes about 5 % of the world
hydrogen production. Methanol is basically used in the chemical industry as an intermediate
product ("Cj chemistry"). It is gaining further attention as a secondary energy carrier
with less CC>2 emission, e.g., as a direct vehicle fuel or as a basis for the production of
hydrogen-rich gas to feed fuel cells.
The formation of methanol from synthesis gas can be described by the two indepen-
dent reverse reactions that were given in section 5.1.1.6. for the methanol splitting process
by high temperatures or steam reforming. In the conventional method, a mixture of CO,
CC>2, and H2 is compressed to about 10 MPa and introduced into a fixed-bed catalytic
reactor at temperatures of 220 - 280 °C and pressures of 5 - 20 MPa [27].
CO + 2 H2 -* CH3OH + 91 kJ/mol
CO2 + 3 H2 -* CH3OH + H2O + 50 kJ/mol
The reactions are exothermal and volume-reducing, thus low temperatures and high
overpressures are desirable. A catalyst is required to maximize methanol output. The
specific consumption is 2300 Nm3 of CO and H2 per ton of methanol. A processing
scheme has been proposed by Lurgi (see Fig. 7-1). Nowadays conventional synthesis
reactors have a capacity of up to 3000 t/d of methanol.
Methanol production generates a surplus of hydrogen which can, by adding CO2, be

utilized to increase the methanol yield and thus reduce CO2 emission into the atmosphere:
3 CH4 + 2 H2O + CO2 -* 4 CH3OH
In a subsequent step, hydrocarbon fuel synthesis can be made:
n CH3OH -* (CH2)n + n H2
A CO2-neutral solution is obtained if the CO2 released during combustion of the

methanol is recovered from the ambient during the methanol production step. A CO2-free
coal-to-methanol production is given by taking H2 as a supplemental feed and a non-fossil
source of high temperature heat (e.g., nuclear).
170
Synthesis gas
Purge gas «— Startup vapor
Purification systems
Heating gas
Pure methanol •<
Waste water
Fig. 7-1: Chemical processing flow chart of the Lurgi low-pressure methanol
synthesis, from [38]
In the low-temperature liquid phase process, the use of a highly active liquid catalyst
results in an efficiency of 90 % compared with 40 % in the conventional process. A 10 t/d
unit has been constructed and continuously operated for 320 h. Next step is the realization
of a 100 t/d pilot plant [2].
The development of a new catalyst allows the methanol synthesis to take place at
lower temperatures (80 - 120 °C) and lower pressures ( 1 - 5 MPa). A drawback is the
rapid deactivation of the catalyst in the presence of strong lyes [33].
In Japan, the conventional method of methanol synthesis has been modified by

adopting a fluidized granular catalyst bed rather than a fixed bed which allows the possibility
to enlarge the scale of the reactor, to reduce the power required, and to exchange the
catalyst during operation [2].
7.13. Direct Reduction of Iron Ore
Blast furnace technology is used for more than a century for raw iron production. The
stack gas generated during the process was usually considered a waste gas. Its constituents
and typical fractions are CO (« 40 %), N2 (« 40 %), CO2 (« 12 %), H2O (« 6 %),
and H2 (« 2 %).
171
Direct reduction of iron ore to sponge iron which can then be converted to steel in
an electric arc furnace is a process that takes place outside a blast furnace avoiding the use
of coke. This process is more favorable than the traditional raw iron production by means
of coke. The chemical reactions are:
(a)______Fe2O3 + 3 H2 -» 2 Fe + 3 H2O - 91 kJ/mol_______

~ ( D ) F e 2 O 3 + 3H 2 ~ 2 F e + 3 H 2 O - 9 1 kJ/mol
Fe2O3 + 3 CO -» 2 Fe + 3 CO2 + 32 kJ/mol
translated into demands of 610 Nm3 of hydrogen or 604 Nm3 of carbon monoxide per
ton of iron. Process (a), the H-iron process, operates at 500 °C temperature and 3-4 MPa
pressure and needs pure hydrogen (98 %) and a high dilution of the reformer feed gas with
steam. Process (b), the Korf-Midland-Ross process, operates at 800 °C temperature and
0.3 MPa pressure and requires low steam dilution ratio to prevent carbon deposition and a
low methane level (< 2 %) in the reformer product gas [27]. The optimal H2 to CO ratio
is a compromise between a favorable energy balance of the endothermal and exothermal
reaction, and reaction kinetics as well as environmental impact by the CO2 produced [32].
Heat treatment of ferrous metals is made in a hydrogen-nitrogen atmosphere with 3 -

30 % H2 to increase metal ductility, improve machinability, and alter electric and magnetic
properties.
7.1.4. Fossil Fuel Processing
Steady utilization of hydrogen in refineries commenced more than 50 years ago. It is

mainly used for hydro-cracking and fuel burning of excess byproduct and hydro-treating
comprising pre-treatment of reformer feed and treatment of heavier streams in upgrading
processes including the removal of sulfur compounds, halides, metals, nitrogen or oxygen,
the saturation of olefins, diolefins, cyclo-olefins, or aromatics, the decyclization or ring-
opening. The cracking of heavy crude oil is made to produce lighter hydrocarbons or
refined products such as high-octane gasoline [41].
In the process of coal hydrogenation which is a high-pressure (30 - 70 MPa) catalytic

process at a temperature of 500 °C, hydrogen is used to convert coal to gasoline
C + H2 -» - CH2 -
and to remove sulfur, oxygen, and nitrogen. Rigid specifications concerning the
contents of the catalyst poison CO (< 10 ppm) have to be met. A purity of 98 % hydrogen
is desirable. With an input of 2000 - 2600 Nm3 of hydrogen and 1.7 t of coal, an output
of 1 t of gasoline is obtained. In addition, 600 kWh of electricity are necessary for H2
compression and coal preparation [27].
The oxo synthesis is an exothermal catalytic process in which a H2-CO mixture, as

the product gas of the steam reforming process, is used to produce aldehydes which are then
hydrided to the respective alcohols. The oxo process works at 3 - 30 MPa and 100 -180 °C.
172
The hydrogen demand in cleaning and upgrading of coal and oil is listed in Table
7-1 for various processes.
Table 7-1: Hydrogen demand for coal upgrading or product

improvement, from [6]
Hydrogen demand
Process [mol Hj per mol CH2
(CH3OH)]
Crude oil desulfurization 2-5
Hydrotreatment 5 - 10
Hydrocracking 15-30
Residue hydrogenation 25-30
Very heavy oil —» syncrude 30-40
Oil sand -»• syncrude 30-45
Oil shale —> liquid products 80 - 125
Coal gasification (CH3OH) 100 - 160
7.1.5. Fischer-Tropsch Synthesis
The catalytic hydrogenation of carbon monoxide resulting in the formation of paraffin-

olefin mixtures of different C-numbers is what is called the Fischer-Tropsch synthesis. The
product spectrum depends on the operating conditions and the catalysts applied and on the
partial pressures of CO and H2 in the synthesis gas:
CO + 2 H2 (- CH2 -) + H2O + 165 kJ/mol

for Co catalyst
2 CO + H2 (- CH2 -) + CO2 + 205 kJ/mol
for Fe catalyst
The resulting hydrocarbons are characterized by the strong arrangement of the C-

atoms used as a raw material in the petrochemical industry or for the production of synthetic
hydrocarbons [13].
Until now the Fischer-Tropsch technology has never been able to compete econo-
mically with conventional fuel and was limited to very isolated cases. Recent process
innovations, however, give hope to a comeback for the exploitation of untapped gas fields
in remote regions as an alternative to LNG processing. Shell and Sasol are currently the
only companies to operate Fischer-Tropsch plants on a commercial basis.
173
7.1.6. Others
A significant application of hydrogen with a large-scale technological importance

is in the food industry the hardening of fats and oils by catalytic hydrogenation for the
purpose of extended durability. The process takes place at pressures of 20 - 40 MPa and
at temperatures of 200 - 400 °C.
Other fields of hydrogen applications are in the metallurgical industry as the reducing
medium for the nickel production, for autogeneous welding and cutting, in the glass industry
when clean combustion is required, or in the electronics industry for the epitaxial growth
of polysilicon. For the fabrication of semiconductor elements, vaporized liquid hydrogen
is used as a doping gas because of its cleanness.
A field with growing interest is the ability of hydrogen of scavenging oxygen, e.g.,
in boiling water reactors where traces of oxygen in the feed water which were found to
cause intergranular stress corrosion cracking, are removed by the injection of hydrogen
down to a level of < 100 ppb [41].
7.2. HYDROGEN AS A FUEL
The first application of hydrogen as a "fuel" was starting in the late 18th century
when it was tried to utilize the physical property of low density for flying balloons. The
Frenchman Charles realized in 1783 the first liftoff of an H2 balloon ("Charliere") filled
with 40 m3 of H2 which he produced by spilling sulfuric acid onto iron. The balloon
traveled a distance of 25 km in a height of up to about 1 km.
7.2.1. Heating
Catalytic or flameless combustion of hydrogen exhibits many advantages in com-

parison with flame combustion. It occurs at low temperatures (ambient - 800 K), is safe and
leads to a very high conversion of the burning gas (99.9 %). The NOX formation which
usually occurs in conventional combustion at temperatures of about 1700 K, is here almost
completely suppressed. The catalytic combustion in diffusion burners (Fig. 7-2) occurs
in the presence of small amounts of Pt or Pd catalysts. It is adequate, e.g., for kitchen
appliances such as cooker, oven, water heater, space heater. Drawbacks are the possible
non-uniformity of the temperature distribution at the catalyst surface, rapid changes in the
operational state, and relatively small heat flux densities [47].
7.2.2. Fuel Cells
7.2.2.1. Principle
The fuel cell is an energy conversion device that works in reverse of electrolysis.
In the cell, the energy of a fuel is directly converted by an electrochemical reaction
174
Water vapor Air
Reaction zone
\\UM
Catalytical
body
Water vapor
Fig. 7-2: Schematic of a hydrogen diffusion burner, from [14]
into D.C. electricity. Streams of hydrogen and oxidizer pass through porous metal plates
separated by an electrolyte with partial pressure differences as the driving force. Outside
the electrolyte, the plates are electrically connected. The hydrogen plate (anode) converts
hydrogen molecules into ions and electrons where the former migrate into the electrolyte.
The oxidizer plate (cathode) separates oxygen molecules into oxygen atoms^ They also
migrate into the electrolyte where they recombine with the hydrogen ions and anode
electrons to create water and heat. Electricity can be captured from the circuit and put
to useful work. Water and heat are_ expelled from the electrolyte to be further used or
recycled. The principal advantages of a fuel cell in comparison with the indirect energy
conversion (which goes via heat first) are its high efficiency and its cleanness. Practical
cells typically generate a voltage of around 0.7 - 0.8 V per cell and a power output of a
few tens or hundreds of watts.
Of all fuels that have been tested in fuel cells, hydrogen has the highest reactivity.
At temperatures < 300 °C, most hydrocarbons and alcohols show low catalytic reaction
rates and often unwanted secondary products. At high temperatures, direct electrochemical
oxidation may be feasible. Fuel cells convert stored energy to electricity with an approx.
60 -70 % efficiency (based on lower heating value), a higher is thought to be theoretically
possible [11]. The efficiency is nearly constant independent of size, thus configurations of
power plants could range from watts to megawatts. Internal reforming gives an increase
in efficiency as compared with external reforming. Heat from the electrochemical reaction
is available for cogeneration application.
A detailed description of all different kinds of fuel cells, their techniques and
applications can be found in [30].
175
-J
ON
Table 7-2: Fuel cell characteristics, from [25, 29, 49]
Operation
Typical
Type of tempera-
Fuel efficiency Anode reaction Cathode reaction
fuel cell ture
[%]
PC]
H2,
PEFC(1) 50- 140 50-60 H2 -» 2 H+ + 2 e' O 2 + 4H + 4 e ' ^ 2 H 2 O
(no CO)
H2
AFC<2> 25 - 120 50-65 H2 + 2 OH' -+ 2 H2O + 2 e- 02 + 2 H2O + 4 e- -> 4 OH-
(no CO2)
DMFC or
80 - 130 CH3OH 24-40 CH3OH + H2O -> CO2 + 6 H+ + 6 e- 3/2 O2 + 6 H+ + 6 e' -» 3 H2O
SPFC(3)
H2 O2 + 4 H+ + 4 e' -» 2 H2O
PAFC<4) 60 - 210 35-45 H2 -» 2 H+ + 2 e'
(no CO)
H2 H2 -i- CO3- ' -» H2O + CO2 + 2 e"
MCFC(5) 620 - 660 45 -60 O2 + 2 CO2 + 4 e- -> 2 CO3- '
CO CO + CO3- -» 2 CO2 + 2 e
H2 H2 + O- ' -» H2O + 2 e-
SOFC(6) 800- 1000 CO 50- 60 CO + O' • -> CO2 + 2 e' O2 + 4 e" -» 2 O- '
CH4 CH4 + 4 O- - -* 2 H2O + CO2 + 8 e-
(1) PEFC - Polymer Electrolyte Fuel Cell or PEM-FC - Proton Exchange Membrane Fuel Cell
(2) AFC = Alkaline Fuel Cell (4) PAFC - Phosphoric Acid Fuel Cell
(3) DMFC = Direct Methanol Fuel Cell or SPFC = Solid Polymer Electrolyte Fuel Cell
(5) MCFC = Molten Carbonate Fuel Cell (6) SOFC - Solid Oxide Fuel Cell
7.2.2.2. Fuel Cell Types
Various fuel cell types have been developed usually classified according to the
electrolyte used. In Table 7-2, fuel cell types are listed with their operating temperatures
and anode / cathode reactions.
Certain constraints to the fuel cell do apply depending on the type of fuel cell. In
low-temperature fuel cells, noble metal electrodes are necessary to increase the reactivity.
Poisoning of electrodes can be avoided by purification of natural gas. Various methods
have been developed to remove sulfur which is a poison to nickel steam reforming catalysts
and to many anode catalysts incl. platinum: high-temperature based hydro-desulfurization
for a large-scale hydrogen production (with a small quantity of H2) and low-temperature
based desulfurization adsorption processes that do not need Hk [11]. The high-temperature
fuel cell designs avoid expensive noble metal catalysts, thermal activation is insensible to
pollutants. Drawbacks are the short lifetime and expensive plant components. To avoid
dilution of the electrolyte by the product water, it must be removed by evaporation or
vaporization. In high-temperature fuel cell systems, the water is steamed off anyway.
If carbon-containing fuel gases are used instead of pure hydrogen, oxidation must
be realized first in a reforming reaction. Internal reforming has the advantage that there
is no requirement for hydrogen storage. Incomplete reaction at lower temperatures results
in presence of carbon monoxide which is a catalyst poison and must be removed, at the
expense of efficiency. CO causes no trouble at high operating temperatures since reforming
can take place internally. Attempts of a direct conversion of coal or oil in an integrated
gasification fuel cell system are presently at an early R&D stage, but they are considered
potential primary fuels in the long run.
The Polymer Electrolyte Fuel Cell, PEFC, or Proton Exchange Membrane Fuel
Cell, PEM-FC, is an efficient, compact, robust, and quiet method of generating electricity.
A diagram is shown in Fig. 7-3. It has a 0.1 mm thick proton conducting foil as a solid
electrolyte. It has the highest power density of all cell types, a relatively low operating
temperature and thus long cell lifetime, good load change behavior and good efficiency at
partial load, and a good standby performance, thus being ideally appropriate for mobile
applications. A problem in PEFC is the necessity to keep the polymer electrolyte membrane
in a wet state and at higher operating temperatures to prevent a dry-out by means of external
wetting [37]. Applicable fuels are hydrogen, reformed methanol or methane with purity
requirements of < 10 ppm of CO, < 1 ppm of NH3, and < 0.1 ppm of H2S. The development
is concentrating on improvement of catalysts and reduction of CO poisoning effects [15].
The Alkaline Fuel Cell, AFC, is the most efficient low-temperature fuel cell pre-
sently available with a very high power density, therefore ideally appropriate for mobile
applications. The AFC shows a similar performance as the PEFC, but with a much more
demanding process control which is complicated because of the requirements of fuel purity
(no CO2) and of the corrosive liquid electrolyte. Efficiencies of more than 60 % have been
achieved with clean hydrogen and oxygen and noble electrode materials. The AFC was
demonstrated to also work with a hydrogen-air system, in a 1 kW stack in Japan and in a
6 kW Russian system with more than 5000 h lifetime [34]. PEFC and AFC may play a
177
role where pure and inexpensive hydrogen is available [29]. Unsatisfactory is its observed
power decrease with operation time, by 20 % after 5000 h [51].
The Direct Methanol Fuel Cell, DMFC, (see Fig. 7-6 in section 7.2.2.4.) is another
low temperature fuel cell enjoying a renaissance after significant improvements in current
density. The DMFC runs on either liquid or, with better performance but higher system
complexity, on gaseous methanol and is normally based on a solid polymer electrolyte
(SPFC). Pt-Ru catalysts were found to produce best oxidation results at the anode, still
the power density is relatively low [5, 29]. Conversion rates up to 34 % of the energy
content into electricity were measured, an efficiency of 45 % is expected to be feasible in
the future. SPFC in the power order of several kW to be used in automobile applications
are currently in the development phase.
Most practical experience has been gained in the operation of Phosphoric Acid Fuel
Cells, PAFC. Because of its higher operating temperature, it has a minor CO poisoning
problem plus it offers operation in the CHP mode providing good quality steam of approx.
200 °C. Fuel is hydrogen or reformed methane plus CO converter. The principal poison
for the Pt catalysts is sulfur carried in the reformate stream as H2S.
A Molten Carbonate Fuel Cell, MCFC, is being operated at > 600 °C temperatures
with porous Ni catalysts. Carbon dioxide which is yielded at the anode, needs to be fed back
to the cathode where it is consumed. Besides hydrogen, the MCFC can be operated with
methanol or methane or coal gas with external or with partial or full internal reformation,
the possibility of internal naphta reforming is also investigated. In the future the synthesis
gas from (internal) coal gasification systems may be linked to MCFC electricity generation
inducing a further increase of the coal conversion efficiency up to 55 % with 60 % expected
to be possible [37, 46]. Drawbacks of this fuel cell type are its low current and power
densities. A problem often observed in MCFCs is the strong corrosivity of the electrolyte
and its leakage through gaskets. Also the high serial resistance of the electrolyte limits
the efficiency at low 40 % [34].
The Solid Oxide Fuel Cell, SOFC, (Fig. 7-4) is attractive - and difficult at the
same time - because of its very high temperature which allows fast chemical reactions, but
on the other hand implies stringent requirements for materials and construction. Without
any further modifications, the SOFC can be operated either in the electrolysis or in the
fuel cell mode. Two different stack design are currently pursued using tubular or flat cell
geometry. Typical dimensions of the tubular design are: support (= cathode) with inner
diameter of 12 - 22 mm, 2.2 mm wall thickness, and a length up to 1500 mm, an electrolyte
layer of 20 - 40 fim, and outside the anode with 100 fim thickness. For a flat cell, the
typical data are: a gaslight electrolyte of 150 - 200 /*m thickness, electrodes with 50 //m
thickness and 30 - 50 % porosity. SOFCs convert gaseous hydrocarbons either directly or
after internal reforming with a very low emission level. Utilization of the high-temperature
waste heat raises the overall efficiency to a maximum. Furthermore the waste gas at
the anode consisting of COi, H2O, CO, and residual H2 could be utilized for methanol,
ethylene, or ammonium synthesis. On the other hand, heat release during internal reforming
creating sharp temperature gradients, is difficult to control. A combination of 25 % internal
and 75 % external reforming was found to work best for the (Westinghouse tubular) SOFC
design [29]. Goal of the future development is the reduction of the operating temperature.
178
Little fundamental knowledge is yet available on the mechanisms of degradation in the
ceramic and metallic components during cell operation. Long-term stability is currently
subject to research activities within the EA SOFC cooperation [36].
7.2.2.3. Stationary Fuel Cell Applications
The complete fuel cell power generation system consists of a fuel processing system
to convert raw fuel gas into a hydrogen rich gas, the cell stack assembly for electric energy
production, a power conditioning system to connect the electricity to the regulated A.C. grid
supply, plus some form of heat recovery if used in the CHP mode. The core part of a fuel
cell system is the fuel cell stack which consists of layered unit cells and in which hydrogen-
rich gas and air are continuously supplied. Stationary applications embrace a large-scale
central power generation, distributed generation, and cogeneration. The potential of fuel
cell power plants is strongly related to the global demand for new power and heat capacity.
A cogeneration system with waste heat recovery can achieve efficiencies of up to 80 %.
Both high-temperature and low-temperature fuel cells are suitable for stationary applications
and also convenient in all sizes from medium-sized units on the order of MW down to
space heat supply for single rooms on the order of W.
State-of-the-art AFC units are in the range of 50 to 100 kW. Within the smaller-scale
commercial sector, considerable potential is believed to exist for SPFC based systems
because of their advantages in terms of low noise and vibrations and favorable emission
characteristics [5].
PAFC, MCFC, and SOFC are currently the strongest contenders for commercial
stationary power generation. PAFC power plants are presently at the pre-commercial
demonstration stage for small, decentralized on-site power plants of 200 kW. When
operated in the cogeneration mode, the overall efficiency could amount to 80 %. A
fleet of 67 phosphoric acid systems has already accumulated in excess of 1,000,000 fleet
hours of in-service operation at a sustained high level of availability [19] including use of
external reforming of either methane or naphta. State-of-the-art is an 11 MW power plant
operated in Ichihara, Japan, since 1991. It was shutdown in March 1997 having attained
16,000 h of operation time up to 1995 with an efficiency of 49 % and demonstrating a
smaller performance decay rate than expected (7 % loss of efficiency over 40,000 hours).
"Bestseller" on a commercial basis is the 200 kW(e) unit, PC25, by the US company ONSI,
the only commercially available plant with an availability of 95 %. Operating times have
reached 20,000 hours for a single unit [4, 50]. Worldwide 150 demonstration power plants
are currently (mid 1997) being operated with a total power of 40 MW(e).
MCFC have reached a high developmental level and are expected to be ready for
market entry before 2000. The MCFC is approaching the stage of the first, pre-commercial
demonstration units [5]. Present prototypes have a power of 300 kW; large units have been
operated for more than 40,000 hours. Power plants in the range of 0.25 - 1 MW are in
development in Japan and the US. The world's largest MCFC demonstration power plant
with 2 MW has begun its operation in Santa Clara, California, in April 1996 [19]. The
construction of a 3 MW unit in New York State has been initiated [4]. MCFC are unlikely
to be commercialized below « 500 kW because of system complexity [29].
179
950 8C Net Efficiency (LHV)
Natural Gas (2bar)
3,65 g/s; 20 °C
Total Efficiency (Steam)
)Q=H T)ges = 84,7 %
Total Efficiency (District Heat)

t|ges = 92,9 %
Pel. Gross = 100 kW I
Waste Gas 74,5 g/s; 100 6C
Process Steam
Fig. 7-5: Siemens design of a 100 kW SOFC combined heat and power plant, from
[12]
SOFC-based power plants appear to be the most interesting alternative because of a

simpler system technique, high efficiency, and the option of utilizing the waste heat in a
gas turbine for electricity generation with an effective system efficiency of about 77 % [53].
The most advanced concept is the tubular version by the US company Westinghouse, which
has developed a large SOFC unit of 44 kW. A growing number of SOFC is being tested
in different cell configurations, different materials and different experimental conditions.
More and more cells are being operated for longer testing times, i.e., > 1000 h [36]. A
world record19 has been achieved by Westinghouse with a 25 kW stack operated for 13,200
hours including a 6500 h period of nonstop operation with 10 cycles and a 92 % availability
[3]. Although several demonstration units in the 10 to 25 kW(e) size range are existing
already, the solid oxide system is still essentially at the research stage. The main risks
of SOFC at present are unknown degradation of the fuel stack and operating performance
of the balance of plant. Startup and shutdown operations impact the stack because of
thermo-mechanical tensions [29].
SOFC are foreseen to phase out the classical Combined Heat and Power (CHP)
for stationary applications with flexible power to cogeneration heat ratios, although it is
believed to have no significant chance on the central power generation market up to 2010
[5]. The SOFC still needs to demonstrate its competitiveness in power generation. Their
most pertinent competition are gas and diesel engine and gas turbine based systems in
the 1 - 3 MW power range which find wide spread application in industrial cogeneration
schemes, where their characteristics are in line with process requirements [5]. A significant
chance is seen for SOFC and MCFC in distributed generation systems. The general trend
19
As can be seen world records are moving targets: The most recent information from October 1997 considered an SOFC world record
is about Westinghouse's generation of 120 kW on their SOFC cogeneration unit [28].
180
towards smaller units and increasing cost of transmission and distribution capacity make an
advanced SOFC power plant (Fig. 7-5), owned and operated by the utility likely to offer
significant advantages over the existing slow speed diesel engine systems in this sector [5].
For the cogeneration or combined heat and power application, the market uptake was found
to be highly dependent on the national legislative / regulatory framework and thus may be
strongly country, site, and situation specific. The SOFC offers only a poor heat to power
ratio of less than 1, whereas the industrial sectors are typically heat demand led with a heat
to power ratio required of at least 2 or greater [5]. A potential niche market for SOFC may
be the so-called micro-CHP for household applications in the order of 1 kW [29].
The integration of fuel cell and coal gasification technologies is expected to raise
efficiencies significantly by utilizing hot exhaust and unused fuel gas in a steam turbine.
The US DOE has launched an Integrated Gasification Fuel Cell (IGFC) Program with the
goal to achieve the operation of an IGFC commercial-scale power plant with a net system
efficiency of greater than 60 % by 2010. Pilot-scale gasification and gas stream cleanup
facilities are planned to be completed by 1998 [52]. In Japan, a study is underway on
a coal gasification combined power generation system based on MCFC. The preliminary
target is a single 500 MW coal gasifier facility, an 18 stacks arrangement, four expansion
turbines and two steam turbines.
7.2.2.4. Mobile Fuel Cell Applications
The first successful applications of fuel cells were made in the US space missions
Gemini in 1963 (PEFC) and Apollo in 1969 (7 kW AFC with a life time of more than 7000
h). Due to this experience, the most advanced AFC, however, not commercially viable, are
found in space and military applications (satellites) with requirements of 65 % efficiency for
10,000 hours. The Space Shuttle employs three stacks with 12.5 kW each. Nevertheless,
the AFC has recently lost its important role in the context of terrestrial development [29].
Fuel cell systems have gained increased interest as an automobile propulsion system
fostered by the new regulations in California ("Clean Air Act") for emission reduction.
Fuel cell-based propulsion is efficient, has a high energy density, consumes comparatively
less energy, and, in particular, reduces the emission levels of all relevant pollutants.
Hydrogen powered fuel cell cars would represent true zero-mission vehicles (ZEV). The
most significant development goals which need to be reached for market introduction is
a wider cruising range, larger long-term stability and lower cost. A large-scale fuel cell
application in the road traffic needs, in addition, a grid of refueling stations with hydrogen
or methanol. From all low-temperature fuel cells, the PEFC is favored.
Using methanol as the energy carrier and hydrogen as the feed gas for a fuel cell,
methanol combined with a reforming process offers better capabilities for storage in mobile
systems with respect to tank size and weight. The entire unit should be small and light
and provide a high conversion rate. Methanol reforming with/without water vapor is used
on an industrial scale. The thermal splitting reaction does not suit the low-temperature
fuel cells' need of a very low CO content in the feed gas [22]. Two fueling systems
are being developed: hydrogen and methanol reformate or indirect methanol fuel cell
(IMFC). Compared with hydrogen, methanol has the higher storage density and is easier
to handle. Also the existing infrastructure for fuel transport, storage, and distribution could
181
air
water
(a)
H2/COV H2/C02 ^
UL
liquid^ Gas
Methanol Reformer Fuel Cell -electricity
1-2% CO Treatment <250 ppm CO
i j heat
i
i Catalytic «————————
Converter "*" anode gas
I
exhaust gas
water
(b)
CH3OH/H2O
i
i
i
1 liquid
———————————»• Evaporator
ik
heat
heat/qas
* Fuel Cell
.
-electricity
i Catalytic ^ anode gas

* Converter
I
exhaust gas
Fig. 7-6: Schematics of the processes in a PEM fuel cell system using methanol
(a) indirect methanol fuel cell and (b) direct methanol fuel cell, from [43]
be utilized. The methanol option appears to offer the safer concept, but does tolerate
only a very low CO level. Procedures for CO reduction are CO-methanation, selective
CO-oxidation, and membrane separation [8]. An analysis of the emissions of an IMFC
propulsion system at the Research Center Jiilich exhibited its advantages compared with a
conventional combustion engine by meeting the ULEV standard [35]. Works have started
on ethanol fuel as an alternative to methanol with the advantage of an easy bioproduction.
A serious candidate for transportation application is also the "direct methanol fuel
cell (DMFC)" which has been realized already on a laboratory scale. A catalytic burner is
required to evaporate the methanol/water mixture and to burn the exhaust gas at the anode
[43]. Considering the complete energy chain, a PEFC is by 50 % more efficient than a
diesel engine which consumes 4 1 per 100 km; this is also valid for a natural gas driven
engine [37]. Fig. 7-6 presents the processing schematics of both IMFC and DMFC. The
DMFC offers a much simpler system than the PEFC. The DMFC is currently at an early
development stage. It is perceived to offer improved solutions to the need for a small-scale
power supply. A program for the construction of a 30 kW stack has recently started [29].
Also a gasoline driven fuel cell has been developed recently in a joint effort between
leading car and oil companies in the USA demonstrating a much lower CO2 emission level
182
than a conventional engine. Vehicles with a gasoline fuel cell drive are hoped to double
the cruising range compared with conventional drive with emissions lower than ULEV
standard [21].
The critical issues in designing a vehicular fuel cell power system are the selection
of an optimal operating pressure, oxidant flow rate and the choice of an adequate control
strategy. Computer models have been developed to simulate fuel cell performance under
various operating conditions.
An interesting statement to fuel cell technology was made by a fuel cell sales director
on a seminar in November 1996 telling that fuel cells have the potential of changing
the accepted pattern of the electric utility and other industries and fuel cell technology
could lead to the demise of combustion technology as a means of powering land and sea
transportation, with the potential of disrupting automotive and petroleum industries as they
exist today [20].
1.23. Hydrogen Powered Vehicles
The properties of a high heat of combustion and a high specific heat as well as the
clean combustion product has made hydrogen very attractive to be used as a fuel in mobile
applications.
7.2.3.1. Cars
Internal combustion engines (ICE) that run on hydrogen are on the order of 25 - 30 %
more efficient than gasoline ICE, because they take advantage of the fast-bum and far-lean
combustion characteristics of hydrogen. ICE for earth-bound vehicles is seen by many to be
superior to battery or fuel cell powered vehicles in terms of range, acceleration, and power-
to-weight ratio with no substantial improvement in the near future [40]. Internal combustion
engines extract about 30 % of the stored fuel energy. Consideration of hydrogen as an ideal
fuel for the internal combustion engine has started in the early 1920s. The operation with
hydrogen is possible in a wide range of mixtures with air. At extremely lean conditions,
it provides a low emission level and a high thermal efficiency in the low power range. In
addition, cryogenic hydrogen provides a high density storage and a considerable cooling
effect. From the works done so far it can be concluded that successful hydrogen fueled
direct injection engines are feasible (see Fig. 7-7). Limitations are given by the present
storage technology and, in principle, by lack of suitable applications [7].
A Ford pickup truck is reported to be the first LH^/LOX-fueled car by Perris Smogless
Automotive Association in 1971 in the USA. The first European car running on LH2 was
demonstrated by the German DFVLR (now DLR) in 1979, who also built the first LH2
system for vehicle refueling [48]. Many automobile companies have designed and operated
prototypes of cars or trucks. This is a rapidly changing field. A recent study by the
Japanese Musashi Institute of Technology has shown that the most practicable solution for
cars, trucks or buses in terms of onboard H2 storage is liquid hydrogen compared with
hydrides or electric drive [54]. Detailed description of LH2 use in vehicles can be found
in [39]. Major hydrogen fueled vehicle R&D programs in the world are listed in [26, 31].
183
Mechanic Electronically H2 sensors • LH2 tank Relief
gyro charger controled for automatic with vacuum and
with revolution throttle flap leakage jacketed safety
variator for gasoline control super- valves
operation insulation
H2 Electronically LH2 Vacuum Vacuum -

injection regulated vaporizer insulated insulated
nozzles injection valve LH2 valve block
for power transfer for LH2
control lines refueling /
retrieval
Fig. 7-7: Arrangement of main components in the BMW 735i LH2 test vehicle with
internal combustion engine, from [42]
Computationally assisted engine development is conducted focusing on optimization

of fuel efficiency and low emissions with respect to operating conditions and combustion
chamber geometry. Fluid dynamics in lean premixed combustion are examined for complete
oxidation of the fuel in a safe and clean way [24].
Ground-based vehicles with PEM fuel cell systems are foreseen to appear on the
market by the year 2004. Various prototype cars and buses have been presented already,
the first in the USA in 1991 (see section 9.5.2.). The critical aspect is the on-board storage
of the hydrogen fuel which can be provided as compressed gas, cryogen, or metal hydride.
A mid-term goal for fuel cell cars is the employment of a reformer in which any
hydrocarbon, including gasoline, is processed by steam reforming or partial oxidation to
generate hydrogen, which is then consumed in the fuel cell. One appropriate, easy-to-
handle energy carrier is methanol. A comparison of representative vehicle fuels in terms
of mass and volume requirements yields the amount of 30 kg (40 1) of gasoline to be
equivalent to 60 kg (70 1) of methanol or 24 kg (57 1) of LNG or 24 kg (110 1 at 30 MPa)
of CNG or 10 kg (141 1) of LH2 or 10 kg (370 1 at 30 MPa) of compressed hydrogen.
184
A hydrogen hybrid electric vehicle represents a transitional step between a gasoline
ICE vehicle and a fuel cell vehicle. Hydrogen is burnt in a small optimized engine which
runs a generator to charge a battery. The total efficiency is estimated to be 39 %. Not
more than 3.3 kg of hydrogen would allow a theoretical cruising range of 480 km [31].
7.2.3.2. Aviation
Liquid hydrogen was early recognized to be an important rocket fuel for use in space
flights. The H2 / C>2 system together with the H2 / F2 system belongs to the most energetic
propellants allowing for exhaust velocities sufficiently high to escape the earth gravity or
reach earth orbit [39]. The Japanese LE-7 is a two-stage combustion engine (see also Fig.
9-9). LH2 and LOX are pressurized by a pump and burnt in a prebumer chamber (stage
1). The resulting high-temperature high-pressure gas operates the pump. The exhaust gas
from the pump is routed to the main combustion chamber where it is burnt once again
using remaining LOX (stage 2). Ram engines for future reusable space transportation can
be operated with atmosphere oxygen up to a height of 40 km. Air-breathing engines with
supersonic combustion, scramjet, offer an excellent thrust potential. Fig. 7-8 shows the
schematic of a scramjet. Hydrogen in space transportation is being used by the USA,
Russia, Europe, China, Japan, and India.
Considering weight and volume restrictions in aviation, liquid hydrogen has been
shown to be particularly attractive as an aircraft fuel. The wide range of flammability of
hydrogen in air enables a stable combustion chamber operation far beyond the limits of
hydrocarbons. The lighter fuel load compared with conventional fuel results in a gross
weight reduction allowing the use of a smaller engine. Together with its high combustion
heat per mass unit, fast ignition and high heat sink capacity, hydrogen is a perfect candidate,
in particular, for supersonic applications. In addition, it avoids all pollutants of fossil fuels
except for NOX. Compared with kerosene, the flight-related energy content of LH2 is nearly
2.8 times larger on a mass basis and it could be used as a coolant Also the engine life time
is expected to be higher and maintenance requirements to be reduced. The main penalty,
however, is the storage volume of the hydrogen tanks which is by a factor of about 4 larger
than kerosene [9, 39].
Also in terms of safety, liquid hydrogen is superior to kerosene. A fuel fire is

expected to be less dangerous, the endangered area is much smaller, a kerosene fire lasts
much longer [10].
7.2.3.3. Ships
A specialist field of application of hydrogen is seen in ship propulsion using low-

temperature fuel cells. The operation of submarines powered by SPFC systems has been
successfully demonstrated. The German submarine class U212 is equipped with an air-
independent Siemens PEM fuel cell drive. The hydrogen is stored as titanium hydride in
cylindric vessels attached to the inner hull. The first four units were ordered in 1994 and
are planned to go into operation from 2003 [18]. Japan is investigating a two times 500
kW PEFC system operating on methanol reformate gas. US research is focusing on 250
kW and 2.5 MW shipboard power units [20]. Naval applications are being realized in
Australia, Canada, and Germany.
185
Within the Euro-Quebec project, a passenger boat is currently under construction
equipped with a PEFC electric drive and three 200 1 LH2 storage tanks [15].
The German Howaldtswerke Deutsche Werft AG, Kiel, has recently investigated in
an internal feasibility study the use of fuel cells in powering merchant ships [44].
7.2.3.4. Locomotives
A fuel cell propulsion system for locomotives is being reconsidered one option
in future railroad technology. A new emission standard for locomotives, which will be
effective in the year 2000, has led to a revival of respective research activities from the
past [1].
7.2.4. Stationary Gas Tbrbines
Hydrogen is an ideal fuel for gas turbines. Due to its rapid mixing with air, a smaller
combustion chamber is sufficient and the efficiency is higher compared with conventional
fuels. Gas turbines modified for liquid hydrogen operation yield an up to 10 % higher
thermal efficiency and output compared with fossil-fueled turbines. For systems with
advanced heat exchange, efficiencies of more than 50 % are estimated to be achievable.
The remainder-free combustion is stable and favorable for lifetime and maintenance. Of
disadvantage is NOX production. No particular difficulties are expected for a conversion
of a stationary gas turbine to H2 fuel [51].
The schematic of a hydrogen-oxygen gas turbine system with recirculating argon

gas as the working fluid is shown in Fig. 7-9. The internal stoichiometric combustion
of hydrogen and oxygen heats up the argon gas which actuates the turbine for power
generation. The water vapor produced is condensed and removed from the circuit while
the argon is returned to the combustor. Innovative developments of gas turbines capitalize
on the experience of aerospace propulsion systems. Performance goals are control of
combustion temperature by pre-mixing steam and operational flexibility [17].
A modified gas turbine cycle, a H2-fueled chemical-looping combustion, has been

proposed in Japan. The oxidizer air or oxygen is replaced by a solid metal oxide, e.g.,
NiO. An increase in the efficiency up to 66.8 % is predicted compared with 61.6 % as the
best figure for H2/O2 cycles [23].
The hydrogen-oxygen steam generator is a novel power plant component derived

from rocket technology to provide instantaneously spinning reserve capacity upon demand.
H2 and 02 stored in high-pressure tanks are injected into a combustion chamber and ignited.
Combustion takes place at pressures of 2 - 20 MPa. The reaction heat is directly transferred
to additional feedwater introduced. Steam of the required temperature of 500 - 1000 °C
is obtained and routed to the intermediate pressure steam turbine, thus being capable of
increasing the electric power by 20 MW within seconds [16]. However, lack of demand
for such direct power reserve and relatively high cost have made DLR to abandon this
development. Respective research projects are presently pursued only in Japan and Russia.
186
Air temperature • Supersonic Combustion
1500K(Mach = 5.5) nozzle chamber
Hydrogen
inlet
Fig. 7-8: Schematic of a model combustion chamber of a scramjet, from [DLR]
H2 ——
=T Combustor
Inert Gas
Compressor Turbine
Power Axle —
Inert Gas + H2O

Water
Cooler
Fig. 7-9: Inert gas recirculating hydrogen-oxygen gas turbine system, from [17]
[1] ANON, EPA legt Emissionsgrenzwerte fiir Lokomotiven fest, neues Interesse an
Brennstoffzellen fiir Lokomotiven, Wasserstoff-Gazette 2 (1998) I. Quarter, World
Wide Web, http://www.hyweb.de/gazette/, Ludwig BOlkow Systemtechnik, Ottob-
runn (1998).
[2] ANON, New Low-Cost Methanol Fuel from Natural Gas, CADDET Energy Effi-
ciency Newsletter, 2/96 June (1996).
187
[3] ANON, Brennstoffzellentest war erfolgreich, Newspaper article in Blick durch die
Wirtschaft, Frankfurt, September 9, 1997.
[4] APPLEBY, A.J., Issues in Fuel Cell Commercialization, J. Power Sources 69 (1996)
153-176.
[5] BAKER, J.N., EVANS, RJ.D., Initial Studies on the Market Potential for SPFC
and SOFC Systems (Summary Report), Report No. 3719, ETSU F/03/00080/REP,
EA Technology (1996).
[6] BARNERT, H., Wasserstoff zur Veredelung fossiler Brennstoffe, KFA Annual
Report 1982/83, Research Center Julich (1983) 9-16.
[7] BECK, N., Review and Assessment of the Developments in H2-Fueled Engines,
in: IEA, Hydrogen - Production and Utilization, Task Vn Storage, Conversion,
Safety, (Workshop, Denver, USA, 1991), Deutsche Forschungsanstalt fur Luft- und
Raumfahrt, Stuttgart (1991).
[8] BENZ, U., REINDL, M., TILLMETZ, W., Brennstoffzellen mit Polymermem-
branen fiir mobile Anwendungen, Spektrum der Wissenschaften, 18 (1995) July
97-104.
[9] BREWER, G.D., Hydrogen Aircraft Technology, CRC Press Inc., Boca Raton
(1991).
[10] DAS A, CRYOPLANE - Deutsch-Russisches Gemeinschaftsprojekt zum Einsatz
kryogener Treibstoffe in der zivilen Luftfahrt - Realisierbarkeitsstudie 1990/91/92,
Deutsche Aerospace Airbus GmbH, Hamburg (1992).
[11] DICKS, A.L., Hydrogen Generation from Natural Gas for the Fuel Cell Systems
of Tomorrow, J. Power Sources, 61 (1996) 113-124.
[12] DRENCKHAHN, W., SOFC Entwicklung bei SIEMENS, (Zweites Industriesemi-
nar, Julich, FRG, 1995), NABffiLEK, H. (Ed.), Solid Oxide Fuel Cells, Research
Center Julich (1995) 25-39.
[13] EICKHOFF, H.-G., Technologische und wirtschaftliche Moglichkeiten, die sich
durch den Einsatz des Hochtemperaturreaktors fiir die kunftige Mineralolversorgung
der BRD ergeben, Report Jul-1017-RG, Research Center Julich (1973).
[14] GARCHE, J., ALBRECHT, H., SPECHT, M., Wasserstofftechnologien - Aktivita-
ten in Baden-Wurttemberg, Auf dem Weg zur Wasserstoffenergie - Wie kommen
wir weiter?, (BAM-Seminar, Berlin, 1995), Federal Institute for Materials Rese-
arch and Testing, Berlin (1997) 37-35.
[15] GRETZ, J., WURSTER, R., Hydrogen Activities in Europe, VDI Berichte No.
1201, VDI-Verlag, Dusseldorf (1995) 229-243.
[16] HAIDN, O.J., et al., Improved Combustion Efficiency of a H2/O2 Steam Generator
for Spinning Reserve Application, (llth World Hydrogen Energy Conf., Stuttgart,
FRG, 1996), VEZIROGLU, T.N., et al., Hydrogen Energy Progress XI, Internatio-
nal Association for Hydrogen Energy (1996) 1418-1427.
[17] HAMA, J., Controllability of Hydrogen Combustion Gas Turbine with Inert-Gas
Recirculation, Hydrogen and Clean Energy (Int. Symp., Tokyo, 1995), NEDO
(1995) 349-352.
[ 18] HDW, HDW Wasserstoff-Energietechnologie, Howaldtswerke Deutsche Werft, Kiel
(1994).
[19] HOFFMANN, P., (Ed.), Hydrogen & Fuel Cells Letter 11 (1996) May.
[20] HOFFMANN, P., (Ed.), Hydrogen & Fuel Cells Letter 11 (1996) December.
188
[22] HOHLEIN, B., GANSER, B., Hydrogen from Methanol for Fuel Cells in Mobile
Systems, Hydrogen Production, (2nd IEA Technical Workshop, Julich, FRG, 1991),
6, Research Center Julich (1991) 159-170.
[23] ISHIDA, M., JIN, H., OKAMOTO, T., Hydrogen Fueled Gas Turbine Cycle with
Chemical-Looping Combustion, Proc. New Energy Systems & Conversion, (3rd
Int. Conf., Kazan, 1997) 69-73.
[24] KELLER, J., Development of a Clean, Premixed Burner for Hydrogen Utilization,
World Wide Web, http://www.eren.doe.gov/hydrogen/hypremx3.htm, US Depart-
ment of Energy (1997).
[25] KINOSHITA, K., MCLARNON, F.R., CAIRNS, EJ., Fuel Cell Handbook, La-
wrence Berkeley Laboratory (1988).
[26] KORDESCH, K., SIMADER, G., Fuel Cells and Their Applications, VCH Ver-
lagsgesellschaft, Weinheim (1996).
[28] KUIPERS, J., Status of the Westinghouse, EDB and ELS AM 100 kWe SOFC Field
Unit, European Fuel Cell News 5 (1998) January 19.
[29] LANE, R.M., FRY, M.R., BAKER, J.N., IEA Greenhouse Gas R&D Programme
- Fuel Cell Study, Report No. PH2/1, International Energy Agency (1995).
[30] LEDJEFF, K. (Ed.), Brennstoffzellen, C.F. Muller Verlag, Heidelberg (1995).
Design 17 (1996) 562-589.
[32] MANTHEY, C., Einsatz von Hochtemperaturreaktoren in der Eisen- und Stah-
lindustrie unter besonderer Berucksichtigung des Einflusses auf die Standort- und
Umweltprobleme dieses Industriezweiges, Report Jill-1180, Research Center Julich
(1975).
[33] MARCHIONNA, M., LAMI, M., GALLETTI, A.M.R., Synthesizing Methanol at
Lower Temperatures, ChemTech 27 (1997) No. 4 27-32.
[34] MEISSNER, D., Grundlagen der Brennstoffzelle, in: HAKE, J.-F., et al. (Eds.), 2.
Part I, Research Center Julich (1996) 177- 196.
[35] MENZER, R., HOHLEIN, B., Verfahrensanalyse der Stromerzeugung fur Fahr-
zeugantriebe mit Methanol als Energietrager und Brennstoffzellen als Energieum-
wandlungssystem, Report Jul-3445, Research Center Julich (1997).
[36] NABffiLEK, H., SOFC Research in International Cooperation: The IEA SOFC
Annex, (5th Int. Symp., Aachen, 1997), STTMMING, U., et al. (Eds.), Solid Oxide
Fuel Cells V, Volume 97-40, The Electrochemical Society, Inc., Pennington, USA
(1997) 26-34.
[37] NOLSCHER, C, Brennstoffzellen - Energiequelle fur das 21. Jahrhundert?, Physik
in unserer Zeit, 27 (1996) 52-59.
einem modifizierten Hochofengichtgas unter Einsatz von Kernenergie aus dem
Hochtemperaturreaktor, Report Jiil-2031, Research Center JQlich (1985).
189
York (1992).
[40] PESCHKA, W., Hydrogen the Future Cryofuel in Internal Combustion Engines,
(1996) 1453-1481.
[41] RAMACHANDRAN, R., et al., An Overview of Industrial Uses of Hydrogen,
(1996) 1407-1417.
[42] REGAR, K.-N., FICKEL, C, PEHR, K., Der neue BMW 735i mit Wasserstoffan-
trieb, VDI Berichte 725, VDI-Verlag, Dusseldorf (1989) 187-196.
[43] SCHMIDT, V.M., STEMMING, U., Hydrogen and Methanol Proton Exchange
Membrane Fuel Cells, (11th World Hydrogen Energy Conf., Stuttgart, FRG, 1996),
[44] SCHEUFELE, M., Howaldtswerke Deutsche Werft, Kiel, Personal Communication
(1998).
stoffeinsatz", TAB Working Report No. 13, Euro fur Technikfolgen-Abschatzung
[46] STARR, C., SEARL, M.F., ALPERT, S., Energy Sources: A Realistic Outlook,
Science 256 (1992) 981-987.
[47] STEPHAN, K., DAHM, B., Catalytic Combustion of Hydrogen in a Diffusive
Burner, (11th World Hydrogen Energy Conf., Stuttgart, FRG, 1996), VEZIROGLU,
Energy (1996) 1483-1492.
[48] STEWART, W.F., Operating Experience with a Liquid-Hydrogen Fueled Buick and
Refueling System, Int. J. Hydrogen Energy 9 (1984) 525-538.
[49] TAKEHARA, Z., Fuel Cells in Japan: Development and Future Prospect, Hydrogen
and Clean Energy (Int. Symp., Tokyo, 1995), NEDO (1995) 107-114.
NATIONAL ENERGY AGENCY, Comparing Energy Technologies, OECD/D5A
(1996) 237-296.
[51] VOIGT, C., Techniken zur energetischen Nutzung von Wasserstoff, BWK 38 (1986)
347-350.
[52] WILLIAMS, M.C., Integrated Gasification Fuel Cell System, World Wide Web,
http://coal.fe.doe.gov/coal_power/igfc_sum.html, US Department of Energy (1998).
[53] WINKLER, W., Konzepte fur Kraftwerke mit oxidkeramischen Brermstoffzellen,
Spektrum der Wissenschaft, 18 (1995) July 107-109.
[54] YAMANE, K., FURUHAMA, S., A Study on the Effect of the Total Weight of
Fuel and Fuel Tank on the Driving Performances of Cars, (llth World Hydrogen
Energy Conf., Stuttgart, FRG, 1996), VEZIROGLU, T.N., et al., Hydrogen Energy
Progress XI, International Association for Hydrogen Energy (1996) 1053-1062.
190
Chapter 8
SAFETY RISKS OF A LARGE-SCALE HYDROGEN APPLICATION
With increasing significance of hydrogen as an energy carrier, the risk of an accident

in a production plant or during storage and transport or while being applied will rise as well.
The broad introduction of hydrogen thus requires a concomitant and prognostic evaluation
of safety and risk. The inadvertant release of hydrogen caused by a leak or spillage creates
a local atmosphere that is depleted in oxygen and, thus, does not support life. Furthermore,
the explosion hazards associated with hydrogen when mixed with air, the assessment of
conceivable risks and its hazard potential after an inadvertant release is mandatory. The
tools for such a deterministic analysis of the risk are calculation models to simulate all the
steps of an accident scenario as realistically as possible. Model validation is based on the
evaluation of experiments and postexamination of previous accidents [120].
8.1. PROPERTIES OF HYDROGEN
Hydrogen gas is colorless, odorless, tasteless, non-toxic, and undetectable for human
senses. If released in a confined area, hydrogen can cause suffocation by dilution of
the oxygen content Gaseous hydrogen at its boiling point (20 K) is heavier than air.
At a temperature > 22 K, it becomes buoyant and tends to rise in the ambient air.
Hydrogen coexists in two phases, para and ortho hydrogen, whose partition depends on the
temperature. At low temperatures, < « 80 K, the para phase presents the more stable form.
Hydrogen exhibits in part a positive "Thomson-Joule effect" meaning a positive temperature
change upon pressure decrease. The effect is found for hydrogen at temperatures > 200 K,
for example, an increase of 6 degrees when released from 20 MPa to ambient conditions.
Mixtures of hydrogen with oxygen are flammable over a wide range of concentrations,
4 - 7 5 vol%. A stoichiometric hydrogen-air mixture contains 29.5 vol% H2- Despite
its relatively high autoignition temperature, the minimum energy required for an ignition
(0.02 mJ) is very low, further reduced by increasing temperature or pressure or oxygen
content Catalytically active surfaces can ignite hydrogen-air mixtures even at much lower
temperatures. The hydrogen flame is nonluminous, comparatively hot, but hardly radiates
any heat.
Liquid hydrogen (LH2) has the advantages of economic transportation in large amo-
unts, simple conversion technology, extreme cleanness, convenient consumption. Major
drawbacks are given by the enormous energy consumption of liquefaction which is around
30 % of its heat of combustion and the unavoidable boiloff losses during storage and hand-
ling, which can total 40 % of its available combustion energy [105]. Most experience in
handling LHa has been gained for decades in the space programs of the USA, the Russian
Federation, Europe, Japan, and China.
Slush hydrogen (SLH2) is a homogeneous mixture of liquid and solid H2- A

fraction of 50 wt% results in a 16 % higher density and an 18 % higher heat capacity
compared with the liquid phase. Both effects reduce volume and thus tankage weight
in possible applications, e.g., in aircraft,- improving transportation efficiency. A reduced
vaporization loss is given at the expense of a storage at higher pressures up to 1.3 MPa.
191
Table 8-1: Properties of hydrogen, from [37, 88, 101]
Molecular weight 2.01594 g/mol

Gas constant 4.1245 kJ/(kg K)
Normal boiling point 20.268 K
Heat of combustion, gross (including thermal energy 286.0 kJ/mol
of water vapor) 141.86 MJ/kg
Heat of combustion, net 241.8 kJ/mol
119.93 MJ/kg
Latent heat of vaporization 445.59 kJ/kg
31.78 MJ/m3 liquid
Vapor density @ Standard Temperature Pressure (STP) 0.083764 kg/m3
Vapor density @ Normal Boiling Point (NBP) 1.338 kg/m3
Liquid density @ NBP 70.78 kg/m3
Slush density @ Triple Point (TP) (50 wt% solid) 81.50 kg/m3
Solid density @ TP 86.50 kg/m3
Liquid-to-gas expansion ratio 845
Specific heat capacity of gas @ STP 14.89 kJ/(kg K)
Dynamic Viscosity of gas @ STP 87.5 10-7 kg/(m s)
Thermal conductivity of gas @ STP 189.7 10-3 W/(m K)
Diffusivity of gas @ STP 6.1 10-5 m2/s
Adiabatic speed of sound in air @ STP 1294 m/s
Stoichiometric composition in air 29.53 vol%
Limits of flammability in air 4.0 - 75.0 vol%
Limits of detonability in air 13. - 65. vol%
Minimum ignition energy in air (stoich.) 0.019 mJ
Auto-ignition temperature in air 858 K
Flame temperature (stoich.) 2318 K
Flame temperature (maximum) 2403 K(at31.6vol%)
Emissivity <0.1
Quenching distance in air @ STP (stoich.) 0.64 mm
Laminar burning velocity in air (stoich.) 2.37 m/s
Laminar burning velocity in air (maximum) 3.46 m/s
Flame front velocity (spherical gas cloud) 18.6 m/s
Detonation velocity (Chapman- Jouguet) 1968 m/s
TNT equivalent 2.22 kg TNT/Nm3 gas
192
The subatmospheric vapor pressure of (slush) hydrogen at 13.8 K can cause the total
system pressure to drop below that of the local atmosphere, requiring precautions to avoid
air entrainment into the system, which would solidify the air and warm up the system.
Another safety concern is electric charge accumulation in SLH2 flows since it involves solid
particles; this effect needs further investigation. NASA has studied SLH2 characteristics
under conditions of production, storage, and transfer [81, 109].
Various property data of hydrogen are summarized in Table 8-1. A more detailed
description of the properties is given, e.g., in [45, 120].
8.2. SAFETY MEASURES IN HANDLING (CRYOGENIC) HYDROGEN
8.2.1. Physiological Hazards
Direct contact with liquid hydrogen or surfaces on very low temperature causes
cryogenic "burns", i.e., freezing of living tissue, except for very brief contact periods where
the temperature difference between cryogen and skin is still high (film boiling regime) and
heat transfer small. Prolonged exposure to cold temperatures after a large spill lowers the
body temperature resulting in hypothermia, organ dysfunction, and respiratory depression
[43].
Vaporization of released liquid hydrogen affects the local atmosphere by diluting the
oxygen in the air. A fraction of less than 19.5 % is considered by NASA to be dangerous
to humans causing asphyxication; less than 8 % will be lethal within minutes.
Physiological effects as anticipated upon the explosion of a flammable gas mixture

are listed in Table 8-2; the effects arising from the accidental release and combustion of
hydrogen are treated in section 8.4..
Table 8-2: Physiological Effects of blast overpressures, from [43]
Maximum overpressure
Effect on personnel
[kPa]
7 Knock personnel down
35 Eardrum damage
100 Lung damage
240 Threshold for fatalities
345 50 % fatalities
450 99 % fatalities
8.2.2. Safety Measures
Safety of a system must be considered in all phases from design to fabrication and
its operation, sufficient instrumentation, safety analysis to assess consequences of untoward
193
Primary Secondary Tertiary
Safety Measures Safety Measures Safety Measures
Basic Safety by Design, Material, Manufacture,

Quality Assurance, Repeated Non-Destructive Examination,
Operation-Manuals. Qualified Personal, Accident Management
Prevention Prevention Mitigation

of of of
[nciderits Accidents Acc.Consequences
Fig. 8-1: Safety approach for handling hydrogen in the industries, from [10]
events, plant maintenance and documentation, well-trained and knowledgeable personnel

[43]. The investigation of the combustion behavior of hydrogen connected with the pressure
development under accident conditions is the basis of all safety-related considerations.
NASA experience with hydrogen began in the 1950s, when hydrogen was considered
the principal liquid rocket fuel. Safety engineering soon started to address hydrogen hazards
and to develop procedures for the safe operation of equipment and facilities. NASA also
determined the need for rigorous training and certification programs for personnel.
The large-scale industrial application of hydrogen has led to a safety concept with
a scale of safety measures aimed at minimizing risks by preventive safety measures and
by mitigation measures. A basic safety approach that can be applied to hydrogen energy
technologies is outlined in Fig. 8-1. Some examples are [45]
1. for primary measures (design):
welding as a basic principle of construction, metal gaskets for connectible parts,
selection of proper materials, waste or leakage or relief valve piping leading into
the open atmosphere, sufficient ventilation in closed buildings, handling of LH2 in
closed systems only, leak detection, regular check, training of personnel;
2. for secondary measures (avoidance of ignition sources):
prohibition of open fire, avoidance of electric or mechanically generated sparks,
avoidance of electrostatic charge, flame barrier or N£ inertialization system in
large plants;
3. for tertiary measures (mitigation of consequences):
quick shutdown system with cutoff of H2 feed lines, explosion-proof construction of
194
components (e.g., stack), avoidance of a transition from deflagration to detonation,
safety distance, flare rather than extinguish hydrogen fire.
Measures to be applied are defined and fixed in rules and regulations or in respective
recommended industrial standards or company internal rules.
8.3. EFFECT OF HYDROGEN ON MATERIALS
The stress which a structural material is able to withstand is determined by its ductility
(see Fig. 8-2). A material is elastic if, after being elongated under stress, it returns to its
original shape as soon as the stress is removed. At a certain strain, it departs from linearity
meaning that the material will retain a permanent elongation; the applied stress is called
"yield stress". A further increase of the strain eventually reaches the "ultimate or tensile
stress" beyond which the stress decreases finally leading to rupture. In contrast, a brittle
material does not exhibit the phase of permanent elongation. The material abruptly breaks
without any warning as soon as it is exposed to its tensile stress [43].
Metals that work successfully at low temperatures include aluminum and its alloys,
copper and its alloys, nickel and some of its alloys, as well as austenitic stainless steels.
I—02% Strain
Fig. 8-2: Ductile and brittle behavior, from [43]
195
8J.I. Hydrogen Embrittlement
8.3.1.1. Phenomena
Hydrogen has long been recognized to have a deleterious effect on some metals by
changing their physical properties. It is basically due to the presence of hydrogen atoms
dissolved in the metal grid and accumulating in disturbed lattice regions. Research on this
embrittlement effect has led to the definition of three different categories [39, 43]:
a. Hydrogen Reaction Embrittlement

Hydrogen reaction embrittlement is a phenomenon in which the hydrogen chemi-
cally reacts with a constituent of the metal to form a new microstructural element
or phase such as a hydride or gas bubbles ("blistering"), e.g., methane gas if combi-
ned with carbon, or steam if combined with oxygen. These reactions usually occur
at higher temperatures. They result in the formation of blisters or expansions from
which cracks may start to weaken the metal.
b. Internal Hydrogen Embrittlement
Internal hydrogen embrittlement means that hydrogen is introduced into the metal
during its processing, e.g., chemical reactions with water to form metal oxide
and liberate hydrogen. It is a phenomenon that may lead to the structural failure
of material that never has been exposed to hydrogen before. Internal cracks are
initiated showing a discontinuous growth. Not more than 0.1 -10 ppm hydrogen in
the average are involved. The effect is observed in the temperature range between
-100 and +100 °C and is most severe near room temperature.
c. Environmental Hydrogen Embrittlement
Environmental hydrogen embrittlement means that the material was subjected to
a hydrogen atmosphere, e.g., storage tanks. Absorbed and/or adsorbed hydrogen
modifies the mechanical response of the material without necessarily forming a
second phase. The effect occurs when the amount of hydrogen that is present, is
more than the amount that is dissolved in the metal. The effect strongly depends
on the stress imposed on the metal. It also maximizes at around room temperature.
8.3.1.2. Mechanisms
•
To which extent metal is degraded by hydrogen is influenced by strain (tension,

frequency, stretch velocity), specimen geometry (notches, bad welds, geometric inhomoge-
neities), the medium (pressure variation, temperature, impurities), and material (chemical
composition, fabrication, heat treatment, welding joints) [67].
Several mechanisms have been proposed which might explain at least partially the
degradation of metal by hydrogen embrittlement and which might act simultaneously:
- The formation of hydrides can lead to new hydrogen-related phases which may be
brittle and also may have a lower density than the pure metal leading to internal stress.
- The hydrogen distribution in a metal under stress is highly non-uniform which can
lead to locally increased hydrogen-enhanced plasticity causing local microscopic
deformation and eventually a failure.
196
- The lattice decohesion effect is presumed to cause embrittlement by a decrease in
the atomic bonding strength in the presence of hydrogen. A fracture occurs when the
stress exceeds the cohesive stress.
Molecular hydrogen precipitation forming high pressures and compound formation

are other mechanisms identified.
The above ideas help understand the observations that whether a metal is susceptible
to embrittlement by hydrogen or a hydrogen compound or whether not, depends on the metal
and also its metallurgical history which affect the migration behavior of hydrogen within the
metal. The embrittlement is strongly connected with locally high hydrogen concentrations
which can be caused by stress-enhanced diffusion rates to lattice defects and reaction sites
to initiate cracks [62]. Cracks grow when hydrogen concentrations reach a critical level;
crack growth stops when the crack has grown through the H2-enriched region. The ability
to explain the observations, however, was not accompanied by the development of non-
destructive methods to detect early stages of degradation and the ability to predict with
certainty the susceptibility of new alloys to hydrogen [71].
8.3.2. Low-Temperature Influence
Cryogenic temperatures can affect structural materials. With decreasing temperature,

ultimate stress and yield stress increase for most metals, generally connected with a
corresponding drop in fracture toughness which is a measure of the material's ability
to resist crack propagation. The lower the toughness, the smaller is the tolerable crack
length. A material can change from ductile to brittle behavior as soon as the temperature
falls below its "nil-ductility temperature" which is sometimes considerably higher than the
temperature of the cryogen. It is a particular problem in cryogenic equipment exposed to
periodic temperature changes. Several accidents with failure of a cryogen storage tank have
been traced to originate from cold embrittlement, for example the severe accident with the
rupture of a 4250 m3 LNG tank in Cleveland, USA, in 1944 [131].
Low temperatures can also affect materials by thermal contraction. The thermal
expansion coefficient is a function of temperature. For many materials, which are cooled
down from room to cryogenic temperature, more than 90 % of the total contraction
experienced will have already taken place at 77 K. Rule-of-thumb figures of thermal
contraction are 0.3 % in iron-based alloys, 0.4 % in aluminum, or over 1 % in many
plastics [43]. Cryogenic vessels or piping systems must account for this contraction to
avoid large thermal stresses.
The components of a cryogenic system usually undergo a thermal gradient, some

only during cool-down or warm-up phases, others even at steady state of operation. Strong
gradients, particularly if non-linear, result in stresses which under certain circumstances
may lead to rupture. Thermal gradients are of particular significance in systems with
stratified two-phase flows of cryogenics.
197
83.3. Materials Questions
Materials for hydrogen containing components have to meet the requirements of high
strength and high ductility to enable a high system pressure and have to be appropriate for
welding. Respective materials are ultrafine grain steels and thermomechanically treated
steels. Usually applied dense coatings on the surface of the ferritic steels containing
gaseous hydrogen prevent the adsorption of hydrogen unless they are damaged by plastic
deformation. Other sensitive areas are welding seems and surface quality. Also a high
degree of purity of the hydrogen favors embnttlement which could be influenced by adding
gaseous promoters (e.g., H2S) or inhibitors (e.g., 02 and CO) or inert gases (N2 and noble
gases) depending on the specific operating conditions [67].
An additional requirement for materials containing liquid hydrogen is a sufficient

fracture viscosity. Adequate materials are aluminum alloys, austenitic steels and steels
with a high Ni fraction. The need for weight reduction in mobile applications has led to
Boiling
Gas Poini "C Applied Steels
o - - Al-killed
Steels
Ammonia -33.4 -
Propane -42.1 -
50 -
•^\J
2.5%Ni Steel
Carbonic Acid Gas -78.5 —
Acetylene -84.0 — 3.5%Ni Steel
--100 -
Methane -161 .5>~ --150 - - 9 % N i Steel

Oxygen -182.9
Argon -185.9^
Nitrogen -196.0-
--200-
Aluminium Alloys
Austenitic Stainless
. Sleels
Hydrogen -252.8- --25Q- 36% Ni Steel (Invar)
Helium -268.9-
Fig. 8-3: Materials for containers of cryogenic liquids, from [45]
198
the development of composite materials, a matrix of plastic or metal in which glass or
carbon fibers are embedded. Qualities of the composites are a higher strength compared
with steel, a lower heat conductivity and almost no thermal expansion in fiber direction.
With decreasing temperature, however, composites are prone to enhanced embrittlement and
lower strength as well as a higher thermal expansion perpendicular to the fiber direction.
Fig. 8-3 shows which materials are appropriate as containers for various media at different
temperatures.
Two special materials testing systems have recently gone into operation at the
University of Stuttgart, FRG, to investigate storage vessel materials behavior at cryogenic
temperatures down to the helium boiling point of 4 K. The systems are suitable for tensile,
fracture mechanics and fatigue tests which are conducted at pressures up to 20 MPa [32].
A materials test equipment has been developed in Japan within the WE-NET project
to investigate the environmental hydrogen embrittlement of specimens in high-pressure
hydrogen up to 10 MPa at different temperature levels between 20 K and 1500 K. Additions
of various gaseous inhibitors have been examined as a preventive technique to arrest crack
growth in the materials [129].
Based on many years experience, a material related safety concept for hydrogen tech-
nology has been developed in Germany applying so-called "basis safety" and independent
redundancies. Goal of the "basis safety" concept as has been successfully applied in the
field of nuclear energy, is the evaluation and validation of information on critical leak
size, brittle and ductile behavior including the leak-before-rupture criterion of pipelines
and pressure vessels in accordance with the state-of-the-art. With respect to a large-scale
hydrogen economy, material development and optimization must be continued [67].
8.4. THEORETICAL AND EXPERIMENTAL INVESTIGATION OF ACCIDENTAL

RELEASE OF HYDROGEN
The technical risks which are existent when handling hydrogen, also include the
uncontroled release into the environment under accidental conditions. Three categories of
release types are distinguished:
• Vaporization of a liquid hydrogen pool on the ground

As a rule, the vaporized hydrogen escapes upwards, away from the site of release
and from ignition sources, an advantage which, however, does not hold for confined
or obstructed areas. The smaller the specific vaporization rate, the larger is the pool
and the more extended and shallow is the evolved gas cloud. An LH2 pool exhibits
a decrease in depth of 0.4 - 0.8 mm/s which is high compared with other liquid fuels
[41].
• Jet stream of liquid or gas through a small opening in a pressurized system
A liquefied gas is widely sprayed from the exit developing a multi-phase mixture. In
the opening, a narrow, so-called "flashing zone" is formed in which a certain fraction
of the liquid depending on its thermodynamic conditions is spontaneously (flash-)
vaporized. Adjacent zones are the zone of flow establishment characterized by a
dilution of the jet stream from its boundaries and the zone of established flow with
the full development of the plume interacting with the ambient [84].
199
A
Small hole in vapor Catastrophic failure of
space-pressurized tank pressurized tank
Jets
Intermediate hole in vapor Escape of liquefied gas from

space-pressurized tank a pressurized tank
Liquid
Evaporating
o cloud
Boiling pool
Spillage of refrigerated Spillage of refrigerated
liquid into bund liquid onto water
Fragmenting
jef
V = vapor
PL = pressurized liquid
RL = refrigerated liquid
High velocity fragmenting

jet from refrigerated
containment
Fig. 8-4: Conceivable release mechanisms from tanks containing liquefied gas, from
[50]
Catastrophic failure of a pressure vessel

The explosion-like expansion of a gas stored under pressure is characterized by its
rapidity, the high energy release, and the high chance of ignition. The positive
effect with respect to hydrogen is that a rapidly rising cloud is formed with a strong
release-induced turbulence.
200
In Fig. 8-4, illustration of some conceivable release mechanisms is given. Release
types include venting, leaks, or pipeline ruptures showing that the medium could escape
as a gas, liquid, or two-phase mixture. The figure also demonstrates that the specification
of the source term for an accidentally released liquefied gas is of particular importance
for the accurate modeling of the gas cloud evolution and its atmospheric dispersion. The
conditions and configuration of the source determine such features as release height, initial
plume distribution, gas cloud dimensions as well as temperature, composition, and energy
balances [50, 56].
8.4.1. LH2 Pool Formation, Spreading, and Vaporization
8.4.1.1. Phenomena
If hydrogen is released in its liquid form, it can accumulate and form a pool on
the ground and immediately starts to vaporize. A growing pool surface area implies an
enhanced vaporization rate. After the initial phase, a state is reached which is characterized
by the incoming mass to equal the vaporized mass. This, however, does not necessarily
mean a constant pool radius: on solid grounds, the pool surface area will continue to rise
due to the fact that the heat transport to the pool becomes smaller, the lower the ground
temperature, thus decreasing the vaporization rate per unit area. The temperature of the
cryogen itself corresponds to its boiling point and remains unchanged. A cutoff of the
mass input results in a breakup of the pool from the central release point creating an inner
pool front. The ring-shaped pool then recedes from both sides until it has completely died
away. The physical phenomena arising with LH2 spillage are depicted in Fig. 8-5.
LH2 vaporization is determined by heat transport into the pool. The principal
heat sources are conduction from the "hot" ground, atmospheric convection (wind), and
radiation from the ambient (sun, fire). Of all contributions, heat transfer by conduction
plays the dominant role; the error introduced by neglecting all other heat fluxes is in the
order of 10 - 20 % for liquid hydrogen [34]. Due to the large temperature difference
between ground and cryogen during the initial phase, a vapor film is formed inhibiting heat
transport (film boiling regime). Vapor that cannot escape to the sides due to friction forces
rises as bubbles disturbing the liquid layer. Cooling of the ground reduces the generation
rate of vapor leading to an unstable and partially collapsing vapor film which immediately
improves the heat transport between ground and liquid (nucleate boiling regime) connected
with a rapid cooling of the ground. The processes are similar if the LH2 is spilled on a
liquid surface, e.g., water, except that the pool penetrates the surface to a certain degree,
thus reducing the effective height responsible for the spreading; for instance, 7 % of an LH2
pool on water are underneath the water surface while it is 80 % for an LN2 pool. Cooling
of the liquid ground leads to a natural convection flow transporting warmer water to the
surface. A further decrease of the water temperature eventually leads to ice formation.
Thermal interactions between the cryogen and water at ambient temperature can
induce vapor explosions or so-called rapid phase transitions. A spontaneous phase change
occurs due to a sudden improvement of heat transfer between the two liquids. Under
certain circumstances, an essential fraction of the thermal energy could be transformed into
201
Water
Rapid phase
transformation (rpt)
Fig. 8-5: Physical phenomena after the accidental spillage of liquid hydrogen, from
[121]
mechanical energy. Besides, large amounts of gaseous hydrogen could be released into the
atmosphere within a short period of time.
The result of most interest in this initial stage of an accident is the temporal and
spatial production rate of gaseous hydrogen to be a boundary condition for describing the
dilution by atmospheric dispersion of the hydrogen-air gas mixture and the determination
of the flammable fractions within the concentration distribution of the gas cloud.
202
8.4.1.2. Experimental Investigation of Cryogenic Pool Spreading and Vaporization
Starting from a large-scale liquid fuel (LNG, LPG) maritime transportation and
application in the 1970s, interest increased in the assessment of the consequences of spill
accidents.
Many experiments have been conducted in the past with cryogenic liquids, mostly
LNG, LPG, or the non-hazardous LN2, to examine transient cryogenic pool behavior on both
solid and liquid surfaces [e.g., 24, 25, 34, 38, 64, 65]. Main goals were the examination of
the transient pool behavior under different release conditions on different surfaces and the
verification of the thermal phenomena. Only a few data are available on LHz pool behavior.
The NASA hydrogen release tests with the spillage of 5.7 m3 LH2 with the purpose
of investigating dispersion behavior [26] revealed a few coarse data about the pool size on
the ground which was compacted sand. The data revealed a much smaller pool size than
predicted with a simple model explained by a strong release-induced vaporization and by
ignoring the flash vaporization effect and possible percolation into the sand [26, 120].
The most detailed investigation of LH2 pool spreading so far was conducted by the
Research Center Jiilich [34] while conducting the hydrogen dispersion experiments under
BAM direction in 1994 near Cottbus, Germany [102]. In four spill experiments, two on
water and two on an aluminum sheet, the LHj pool size was measured as a function of
time and location. Release rates were 5 1/s over 62 s for the water tests and 6 1/s over 62
s for the Al tests. A pulsation-like release behavior resulted in a pool front moving in the
range between 0.4 - 0.6 m for the tests on water and between 0.3 - 0.5 for the tests on
aluminum, respectively. LH2 spillage onto the water created a massive ice layer with radial
ice tracks attached that resembled very much the structures as observed in continuous LN2
release experiments [34].
Within the WE-NET project (see section 9.2.1.), Mitsubishi Heavy Industries is
planning to conduct small-scale LH2 spill experiments for a detailed investigation of the
vaporization phenomena with the purpose to develop and validate respective computer
models [29].
8.4.1.3. Modeling of Cryogenic Pool Spreading and Vaporization
A concomitant effort of all experimental work on cryogenic pool spreading behavior

has always been its simulation by calculation models. The development of mathematical
models has passed several qualitative steps over the years. First approaches were of pure
empirical nature correlating pool surface and maximum diameter with time or mass released
[e.g., 24]. These model equations are only applicable to the material released. In an
improved class of models, a mechanistic approach is used where the pool is represented
in a cylindrical shape and the conservation equations for mass and energy were applied
[e.g., 23, 47]. This kind of model is mostly used at present for safety analyses purposes.
State-of-the-art modeling, however, is based on the solution of the so-called shallow-

layer equations, a non-linear differential equation system which takes into account the
conservation of mass and momentum. This allows the tracking of the dynamic behavior
203
O Trial #5,1st branch
V Trial # 5, 2nd branch
D Trial #6,1st branch
A Trial # 6, 2nd branch
Time
Fig. 8-6: Comparison between experimental data of the continuous LH2 release at a
rate of 0.006 m3/s onto an aluminum sheet and corresponding postcalculation,
from [121]
of pool growth and recession and calculates pool height and velocity as a function of
time and location. Respective calculation models with successful performance have been
implemented in the British GASP code [125, 126] and recently in the LAUV code [34] of
the Research Center Julich. The latter contains, in addition, a submodel for ice formation
on a water surface. Validation of the LAUV model was made by postcalculation of LN2
spill expenments conducted in Julich and the above described LH2 spill experiments and
of examples from the literature [34]. The calculation of a Julich experiment is shown in
Fig. 8-6.
8.4.2. Atmospheric Dispersion of Hydrogen
8.4.2.1. Phenomena
The accidental release of gaseous hydrogen or the spillage of liquid hydrogen leads to
the evolution and dispersion of a gas cloud whose shape is influenced by the type and rate
of release and by the atmospheric conditions as well as by topography. Liquid hydrogen
at 20 K which immediately starts to vaporize upon release, behaves like a heavier-than-air
gas in the very first phase ("cold sink effect"). This short phase of negative buoyancy
with a reduced heat and mass transfer due to the stable stratification effects is replaced
by an enhanced positive buoyancy as soon as, due to continuous air entrainment from
outside into the gas cloud, the cloud temperature has been raised to > 22 K. The dilution
204
with ambient air makes the hydrogen cloud pass through the whole broad flammable range
from high to low concentrations. The density of the hydrogen-air gas cloud is a function
of time. Condensation of the always available humidity in the ambient air, which makes
the hydrogen cloud visible, reduces the cloud density and liberates latent heat, both effects
supporting the positive buoyancy. The strong buoyancy of hydrogen connected with a rapid
mixing with the ambient air significantly reduces the lifetime of the flammable gas cloud.
While for small release amounts, the dynamics of the atmosphere is dominant, a large gas
cloud is observed to be capable of affecting the convective conditions, in particular, if
wind velocities are small [76].
8.4.2.2. Experimental Studies
Pioneering experimentation with liquid hydrogen started way back in the late 1950s
to examine hydrogen dispersion behavior as a consequence of its increasing application in
space aviation. Test series conducted by Arthur D. Little [7] confirmed the decelerated
buoyant behavior of vaporized but still cold hydrogen and its tendency of a horizontal
spreading and revealed strong concentration fluctuations.
The first significant experiments to study the dispersion behavior of a hydrogen gas
cloud was the NASA series of seven spill tests in 1980 with the release of 5.7 m3 LH2
per test within a period of 35 - 120 s [26]. Instrumentation allowed for concentration
measurements at various positions (see Fig. 8-9). High release rates in connection with
release-induced turbulence resulted in intensive cloud formation and rapid mixing with
ambient air. The tests demonstrated that the hydrogen cloud was existent for a couple of
100 m downwind, especially if the ground was able to cool sufficiently.
In 1994, the Federal Institute for Materials Research and Testing (BAM), Berlin,
conducted some spill tests with LH2 and LPG to compare the dispersion behavior of both
liquefied gases. Of special interest was the influence of adjacent buildings on the dispersion.
The amounts released were 0.5 - 1 m3 LH2 at a rate of approx. 0.8 kg/s and about 0.2 m3
LPG at a rate of approx. 0.25 kg/s. The tests revealed the completely different dispersion
behavior of the "light" hydrogen and the "heavy" propane and demonstrated that hydrogen
rapidly disappears upwards away from most ignition sources and does not accumulate
around the location of release [102]. Typical hydrogen gas cloud formations under the
given somewhat fuzzy wind conditions are shown in Fig. 8-7.
Test data have been collected for mixing and dispersion characteristics of methane
and hydrogen leaking into the interior of a passenger car. Non-vented cars were found to
hold flammable gas mixtures about 10 times longer than vented cars. Potential flammable
ranges were identified as a function of leak rate and (roof) vented area. An integral design
of the fuel system isolated from the passenger compartment incl. gas warning system was
recommended for future gas-fueled vehicles [8].
A whole variety of other experimental series involving many liquids or gases of

different densities, both on a laboratory scale in wind tunnels and on a large scale in field
trials, instantaneously or continuously, have been conducted in the past with the prime
goals to observe their behavior and to validate the performance of respective simulation
models [120].
205
Fig. 8-7: Hydrogen cloud formation during the BAM LH2 spill experiments as
indicated by condensed water vapor in the atmosphere, from [Research Center
Julich, ISR]
Fig. 8-8: Three-dimensional calculation of the 4 % hydrogen concentration isosurface

for trial #6 of the BAM LH2 spill experiments with the JRC Ispra code ADREA-
HF, from [111]
8.4.2.3. Calculational Methods
The need to simulate the atmospheric dispersion of a gas cloud in order to determine
extension and lifetime the flammable portions as realistically as possible has led to a
lot of approaches, however, with different degrees of success in reaching the ideal profile
of a mathematical model.
206
A classical approach is being made in the Gaussian plume model to simulate the
spreading of neutral (sufficiently diluted) gases or pollutants, e.g., radioactivity. It is a
simple model describing the concentration profile as a solution of the diffusion / advection
differential equation with empirical spreading coefficients depending on the atmospheric
conditions [87]. The Gauss model is inappropriate for treating the buoyant behavior of
light or heavy gases.
In box or slab models, the released gas cloud is assumed to be of cylindrical shape.
The processes of advection, i.e., the transport by the mean wind field, of air entrainment,
and of gravitational spreading are implemented in empirical correlations derived from
experiments. Box models were basically developed to simulate the behavior of a heavier-
than-air gas cloud with averaged (in later developments vertical profiles of) temperature
and concentration. Acknowledged (extended) box models are, e.g., the US code DEGADIS
[57] and the British code HEGADAS [75].
Many dispersion models, such as, for example, the software packages HGSYSTEM
or DEGADIS are accompanied by a jet model to calculate the dispersion of releases with
significant momentum flux (the dominant parameter for jets). The jet behavior falls into
two main zones: a region of adjustment from storage conditions to atmospheric pressure
followed by "conventional" jet dispersion at atmospheric pressure. As noted above, many
failure scenarios produce jet releases, which may be liquid, gas, or two-phase. If the
storage conditions are pressurized, an initial zone of adjustment to atmospheric pressure
would be expected possibly including flashing for a liquid or choked two-phase jet. The
conventional jet dispersion region begins with a so-called region of flow establishment,
in which similarity profiles for the concentration and axial velocity evolve; following this
the jet evolves with self-similar profiles. The main features distinguishing the various jet
models in this part of the flow are the treatment of the entrainment of ambient air and the
choice of similarity profile (top-hat, Gaussian).
State-of-the-art modeling of the transient behavior of gases with positive or negative

buoyancy in the atmosphere is provided by computer fluid dynamics (CFD) models which
simulate complex flow processes by solving the Navier-Stokes equations in an Eulerian
three-dimensional calculation grid structure. This approach comprises the conservation
equations of mass, momentum, and energy. Besides their need for immense computer
time and storage capacity which can only be realized by the current computer generation,
these models require a detailed input of initial and boundary conditions for the calculation
grid. In the two-equation k-€ turbulence model, special partial differential equations are
solved to describe the transport of turbulence as well as its generation and dissipation.
The standard k-e model has been modified in [97] such that the influence of thermal
stratification can be taken account of by introducing and additional expression of turbulence
generation/dissipation.
Of all the approaches, the k-e model offers the highest relative independence of
empirical relations. It appears to be the only one to allow a proper simulation of hydrogen
dispersion, because it meets the requirements of describing effects such as turbulence energy
in the gas cloud, interaction of the cloud with the atmospheric wind field, the characteristic
positive buoyancy, transient sources with initial momentum, and last but not least, gas flow
in a complex geometry (buildings, terrain). K-e modeling has been realized in a variety of
207
computer codes distinguished by the choice of the numerical solution method which was
found to have a significant effect on the calculation procedure.
Only a few theoretical efforts have been made to simulate the dispersion of hydrogen
gas mostly due to scanty experimental data. Early efforts were made in the late 1970s by
the Los Alamos National Laboratory on a box model for hydrogen taking account of heat
transport from the ground into the cloud [42] and later applying the Gaussian model for
neutral and buoyant dispersion as a part of the World Bank software package WHAZAN
[112].
The NASA has developed the dispersion code AFGASDM applicable to LHa and
other aviation fuels [22], which is in a sort of interim stage between a Gaussian model
and an Eulerian grid model. The model solves the conservation equations following a gas
"parcel" released as a puff until its dilution below the flammability limits. It takes account
of air entrainment and phase changes.
A 2D version of the BITC model [122] developed by the Solvay company, Belgium,
for heavy gas dispersion has also once been used for hydrogen to demonstrate its basic
applicability for light gases. It uses a Lagrange approach of tracking "marked" particles
representing the statistics of turbulence transport.
The 3D finite volume code ADREA-HF was developed for the computation of the
atmospheric dispersion of heavy gas clouds in complex terrain. It contains a one-equation
turbulence submodel taking account of two-phase processes [3]. The code has been applied
to ammonia, propane, chlorine, and also buoyant releases. It has recently been tested against
the BAM hydrogen release experiments. The calculational results for one of the trials are
given in Fig. 8-8 showing the hydrogen dispersion near buildings [111].
Mitsubishi Heavy Industries' computer code CHAMPAGNE is a multi-phase, multi-

component thermodynamics model originally created for the assessment of severe accidents
in fast breeder nuclear power reactors. It was recently modified to also treat the formation
and spreading of hydrogen gas clouds. CHAMPAGNE has been successfully applied both
as a 2D and 3D version to the NASA LH2 spill tests from 1980 (Fig. 8-9) [30].
A k-e atmospheric dispersion model POLLUT was developed at the TU Munchen

[93] to describe hot gas plumes escaping from stacks of power plants. The code was used
in a DLR study [45] to investigate hydrogen dispersion upon accidents with LH2 powered
cars releasing their tank contents both in open terrain and in a road tunnel.
The Battelle k-e model BASSEM originally designed for hydrogen combustion in
nuclear containments (see section 3.4.5.) was also applied to the above mentioned LH2
spill experiments conducted by the BAM, Berlin, for predictive calculations providing
reasonable qualitative results for 3D effects of hydrogen dispersion behavior [94].
The commercial CFD code CFX by AEA Technology [1] (see section 3.4.5.) has
been taken to predict the hydrogen gas distribution inside the ELBA fuel cell demonstration
vehicle (see section 9.5.2.2.) under H2 leakage conditions. It has, however, not (yet) been
applied to H2 dispersion in the open atmosphere, but according to its profile, it should be
an appropriate tool of simulation.
208
ALTTTUDEm
20
15
10
10 15 20 25 30 35 40
20 r
15
10
0
0 5 10 15 20 25 30 35 40
DOWNWIND DISTANCE, m
Fig. 8-9: Hydrogen concentration measurements during the NASA LH2 spill experi-
ment #6 (top) and corresponding 2D postcalculation with the Japanese CHAM-
PAGNE model (bottom), from [30]
8.5. COMBUSTION BEHAVIOR OF HYDROGEN
8.5.1. Flame Propagation, Deflagration
8.5.1.1. Phenomena
According to the evaluation of explosion accidents in the past, only a small portion
of the energy of a hydrogen cloud is expected to be liberated in an open air explosion; it
is estimated to be in the range of 0.1 - 10 %, mostly < 1 % [68].
The lower flammability limit of a hydrogen-air mixture was observed to be a function

of temperature expressed by the empirical Burgess-Wheeler law [131]. The respective value
near the boiling point (20 K) is 7.7 vol% and as low as 2.3 vol% at a temperature of 723
K. Because of its strong buoyancy connected with a rapid mixing with air, a hydrogen gas
209
cloud soon reaches concentrations in the flammable range, but is on the other hand also
soon diluted below its lower flammability limit In most accidental situations, the lean
limit is the most relevant one.
The ignition of a gas cloud always requires an ignition source. Under normal
circumstances, a spreading cloud is ignited at its perimeter. According to accident statistics,
the most probable reasons for the ignition of hydrogen gas clouds are autoignition on hot
surfaces, an electric spark, or an electric discharge. The positive Thomson-Joule effect
does, in principle, not lead to an ignition. In practice, however, ignition is observed in
almost all cases of sudden expansion due to electric charging by dust particles during the
release process [61].
The laminar burning velocity which is defined by the reaction rate of volume of
unbumt mixture per square unit surface or the velocity of the reaction front relative to the
coordinate system that follows the fluid motion, is a material property which increases with
increasing temperature. For a stoichiometric hydrogen-air mixture at room temperature, the
burning velocity is 2.37 m/s with a maximum of 3.46 m/s for a 42.5 vol% concentration.
In an H2/C>2 mixture, it is even as high as 9.4 m/s. The flame front velocity is defined by
the burning velocity multiplied with the expansion ratio leading to a figure for hydrogen
in air of about 24 m/s.
The turbulent burning velocity is significantly higher due to small-scale turbulence

and flame instabilities reaching the order of several hundred m/s. Most realistic flames are
turbulent. Turbulence causes an increase of both energy transfer and flame surface, thus
enhancing the burning process (positive feedback), plus it forms new ignition centers. At
present, the turbulent velocity cannot be sufficiently determined on a theoretical base, but
must be deduced from experience [2, 45].
Deflagration is the type of combustion with a subsonic burning velocity level. It

occurs under non-adiabatic conditions and at densities lower than that of the unburnt
mixture. Its propagation mechanism is conduction heating and free radical diffusion. Only
fuel-air gas clouds at around stoichiometric mixture imply high flame front velocities and
maximum expansion upon ignition, i.e., the ratio of the density of unbumt mixture to that
of the reaction products, which is about 7 - 8 for hydrocarbons (in reality smaller by up to
50 % because of non-ideal micro-mixture), but < 1 for hydrogen.
As a rule, the ignition of an unconfined vapor cloud leads to flame front velocities
small enough to preclude a fast deflagration and a pressure wave. In such a flash fire,
the burning energy is almost completely transferred into thermal energy. The so-called
"Unconfined Vapor Cloud Explosion" (UVCE) is the explosive type of deflagration where
about 60 % of the burning energy is expressed as kinetic energy in the form of pressure
waves [54]. The flame front consists of a preheating zone and a reaction zone. Heat
transport by conduction and diffusion from the reaction zone into the preheating zone
enables the propagation of the flame.
Experience with explosion accidents has shown that whenever hydrogen was involved,
a shock wave was created, however, a comparatively weak one because only a portion of the
gas cloud actually did explode. One reason might be the pinch effect, a flame deceleration
210
in a less dense mixture causing an isolation of the flame from the main cloud [54]. A near-
ground deflagration results in a hemispherical fireball which subsequently gains a spherical
shape due to buoyant forces. Depleting fuel terminates the radial spreading and the buoyant
forces allow for a liftoff of the fireball resulting in the typical mushroom shape.
"Boiling Liquid Expanding Vapor Explosion" (BLEVE) characterizes the failure

mechanism of a liquid gas container where the major part of the contents are released by
flash vaporization and explode while the remainder forms a pool which bums down. Three
conditions need to be met (i) the liquid is superheated, (ii) a sudden pressure drop occurs,
(iii) pressure and temperature conditions are such that flash vaporization is possible. This
type of explosion is caused if a storage tank is exposed to a heat source from outside. A
tank structure weakened by the high temperatures, a high inner pressure, and tensions in
the container wall near the gas / liquid boundary layer eventually results in the catastrophic
failure, however, merely causing pressure waves. An intact safety valve could delay, but
possibly not prevent the vapor explosion [73].
An even stronger explosion is represented by a "Boiling Liquid Compressed Bubble

Explosion" or BLCBE which runs in two steps. Starting with a partial failure of the
container by a crack growth, there is a spontaneous expansion of a two-phase mixture
which makes the pressure rise again up to its starting value. This then causes the vapor
bubbles to collapse creating pressure waves of 100 - 300 MPa. It finally results in a
rapid crack growth and a total failure of the tank with the whole contents to be violently
and uniformly distributed as a fine, high-turbulent aerosol. This failure mechanism was
experimentally verified [119].
The thermal emission of a burning gas cloud can induce further damages in the
neighborhood by igniting other objects. The thermal radiation of a hydrogen flame is low
(emission coefficient e < 0.1) compared with that of conventional gaseous fuels (e w 1)
leading to a much smaller thermal impact.
If an ignition occurs above an LHa pool, heat transport develops from the burning
gaseous cloud by conduction, convection, and radiation enhancing the vaporization rate.
The pool regression rate is dependent on the radius: For small radii, heat conduction from
the perimeter dominates meaning that the burning rate decreases with incresaing radius.
For large radii, radiation is dominant causing the burning rate to increase with increasing
radius. Burning rates have been measured in the range of about 0.3 - 1.1 mm/s [131].
If liquid hydrogen is released in ambient air with a high humidity content, the
flammable fraction of the gas cloud is generally enclosed in the visible cloud, while under
conditions of a dry atmosphere, flammable fractions could also exist outside the visible
cloud. These observations show that the visible cloud is not a precise indicator for the
flammable zone of the gas cloud [130]. The 4 % lower limit coincides with the perimeter
of the visible cloud at temperatures of about 270 - 300 K and humidities of about 50 - 57
%, but with strong fluctuations in the concentration which affect the capability of flame
maintenance.
Condensation and solidification of air in liquid hydrogen is always connected with an

oxygen enrichment. It leads to local combustion if ignited. If the oxygen fraction exceeds
40 %, the ignition immediately results in a detonation [108].
211
The first experiments conducted in the late 1950s by Arthur D. Little Inc. [7],
Lockheed-California Co. [96], and the US Federal Bureau of Mines [130] with the release
of up to 19 m3 liquid hydrogen and subsequent ignition have been used to find empirical
correlations for flame height and width in order to define safety distances. The more
delayed the ignition, the larger was the fire ball. The hydrogen flames were restricted to
the pool area, but far extended in vertical direction with flame height up to 50 m. It was
also observed that the fire ball could ignite flammable portions of the gas mixture separated
from the main cloud.
Ballistic tests and lightning strike tests have been conducted in the USA to investigate
the impact on LH2 and conventional JP-4 fueled areas of an aircraft. Test specimen with
dimensions of 40.6 x 2.5 x 30.5 cm3 and 5 cm thickness filled with fuel were subjected to
projectiles and artificial arcs. In the case of LH2, moderate fire balls were observed and a
deflagration of hydrogen vent gases with no evidence of a detonation. Hazards associated
with LH2 fuel appear to be less severe compared with JP-4 [70].
In Japan in the early 1980s, tests with ignition of LH2 in an open dewar of 0.3 m
diameter were conducted. A steady flame above the pool was observed higher and wider
than a respective gasoline flame. The burning rate expressed as a liquid level regression
was measured to be 0.4 mm/s or 6.5 times larger than for gasoline. Abnormal burning with
very high flames was observed when the burning LH2 came into contact with solidified
oxygen from the air in the LH2 [116]. Tests with burning LH2 pools have also been used
to examine different methods of extinguishing pool fires [59].
Within the so-called "Hydrogen Oxygen Vertical Impact", HOVI, project, the national
space institutions of the US (NASA) and Japan (NASDA) have recently conducted a series
of explosion experiments on the NASA White Sands Test Facility. In these tests, LH2 and
LOX tanks were dropped from a 70 m tall tower and exploded upon touchdown. Explosion
pressures have been measured to derive a TNT equivalent.
Ignition of the cone-shaped hydrogen gas cloud evolving from the leakage off a
compressed gas tank for vehicles is being investigated at the University of Miami with
the goal to improve the design of vehicle storage systems. Resulting overpressures were
greater than in the case of methane, however, still relatively low [114].
8.5.1.3. Modeling
Early explosion tests with gas clouds were used to derive simple empirical correlations
concerning flame extension or lifetime as a function of fuel mass released, e.g., WHAZAN,
a software package of the World Bank [115]. Other approaches are subjected to a turbulent
jet flame [74] or the dynamics of a rising and expanding fire ball or the impact by the
heat radiation flux [31]. Common to all models is the fact that they are based on a more
or less empirical approach with a contentious capability of predicting the consequences of
an explosion.
212
Detailed chemical mechanisms for combustion processes, in particular in gas mixtures,
comprise a large number of species and elementary reactions, a respective detailed coding
would not be practicable. Therefore a systematic reduction of the chemical mechanisms
down to a three-step or two-step reaction process is being made with different degrees of
approximation, e.g., the assumption of partial equilibria or the neglecting of highly reactive
species whose concentrations thus are small [86].
The flamelet model for diffusion flames has been developed at the RWTH Aachen
describing in a simplified manner the flame front of a multi-dimensional flow. The flame
front is assumed to be locally one-dimensional. By introducing an appropriate coordinate,
e.g., fuel-mass ratio, the determination of flame structure and of flame front propagation
can be separated [79].
The above mentioned CFD code BASSIM of the Battelle Institute was originally
written to simulate hydrogen deflagration in confined 2D geometries with combustion in a
one-step process. It has been successfully applied to calculate jet ignition and the influence
of water vapor on the deflagration [29].
The German DLR has been using a computer code to simulate 3D turbulent reacting
flows by solving the Navier-Stokes equations and using the k-e turbulence model in a
Lagrange approach. The modeling of the chemical kinetics of hydrogen-air combustion
was realized by considering 12 elementary reactions with 9 species including the NOX
formation mechanism. The code is intensively used for testing and optimizing hypersonic
propulsion concepts (Fig. 8-10) [80].
10-
1
§ A\
2-
O-
-2 0 2 4 6 B 1O 12 14 16 1B SO S2 24 26 2B
normalized length
Fig. 8-10: Phenomenological diagram of the complex flow field in an advanced

rocket thrust chamber and nozzle, from [DLR]
213
8.5.2. Flame Acceleration and Detonation
8.5.2.1. Phenomena
In contrast to a deflagration, the chemical reaction in a detonation is faster by several

orders of magnitude where the reaction front propagates at a supersonic speed in the order
of several km/s. The flammable gas mixture is compressed connected with a temperature
increase and autoignition of the still unbumt mixture. The release of the chemical energy
keeps the shock wave going. The detonation is a rare event, but the damage potential is
much stronger than in a deflagration.
The shock front possesses a 3D cellular structure characteristic of the fuel-air mixture.
The higher its reactivity, the smaller is the cell. In general, the detonation cell size, A,
is proportional to the characteristic chemical reaction time for the mixture. It can also be
related to other key quantities, such as detonability limits, confinement, geometry, critical
tube diameter20, chemical induction length, whereas the run-up distance does not appear
to be of fundamental importance. With 15 mm, the A of hydrogen is smaller compared
with other fuels meaning that detonation of a hydrogen gas cloud can occur even in small
geometries. The critical diameter in a stoichiometric mixture corresponds to w 13*A or
about 0.2 m. Deviation from stoichiometry increases the detonation cell size, e.g., A =
50 mm at 20 vol% H2 [55, 77]. The same effect results from the addition of diluents; a
10, 20, 30 % steam fraction was found to result in a A enlarged by a factor of 6, 20, and
60, respectively [15]. Also the addition of 8 % methane to a stoichiometric hydrogen-air
mixture was found to act as an inhibitor for DDT [44].
The range of detonability of a hydrogen-air mixture is from 18 to 59 vol% with the

most hazardous range between 35 and 40 %, measured in detonation tubes with diameters
of the order of a few centimeters. In larger experimental facilities (see section 8.5.2.2.),
even values as low as 15 % (FLAME) and 12.5 % (RUT) as the lower detonability limit in
air have been observed [21]. Depending on the strength of the ignition source, the limits
of detonability can even coincide with the flammability limits [98]. In pure oxygen, the
detonation range of hydrogen is 15-90 vol%. A mixture of gaseous hydrogen with solid
oxygen has the quality of an explosive and can easily detonate.
The detonation of a gas cloud occurs either directly by detonative ignition (explosive,
turbulent jet) or starting from a deflagration by flame acceleration. During the transition
from deflagration to detonation (DDT), two phases are distinguished:
1. Occurrence of a local explosion (hot spot) leading to a shock wave
The area in front of the reaction zone is ideally prepared for autoignition. Hot
spots are being created by focussing of weak waves behind obstacles.
2. Amplification of the shock wave until autoignition and transition to detonation by
means of an effective mechanism
One DDT mechanism is SWACER ("Shock Wave Amplification by Coherent
Energy Release") based on a stimulated release of the chemical energy which
goes synchronous with the shock wave.
30
Critical lube diameter is the minimum diameter of a tube that will permit a detonation wave to continue propagating into an open
volume as a spherically expanding wave.
214
The decisive criterion for flame acceleration is the state of turbulence in the gas cloud
defined by intensity and characteristic length and time of the spectrum. Main effects on
the flame speed are given by:
• partial confinement
Geometric arrangement, e.g., a lane, or interaction with (a series of) obstacles might
induces high turbulence. A positive feedback effect (enlarged flame surface area —»
enlarged reaction rate —* enlarged flame front velocity) leads to a continuous flame
acceleration. An upper limit for the flame front velocity is given when the effect of
rapid mixing levels the effect of temperature decrease due to the mixing. Hydrogen
flames have been observed to soon reach the stage of a quasi-steady velocity.
• Rayleigh-Taylor instability
If the boundary layer between two fluids of different densities is accelerated
towards the higher density fluid (like reaction products towards unburnt mixture),
disturbances are amplified causing an enhanced reaction rate (and vice versa).
• Markstein-Taylor instability
Pressure waves of local explosions can accelerate a preceding slower flame front
leading to a sudden increase of burning rate and flame velocity.
turbulence prior to ignition
Turbulence created during a release process does not have a long lifetime. However,
if an ignition hits this narrow time window of ideal turbulence and concentration
distribution, a high flame speed can be reached.
• type of release
If ignition in a confined area leads to an expelling of the unbumt mixture through
a relief opening into a gas mixture outside, the result could be a spontaneous (jet)
ignition, which is even more probable, if obstruction is available inside or outside.
• dust
whirled-up dust can absorb thermal radiation from the flame front increasing the
flame speed. The heatup effect can then lead to autoignition with initiation of
a detonation. This effect, however, needs a certain size of the gas cloud and a
certain loading with dust.
• gas cloud shape and size
The velocity of the flame front becomes smaller when it intersects the cloud
perimeter. Larger clouds exhibit a slightly higher flame speed.
All above effects have in common a generation of turbulence as the dominant

mechanism for flame acceleration. In addition, large-scale experiments have revealed that
also effects of scale and geometry have a strong influence on the probability of DDT. In
the whole detonation range, hydrogen appears to be much more sensitive to a transition
to detonation than the hydrocarbon fuels. DDT behavior can be influenced by introducing
inhibitors, which are non-reacting components such as nitrogen or water vapor or even
slow-burning hydrocarbons.
The explosion of an unconfined hydrogen gas cloud in the presence of a weak

ignition source was never observed to have turned into a detonation [45]. The ignition
energy required for a direct detonation can only be provided by either an intensive turbulent
hot gas jet or by an explosive. For a stoichiometric mixture, the minimum energy was
experimentally determined to be 4.7 kJ (« 1 g TNT). A methane-air mixture even requires
215
at least 22 kg of an explosive for detonation [89]. This type of explosion, however, is
irrelevant under realistic accident conditions. A real gas cloud evolved after a release,
normally has a concentration gradient which tremendously diminishes the strength of an
explosion as it were in a well-mixed cloud. Only an estimated 17 % of a real gas cloud
behave like a stoichiometrically premixed cloud [89].
Numerous large-scale and small-scale test facilities have been used to investigate
flame acceleration and the deflagration-to-detonation transition as well as the relevant
detonation parameters (e.g., blockage ratio, obstacle distance). The most dangerous
configurations were, as expected, those with (partial) confinement and obstruction. But
spontaneous transition has also been observed with a weak ignition source and at H2
concentrations relatively far from stoichiometry (0.6 -1.6 equivalence ratio). Key similarity
is the turbulence generation. A good survey on the worldwide experimental work is given
in [79].
The initiation of a detonation by means of a jet ignition has been demonstrated in

experiments for the first time at the end of the 1970s. An essential result was the rule of
thumb according to which the high flame front velocity and the overpressure have decayed
within the distance of 10 times the characteristic length of the vessel [113].
Within the nuclear power plant safety program and the PNP gas cloud program, the
German Fraunhofer Institute for Chemical Technology (ICT) has conducted different
test series involving mixtures of propane, ethylene, methane, and hydrogen with air
to investigate detonation and DDT in spherical, hemispherical, and tube geometries.
Unconfined, hemispherically shaped H2-air mixtures were investigated at volumes between
7.5 and 2100 m3 with a measured maximum overpressure of 6.3 kPa corresponding to 84
m/s flame velocity [35]. A series of six balloon tests was conducted in 1991 with hydrogen-
air mixtures at concentrations of 20 and 29.6 vol%, respectively [91]. The hemispherically
shaped gas cloud of 50 m3 volume was ignited in the center on the ground by an explosive
to trigger a detonation. Pressures were measured at various positions inside and outside the
balloon. Visually measured detonation speed agreed well with the theoretical value [20].
Tests with two interconnected balloons have shown that the detonation in one balloon
did initiate not more than a fast deflagration in the other, unless a tube connection was
chosen which was shown to pass on a detonative combustion. The geometry of the gas
cloud seems to play an essential role. DDT was examined using various igniters in balloon
volumes up to 500 m3. Plastic tubes with up to 25 m length were found to have the higher
degree of turbulence. Flame acceleration in all tests was far away from yielding a DDT.
Other balloon tests with up to 15 m3 volume and tests in plastic tubes with ethylene-
air gas mixtures were found to detonate by means of 8 g of an explosive, however, not
with a 100 J heated wire. Minimum mass of explosive is 80 g for propane-air mixtures,
and as much as 22 kg for methane-air mixtures. For real gas clouds, the required explosive
mass is even higher 300 g did not initiate a detonation of a flash-depressurized propylene
gas cloud, where 20 g would suffice for a stoichiometric mixture.
216
I Head-On
Fig. 8-11: Measured flame front profiles for the ICT test "GHT41". The rectangular-
shaped body in the center represents the balloon with a 40 % H2-air mixture.
Ignition point is near the left wall. The vertical rectangular box on the left-hand
side describes the position of the fan, the crosses are measurement positions.
The numbers on the contour lines represent time in milliseconds. DDT occurs
approx. 2 m past the fan, from [15]
A series of 17 tests was carried out by the ICT with ignition of pancake-shaped, free,
homogeneous, near-stoichiometric methane-air or propane-air gas clouds of up to 13,000
m3 volume. The generation of additional turbulence due to obstacles or entries/exits was
observed to result in locally higher flame speed (up to 80 m/s) and overpressures (up to
8 kPa) compared with undisturbed clouds with respective figures of 6 - 8 m/s and 0.1 -
0.2 kPa [90].
More ICT tests were conducted under conditions of partial confinement (two parallel
walls forming a ! 0 x 3 x 3 m 3 lane) and with rich (36 - 41 vol%) H2-air mixtures. With
full fan-induced turbulence, flame acceleration was observed until DDT occurred resulting
in pressures and velocities exceeding the Chapman-Joguet values (see next section) [35].
Flame and detonation front profiles for one of the tests are plotted in Fig. 8-11 clearly
showing the sudden transition to detonation shortly past the fan.
Shock tube experiments with hydrogen-air mixtures conducted at the RWTH Aachen
have shown different modes of self-ignition. At lower temperatures (< 1150 K), randomly
distributed hot spot mild ignition leads to a deflagrative type of combustion [16].
The Research Center Karlsruhe has conducted combustion test series with premixed
stoichiometric hydrogen-air mixtures both in a 12 m long cylindrical tube of 0.35 m diameter
and in a rectangular channel with 0.1 x 0.1 m2 cross section and 3 - 6 m length. Different
217
degrees of obstruction allowed the examination of the whole spectrum of turbulent flames
and their reflection and diffraction at obstacles and system boundaries [20].
DDT experiments at the Technical University Munich have been carried out in a 66
mm diameter and 6 m long closed test tube with a 3 m acceleration section. The degree
of obstruction was varied between 30 and 70 %. Mixtures with < 10 vol% H^ could not
be ignited reliably. High flame speeds were observed at 10 - 13 % H2 concentrations in
the obstructed section. At > 20 vol%, detonations occurred; a lower H2 concentration is
sufficient the smaller the initial temperature. For high blockage ratios, the flame was found
to be strongly accelerated at the beginning, but then eventually quenched [11].
Another facility to investigate DDT in hydrogen-air-steam mixtures has been operated

at the Battelle Institute, Frankfurt, with the goal to examine the influence of steam as a
possible inhibitor of a detonation. Two series of experiments have been conducted under
conditions simulating a core meltdown accident in an LWR: (i) at 100 °C and 0.1 - 0.2
MPa with a homogeneous mixture (9 tests) and (ii) at 120 - 130 °C and 0.2 - 0.3 MPa with
a high steam fraction and H2 concentration gradient (12 tests). The experimental setup
consisted of a tube of 10 m length and 0.4 m diameter installed in a closed steel vessel
with 5.4 m3 total volume. The flame front passes a 2 m acceleration section with obstacles,
propagates upwards through the inner tube, and then after exiting u-turns down in the outer
vessel. The significant change of geometry in the upper part makes the combustion process
rather complex enhancing turbulence in connection with a quenching of the flame. In four
out of the 21 tests, DDT was observed [14].
A small-scale test facility, FAST, was erected at the TNO research center, The
Netherlands, to investigate the influence of obstruction, inerting, and multicompartment
arrangement on hydrogen explosions under LWR accident conditions. It consisted of an 8
m long duct with a 0.5 x 0.25 m2 cross section. A total of 450 tests with a H2-air mixture
were conducted. A higher blockage reduced the flame speed. Partial inerting was found to
increase pressure wave duration and to decrease the amplitude [123].
With the HOT (Heated Detonation Tube) facility, the Sandia National Laboratory
has constructed a 0.43 m inner diameter tube capable of operating at temperatures above 100
°C. Detonations were observed in the H2 concentration range between 13.5 and 70 vol% at
room temperature confirming the tendency towards lower figures in scaled-up experimental
arrangements. At 100 °C, the lean detonability limit was down to 13 %. Detonation was
even observed at a 30 % steam diluted mixture [15].
More large-scale tests were conducted at Sandia in the two channels, FLAME (Flame
Acceleration Measurements and Experiments) [110] with the dimensions 30.5 * 1.8 * 2.4
m3 and MINI-FLAME, reduced in size by a factor of 8, with one open end and movable
top and including obstacles. The FLAME facility had an open end; its top could vary from
fully closed to 50 % open with intermediate open spaces. A total of 28 tests were carried
out in FLAME with variation of H2 concentration (12 - 30 vol%), obstacle blockage (0 -
33 %), top venting. The observed lower detonability limit was at 15 vol% H£ with and
at 25 vol% without obstacles. DDT also occurred for 24.8 vol% H2 and a 13 % open
top. Top venting was found to have two counteracting effects: increase of burning rate
218
due to turbulence dominant at low venting rates, flame speed reduction due to loss of gas
for large venting rates [128].
Hydrogen-air detonation in a 3D geometry has been investigated in the Russian

facility RUT (Rocket Shock Tube). The channel used has a variable cross section with a
length of 69.5 m and a total volume of 480 m3, representing presently the largest facility for
such experiments in the world. Ignition was made by detonators. Test series conducted were
with a premixed hydrogen-air atmosphere (1993), with dynamic H2 injection in air (1994),
and, relevant for nuclear containments, with a premixed hydrogen-air-steam atmosphere at
elevated temperature level (100 °C). One of the key results in RUT was the observation
of a 12.5 vol% Ha lower detonability limit, demonstrating that larger-scale facilities seem
to extend the detonability range of a flammable gas mixture [20, 79]. The Russian UTR
facility, an up to 3 m long tube with 66 mm diameter was taken for systematic studies of
the effect of different DDT locations on peak pressure and impulse using a 2H2-O2-1.75N2
gas mixture [20].
8.5.2.3. Modeling
The classical theory of detonation is formulated in the ZND model (Zeldovich, Von
Neumann, Doring) assuming that the propagation mechanism of a detonation is ignition
by shock-induced adiabatic compression.
The Chapman-Joguet model [113] offers the opportunity to approximate the pa-
rameters of the complex dynamic process of a detonation such as detonation velocity,
pressure, and transformed energy. For a stoichiometric hydrogen mixture, these parameters
are: 1968 m/s, 1.6 MPa, and 2.82 MJ/kg, respectively.
The TNT equivalent model is often used as a simple method of estimating the mass
of TNT per mass unit fuel gas whose detonation results in the same blast wave at the same
distance. One kg of TNT translates into an energy of 4520 U. The equivalent for hydrogen
is 2.22 kg TNT per Nm3 gas. The weakness of this model is to ignore the pressure-time
characteristic differences between a gas cloud and a detonative TNT explosion. In the
short range, the model overestimates the pressure. Furthermore the model does not take
into account the influence of turbulence and confinement [113].
A better approach is given by the numerical calculation models. Detonation codes

have been developed at the Research Center Karlsruhe, DET1D and D3D, to determine
the characteristic parameters within the reaction zone and outside in the unbumt mixture
[20, 95, 100]. These models consider a homogeneous mixture of H2, 02, N2, H2O and are
mainly applied to examine the load on a nuclear reactor containment. Validation was made
against the above mentioned balloon tests of the Fraunhofer Institute and the Russian RUT
experiments. Parameter calculations of 3D detonation have shown that the 3D structure
is unimportant for the pressure load and that a relatively coarse grid provides sufficient
accuracy [20].
The Multi-Energy method developed by TNO and coded in REAGAS [117] con-
siders a vapor cloud explosion to be composed of a number of sub-explosions. Portions
of the gas cloud which are confined up to a certain extent, are correlated with an initial
219
(a) Numerical (b) Schlieren record
Fig. 8-12: "Tulip" flame in a closed channel, from [48]
pressure value. The strength of the explosion blast is expressed as a number between 1
and 10 representing categories of "insignificant" (= 1) to "detonative" (= 10). These cloud
portions are responsible for the pressure buildup. The model simulates the feedback in
the interaction between combustion and reaction products flow as a consequence of heat
transfer and turbulence. Turbulent combustion is realized in a one-step process.
State-of-the-art are the CFD models, which are similar to the dispersion models, based
on the Navier-Stokes equations for momentum conservation with a k-e approach to deal with
turbulence. They need, however, to be adapted to the characteristic spatial and temporal
lengths for chemical reactions. Typical time steps for hydrodynamic flow calculation are
in the order of 10"6 s and for the H2-C>2 reaction mechanism IO"10 s occurring on the 10"4
m scale [20]. Most approaches are based on the assumption of incompressible flows or are
restricted to slow deflagrations or to pure detonations. The assumption of incompressible
gases in fast deflagrations with Mach numbers > 0.3, however, is not applicable since the
pressure waves caused by the flame front are no longer negligible. Numerical simulation
methods are applied to study turbulent combustion, flame acceleration processes, and (re-
initiation of) detonations. Today's modeling and simulation show a good agreement with
a variety of fast combustion phenomena [48].
The DLR Stuttgart has developed a 2D algorithm to model the flame folding in
the early phase of the process after ignition and the acceleration process in obstacle areas
which can lead to detonation. The method of approximate mapping functions is introduced
to allow the simulation of more complex geometries. Also the detonation propagation
and the quenching of a detonation and an eventual re-ignition are numerically simulated.
The numerical analysis of flame acceleration is based on the random vortex modeling
the turbulent eddy movement. What is not yet possible is the adequate modeling at the
link between deflagration and detonation (DDT). The results obtained with the calculation
model are in quite good agreement with the Schlieren pictures taken from experiments (see
220
Fig. 8-12) [66]. The results of CFD calculations were used to optimize the design of a
detonation arrester geometry [48].
A 2D calculation code has been developed at the RWTH Aachen to simulate fast
turbulent flame propagation in hydrogen-air explosions in large-scale geometries. The
trubulent flame front is considered as a reactive discontinuity. The model is based on
the solution of the compressible Navier-Stokes equations with chemical source terms and
buoyancy forces using a k-e turbulence submodel. Validation work was made against the
hydrogen combustion tests in the Russian RUT facility showing good agreement between
experimental and calculational results for the flame propagation in the obstructed channel
section, except for the fact that the DDT observed in the expanded part of RUT could not
be reproduced. This is explained by a still insufficient grid resolution and/or inaccurate
initial turbulence distribution in the simulation [79].
The 3D CFD computer code FLAGS (Flame Acceleration Simulator) developed at

the Christian-Michelson Institute in Bergen, Norway, is able to handle complete chemical
processing plants down to the level of piping. The code treats spreading and combustion
of flammable gases and determines concentration distributions and the impact on structures
from explosions. Its main application is the prediction of explosion effects in offshore
geometries [118].
The Harwell code CFX has been described already in the section on dispersion
models. The chemical kinetics submodel specifies a chemical system of several species
and reactions with an offered selection of different combustion models. Heat transfer is
accounted for by convection, conduction, and radiation [1]. Examples calculated at the
Research Center Julich include Russian RUT tests and combustion experiments conducted
in a Swiss railroad tunnel.
8.53. Pressure Impact and Damage Analysis
8.5.3.1. Phenomena
Combustion waves generate pressure waves. In a deflagration, the volume expansion

of the burnt products reacts like a piston pushing the unburnt gas mixture aside. The
pressure is in principle uniform and static. At the cloud perimeter, expansion waves are
created which both run backwards into the cloud or follow the shock wave. Pressure and
density decrease in the wave direction.
The static pressure loading in slow deflagration processes is conservatively described

by the adiabatic, isochoric, complete combustion (AJCC) pressure representing an upper
bound in a closed volume. Analytical load functions approaching the AICC pressure in the
long term are presented in [19]. Mitigation of the (theoretical) AICC pressure is given by
incomplete combustion, venting, heat losses, and/or the addition of diluents. Therefore, the
maximum static pressure will be generally lower than the AICC pressure. For combustion
at atmospheric pressure, the AICC peak pressure is 0.8 MPa enough for many building
structures to exceed their failure limits.
221
In contrast, a detonation generates a compression wave. Its pressure-time behavior is
characterized by a distinct increase ("Von Neumann spike") and a subsequent exponential
decay. Behind the shock wave, a pressure discontinuity is observed where the pressure
drops below the atmospheric pressure evoked by the so-called "molecular collapse" due
to the much denser oxidation product water upon hydrogen combustion. The peak over-
pressure increases with increasing flame speed. Transient pressures can be locally higher
thean the AICC pressure. Inhomogeneities can result in local detonations decaying to de-
flagrations. When the detonation wave leaves the cloud, it will become an expanding,
decaying wave. Overpressures drop fairly rapidly in a spherical expansion. In the far
distance, deflagration and detonation show about the same pressure shape. An example
from nature is the flash in a thunderstorm with a typical peak overpressure of 50 MPa.
In the intermediate stage of a fast deflagration, the peak overpressure is not so high,
but the pressure drop is over a larger time period making the impulse, i.e., the integral
of pressure over time, which is a measure for the load on a building structure, about the
same in both cases [12].
Dynamic loads of fast transient pressures are imposed in the presence of inhomo-
geneously distributed combustion energy and are specific to the structure geometry. The
response to an incident shock wave could be a movement of solid objects [60].
The essential parameters are peak overpressure and positive / negative phase of the
specific impulse which are dependent on the liberated explosion energy which itself is
dependent on pressure and volume of the container, on the type of gas, and on the distance.
Deflagration overpressures for an unconfined hydrogen gas cloud are in the order of 10
kPa. As a rule, an overpressure of 2 kPa is considered safe although about 10 % of the
window panes could be ruptured [108]. An 11 vol% Ha mixture suffices to cause damage
to human beings and buildings [46].
Gas explosions in a vessel which is connected by a small opening to another vessel

produce peak pressures and pressure increase rates in the second vessel much higher than in
a single vessel explosion, a phenomenon referred to as pressure piling. A pressure in excess
of 3.5 MPa was measured in a two-chamber vessel for a stoichiometric hydrocarbon mixture,
where 0.8 MPa were expected for a single vessel. From experiments with stoichiometric
methane and propane-air mixtures, it could be concluded that in front of the propagating
flame incl. the secondary chamber, a higher turbulence level is created which leads to an
accelerated flame speed into the tube and the second chamber with a large overpressure.
Unlike the length of the interconnecting tube, its diameter has a significant influence on
the overpressure. An empirical model has been developed to predict explosion pressures
in interconnected vessels up to a volume of a few hundred liters [106].
Series of tests concerning the simulation of heat fluxes and temperature loadings
on equipment, pressures, and moisture conditions during multiple hydrogen burns were
conducted in CRTF (Central Receiver Test Facility) and in SCETCH (Severe Combined
Environment Test Chamber [60].
222
Characteristic destruction lines for explosions Destruction pressures generated by nuclear
produced by detonating conventional explosives,, ^K- -•»•
weapons, according to Glasstone
according to Schardm (ground explosion] [recalculated for ground explosion)
!
y
5 X Ul ft
\ \
rjr>9« <f ttx BASF r»»lu«hon self-supporting 2-block wall {t8cm)
\
—N
1 light steel frame, heavy damage
-10 ,-„,, ——— „,
_,. - 2 -block Rlrnl
m
ur i \ "'
•a
<u
10 —— . <* Mf-H»bff- •\ ^
^ 'v HL , <c«l
•2-storey brick house, 1-block walls
(ZCcm) completely destroyed
——— true tyre -- ——^-bl oclK _A
v^ .
multi-storey buildings with load-
s T
tflffmY^,
•JI i^ -I// -block . jll 11?
,
05 : ll ^^ Tt ———^ bearing brick walls, heavy damage

1-0 t- K^;" ~r>((^ ^*\* wft
k——//
• • — —^
-X= ^c\
v
N k VS
k
^ s
N
y
"
-light steel-framed structures,
moderate damage
^s ^< .;
^
û ^v - ^ multi-storey buildings with load-
010
S.^ - i_ _. * ;>x^- ,SO' ', win do r- -•xS^-—- ?f^ bearing brick walls, moderate damage
s
lâft ^*ii,
^*-, *• . .
-_=\
005
*"<
^ -^^~-— I ~J
^
=v "*• - -. half-timbered houses, moderate damage
I--0 » f >,
V« window-- *
i —1
b__
--En tf-fv 50% of window crossbars pushed out
om
1.0 10
^
50 100
•1, ^
^ "V
slight damage to window frames
window panes
Distance r [mj [—- iarge distance
Fig. 8-13: Characteristic destruction lines for explosions produced by detonative

conventional explosives according to Schardin (ground explosion), from [52]
8.5.3.2. Modeling
In the post-combustion modeling, the partial transformation of combustion energy

into mechanical energy is treated reproducing the complicated structure of a shock wave
and determining the pressure distribution on buildings. Pressure loads depend on the mutual
orientation of the incoming wave and the loaded surface.
In general, it is possible in a first approximation to deduce the strength of the destroy-

ing pressure waves from the evaluation of damage. Charts with characteristic destruction
lines give peak overpressures of a damage-equivalent pressure wave as a function of the
distance ("Scharding criteria") [52] as shown in Fig. 8-13. "Experimental" data were
derived from explosives and nuclear weapon explosions (which both are detonations). The
Scharding criteria, however, are only partially applicable to a deflagration. Here for a defla-
gration to be expected in an unconfined explosion, the relation between mass of flammable
mixture and pressure wave is derived from an evaluation of damages that occurred in
accidents in the past
In a joint French/Italian benchmark study, measured pressure histories from detonation

tube tests with stoichiometric H2-air mixtures have been compared with the codes DET1D,
D3D, and PLEXUS. The agreement in pressure amplitudes and speed was quite satisfactory
[20].
223
Computer simulation techniques have been developed by SRI International in the USA
for analyzing the explosion of a pressurized hydrogen tank and the response of structures and
humans exposed to the resulting blast environment. The SRI equilibrium thermodynamic
code TIGER is used to calculate explosion pressures and expansion characteristics of the
explosion products. After obtaining the pressure contour plots, the DYNA3D model is
taken to determine structural response. A simplification of the blast load damage analysis
has led to the development of the Pressure-Impulse (PI) methodology. Its basis is the
observation that for a broad range of structures subjected to simple pulse loads, the final
damage state depends on peak pressure (P) and total impulse intensity (I) only. For a specific
level of damage, the corresponding load varies between high pressures with short duration
pulses and low pressures with long duration pulses. Focus is laid upon the development
of a PC-based algorithm, AACA (Accident Analysis Computer Algorithm), for quick and
easy evaluation of accident scenarios, which includes a continuously growing library of
pressure-impulse load and damage curves that are based on either experimental data or
computer simulations [99].
Combustion of unconfined hydrogen-air clouds have shown to result in simple

deflagrations with typical overpressures no higher than 7 kPa, but high enough to knock
down personnel [43] (see also Table 8-2).
8.5.4. Safety Distance
The safety distance as defined in Germany is the required distance between the
location of a gas leakage and the object to be protected which takes account of the
evolving flammable atmosphere as well as of the pressure and heat wave resulting from
a possible ignition. The distance can be fixed according to physically defined criteria,
e.g., the dose of thermal radiation or peak overpressure, to have reached a certain threshold
value. A minimum safety distance is desirable for economic purposes. A basic prerequisite
is knowledge of the source term which is dependent on leak size and thermodynamic
conditions of the leaking material. A problem is given by non-quantifiable leakages,
e.g., from cracks in welding seams. Minimum safety distances between LH2 storage
systems and inhabited buildings are shown in Fig. 8-14 as a function of LH2 mass. The
curves illustrate the variation in conservativism of different institutions that generate safety
criteria. The safety distances are different for people and for less demanding equipment,
e.g., adjacent storage tanks, working buildings, or distinguished with respect to fireballs,
shrapnel, structural response or physiological effects (Fig. 8-15).
A formula for the safety distance is generally acknowledged to have the form
R = k * M1/3
with R the distance in m and M the mass in kg. The formula is extended by damping
parameters, if some protective measures, e.g., wall or earth coverage, are being applied. In
Germany, k values have been fixed as listed in Table 8-3, depending on what object is to
be protected. For the more relevant cases where the safety of people is of primary concern
and material-related damage of secondary order, the k-factor of the above formula has been
defined based on a pressure shock wave of 15 kPa for a human to be tolerable [35].
224
1 US Dep. Defense Instruction No. 4145.21 (1964)
2 National Fire Protection Association (NFPA), Boston (1973)
3 US Army Material Command Safety Manual No. 385-224 (1964)
4 Burean of Mines, Pittsburgh (1961)
5 German Fed. Ministry of the Interior, Bonn (1974) (nuclear power plants)
for liquefied gases
6 High-Pressure Gas Control Legislation, Japan (category I)
2000
LH2 mass
Fig. 8-14: Safety distances diagram (please note the scale change on the ordinate!)
#1 and #3 from [40], #2 and #6 from [4], #4 from [130], #5 from [36]
10 'FT I r t i T i I 11 I I I M I
STORAGE AREAS '
1* Data from Bureau of Mines
2 Data from ASESB-LeRC
EXPERIMENTAL AREAS
3 Equivalent spherical fireball radius

HYF=1. 0 < W f < 3 3 . 3
Based on
10'
rl) HYF=3.22Wf^.Wf>33.3
6 Protection from shrapnel in TNT
Explosions & Space vehicle explosions
^ 10
•Undiked
QUANTITY vs DISTANCE
for INHABITED BUILDINGS.
ROADWAYS, etc.
" 10*
(Unfaarricaded)
10
i I i i i l iiI I i
10'
10° 10' 102 103
0 = DISTANCE, m
Fig. 8-15: Safety distances diagram, HYF = Hydrogen Yield Factor in g TNT/g H2,
W f - Fuel Weight, from [58]
225
Table 8-3: K-factor of the safety distance formula according to official regulations
in Germany, from [35]
K-factor in safety distance

Object formula R = k * M173
Working building 2.5 - 8

Neutral building 8 - 22
Residential building 22
No damage 200
The German regulation TRB 810 No. 25 proposes a safety distance of 120 m for
> 300 tons of liquid gas storage tanks. The difficulty arising with this figure is the
enormous consumption of land which turned out to be not practical in all cases. Therefore
an additional rule was released saying that the safety distance could also be determined
by considering the traveling distance of an accidentally released flammable (heavy) gas
cloud [9]. The computational model VDI 3783 is an acknowledged guideline in Germany
to allow for a conservative estimation of such a safety distance [53]. Another rule says
that the safety distance can be cut to half, if additional measures are applied, e.g., earth
coverage, rupture discs, far-distance emergency shutdown, or warning systems to initiate
deluge or spray devices [9].
The deployment of protection walls could have a counteracting influence on the

accident consequences. The effects of obstruction and turbulence generation can usually
lead to enhanced flame speed and pressure increase. Therefore gaslight fences or walls
should not be generally applied and should be restricted to special cases only [53].
The safety distance between LH2 and LOX tanks at the Tanegashima Space Center
was set to be 167 m according to the US Air Force Standard for blast wave pressure [83].
8.6. EXPERIENCE IN HANDLING HYDROGEN
Within the distribution system of pressurized hydrogen, cases of damage were ob-
served in terms of leakage due to structural defects on the inside surface of the vessel, or
of catastrophic failures as a consequence of frequent pressure fluctuations due to refilling
procedures.
The first spectacular damage reported was as early as 1894 where at least 70 out
of 1000 H2 pressure bottles exploded. Little knowledge was available at that time on
the embrittlement phenomena. Further failures occurred in NASA pressurized hydrogen
storage tanks at 35 MPa occurred as well as in hydrogen tube trailers in Germany in the
1960s and 1970s.
The Hi- '^nburg accident in 1937 at Lakehurst is very familiar to the public if the
talk is on hydrogen. In fact, the 220 tons Zeppelin LZ-129 Hindenburg, the largest airship
226
ever built filled with 200,000 Nm3 of hydrogen in 16 gas cells, did not explode, it tragically
caught fire when docking. At that time practice was to allow hydrogen to be stored in highly
combustible materials [33, 82]. This accident has put a preliminary end to the development
of airships as a means of transportation.
The accident of the Challenger Space Shuttle evoked a wide range of deeply felt
public responses. The catastrophe was said by NASA officials to be caused by the failure
of an O-ring rubber seal in one of the two solid rocket boosters due to erosion by hot
gases. Later a steady flame developed which was directed towards the surface of the LH2
and LOX containing external tank (ET). The flame eventually breached the tank resulting
in an explosive bum of the hydrogen and oxygen [6]. The above explosion root theory
has been questioned recently. According to [27, 28], so-called "phantom" fires resulting
from undetected small cryogenic hydrogen leakages (< 1.4 kg/s) from the ET and their
early (i.e., at low altitude) ignition by the anomalous plume from the rocket booster have
caused the quick deterioration of the ET structural elements. The analysis indicated that
LH2 leaks occur in all missions meaning that the shuttle may explode again, if the leaking
LH2 were ignited again by unforeseeable ignition sources.
The spectacular accident in Hanau in 1991 demonstrated the consequences to be

expected from the catastrophic failure of a tank with pressurized hydrogen. The 100 m3
cylindrical hydrogen pressure vessel with 2.8 m diameter, 16 m total height, and 22 mm
wall thickness was exposed to a cyclic inner pressure between 1.5 (empty) and 4.5 (full)
MPa. The explosion was caused by a fatigue rupture near a longitudinal welding seam
[13]. Clarification of the causes resulted in a significant progress in terms of safety [103].
The evaluation of reports and statistics of accidents with hydrogen in industrial

facilities in the period 1965 - 1977 exhibit that [132] (see Tables 8-4, 8-5, and 8-6)
accidents were mainly caused by leakage or insufficient purging or venting,
• in 38 % of the accidents, the evolution of a flammable hydrogen-air mixture
remained undetected,
The average damage per accident was independent of the state of the hydrogen,
• most releases finally lead to ignition,
• in partially obstructed areas most cases result in an explosion caused by a fast
deflagration or even a detonation,
• spontaneous ignition is also possible by jets escaping from a burst disk or a safety
valve.
From a total of 409 accidents investigated, 78.5 % were related to gaseous hydrogen,
20.8 % to liquid hydrogen, and 0.7 % to hydrides.
Another study evaluating 96 mishaps in NASA hydrogen operations has come to the
following conclusions [49]:
• Valve malfunctions or leaks caused more than 20 % of the mishaps.
• Leaks were mainly due to personnel not following prescribed procedures.
• Ignition occurred in 60 % of the atmosphere releases and in all cases of release
into enclosed areas.
• Improper system purging was the cause in 25 % of the mishaps.
227
Table 8-4: Statistics of hydrogen accident reports between 1965 and 1977 on primary
causes of combustible mixtures, from [132]
Number of
Percentage
incidents
External leak 71 17.4
Internal leak 30 7.3
Inadequate purging 26 6.4
Inadequate ventilation 29 7.1
Vessel or pipe rupture 69 16.9
Venting incident 54 13.2
Exposure to fire 4 1.0
Collision or puncture 5 1.2
Excessive vacuum 3 0.7
Intentional operation 33 8.1
Electrolysis malfunction 18 4.4
Battery charging 5 1.2
Operator negligence 35 8.6
Other 7 1.7
£ 409(D
(1) The above table is as given in [132]. Although not explained, the total sum of
incidents in the table is 389, not 409 as indicated. Therefore the overall percentage
does not wind up to 100 %.
Table 8-5: Consequences of accidents involving hydrogen in industrial applications,

from [132]
Number of
Casualties Percentage
cases
Fire 117 16 28.6
Exlosion 211 90 51.6
Pressure rupture 25 1 6.1
Unignited release 39 5 9.5
Other 17 11 4.2
£ 409 123 100
228
Table 8-6: NASA statistics on accidents with release of hydrogen, from [85]
Number Percentage
Accidents with H2 release 80 83
Release to atmosphere 66« 69
Release to containment 20^ 21
Ignition of release to 41(2) 62
atmosphere
Ignition of release to
20 21
enclosure
(1) In 6 mishaps, hydrogen was released to both atmosphere and enclosure

(2) In 25 mishaps, hydrogen released to atmosphere did not ignite
• Some cases were due to air entrainment into LHi systems.
Hydrogen leaks typically originate from valves, flanges, diaphragms, gaskets, seals,
fittings.
Considering hydrogen accidents during transportation, in 71 % of the inadvertant

hydrogen release cases recorded, no ignition was observed, a trend that is also confirmed
by NASA experience [104].
A couple of hydrogen accidents were associated with the chlorine production by

chloralkali electrolysis, which are included in Table 8-4 under "Electrolysis malfunction".
Fire or explosion occurred when for some reason, for example a reversal of cell polarity,
the hydrogen entered the chlorine processing equipment downstream of the electrolysis
cells [104].
A recent inquiry of accidents involving hydrogen by Mitsubishi Heavy Industries

within the WE-NET project has revealed a number of accidents in the order of a few
cases per year. Most accidents were caused by electric sparks. The maximum amount of
hydrogen released was 4000 Nm3 (Hanau accident), and 1300 Nm3 referring to accidents
in Japan [29].
Although hydrogen has a reputation as a particularly dangerous fuel, the overall

statement was made that despite the fact that some metals can become brittle to hydrogen
or hydrogen compounds, the technology of handling hydrogen as a high-pressure gas or
as a liquid is successful and safe [43]. Explanations of this good record are possibly the
facts of conservative assumptions for the design stress and of a scheduled retesting of the
vessels. Further measures for accident prevention are the appropriate choice of materials,
coatings, adsorption layers, the deployment of stainless steel liners in composite vessels
and piping, and proper welding procedures. Although the excellent safety record should
not be extrapolated to a large-scale use in a hydrogen economy, since experience so far is
229
based mainly on the existence of well-trained personnel and extreme concern about system
reliability. If handled properly, hydrogen will be a safe alternative to fossil fuels. The
hazards of hydrogen are generally recognized to be different than those of petroleum fuels,
but not necessarily greater [69].
8.7. SAFETY AND RISK ASSESSMENT FOR HYDROGEN CONTAINING SYSTEMS
8.7.1. General Methodology
Technical systems which contain material representing a hazard potential must be

treated under controled, safe conditions. A respective design of the facility and its
operational modes is checked by different kinds of safety considerations. In addition to
the consequences of normal operation, the system behavior under accident conditions has
to be considered in terms of sufficient measures for controling accidents. Assumptions are
being made based upon experience which has been achieved in the course of the technical
development and which has resulted in a system of laws, regulations, rules, guidelines,
and recommendations, e.g., for the design of components, for quality assurance, for fire
protection and also for accident prevention measures.
Not only is the selection of accident scenarios made on the basis of experiences, but
also on the basis of intuition. It comprises both the reliability of the technical systems and
the consequences of uncontroled accidents. Problems arise if conceivable accident scenarios
start to leave the realm of direct experience and if the possible damage is getting high.
It was typically in the nuclear technology that systematic prognosis procedures for
the event and the consequences of uncontroled accidents were applied. They have been
significantly improved over the last 25 years and are now an essential part of the valuation
process of nuclear power plants in the form of probabilistic safety and risk assessments
(PSA and PRA).
The methodology of PSA is in principle a combination of event tree and fault tree
analyses. The target of the analysis decisively determines the selection and the definition
of modes in which the technical system is at the end of an event sequence. These modes
comprise both the safe ones resulting from successful safety measures and also those
unwanted ones characterized by a certain release of hazardous materials. A PSA is extended
to a PRA, if the consequences of the different system modes are determined and linked
with the event frequencies of those modes.
The single steps of a PSA are:

(i) the identification of hazardous potentials by searching for sensitive areas and
processes which contain toxic or flammable materials;
(ii) the identification of accident-initiating events, e.g., the failure of components.
Starting from such an event, the (non-) availability of safety installations and
measures are questioned in an event tree resulting in a ramification, where every
branch represents an event sequence. Sequences with similar consequences are
summarized in categories.
230
For the initiating event, frequencies per operation year are determined, as a rule, by
evaluation of statistics, or, if no experience is available, by justified assessments. The
probability of non-availablities of safety installations or measures is determined by fault
trees or immediately by values derived from operational experiences. The total frequencies
of the event sequences are then calculated by multiplication of the frequency of the initiating
event with those of the non-availabilities.
8.7.2. Examples
The spectrum of hydrogen quantities that might be released into the atmosphere
varies from small amounts in a vehicle tank to the large capacity of a stationary storage
tank. Table 8-7 lists typical LHj quantities together with the corresponding volumes of a
(stoichiometric) mixture gas cloud and the (theoretical) TNT equivalent.
8.7.2.1. Liquid Hydrogen River Transportation
Ship transportation is considered an economic method to carry large amounts of

hazardous cargo. In case of an accident, however, it turns into a major safety problem
with possibly disastrous consequences for the environment. The coarse procedure for a
risk assessment focusing on the inland navigation of a large-scale LH2 cargo on a 15,000
m3 capacity barge carrier has been demonstrated in [17, 72].
Table 8-7: Typical liquid hydrogen quantities and hazardous potential, from [120]
Liquid hydrogen Hydrogen-air

TNT equivalent
contents mixture
[tTNT]
[m3] [m3]
Passenger car tank 0.15 433 0.28
Bus tank 0.6 1,730 1.13
Tank truck load 40 115,400 75.2
Railway transport
100 288,300 188
tank
Aircraft tank
240 692,000 451
(Airbus)
Ship barge 3800 10,950,000 7,140
Ship barge
33,000 95,100,000 62,000
(WE-NET)
Stationary storage
130,000 375,000,000 244,400
tank
231
The initiating event of an inadvertent LH2 release from a ship could be of internal
or external nature. Internal initiating events are caused by technical failure (tank rupture)
or by human error. External events comprise maritime events, i.e., a collision with any
obstruction (ground, other ship), and general hazards such as natural hazards or sabotage.
A collision may result in a sinking of the barge carrier by which the barges could be set
free and float on the water under no control possibly risking a subsequent damage.
For the period 1983 -1989, the number of ship movements on the Elbe river amounted
to about 65,000 per year with an approximate 30 % fraction (w 19,500) assumed to be large
enough to have the potential of significantly damaging the carrier ship. The total maritime
accident probability is then approximated by summing up the single probabilities for
different river sections characterized by geometry, traffic volume, and the nearby population
density. Multiplication with the number of trips of the barge carrier per year finally yields
the frequency of an initiating event. It was found to be largest for grounding/stranding,
followed by an encountering collision, a crossing collision, and the extremely unlikely
event of technical vessel failure.
Conceivable accident scenarios identified result in consequences which can be grouped

in different release categories:
- continuous release of gaseous hydrogen through safety valve, e.g., due to loss of
insulation or failure of safety valve in open position
- quasi-continuous or instantaneous release of a small fraction of the tank contents,
e.g., due to pipe rupture
- continuous release of a large amount of the tank contents, e.g., due to a leak in
the vessel
- instantaneous release of the complete tank contents, due to catastrophic vessel
failure
The spill of a small amount of LHi is more likely to occur than the release of the
complete vessel contents (leak-before-failure construction rule). The conceivable ignition
of the flammable gas cloud reveals the actual hazardous potential for the environment. A
delayed ignition represents the worst case with a certain chance for deflagration or even
detonation, connected with the formation of a blast wave and heat radiation. Corresponding
relations connect received radiation dose or pressure impact with the probability for a
fatality resulting in certain individual risk levels.
Qualitative conclusions can be drawn:

• The highest societal risk is found on-shore for the population of Hamburg city, the
highest individual risk is found on water for the ship crew
• Collision with an encountering ship causes the highest total risk.
• Collision with a crossing ship causes the highest risk for the on-shore population
The quantitative evaluation given in [72] is deemed preliminary, because many

assumptions were based upon either conservative estimates where no better data could
be acquired or too simple modeling methods which do not realistically account for the
specific behavior of gaseous or liquid hydrogen. Nevertheless, the study has given a
feeling for the specific aspect of a liquid hydrogen river transportation and demonstrated a
viable procedure for a probabilistic risk assessment.
232
8.7.2.2. PHOEBUS
The prognostic procedures for a risk assessment as have been used at the Research
Center Julich for safety analyses for nuclear power plants, have been applied to the
Photovoltaic Hydrogen Fuel Cell Demonstration Facility, PHOEBUS [18].
8.7.2.2.1. Safety Concept
The safety concept of PHOEBUS is composed of several active and passive measures
which are basically directed against hydrogen release inside the building (Fig. 8-16):
• hydrogen gas warning system
with a total of 8 detectors immediately above the sensitive components and, in
addition, at the ceiling of the building. Alarm is given, if the H2 concentration
reaches a level of 0.4 vol%. At an 0.8 vol% level, the gas lines between building
and outdoors storage are automatically closed off as well as gas and energy
provision of the respective component. Also air ventilators in the ceiling are
started. At the same time, alarm is given to the security service of the Research
Center Julich.
Roof
See detailed
scheme
Demonstration Test Facili(>

Photovoltaic-Hvdrogen-Fuel-Celi
Detail: Connection between

H2 - storage and fuel cell
1: position of leak assumed 4: automatic shutoff

2: H2 control 5: automatic venting
3: flow control 6: emergency manual shutoff
Fig. 8-16: Schematic of the safety concept in the PHOEBUS facility, from [17]
233
• gas throughput and pressure supervision system
to disconnect gas lines leading into the building, if limits are exceeded.
• stainless steel piping
with a maximum pressure of 0.7 MPa indoors. High pressure pipes up to 15 MPa
(H2) and 7 MPa (02), respectively, are operated only outdoors.
8.7.2.2.2. Conceivable Accident Scenarios
The final state of an accident sequence with a hydrogen release comprises the
formation of a flammable gas mixture either inside H2 carrying components or inside
the building or outside in the open atmosphere.
• Flammable H2-O2 mixtures in components
are conceivable in the electrolyzer and the fuel cell. Component-internal leakages,
e.g., after a diaphragm fracture, however, do lead to relatively small amounts
of gas mixtures. Only the failure of the protective measures, i.e., an erroneous
continuation of operation, allows a larger amount of the mixture to be routed into
the oxygen storage vessel where, however, there is no ignition source.
• Flammable H2-air mixtures inside the building
is possible after leakage of H2 carrying components. Three categories of leaks
have been denned:
i. small leaks with a leakage rate less than 10 Nm3/h, which is the lower
throughput supervision limit,
ii. medium leaks with leakage rates up to 120 Nm3/h, which is the maximum
escape rate from the high-pressure storage with still working high-pressure
reduction system, and
iii. large leaks with a maximum leakage rate of 600 Nm3/h, if also the pressure
reduction system fails
In the case of small leaks, the natural air exchange in the hall is sufficient to
prevent flammable mixtures. For medium leaks, flammable mixtures on a large
scale can be precluded, however, not on a smaller scale, e.g., in the vicinity of
the leak. Large leaks which are not actively ruled can lead to large flammable
gas mixtures.
• Flammable H2-air mixtures in the open atmosphere
is given upon the catastrophic failure of the outdoor high-pressure storage vessel.
The sudden release of 3000 Nm3 of hydrogen represents in principle the greatest
hazard potential, whereas leaking hydrogen does not cause a serious problem due
to its strong buoyancy.
8.7.2.2.3. Frequencies of Initiating Events
Because of the prototype character of the PHOEBUS facility, hardly any direct
operation experience could be utilized. Therefore generic data, also assessments, have
been used implying a relatively large uncertainty range.
Operational experience has been taken to derive reliability parameters for the various
components and safety systems hi order to determine the frequencies for possible accident
234
Medium H2 Throughput Automatic Manual Automatic Natural Frequency Flammable
Leak Control Control Shutoff Shutoff Venting Venting per year Mixture
1 -103
1
6,2- 10 7
———————1
1,1 • 107
1 (< 1 %)
5-10- 7
- - ——————
1
< 10'9
(2,2 -3,4%)
y 5S
1,5- 10-8
1f - •" (2,2 - 3,4 %)
|no
Fig. 8-17: Event tree for medium-sized H2 leak in the PHOEBUS plant, from [17]
Table 8-8: Classification into groups of hydrogen release events in the open atmos-
phere, from [17]
Reacting
Frequency
Event group Ha mass
[10-5/yr]
[kg]
Rupture of one vessel
1 0.2 2.3
energy conversion 1 %
Rupture of one vessel
2 1.5 - 4.0 1.8
Vessel rupture with subsequent emptying of
3 a battery of nine vessels 15-35 0.2
Vessel rupture with subsequent emptying of
4 both batteries of 18 vessels 65 0.001
sequences. As an example, the event tree for a medium-sized hydrogen leak inside the
experimental hall is presented in Fig. 8-17 showing that for two of the final states reached,
locally flammable mixtures cannot be excluded. The respective frequencies, however, are
extremely small.
For accidents outside the hall, no active measures are planned. Ramification of the
event trees is developed by subsequent failures or conditions for the extent and the effect
of ignited gas clouds. The analysis has shown that hydrogen release from the storage
vessels with subsequent explosion dominates the still very small environmental risk. All
235
sequences have been grouped into four categories with a similar explosive potential, as
listed in Table
listftH Tahlf 8-8.
R-8
The individual risk expressing the probability for a fatality as a consequence of an

event was determined to be 10~5/yr restricted to the immediate neighborhood of the building.
The unusually large volume of the building is of advantage in case of small leak accidents.
The H2 storage in 18 interconnected bottles increases the leak frequency, however, reduces
the consequences [18].
Peculiarities of the PHOEBUS plant such as type, size, venting concept, storage
concept do not allow the results of this study to be transferable to future photovoltaic
systems [18].
8.7.2.3. Liquefaction Plant
Some results from a safety analysis for the French liquefaction plant near Lille with
an LH2 production rate of 10 t/d have been reported showing that in a design basis accident
represented by an LH2 pipe rupture, the maximum escape rate is 1.9 kg/s. Assuming a
wind velocity of 2 m/s, a hydrogen-air mixture cloud will evolve with a predicted maximum
extension of approx. 200 m within the flammability limits. A safety distance of 238 m
around the LH2 facilities for the plant was considered sufficient to protect offsite residential
areas [5].
8.8. ENVIRONMENTAL ISSUES
In principle, a significant reduction in the energy related impact on the environment

is being achieved, if non-fossil hydrogen be substituting conventional fuels. There are no
emissions of carbon oxides, hydrocarbons, dust or ash, heavy metals, lead compounds, or
SO2- Some environmental risks are left concerning, e.g., gas emission during combustion
or the establishment of the infrastructure.
8.8.1. Water Vapor
Water vapor is the most climate-relevant atmospheric gas contributing a portion

of 65 % to the overall greenhouse effect. The (anthropogenic) effect of the tremendous
amounts of water vapor expected to be released into the atmosphere within a large-scale
hydrogen technology is disputed among climate researchers. Critical voices estimate the
liberation of latent heat upon condensation to contribute much stronger to global warming
than carbon dioxide, so that any further water vapor sources should be avoided [63].
Others expect water vapor emissions in a hydrogen energy economy (assumption is

that half of the present energy supply is covered by hydrogen) to be reducible down to the
respective level presently given by the fossil and nuclear economy which is a 0.005 % share
of the total atmospheric water cycle. In order to reach the above reduction factor, hydrogen
losses need to be decreased from currently estimated 10 % over the whole chain to 2 - 3
%. Today's world energy economy emits around 20*1012 kg of water per year [134].
236
The primary product of hydrogen combustion is water which is ejected as water
vapor and may form condensation trails (= contrails) which are ice clouds. Comparing
the greenhouse effects of kerosene and hydrogen at varying cruising levels for mid-latitude
atmospheric conditions, it can be stated that for mid-latitude atmospheric conditions below
an altitude of 10 km, i.e., in the troposphere, the contribution of water vapor is negligible,
the greenhouse effect is almost zero. Above 10 km, the emitted water vapor behaves like
a greenhouse gas exceeding the effect of kerosene at altitudes > 12 km (cruising altitudes
of today's aircraft, however, usually are not higher than 12 km [92]. On the other hand,
water has a residence time in the order of one year for typical cruising levels which is short
compared with CO2 with 100 years residence time in the atmosphere independent of height.
Water vapor is a concern for the lower stratosphere. Its contents is 1.3*1012 kg
of H2O with an average lifetime of one year. The contribution by today's high-altitude
air transportation which is 20 % of the total aviation, is 0.04* 1012 kg of H2O per year.
A complete conversion to H2-fueled aircraft would result in an emission of as much as
500*1012 kg of H2O per year. Therefore it is mandatory to lower the cruising height below
the tropopause [134].
The formation of contrails is dependent on ambient temperature and humidity. It

prevails in the warm sectors of low-pressure areas. An LH2 aircraft needs to fly at a 1 km
lower cruising level to account for the same amount of evolved contrails as a kerosene-
driven aircraft. The formation of contrails is avoided at lower cruising altitudes, however,
at the expense of higher fuel consumption. The greenhouse effect of contrails is still unclear
[92]. NASA scientists have found indications for a warming of the air below contrails, in
particular in areas with high aircraft traffic, e.g., New York [51].
Within the CEC funded project MOZAIC ("Measurements of Ozone by Airbus In-
Service Aircraft"), continuous measurements of water vapor (and ozone) in the upper
troposphere are being made during commercial airbus flights. At 40° - 60° N latitude over
the North Atlantic, contents of water vapor were found to range between 0.07 and 0.12
g/kg depending on the seasonal large-scale circulation and jet stream pattern [107].
8.8.2. Nitrogen Oxides
The only pollutants of concern in combustion processes of hydrogen-air mixtures

are nitrogen oxides (NOX). Emissions of NOX which are generated as a function of
temperature, pressure, and time, however, are small compared with conventional fuels and
even negligible at temperatures < 200 °C. The generation and emission of NOX peaks at
excess air ratios of about 1, but is steeply reduced by applying a lean H2, low-temperature
burning process (see Fig. 8-18) at high combustion rates and using a homogeneous fuel-air
mixture to suppress local hot spots. In experimental tests with hydrogen combustors in
aircraft engines starting in 1991 within the Euro-Quebec project, a reduction of the NOX
level by 70 - 80 % compared with kerosene combustors was achieved (see Fig. 8-19) [78,
92]. Although connected with a power output loss of up to 30 %, the specific emissions
per power unit are still smaller. In car engines, the NOX reduction achieved in practice
is 30 to 60 % [124].
237
Pre-mixcd
15 -
10 -
GO
0.01
1.0 1-5 2.0 2.5 3.0 3.5 4.0

Excess Air Ratio ( A)
Fig. 8-18: NOX emissions in H2 combustion processes in a rotary engine, from [78]
50
Cruise conditions,
35000 ft, M = 0.8
§ 40
U-
cn
-o
^ 20
10
0
0 10 20 30 40 50
Engine Pressure Ratio
Fig. 8-19: NOX emissions for different combustion technologies, from [92]
Experimental data with hydrogen powered ICE cars have shown that if optimized to
low NOX levels by using ultra-lean mixtures, the former US standard of 0.6 g/km could
be met without catalyst [69].
238
8.8.3. Hydrogen
The present concentration of hydrogen in the atmosphere is 510 ppb which translates
into a total mass of approx. 200 million tons, and the annual increase rate is 3 ppb. The
average lifetime is two years before chemically bound or lost into space. The H2 emissions
from the industrial use are with < 1 million t/yr small compared with other natural and
anthropogenic sources. In case of the hydrogen economy based on a 80 % CO2 reduction
scenario, H2 emissions are predicted to be 1.1 - 1,4 million t/yr [108].
Hydrogen, and to a much higher extent, methane take part in a whole variety of
chemical reactions in the atmosphere responsible for ozone depletion [108].
The ecological consequences of an accidental (liquid) hydrogen release are harmless

due to its untoxic character compared with an uncontroled spill of fossil fuels [108].
8.8.4. Others
Emissions of other pollutants, hydrocarbons or CO, from hydrogen combustion in

ICE become significant, if the engine burns excess lubricating oil. These emissions can be
minimized by regular inspection and maintenance [69],
8.8.5. Modeling
An analytical NOX chemical kinetic model has been developed by the DLR to
investigate the influence of various parameters on the formation of pollutants. It has been
coupled with an eddy breakup model for the combustion process [133].
The NASA Goddard Space Flight Center has developed a "Residual Circulation
Model" which uses emission and representative flight profile data to estimate the amount
of stratospheric ozone depletion. Applied to the respective design data for the hypersonic
X-30 engine based on a total of 150 flights within five years, the maximum local ozone
depletion was predicted to be 0.006 % [127].
A respective study for the German SANGER spacecraft design estimates that in the
lower and medium stratosphere accumulation of water vapor and trace gases would occur
due to low vertical air exchange enhancing the greenhouse effect. It is only the planned low
number of flights (24/yr) that makes this disturbance of the stratosphere negligible [108].
[1] AEA TECHNOLOGY, CFX 4.1: User Guide, AEA Technology (1995).
[2] ANDREWS, G.E., BRADLEY, D., LWAKABAMBA, S.B., Turbulence and Tur-
bulent Flame Propagation - A Critical Appraisal, Combustion and Flame 24 (1975)
285-304.
[3] ANDRONOPOULOS, S., et al., Modelling the Effects of Obstacles on the Disper-
sion of Denser-Than-Air Gases, J. Hazardous Materials 37 (1994) 327-352.
[4] ANON, Anzen Kogaku Kyokai (Japan Society for Safety Engineering), March
(1977) (in Japanese).
239
[5] ANON, Fliissiger Wasserstoff, Raketentreibstoff und Rohstoff fur ultrareines Pro-
zessgas, Newspaper article in Neue Zuercher Zeitung, Zurich, Switzerland, March
16, 1994.
[6] ANON, Report of the Presidential Commission on the Space Shuttle Challenger
Accident (1986), World Wide Web, http://www.ksc.nasa.gov/shuttle/missions/51-
1/docs/rogers-commission/table-of-contents.htrnl, NASA (1997).
[7] ARTHUR D. LITTLE INC., Final Report on an Investigation of Hazards Associated
with the Storage and Handling of Liquid Hydrogen, Report C-61092, Cambridge,
USA (1960).
[8] ARVIDSON, J.M., HORD, 1, MANN, D.B., Dispersion of Hydrogen or Methane
Fuels Released into an Automobile Interior, Adv. Cryo. Eng. 21 (1976) 387-398.
[9] BACKHAUS, H., Uber die Bedeutung von Ausbreitungsrechnungen in Sicherheits-
analysen fur die Planung von Fliissiggaslagern, Schwere Gase und Sicherheitsana-
lyse, (4th Symp., Bonn, 1991), HARTWIG, S., (Ed.) Schwere Gase und Sicher-
heitsanalyse - TV, Battelle Eigenverlag, Frankfurt (1992) 252-262.
[10] BAUMGARTNER, K., EWALD, R., Sicherheitstechnik bei der Handhabung von
Wasserstoff, Chem. Ing. Tech. 59 (1987) 296-302.
[11] BEAUVAIS, R., MAYINGER, F., STRUBE, G., Wasserstoffverbrennung im Ver-
lauf auslegungsuberschreitender Storfalle in Kernkraftwerken, VGB Kraftwerks-
technik 71 (1991) 755-758.
[12] BEAUVAIS, R., MAYINGER, F., STRUBE, G., Turbulent Flame Acceleration-
Mechanisms and Significance for Safety Considerations, Int. J. Hydrogen Energy
19 (1994) 701-708.
[13] BEHREND, E., et al., Folgenschwere Explosion - Bericht uber Ursachen und
Hergang des Berstens eines Wasserstofftanks in Hanau - Teil 1/2, TU 34 (1993)
176-179 / 225-229.
[14] BEHRENS, U., et al., Detonationsfahigkeit von Wasserstoff-Wasserdampf-Luft-
Gemischen unter unfalltypischen Bedingungen, Final Report BF-R-66.880-1, Bat-
telle Institute, Frankfurt (1988).
[15] BERMAN, M., A Critical Review of Recent Large-Scale Experiments on Hydrogen-
Air Detonations, Nucl. Sci. Tech. 93 (1986) 321-347.
[16] BLUMENTHAL, R., FffiWEGER, K., ADOMEIT, G., Self Ignition of Hydrogen
/ Air Mixtures, (llth World Hydrogen Energy Conf., Stuttgart, FRG, 1996),
[17] BONGARTZ, R., et al., Selected Aspects of a Safety Analysis in a Hydrogen
Energy Economy, (10th World Hydrogen Energy Conf., Cocoa Beach, USA, 1994),
BLOCK, D.L., VEZIROGLU, T.N., Hydrogen Energy Progress X, International
[18] BONGARTZ, R., JAHN, W., MARX, J., Probabilistische Zuverlassigkeits-
und Risikountersuchung der Photovoltaik-Wasserstoff-Brennstoffzellen-Demonstra-
tionsanlage PHOEBUS Jiilich, Report Jiil-3176, Research Center Jiilich (1996).
[19] BREITUNG, W., REDLINGER, R., Containment Pressure Loads from Hydrogen
Combustion in Unmitigated Severe Accidents, Nucl. Tech. Ill (1995) 395-419.
[20] BREITUNG, W., REDLINGER, R., A Model for Structural Response to Hydrogen
Combustion Loads in Severe Accidents, Nucl. Tech. Ill (1995) 420-425.
240
[21] BREITUNG, W., KOTCHOURKO, A., Numerische Simulation von turbulenten
Wasserstoff-Verbrennungen bei schweren Kemreaktorunfallen, FZK-Nachrichten 28
(1996) 175-191.
[22] BREWER, G.D., et al., Assessment of Crash Fire Hazard of LH2-Fueled Aircraft,
Report CR-165525, NASA, Cleveland, USA (1981).
[23] BRISCOE, E, SHAW, P., Spread and Evaporation of Liquid, Progr. Energy Com.
Sci. 6 (1980) 127-140.
[24] BURGESS, D., BIORDI, J., MURPHY, J.N., Hazards of Spillage of LNG into
Water, PMSRC Report 4177, US Department of the Interior, Bureau of Mines
(1972).
[25] CHANG, H.R., REID, R.C., Spreading-Boiling Model for Instantaneous Spills of
Liquefied Petroleum Gas (LPG) on Water, J. Hazardous Materials 7 (1982) 19-35.
[26] CHIRTVELLA, I.E., WTTCOFSKI, R.D., Experimental Results from Fast 1500-
Gallon LH2 Spills, Am. Inst. Chem. Eng. Symp. Ser. 82 (1986) No. 251.
[27] CHIRIVELLA, I.E., Analysis of the "Phantom" Fires on the Space Shuttle External
Tank Base, 1997 JANNAF 34th Combustion Systems Hazards Subcommittee/ and
Airbreathing Propulsion Subcommittee Joint Meetings, Palm Beach, USA (1997).
[28] CHIRIVELLA, I.E., The Nature of the Space Shuttle "Phantom" Fires: LH2 Leaks,
1997 JANNAF 34th Combustion Systems Hazards Subcommittee/ and Airbreathing
Propulsion Subcommittee Joint Meetings, Palm Beach, USA (1997).
[29] CHTTOSE, K., Activities on Mffl WE-NET Project (1993 - 1995), Presented at the
Research Center Julich, October 13, 1995.
[30] CHITOSE, K., OGAWA, Y., MORJJ, T., Analysis of a Large Scale Liquid Hydrogen
Spill Experiment Using the Multi-Phase Hydrodynamics Analysis Code (Champa-
gne), (llth World Hydrogen Energy Conf., Stuttgart, FRG, 1996), VEZIROGLU,
Energy (1996) 2203-2211.
[31] CROCKER, W.P., NAPIER, D.H., Assessment of Mathematical Models for Fire
and Explosion Hazards of Liquefied Petroleum Gases, J. Hazardous Materials 20
(1988) 109-135.
[32] DEIMEL, P., HSCHER, H., HOFFMANN, M., Materials Investigations for Hy-
drogen Technology, (llth World Hydrogen Energy Conf., Stuttgart, FRG, 1996),
[33] DICHRISTINA, M., What Really Downed the Hindenburg, Popular Science (1997)
70-76.
[34] DIENHART, B., Ausbreitung und Verdampfung von flussigem Wasserstoff auf
Wasser und festem Untergrund, Ph.D. Thesis University of Bochum, Report Jiil-
3155, Research Center Julich (1995).
[35] DffiPOLD, W., PFORTNER, H., HOMMEL, H., Sicherheitsabstande bei der La-
gerung und Handhabung explosionsfahiger Stoffe, Explosivstoffe 18 (1970) 25-39.
[36] DOHRN, K.-J., et al., Transport und Umschlag von tiefgekuhlt verflussigtem Erd-
gas (LNG) und unter Druck sowie tiefgekuhlt verfliissigten Kohlenwasserstoffgasen
(LPG) mit Seeschiffen auf deutschen SeeschiffahrtsstraBen - Risikoabschatzung,
Arbeitsgruppe des Beirates fur die Beforderung gefahrlicher Giiter beim Bundes-
verkehrsministerium, Bonn (1984).
241
[37] DORNIER SYSTEM GMBH, Sicherheitsaspekte einer kQnftigen europaischen
Wasserstoff-Technologie, Final Report EUR 6854 DE, Commission of the Euro-
pean Communities, Luxemburg (1980).
[38] DRAKE, E.M., JEJE, A.A., REID, R.C., Transient Boiling of Liquefied Cryogens
on a Water Surface -1. Nitrogen, Methane, Ethane, Int J. Heat and Mass Transfer
18 (1975) 1361-1375.
[39] DUTTON, R., Materials Degradation Problems in Hydrogen Energy Systems, Int.
J. Hydrogen Energy 9 (1984) 147-155.
[40] EDESKUTY, F.J., Safety of Liquid Hydrogen in Air Transportation, Hydrogen in
Air Transportation, (Int. Symp., Stuttgart, 1979), DFVLR (1979) Paper 18.
[41] EDESKUTY, F.J., Critical Review and Assessment of Problems in Hydrogen
Energy Delivery Systems, Report LA-7405-PR, Los Alamos Scientific Laboratory
(1980).
[42] EDESKUTY, F.J., et al., Hydrogen Safety and Environmental Control Assessment,
Report LA-8225-PR, Los Alamos Scientific Laboratory (1980).
[43] EDESKUTY, FJ., STEWART, W.F., Safety in the Handling of Cryogenic Fluids,
[44] EICHERT, H., Zur Dynamik des Verbrennungsablaufs von Wasserstoff-Luft- und
Wasserstoff-Methan-Luft-Gemischen, Research Report DLR-FB 89-39, Deutsche
Forschungsanstalt fur Luft- und Raumfahrt, Stuttgart, FRG (1989).
[45] EICHERT, H., et al., Gefahrdungspotential bei einem verstarkten Wasserstoffein-
satz, Deutsche Forschungsanstalt fur Luft- und Raumfahrt (DLR), Stuttgart, FRG
(1992).
[46] ENERGIE SPEKTRUM, Hammenforscher, TU Munchen kummert sich um sichere
H2-Verbrennung, Energie Spektrum, June 1992, 34-35.
[47] FAY, J.A., A Preliminary Analysis of the Effect of Spill Size on the Level of Hazard
from LNG Spills on Land and Water, Report DOE/EV-0002, US Department of
Energy (1978).
[48] FISCHER, M., KRATZEL, T., PANTOW, E., Turbulent Combustion and Detona-
tion Processes in Hydrogen-Air Mixtures, (llth World Hydrogen Energy Conf.,
[49] FRAZffiR, W.R., Hydrogen Safety, Cold Neutron Sources, (Int. Workshop, Los
Alamos, 1990), Report LA - -121, Los Alamos National Laboratories, Los Alamos
(1990).
[50] FRYER, L.S., KAISER, G.D., DENZ - A Computer Program for the Calculation
of the Dispersion of Dense Toxic or Explosive Gases in the Atmosphere, Report
SRD R152, UKAEA, Culcheth (1979).
[51] GEORGESCU, V., Kondensstreifen erwarmen Luft, Newspaper article in Frank-
furter Rundschau, Frankfurt, April 3, 1997.
[52] GIESBRECHT, H., Evaluation of Vapour Cloud Explosions by Damage Analysis,
J. Hazardous Materials 17 (1988) 247-257.
[53] GRATZ, R., Sicherheitsabstande als Schadensvorsorge bei der Lagerung brennbarer
Gase, TO 37 (1996) 13-22.
[54] GUGAN, K., Unconfined Vapour Cloud Explosions, The Institution of Chemical
Engineers, Rugby, UK (1978).
242
[55] GUIRAO, C.M., KNYSTAUTAS, R., LEE, J.H., A Summary of Hydrogen-Air
Detonation Experiments, Report NUREG/CR-4961 and SAND87-7128, Sandia
National Laboratory, Albuquerque (1989).
[56] HANNA, S.R., DRTVAS, PJ., CHANG, J.C., Guidelines for Use of Source and
Atmospheric Dispersion Models for Accidental Releases, AIChE Document No.
0-8169-0702-1 G40 (1996).
[57] HAVENS, J.A., SPICER, T., LNG Vapor Dispersion Prediction with the DEGADIS
Dense Gas Dispersion Model, Topical Report (April 1988 - July 1990), University
of Arkansas, Fayetteville, USA (1990).
[58] HORD, J., How Safe is Hydrogen?, Hydrogen for Energy Distribution, (Symp.,
Chicago, 1978) 613-643.
[59] HORIGUCHI, S., et ah, Fire Extinguishment for Liquefied Hydrogen, J. Nat. Chem.
Lab. Industry 81 (1986) 421-425.
[60] IAEA, Hydrogen in Water-Cooled Nuclear Power Reactors, International Atomic
Energy Agency and Commission of the European Communities, Vienna (1990).
[61] ISTING, C, Wasserstoff - Ein Energietrager der Zukunft?, gwf-gas/erdgas 117
(1976) 511-516.
[62] KESTEN, M., Physikalischer Wasserstoffangriff, in: UEBING, D., SCHLEGEL,
D. (Eds.), EinfluBgrSSen der Zeitsicherheit bei technischen Anlagen, F. Vieweg &
Sohn, Verlag TUV Rheinland (1985).
[63] KISCHKA, S., Temperaturschwankungen nur schwer zu beurteilen, Newspaper
article in Handelsblatt, Dusseldorf, January 26, 1994.
[64] KNEEBONE, A., PREW, L.R., Shipboard Jettison Tests of LNG onto the Sea, (4th
Int. Conf. on LNG, Alger, 1974), Session V, Paper 5 (1974).
[65] KOOPMAN, R.P., et al., Analysis of Burro Series 40 m3 LNG Spill Experiments,
J. Hazardous Materials 6 (1982) 42-83.
[66] KRATZEL, T., PANTOW, E., EICHERT, H., Modeling of Hydrogen Combustion:
Turbulent Flame Acceleration and Detonation, (10th World Hydrogen Energy Conf.,
Cocoa Beach, USA, 1994), BLOCK, D.L., VEZTROGLU, T.N., Hydrogen Energy
Progress X, International Association for Hydrogen Energy (1994) 1771-1780.
[67] KUBMAUL, K., DEIMEL, P., Materialverhalten in H2-Hochdrucksystemen, VDI
Berichte No. 1201, VDI-Verlag, Dusseldorf (1995) 87-101.
[68] LIND, C.D., What Causes Unconfined Vapor Cloud Explosions, Loss Prevention
9 (1975) 101-105.
Design 17 (1996) 562-589.
[70] LIPPERT, J.R., Vulnerability of Advanced Aircraft Fuel to Ballistic and Simulated
Lightning Threats, Int. J. Hydrogen Energy 1 (1976) 321-330.
[71] LOUTHAN, M.R., MORGAN, M.J., Some Technology Gaps in the Detection and
Prediction of Hydrogen-Induced Degradation of Metals and Alloys, J. Nondestruc-
tive Evaluation 15 (1996) 113-120.
[72] LOVAS, G.G., Investigation of Risk from Ship Transport of Liquid Hydrogen on
a Waterway, Diploma Thesis, Research Center Julich (1990).
[73] MANAS, J.L., BLEVEs - Their Nature and Prevention, Fire International 8 (1988),
No. 87, 27-39.
[74] MARSHALL, J.G., The Size of Flammable Clouds Arising from Continuous
Releases into the Atmosphere, InsL Chem. Eng. Symp. Ser. 49 (1977) 99-109.
243
[75] MCFARLANE, K., et al., Development and Validation of Atmospheric Dispersion
Models for Ideal Gases and Hydrogen-Fluoride, Part I: Technical Reference Manual,
Shell Research Ltd., Thornton Research Center (1990).
[76] MCQUAID, J., Observations on the Current Status of Field Experimentation on
Heavy Gas Dispersion, Atmospheric Dispersion of Heavy Gases, (Proc. IUTAM
Symp., Delft, 1983) (1983) 241-266.
[77] MOEN, I.O., Transition to Detonation in Fuel-Air Explosive Clouds, J. Hazardous
Materials 33 (1993) 159-192.
[78] MORISHTTA, T., The Development of the Hydrogen Vehicles, Hydrogen and Clean
[79] MOSER, V., Simulation der Explosion magerer Wasserstoff-Luft-Gemische in
groB- skaligen Geometrien, Ph.D. Thesis RWTH Aachen, Aachener Beitrage zum
Kraftfahr- und Maschinenwesen, Vol. 11 (1997).
[80] MtJHLECK, P., STURSBERG, K., Fliegen bei Mach 7: Der Hyperschallantrieb
auf dem Prufstand, DLR-Nachrichten (1993) No. 73 41-47.
[81] NASA, Small Scale Slush Hydrogen Test Facility, World Wide Web,
http://lerc.nasa.gov/WWW/sIush, NASA (1997).
[82] NATIONAL HYDROGEN ASSOCIATION, Handling Hydrogen Safely, World
Wide Web, http://www.ttcorp.com/nha/hhs_2.htm, National Hydrogen Association
(1997).
NET), 1995 Annual Summary Report on Results NEDO-WE-NET-95, NEDO,
Tokyo (1996).
[84] OOMS, G., MAHIEU, A.P., ZELIS, F., The Plume Path of Vent Gases Heavier
than Air, Loss Prevention and Safety Promotion in the Process Industries, (Proc.
1st Int. Symp.), Elsevier Publ. Comp. (1974)211-219.
[85] ORDIN, P.M., Review of Hydrogen Accidents and Incidents in NASA Operations,
NASA Technical Memorandum NASA TM X-71565, Lewis Research Center,
Cleveland (1974).
[86] PACZKO, G., KLEIN, R., Reduzierte Reaktionsmechanismen fur H2-Luft-Dampf-
Detonationen, Final Report KFK-Project No. 325/03179880/0105, RWTH Aachen
(1992).
[87] PANTTZ, H.J., MATZERATH, C., PASLER-SAUER, J., UFOMOD Atmospheric
Dispersion and Deposition, KfK Report 4332, Research Center Karlsruhe (1989).
York (1992).
[89] PFORTNER, H., Zundverhalten von Erdgas/Luft-Gemischen in freien Wolken, gwf-
gas/erdgas 120 (1979) 19-24.
[90] PFORTNER, H., SCHNEIDER, H., Experimental Investigation into the Deflagra-
tion of Flat, Premixed Hydrocarbon/Air Gas Clouds, Progress in Astronautics and
Aeronautics 114 (1988) 488-498.
[91] PFORTNER, H., Ausbreitungsfunktionen detonierender Wasserstoff-Luft-
Gemische, Fraunhofer-Institut fur Chemische Technologic, Pfinztal-Berghausen,
FRG (1993).
[92] POHL, H.-W., Hydrogen in Future Commercial Aviation, Hydrogen and Clean
244
[93] PÜLTZ, G., Mischungs- und Ausbreitungsvorgänge im Nahbereich von Rauchgas-
Emittenten, Ph.D. Thesis, Technische Universität München (1992).
[94] RASTOGI, A.K., MARINESCU-PASOI, L., Numerical Simulation of Hydrogen
Dispersion in Residential Areas, (10th World Hydrogen Energy Conf., Cocoa Beach,
USA, 1994), BLOCK, D.L., VEZIROGLU, T.N., Hydrogen Energy Progress X,
[95] REDLINGER, R., Der ID-Detonations Code DET-1D, KfK Report, Research
Center Karlsruhe (1992).
[96] RICH, B.R., Lockheed CL-400 Liquid Hydrogen Fueled Mach 2.5 Reconnaissance
Vehicle, Working Symp. on LH2 Fueled Aircraft, May 15-16, 1973, NASA Langley
Research Center (1973).
[97] RODI, W., Calculation of Stably Stratified Shear-Layer Flows with a Buoyancy
Extended k-e Turbulence Model, in: HUNT, J.C.R. (Ed.), Turbulence and Diffusion
in Stable Environments, Clarendon Press (1985).
[98] ROSENKRANZ, G., Spiel mit dem Feuer, Newspaper article in TAZ, Berlin,
September 20, 1993.
[99] S ANAI, M., Computer Simulation of Explosion of a Hydrogen Storage Tank, (11th
(1996) 2149-2158.
[100] SAWYERS, D., JAHN, W., REHM, W., Hydrogen Detonation Calculations Using
the Numerical Code DET2D, Internal Report KFA-ISR-IB-17/93, Research Center
Julich (1993).
[101] SCHILDKNECHT, M., STOCK, M., Statusbericht zum Kenntnisstand des
Übergangs Deflagration-Detonation unter besonderer Berücksichtigung der Ziel-
setzung des PNP-Gaswolkenprogramms, Final Report BF-R-66.459-2, Batteile
Institute, Frankfurt (1987).
[102] SCHMIDTCHEN, U., et al., Simulation of Accidental Spills of Cryogenic Hydrogen
in a Residental Area, (15th. Int. Cryo. Eng. Conf., Genua, 1994) (1994).
[103] SCHMIDTCHEN, U., BEHREND, E., Sicherheitsfragen und andere aktuelle For-
schungsaufgaben, Auf dem Weg zur Wasserstoffenergie - Wie kommen wir weiter?,
Berlin (1997) 56-59.
[104] SCHÖDEL, J.P., Hydrogen - A Safety Risk?, Hydrogen as an Energy Vector.
Its Production, Use and Transportation, (CEC Sem., Brussels, 1978), Report EUR
6085, Commission of the European Communities (1978) 567-581.
[105] SHERIF, S.A., ZEYTINOGLU, N., VEZIROGLU, T.N., Liquid Hydrogen: Po-
tential, Problems, and a Proposed Research Program, Int. J. Hydrogen Energy 22
(1997) 683-688.
[106] SINGH, J., Gas Explosions in Inter-Connected Vessels: Pressure Piling, Trans. I.
Chem. E. 72 (1994) Part B 220-228.
[107] SMTT, H., Seasonal Variations of Upper Tropospheric Water Vapor in Relation
to Ozone in the East-West Transatlantic Corridor: Transport Versus Chemistry,
Report in Preparation (1998).
[108] SOCHER, M., RIEKEN, T., TA-Projekt "Risiken bei einem verstärkten Wasser-
stoffeinsatz", TAB Working Report No. 13, Büro für Technikfolgen-Abschätzung
245
[109] SONNTAG, R., et al., Slush Hydrogen, Gelled Hydrogen, and Gelled-Slush Hy-
drogen, NASP Contractor Report 1027, NASA Langley Research Center (1988).
[110] STAMPS, D.W., BENEDICK, W.B., TIESZEN, S.R., Hydrogen-Air-Diluent De-
tonation Study for Nuclear Reactor Safety Analysis, Report NUREG/CR-5525,
SAND89-2398, Sandia National Laboratory, Albuquerque, USA (1991).
[ I l l ] STATHARAS, J.C., VENETSANOS, A.G., BARTZIS, J.G., WÜRTZ, L, BAM
LH2 Experiments - ADREA-HF Modeling, Report, National Centre for Scientific
Research "Demokritos", Athens (1997).
[112] STEWART, W.F., DEWART, J.M., EDESKUTY, FJ., Safe Venting of Hydrogen,
(8th World Hydrogen Energy Conf., Honolulu, USA, 1990), VEZIROGLU, T.N.,
TAKAHASHI, P.K. (Eds.), Hydrogen Energy Progress VHI, International Associa-
tion for Hydrogen Energy (1990) 1209-1218.
[113] STOCK, M., GEIGER, W., Teilforschungsprogramm Gasexplosionen, Zusammen-
fassende Darstellung und Auswertung, Report BleV-R-64.181-4, Battelle-Institute,
Frankfurt, FRG (1984).
[114] SWAIN, M., Safety Analysis of High-Pressure Gaseous Fuel Container Punctures,
World Wide Web, http://www.eren.doe.gov/hydrogen/safeanal.htm, US Department
of Energy (1997).
[115] TECHNICA LTD., Techniques for Assessing Industrial Hazards - A Manual, World
Bank Technical Paper Number 55, Washington D.C., USA (1988).
[116] URANO, Y., et al., Hazards of Burning Liquefied Hydrogen, (1) Rame of Stable
Burning, (2) Flame of Abnormal Burning, National Chemical Laboratory for
Industry 81 (1986) 143-157 (in Japanese).
[117] VAN DEN BERG, A.C., REAGAS - A Code for Numerical Simulation of 2-D
Reactive Gas Dynamics in Gas Explosions, Report PML 1989-IN48, TNO, The
Netherlands (1989).
[118] VAN WINGERDEN, K., SALVESEN, H.-C., Simulation of an Accidental Vapor
Cloud Explosion, Process Safety progress 14 (1995) 173-181.
[119] VENART, J.E.S., et al., To BLEVE or not to BLEVE: Anatomy of a Boiling Liquid
Expanding Vapor Explosion, Proc. Safety Progr. 12 (1993) 67-70.
[120] VERFONDERN, K., Stand der Modellierung zur Dispersion und Explosion freier
Wasserstoff-Gaswolken, Internal Report KFA-ISR-IB-13/94, Research Center Jülich
(1994).
[121] VERFONDERN, K., DIENHART, B., Experimental and Theoretical Investigation
of Liquid Hydrogen Pool Spreading and Vaporization, Int. J. Hydrogen Energy 22
(1997) 649-660.
[122] VERGISON, E., VAN DIEST, J., BASLER, J.C., Atmospheric Dispersion of Toxic
Gases in a Complex Environment, J. Hazardous Materials 22 (1989) 331-357.
[123] VISSER, J.G., MERCX, P.M., VAYSSIER, G.L.C.M., The Quenching of Hydro-
gen/Air Flames in Obstructed and Multicompartment Configurations by Partial Iner-
tization, Nucl. Tech. 105 (1994) 59-69.
[124] VOIGT, C., Techniken zur energetischen Nutzung von Wasserstoff, BWK 38 (1986)
347-350.
[125] WEBBER, D.M., BRIGHTON, P.W.M., An Integral Method for Spreading, Vapo-
rising Pools, Report SRD/HSE/R 390, UKAEA Safety and Reliability Directorate
(1987).
[126] WEBBER, D.M., Source Terms, J. Loss Prev. Proc. Ind. 4 (1991) 5-15.
246
[127] WIERZBANOWSKI, T., An Environmental Study of the National Aero-Space
Plane, Report NASA-ASC-TR-93-5007, Aeronautical Systems Center Wright Pat-
terson AFB (1992).
[128] YANG, J.W., MUSICKI, Z., NIMNUAL, S., Hydrogen Combustion, Control, and
Value-Impact Analysis for PWR Dry Containments, Report NUREG/CR-5662, U.S.
Nuclear Regulatory Commission, Washington (1991).
[129] YOKOGAWA, K., et al., Hydrogen Environment Embrittlement of Materials for
Hydrogen Energy Service, (llth World Hydrogen Energy Conf., Stuttgart, FRG,
[130] ZABETAKIS, M.G., FURNO, A.L., MARTINDELL, G.H., Explosion Hazards of
Liquid Hydrogen, Adv. Cryo. Eng. 6 (1961) 185-194.
[131] ZABETAKIS, M.G., Safety with Cryogenic Fluids, Plenum Press, New York
(1967).
[132] ZALOSH, R.G., et al., Compilation and Analysis of Hydrogen Accident Reports,
Final Technical Report COO-4442-4, Factory Mutual Research Corporation, Nor-
wood (1978).
[133] ZEEMANN, J., et al., Low-NOx Combustors for Hydrogen Fueled Aero Engine,
(1996) 1787-1798.
[134] ZITTEL, W., ALTMANN, M., Molecular Hydrogen and Water Vapour Emissions
in a Global Hydrogen Energy Economy, (llth World Hydrogen Energy Conf.,
NEXT PAGEJS) I
left BLANK I
247
Chapter 9
PAST AND PRESENT ACTIVITIES

ON HYDROGEN IN THE WORLD
A whole variety of hydrogen energy applications of a demonstration character have

been started or are projected worldwide encompassing autonomous or partial power systems
on different scales as well as vehicle research projects. A selection is presented in the
following sections.
9.1. GENERAL OVERVIEW
National hydrogen associations have been founded in many countries over the last
two decades: Argentina, Canada, China, France, Germany, Italy, Japan, Norway, Russian
Federation, South Korea, Sweden, Switzerland, USA. Their principal goals are to foster the
introduction of hydrogen as an energy carrier, to stimulate information exchange among
interested parties (experts, companies, scientists), and to activate public interest in this field.
It appears impossible to describe or even mention all on-going international efforts

in the field of the application of hydrogen as an energy carrier. In the following, just
examples are given which indicate the directions of present research.
The IEA is an autonomous body within the framework of the OECD with the basic
aims to promote international cooperation for the development of a stable world energy
market and to support information exchange. An IEA hydrogen program was initiated in
1977 which has dealt so far with all aspects of hydrogen technology. The current tasks
(1995 - 1998) chosen for joint international research are [93]:
• photoproduction of hydrogen (as a long-term application)
• hydrogen storage in metal hydrides (as a medium-term application)
• integrated systems (as a short-term application)
A general overview of R&D activities on hydrogen in the different European

countries is given in Table 9-1. European fuel cell research is sponsored within the different
EC programs ("Altenes", "Joule", "Brite-Euram", "Thermic"). It is said, however, to be
still 3-4 years behind respective activities in Japan and the USA [10]. According to the
general strategy, focus is laid upon low-temperature fuel cell systems in the power range
of 200 to 300 kW. The aim is the establishment of a grid of fuel cell power plants where
the hydrogen fuel is centrally produced and then distributed via pipelines [32]. A part of
the European activities is explained in more detail in the following sections.
In Germany, R&D activities sponsored by the Federal Government were the advanced
alkaline water electrolysis, and the testing of components of a future hydrogen economy in
projects such as HYSOLAR (see section 9.2.5.). However, the significant CC>2 reduction
potential is presently not seen in the use of hydrogen as a vehicular fuel, but in a reduction
of conventional fuel consumption. Funding is now concentrating on fuel cells while support
of other hydrogen activities has significantly decreased since its high in 1990/91. As of
February 1998, nine 200 kW(e) PAFC power plants are being operated in Germany. The
249
Table 9-1: European hydrogen research and development activities, from [51]
COUNTRIES B CH D DK E F GB 1 IRL N NL S SF
RESEARCH
AREAS
PRODUCTION
Advanced Electro lyis # # # J4
# # #
jj
Hjfrom PV # # # IT #
H2 from Wind # # xi
rr
H 2 from Bio mass # #

_u
Hz via Photo process # # # rr
H 2 from Natural Gas #«

H2 from Concentr Solar # #
STORAGE AND
TRANSPORT
Metal Hydrides 4J
n *< # # #
Compressed Hydrogen # #
Liquid Hydrogen # # # #
Liquid Organic Hydrides # #
Cryo Adsorption #
Seasonal Storage of H2 #
UTILIZATION
SYSTEMS
Intern. Combust Engine jj
fT if # #
1C E Vehicles ft # # #
FC Vehicles/Boats # # # #
Hz Combust. Turbine #
j_
H2 Sensors 4J
rr tr
Embrittle ment # #
Solar Hz House # # #
Integrated RE Hz System # # # # #
FUEL CELLS
AFC # # # #
PEMFC # # # # # # # #
PAFC # # #
MCFC # # # # # #
SOFC # # # # # #
DMFC #
fuel cell lines currently under research and development are the SOFC and the MCFC in
the high temperature range as well as the low-temperature PEFC [94]. The market for fuel
cell power plants in Germany is predicted to be 300 - 800 MW per year at the turn of
the century [66]. Projects currently sponsored by the German utilities are concentrating on
renewable energies solar, wind, hydro, biomass. Hydrogen is merely present with just one
project out of 265 current projects [53].
250
General goal of the present National Hydrogen Program in the USA as expressed in
the "Hydrogen Future Act" of 1995, is to fund R&D activities in hydrogen production,
storage, and use to ensure that the technologies are available for industrial considerations.
The program also uses methods to evaluate and compare hydrogen technologies and integral
systems for industrial, residential, transportation and utility applications. DOE hydrogen-
related funding in the 5-years period of 1992 - 1996 included [91]:
• coal gasification, improved reforming processes, separation of hydrogen from gas
stream impurities
• high-efficiency, low-cost stationary fuel cell systems operating with coal and coal
/ natural gas fuel for electricity and heat cogeneration
technologies for broad use of hydrogen as an energy carrier and fuel
• thermochemical gasification of biomass, anaerobic digestion of solid wastes
• understanding of chemical, photochemical, photosynthesis, and photobiological
reactions
• PEM fuel cells for vehicular power sources
The Ontario Ministry of Energy in Canada uses its EnerSearch program to promote
the development of innovative energy technologies [52]. One project involves simulation
and experimental work on the direct coupling of an electrolyzer with a photovoltaic power
source and its installation and testing as a long-term field experiment in sunny and less
sunny areas. Another project has the goal to use scrap aluminum where in the presence
of an alkali hydrogen is generated. Experimental work resulted in the operation of a
benchmark-scale prototype with a capacity of 5 1 hydrogen per minute. A fuel cell project
serves the purpose to demonstrate that an SPFC could be used to convert waste hydrogen
from a chloralkali plant to electric power for re-use in the plant.
9.2. LARGE-SCALE HYDROGEN ENERGY SYSTEMS
Energy conversion always takes place in a system with many components, where the
weakest one controls the whole performance. Therefore it is necessary to also consider
complete autonomous systems. Other smaller-scale projects, many based on solar energy
as the primary energy carrier, have been or are still being conducted to test parts of a future
hydrogen energy economy and numerous more are still in the design stage.
9.2.1. WE-NET Project
Japan has shown a tremendous economic growth in the post-war period and is now one
of the world's leading industrial countries. It has mainly resulted from private industry's
rigorous efforts to invest in R&D exceeding those levels found in the USA and European
countries. In 1971, the concept of "Industry-Ecology" was initiated to study the global
environment.
In 1974, short after the first energy crisis, the Agency of Industrial Science and
Technology (AIST) in MTTI21 of the Japanese government has initiated the Sunshine
project to develop new energy technologies and four years later the Moonlight project
to focus on energy conversion technologies. Both projects promoted energy R&D in close
MTTI - Ministry of International Trade and Industry
251
cooperation between industry, government, and academic organizations. Due to the strong
interrelations between new energies, energy conversion, and environmental technology, the
New Sunshine project was started in 1993 to contribute to an international system based
on environmentally friendly energy technologies with the goals of [46]
1. innovative technology development
to implement the Global Warming Prevention Action,
2. international large-scale collaborative research
to implement the Earth Regeneration Program "New Earth 21",
3. collaborative R&D on appropriate technologies
to assist neighboring developing countries to promote technologies that suit the
partner countries.
One important project pursued under item 2. is the "International Clean Energy
Network Using Hydrogen Conversion (WE-NET)", directly managed by NEDO22 and
currently representing the most ambitious hydrogen energy research program in the world.
Its ultimate goal is the establishment of a worldwide energy system based on hydrogen.
WE-NET is projected to run over 28 years, 1993 - 2020, in three distinct phases. Phase I
(1993 - 1998) deals with surveys on key technologies and elemental research and system
studies under 9 subtasks [76]:
Subtask 1: Investigation and Study for Evaluation and R&D Review
coordination of subtasks, evaluation of results
Subtask 2: Investigation and Study for Promoting International Cooperation
international information exchange
Subtask 3: Conceptual Design of the Total System
including power generation, hydrogen production, storage & transportation,
safety
Subtask 4: Development of Hydrogen Production Technologies
electrolysis focusing on the alkaline water and the solid polymer electrolyte
membrane water electrolysis
Subtask 5: Development of Hydrogen Transportation and Storage Technology
liquid hydrogen, liquefaction plant, tank ship, large-scale storage
Subtask 6: Development of Cryogenic Materials technology
behavior of structural materials at LH2 temperatures
Subtask 7: Feasibility Study on Utilization of Hydrogen Energy
use of hydrogen in cryogenic power generation, motor power generation,
automobile technology, fuel cells
Subtask 8: Development of a Hydrogen Combustion Turbine
evaluation of optimal design, combustion control, turbine components, design
and construction of pilot plant
Subtask 9: Study on Innovative and Leading Technology
In Phase 2 (1999 - 2005), prototype systems in the order of 50 MW will be developed,

while in Phase 3, the technology is to be demonstrated in sub and full systems.
According to the concept of the WE-NET system, primary energy sources considered
are the renewables hydro, wind, solar, to be used for electrolytic hydrogen production.
12
NEDO (New Energy and Industrial Technology Development Organization) established in 1980 is an implementing agency of the
Japanese government to promote the development of technologies related to new energy sources.
252
Table 9-2: Targets of Phase I and final targets in the WE-NET project, from [47]
Items Items Phase I target Final target

1. Production (SPEWE) Electrode area: 6,000 cm2 Electrode area: 30,000 cm2
Current density: 1 -3 A/cm2 Current density: 3 A/cm2
Efficiency: 80-90 % Efficiency: > 90 %
2.1 Transportation & storage Development of elemental Establishment of
technology for large-scale technology required for
liquefaction plant, LH2 tanker, 300 t/d liquefaction plant,
LH2 storage 20,000 m3 tanker, 50,000
m3 storage tank
2.2 Metal hydride alloys for Development of hydrogen Development of alloys for
distributed storage and absorbing alloys for commercial use
transportation - automobiles
- compact stationary storage
tank
- large-scale storage system
3. Cryogenic Materials Establishment of methods for Development of structural
materials evaluation materials for cryogenic
tanker and storage plants
4. Utilization Investigation and review of Establishment of
utilization technologies technologies for
hydrogen-fueled
automobiles, households,
etc.
5. Hydrogen combustion Development of basic Pilot system on the basis
turbine technology for design and of a 500 MW level
construction of pilot plant commercial system
inlet temperature of
turbine: 1500 - 1700 °C
Efficiency: 60 %_____
6. Ultrahigh temperature Selection of candidates of Development of ultrahigh
materials promising materials, temperature materials
establishment of methods for (2000 °C level)
materials evaluation
Nuclear is excluded as an option in the first phase. Transmission system of choice for
long-distance sea-borne transportation is LH2 (Fig. 9-1), which is advantageous because
of Japan's geographic location, and in second place liquid hydrides including the methanol
variant. On the demand side, hydrogen is utilized as a fuel in the transportation sector, in
fuel cell plants, in combustion turbines, in households as well as chemical feedstock.
253
RelurnGas Glower
iry 19 7Wi
s ( ] GN2
——>~ Mtrogen U^ciier
-^— now 9240Mm 3 'tiN2
— -^
w
oo7
Delivery Pumo
Storage Tank ST*"'?^
Tanker
Cipaoi? 165000m3
^jp4"1' 3 wp»c«y 3 Tor1

„ Hydrogen Production Plant 33*»Nm /h 33«oooHm A Hydrogen Liquefaction Plant
31 zaXIO^Y/h 3
* Parenthesized is the How al the time ol loading
Sea Water pump
Fig. 9-1: Diagram of system flow (liquid hydrogen) in the International Clean Energy Network Using Hydrogen Conversion, from [77]
Table 9-2 summarizes the targets of Phase I (1993 - 1998) and the final targets (1999
- 2020). Investigation and research in the field of safety is the establishment of a compre-
hensive "System Safety Design" for WE-NET including a concept of safety measures for
prevention of accidental LH2 release or the mitigation of accident consequences. This will
be done by an analysis of the risk of either WE-NET subsystem and by review and sy-
stematization of existing safety standards [34]. The main targets are an efficient 500 MW
hydrogen combustion turbine, efficient membrane electrolysis including SPEWE, PEFC
power plants in the 0.2 - 5 MW range, large-scale (300 t/d) H2 liquefaction with a target
efficiency of 40 %, LH2 tank ship, large-scale LH2 storage, metal hydrides, materials for
low and high temperatures [33].
The Japanese cities Toyohashi (population 350,000) and Soga (population 170,000)
have been selected as model cities in WE-NET to investigate the feasibility of a hydrogen
supply system in a hydrogen-oriented society [78].
9.23. Euro-Quebec Hydro-Hydrogen Pilot Project
A study for the concept of a clean energy system based on hydrogen as a transportable
energy medium was presented in 1987 as a result of the cooperation of more than 20
companies and institutions [39]. This study was the first definition phase of a larger
project, the "Hydrogen Pilot Project - Canada", HPPC. The follow-on feasibility study as
the phase n completed in 1990 contained the details of what was then called the "Euro-
Quebec Hydro-Hydrogen Pilot Project (EQHHPP)" [4]. The idea of this project is to take
100 MW electricity from hydropower in Canada for hydrogen production via electrolysis
which would be a technically feasible and representative size. The gaseous hydrogen is
then liquefied (which requires another 30 MW of electricity input) and shipped in 3000
m3 cylindrical barges on a barge carrier ship to Hamburg, Germany, where the barges are
brought on shore for further distribution of the hydrogen to a variety of consumers (Fig. 9-
2). The annual transportation of 14,600 tons of LH2 corresponds to an estimated equivalent
power output of 74 MW(e) plus some 7 MW "cold" energy recovery as shown in the energy
balance in Fig. 9-3. The prime goal of the Euro-Quebec project is the demonstration of the
possibilities of safe transport, storage, and distribution in a future hydrogen economy [4].
The EQHHPP has been designed as a "local" hydrogen energy economy with demon-
strating character. One of the driving forces is to provide the basis for utilization of still
unexploited hydro power (or any other low-cost non-fossil surplus primary energy) which
are far away from the centers of high energy consumption. A very important near-term
aspect of the Euro-Quebec project is the environmental concern about local pollution in
cities.
Originally three options for the seaborne transport of hydrogen were investigated:
(i) ammonia, (ii) methylcyclohexane (MCH)/ toluene, and (iii) liquid hydrogen. The LH2
option was finally deemed the best alternative. The maritime transportation of the liquid
hydrogen is done in barge-mounted mobile tanks (see chapter 5) which can also be used
for on-shore storage. This concept avoids additional transfers which are always connected
with vaporization losses. For completeness, the idea of LH2 aircraft transportation was
also studied, but soon considered an option with an unacceptably high energy requirement
(see section 6.2.2.4.) [39].
255
MCH = methylcyclohexane
LH, = liquid hydrogen LH,
CH2 = gaseous hydrogen
Liquefaction
Electricity Storage
CH,
MCH
C Toluene
i O
Electricity from Transmission Electrolysis
hydropower lines Hydrogenation of toluene
LH
Fig. 9-2: Concept of the Euro-Quebec Hydro-Hydrogen Pilot Project, from [49]
100 MW
el
22.7
Electrolysis
77.3
29.6
29.6
Liquefaction 3.3
(Hydrogen losses]
74 + 6.9 ('Cold 1 energy)
Arrival site
Fig. 9-3: Energy balance for the Euro-Quebec Hydro-Hydrogen project (in MW(e)),
from [49]
256
The LH2 terminal proposed comprises the port for loading and unloading operations
of the barge-mounted tanks, a lifting platform to transfer the tanks after being floated out of
the barge carrier ship to the shore, and the on-shore storage and distribution devices. The
further distribution of the LH2 is done in semi-trailers by tank trucks. Two reference LH2
consumers were selected: a 5 MW(e) hydrogen fuel cell plant with an 80 m3 stationary LH2
vessel and an urban transit bus transportation system including a hydrogen filling station
with an estimated consumption of 65 m3 of LH2 per day [104].
Other possibilities of hydrogen consumption investigated are, e.g., the operation of

an LH2 driven aircraft, heating power plant with gas turbine, the continuous supply of a
steam power plant via a 1 km pipeline, and the leftover hydrogen to blend with natural gas
to feed into the local gas distribution grid.
All safety considerations performed in the framework of the EQHHPP have shown
that the anticipated hydrogen systems do not pose an unacceptable risk on people and the
environment This safety analysis resulted also in a compilation of all existing regulations
to be followed in Germany if the project were existing [92].
With the start of Phase IE (1992 - 1998) of the EQHHPP, pre-approval activities and
a hydrogen demonstration program have been started. Its most important features are:
• city bus projects with internal combustion engine or fuel cell and LH2 storage in
Belgium, Canada, Germany, and Italy;
fuel cell powered passenger boat in Italy;
• development, manufacture, testing and commercialization of advanced super-
insulated transport containers for LH2;
• safety tests for LH2 vehicle tank and model tank for maritime transportation;
• emission testing on sectors of LH2 adapted combustors for Airbus jet engines;
• utilization of hydrogen in the direct reduction of iron ore in a plasma arc furnace
in Ireland;
power plants in CHP mode with piston engine or PAFC;
• study on rules and regulations, safety, and acceptability of hydrogen;
study on social economic effects of vehicle and aircraft emissions compared with
emissions from hydrogen propulsion for the city areas of Los Angeles, London,
and Munich.
Fact is that many single components exist already, so that no decisive technological
breakthrough is necessary. However, further research efforts are required for the develop-
ment of new components and their modification to enhanced application to hydrogen and
for the introduction and demonstration of pilot plants [104].
9.23. NHEG Project
Norway is gaining its energy exclusively from hydro power and was able up to
now to to refrain from using oil, gas, coal, or nuclear power. Large-scale production of
electrolytic hydrogen is made since 1949 with a peak capacity peaking at 100,000 Nm3/h
(which corresponds to 450 MW of hydro power). It has been discontinued, however, since
it could not compete with the hydrogen produced from hydrocarbons.
257
Electric Power
PV Arrays
jiia'ifla ^ JJ j Fuel for Transportation

Desalination Refinement Sector
"TaHlsg 7
J
t
Methanol
Electrolysis .-. CO2Recovery
i Synthesis
4 i T
Raw Gas LCO2 LSQsTaaig? C02
Compression Gasification Liquefaction
Fuel for Transportation

Sector _________
Fig. 9-4: Methanol and LH2 solar energy system, from [56]
Very similar to the Euro-Quebec project, a joint German/Norwegian study, "Norwe-

gian Hydro Energy in Germany", NHEG, has been conducted to investigate the technical
and economic feasibility of a demonstration project on the generation of liquid hydrogen by
Norwegian hydro power and its subsequent transportation and distribution to various cities
in Germany. Similar to the Euro-Quebec project, the reference case is based on a power
of 100 MW; two more cases investigated are related to 20 MW total power. Maritime
transportation is considered to be done in ships with a capacity of 5700 m3 of LHi per
loading and 44 trips per year. The overall efficiency is calculated to be 54 %. The cost
of the imported LHi has been estimated to be approximately the same as if the electricity
were transmitted and used in Germany for HI liquefaction [3].
9.2.4. PORSHE Project
A solar hydrogen energy system has been proposed within the PORSHE ("Plan of
Ocean Raft System for Hydrogen Economy") to substitute oil by hydrogen as a fuel in the
Japanese transportation sector for the purpose of saving oil resources and reducing CO2
emission. The concept is based on photovoltaic cells with a total surface area of 12,000 km2
deployed on 12,000 overseas ocean rafts located in the South Pacific Ocean for electrolytic
hydrogen production. Liquefied hydrogen at an amount of 25 million t/yr (which is « 1700
times the LH2 amount in the Euro-Quebec project) is then shipped in 270 tank ships with
a capacity of 125,000 m3 each to 20 import bases (capacity 120,000 m3 each) for further
distribution to 22 domestic bases (capacity 5,000 m3 each) and LH2 stations (capacity 10
m3 each), respectively [55]. Also the option of an overseas methanol production has been
investigated (Fig. 9-4) which is superior to LH2 in terms of storage and transportation. The
258
CO2 required is shipped in liquid fonn from the site of consumption back to the production
site. The LCCh and the methanol ships are designed to have a tank capacity of 130,900
m3 and 314,000 m3, respectively [56].
9.2.5. HYSOLAR Project
In a joint German-Saudi research, development, and demonstration project, HYSO-

LAR, starting in 1986, solar hydrogen production and its utilization has been investigated.
Three photovoltaic test facilities with powers of 350, 10, and 2 kW, respectively, were
used for the production of electrolytic hydrogen. The 350 kW plant installed near Riyadh,
Saudi Arabia, included the first prototype of an advanced type alkaline solarized pressu-
rized electrolyzer, which was not designed for continuous HI production. Its operation
time was two years. The produced H2 was stored as a compressed gas up to a bottling
pressure of 15 MPa. Two 10 kW(e) and one 2 kW(e) electrolyzers were deployed in Stutt-
gart for the purpose of systems development for advanced hydrogen equipment. Another
2 kW(e) electrolyzer in Jeddah served as an experimental setup. Hydrogen consumption
units were realized in two different types of a catalytic burner for heating purposes, in
several PAFC stacks with powers varying between 100 and 1000 W, and in three internal
combustion engines with 6, 8, and 13.5 kW power [82]. System studies with the energetical
and economic analysis of decentralized hydrogen systems were conducted with the time
step simulation model INSEL-NETZ. The HYSOLAR project could demonstrate that solar
hydrogen production is feasible with a reasonable efficiency.
9.2.6. Solar-Wasserstoff-Bayern Project
A solar hydrogen demonstration plant project on an industrial scale has been imple-
mented since 1985 by the Solar-Wasserstoff-Bayern GmbH (SWB) in Neunburg vorm
Wald, Germany, to test important subsystems of a future hydrogen energy economy on
a long-term basis. Diverse components of novel or advanced design have been occasio-
nally added to replace older ones or to complete the existing facility. Fig. 9-5 outlines a
diagram with all subsystems of SWB:
• Eight photovoltaic fields with a total of 4200 m2 of modules with a total peak
power of 360 kW provide electricity from solar power.
• Two advanced types of electrolyzers of the 100 kW category have been tested
for generation of hydrogen demonstrating an 80 % efficiency and a capacity of
80,000 m3 per year. However, their further development was abandoned, since no
market chances were seen. Now the testing of a pressure-type electrolyzer which
works under 3.2 MPa pressure and at 150 °C has started avoiding the necessity
of compressors for storage. Two H2 storage tanks hold a volume of 5000 Nm3,
another oxygen tank has a capacity of 500 Nm3.
• Two different fuel cell power plants have been tested. The alkaline type demon-
strated good performance during its trouble-free operation, but due to repeated
repair work, it was later stopped. Since 1993, a PAFC power plant of 79 kW(e)
is in operation which also allows cogeneration of 180 °C heat. It is fueled with
either hydrogen or natural gas to be converted to hydrogen in an upstream steam
reformer. A 10 kW PEM fuel cell is operating a standard electric fork lift truck
259
Project realization
Power generation Scctricai powe; Gas generation Storage Applications

condttxmmg Hest
nstjrsl gas network i
20 kWjj, Gas heating boiler {oxidize,- sir)
20 kWg, Gss heating boiier (oxidsc' Oj),
deccrrmisskincc'
17 kW». catclyticslty htxutd

rtfyigcratior. vr.i?
low pressure
eiectroiyss
SOmbar, lil k'A
13 bar. 100 kWd,
siecorrraissionec' Shdncfork lift tntci; unfit
Oj K2 10 kWcl foe! erf.' plant
: celi pis
), oecorrrr.tssoncs
79 kWj- /42 k'A'ts (phosphoric acid)
Protect - Prase 1: ",SS7 - t991

Projccr - j=tee 2 ,'992 - JSSS
(issue os/se; S". d3ts. all p-tssurcs a-e osc'atins pressures (gauge)
Fig. 9-5: Schematic of SWB subsystems, from [97]
with a 300 kg Ti-based hydride battery with a capacity of 23 Nm3 of H2 sufficient

for 8 hours of operation.
• The calorific-value gas-fired heating boiler of 20 kW(th) is a standard unit modified
to also allow for hydrogen fuel and mixtures of both Ha and CHU.
A 10 kW(th) catalytic burner to be fueled with either natural gas or its mixture
with up to 50 % hydrogen has been developed at the Research Center Jiilich. The
fuel gas is routed through a porous ceramic body, whose surfaces are covered with
Pd and Pt as catalysts. NOX emissions were measured to be less than 20 mg/kWh.
Tests were conducted in cooperation with BMW with an LH2 refueling station for
passenger cars. Refueling procedure with 1301 of LHi started with 1 hour duration
and a 30 % loss: it has improved in the meantime to about 5 min duration and
a loss of less than 5 %.
Despite the fact that not all of the systems installed worked satisfactorily, the overall
impression is positive and the results are promising [98]. The SWB project is scheduled
to be terminated by 1999.
260
92.1. PHOEBUS Project
A Photovoltaic Hydrogen Fuel Cell Demonstration facility, PHOEBUS, has been

erected and is currently operated at the Research Center Julich encompassing all components
for a decentralized autonomous, all-year electricity supply from solar energy using hydrogen
as the energy storage medium. Four process steps are being investigated:
1. Conversion of solar energy into electricity
2. Decomposition of water into hydrogen and oxygen by electrolysis
3. Storage of hydrogen and oxygen under high pressure
4. Electricity production in a fuel cell
The characteristic components of PHOEBUS are:

• Photovoltaic modules
with 312 m2 effective surface area.
• Pressure electrolyzer
consisting of 21 cells in bipolar arrangement, using a 30 % KOH electrolyte and
operating at 80 °C temperature and 0.7 MPa pressure. The hydrogen production
rate is 6.5 Nm3/h, the oxygen production rate is 3.25 Nm3/h.
• Fuel cell
on alkaline basis (AFC) with an electric power of 6.5 kW composed of 60 cells and
operating at 80 °C temperature and 0.23 MPa pressure. The hydrogen consumption
is 3.5 Nm3/h. It is projected for 1998 to move over to two 5 kW PEM-FC stacks.
• Short-term storage
in the form of a lead battery with a capacity of 300 kWh.
• Long-term storage
in the form of pressure vessels. For hydrogen, 18 interconnected high-pressure
bottles with a total volume of 27 m3 and a capacity of 3000 Nm3 of hydrogen at
15 MPa are used. The oxygen is stored in one pressure vessel with 20 m3 volume
and a capacity of 1500 Nm3 of oxygen at 7 MPa.
• Energy management system.
The operation of the plant serves the purpose of a grid-independent electricity supply
of the central library of the research center. Surplus solar energy in the summer season
is stored as hydrogen gas and is re-converted into electricity in the wintertime. The daily
balance of demand and supply is realized with the short-term storage battery with allows
an operation for three days. The plant can be operated without personnel [27]. All plant
components (electrolyzer, gas purification system, fuel cell, etc.) except the storage pressure
tanks are located within a large experimental building (total volume - 20,000 m3) on a 7
m x 18 m area with a distance to the roof of 17 m.
Fig. 9-6 presents the plant structure containing the figures from a simulated 1995
scenario. Based on an insolation of 282 MWh/yr, the PHOEBUS solar cells generate 29.0
MWh/yr and 26.3 MWh/yr D.C. electricity, respectively. Two thirds are being directly
consumed or short-term stored in batteries. The remaining third enters the path with the
production of hydrogen in an electrolyzer to be later consumed in a fuel cell. Conversion
losses allow an energy output no higher than 3.2 MWh/yr. The overall energy consumption
is assumed to be 17.4 MWh/yr corresponding to an efficiency of 59 % [27].
261
92 0% (Watt).97 8%(Ampere)
Solar 282 MWh —*•-Gas 209 kWh

I Battery ^ -Aux 2 MWfi (mcl Computer)
10 27 % PV-Fields | Library 1
I 1335 MWh
28 96 MWh 550 kAh 17 35MWh
t I
907 % PV-converterj 12 28
MWh Inverter 1 879%
kAh
26 27 MWh 4
19 74 MWh
I
DC-bus
8 67 MWh 3 19 MWh
f
i
87 3 % Down- .4—Aux 172 KWh Aux 80 kWh- Up- 91 %
858 h converter 4 91 MWh converter 1017 h
I
I Fictitious
7 57 MWh *
3 51 MWh
grid drawing
i (Aux +Protect +Dnve)
882% •«——Protection 652kWh 549%

Electro- Fuel
858 h lyser *—Aux 43 kWh AuxO kWh——» cell 1017 h
150 Activations 90 Activations
Aux 188 kWh. Drive 720kWh
i 68 MWh-
Gas tanks &
compressors i 39 MWh-
957% 1875 h
Fig. 9-6: Simulated 1995 scenario for PHOEBUS system, from [27]
In an advanced plant concept for the low-power range, PHOEBUS 2, high-pressure

electrolysis is employed to avoid the compression step. Other new components are an
air-driven PEM fuel cell and optionally metal hydride as an alternative storage concept.
The goal is a simplification of the whole system.
A system extension is proposed in the PHOEBUS 3 design by including the combined

heat and power concept and wind power as an additional renewable energy source. The
CHP mode is represented by a heat pump to provide thermal energy drawn from the
environment. A simulation of PHOEBUS 3 has shown a more regular annual energy
production, an increase in efficiency, and a reduction of both energy losses and costs.
262
9.2.8. HYPASSE Project
Key goal of the Swiss-German Hydrogen-Powered Applications Using Seasonal

and Weekly Surplus Electricity, HYPASSE, project starting in 1991 as a part of the
European EUREKA program is the establishment of a seasonal energy storage system
based on organic hydrides. Surplus electricity in weak-load periods including nuclear off-
peak (estimated 3200 h/yr, 500 kW(e) electrolyzer) is used for hydrogen production via
electrolysis which is stored as a compressed gas or chemically as MCH to be later used
in a public bus transportation system to operate two MCH buses and one CGH2 bus (13
pressure bottles with 147 Nl at 30 MPa) or for re-electrification. A hydrogen-powered
city bus is being developed. The basis is a Mercedes Benz diesel engine modified to a
Ha-fueled ICE with internal mixture formation. In the first phase focus was laid upon the
improvement of the MCH/Toluene cycle efficiency [96]. Originally planned to run until
1999, the HYPASSE project has been terminated by the end of 1996, after the Daimler-
Benz company has withdrawn its support with the argument that the technology to operate
H2 buses is mature and developed [28].
9.2.9. LH2 Airport
Aviation is of high economic significance for developed countries. The environmental

impact by aviation is much smaller compared with earth-bound traffic in absolute values.
However, the specific emission per passenger and mile in aviation is very high plus it is
the dominant source of anthropogenic pollution at high altitude affecting sensitive layers
of the atmosphere. Pollution in the vicinity of airports is also not negligible.
High-purity hydrogen is required for the aircraft engines. It will be liquefied in

the immediate vicinity of the airport since delivery by truck or pipeline is precluded for
economic reasons. Conceivable designs for the hydrogen gas supply is either delivery to the
airport via pipeline or electrolysis at the airport for on-site hydrogen generation. For one
third of the LH2 production plus reserves, buffer storage capacity needs to be available [38].
A number of conversion proposals for airports to service IJHb fueled aircraft have
been made. These are considered purely theoretical analyses and conceptual design studies.
Investigations have been made for the airports San Francisco [65], Chicago [29], and Zurich
[1] in terms of installations and investment required,
A preliminary assessment of the technical, operational, and economic impact on air

terminals and airline ground operations, if LH2 were used as a fuel for commercial air
transport, was made taking O'Hare International Airport in Chicago as a basis [29].
The assumption was to replace the current fleet by 400 passenger, 10,186 km design range
LH2 aircraft operating present23 routes at present frequencies, which was 112 airplanes per
day and a simultaneous service required for up to 12 airplanes. The fuel capacity assumed
for one airplane was 28 (metric) tons of LH2 (~ 395 m3). Due to efficient airline operation,
preference was given to a baseline concept that permits both conventional jet fuel and LH2
powered aircraft to operate from the same gates.
23
This study was published in 1976!
263
According to the baseline concept, the delivery of gaseous hydrogen to the airport is
done via 0.36 m diameter pipelines at 4.5 MPa pressure into an underground buffer storage.
The hydrogen gas is then routed to a liquefaction plant consisting of three modules with a
liquefaction capacity of 726 t/d including 625 t/d for aircraft, 19 t/d ground system losses
and the remainder of 82 t/d demand variation. The LH2 is stored in four spherical tanks
with a total capacity of 1450 tons or a two-days production. Two tanks serve as active
tanks for distribution to the refueling stations. Defueling of the airplanes is done into tank
trucks which transfer the LH2 to the storage tanks. As a safety distance, 305 m between
runway and LH2 storage and liquefaction facilities was considered an adequate spacing.
Recovery and reliquefaction of all (boiled off) gaseous H2 was considered simple
and cost effective. LH2 distribution to aircraft refueling devices is done via a vacuum-
jacketed pipeline system. In order to meet redundancy requirements, a three-line system
was selected in which normally one is filled with liquid and the other two are used for vent
hydrogen where one vent line could be switched to LH2 supply. The LH2 supply line with
0.4 m diameter allows a flow rate of 227 t/h at 0.3 MPa pressure. The refueling flow rate
is < 14.3 t/h to keep the airplane fuel pipe inner diameter below 0.18 m.
The low ignition energy of hydrogen requires that hydrogen can be vented or leaked
only to controlled areas. The potential for ignition after leakage is always present, the
elimination of ignition sources is not practical. The necessity to minimize leakage and to
avoid any hazard from leak fire is successfully met by the chemical industries. It has to be
assured that H2-air mixtures cannot be formed or remain undetected in a closed area and
that venting outside the building in a safe disposal system is possible.
The process of cooling down a system usually evolves a two-phase flow causing flow
and pressure oscDlations and different cooling rates and thus imposes excessive stresses on
the piping. If venting is necessary, hydrogen can easily be burnt or flared at low venting
rates (< 0.3 kg/s), whereas flare stacks or burn ponds in remote areas should be used for
high flow rates. Experience is available for disposal rates up to 135 kg/s [42].
Safety considerations are

• separation of LH2 facilities from roads, buildings, runways,
• ventilation for enclosed areas,
• preclusion of air ingress into LH2 systems,
• automated system shutdown in case of malfunction,
• confinement and control of large LH2 spills in critical areas,
• use of non-sparking electric devices.
A first approach of a fault tree analysis has been made as far as information was
available. The analysis included the initiating events loss of fuel supply for one or both
aircraft engines, failure of disconnecting the aircraft from the refueling station, leakage of
loop piping for fuel supply for several aircraft, leakage of ground storage tank. Leakages
are usually announced by a loss of insulation, which leads to high reliability of the LH2
lines, since this is detectable and respective measurement devices could be sufficiently
often checked. Many more data from operational experience are required to conduct a
quantitative analysis including human reliability. But despite lack of information, the
264
qualitative analysis has shown that no major problems with the existing design may be
expected.
A comprehensive safety and risk analysis cannot be made yet due to the still
incomplete design. From a first-stage analysis conducted by the BAM, it was concluded
that a LHi powered airplane as well as the respective infrastructure is feasible from the
safety point of view. New aspects, however, arise with the significantly increasing size of
the components compared with existing technologies [23].
Numerous LH2 refueling processes increase the possibility of a potential accumulation

of impurities, solid N2 or Oz, which enhance the risk of fuel system component damage and
explosion, respectively. The conventional warmup to vaporize impurities is not practicable
for frequently used tanks.
9.2.10. CRYOPLANE Project
Aviation could take over the leading position of introducing hydrogen on a larger
scale, because [38]
• its oil consumption is currently only 3 %, which is easier to be substituted by
hydrogen than the oil in other traffic branches,
• it is concentrated to a relatively small number of airports,
• it is handled by trained personnel, and
• it capitalizes on the weight-related energy density of LH2-
In a joint German-Russian cooperation starting in 1990, a study has been worked

out about the feasibility of an aircraft propelled by cryogenic fuels [38]. The goal of this
"CRYOPLANE" project is to check whether the use of liquefied natural gas or liquified
hydrogen as an aircraft fuel is technically possible and reasonable in terms of ecology and
economy. While the German side is fully concentrating on LH2, the Russians are pursuing
also the LNG option because of its huge natural gas resources as an intermediate step. The
CRYOPLANE project is coordinated by the German "Deutsche Aerospace Airbus".
Topics covered by the project are scenarios for the transition to an alternative fuel,
infrastructure, aircraft design, fuel system layout, engine modifications, and ecological
issues. A plausible scenario for the introduction of an alternative aircraft fuel is to start the
transition for short / medium range aircraft serving routes in or between leading industrial
nations with strong environmental concern, e.g., Europe, USA, Japan. The phases leading
up to commercial LH2 aviation will be the development of the appropriate infrastructure and
the operation of a demonstrator airplane. The first generation of LH2 aviation is foreseen to
require a fleet of 400 - 500 airplanes with about 70 European airports to be equipped with
an LH2 infrastructure. The consumption of fuel is predicted to be two million t/yr of LH2-
A technological development program for the fuel system components has been started
in 1993 comprising materials for tanks and piping, control system and sensors for hydrogen
leak detection, fuel pumps, LH2 gasifier, combustion chamber. The aero engine running
on H2 requires a good mixture homogeneity to avoid high-temperature flame pockets. In
contrast to kerosene, the hydrogen combustion process can be realized at lean mixtures
(low temperatures) thus offering a much wider control range plus reducing the NOX
265
Fig. 9-7: General arrangement of the Do-328 Demonstrator. Shaded areas indicate
the LH2 fuel system, from [88]
emissions [87]. The most advanced low NOX concept is the lean premix combustion [106].
Experiments at the Fachhochschule Aachen on an ancillary gas turbine of the Airbus A320
converted to hydrogen have shown a 65 % reduction of NOX emissions. Research is now
focusing on cryo pumps [11].
In the first stage, an Airbus A310-300 was selected as the baseline aircraft. Its
conversion into an LH2-fueled aircraft required a new fuel tank concept dropping the
conventional wing tank design. The most favorable design was deemed to be the top-
mounted tank configuration with four tanks, two actives of 40 m3 each for either engine
and two passive tanks of 80 m3 each (which refill the active ones). The total fuel weight
is 15,600 kg of LHi compared with 27,000 kg of kerosene for the same flight range.
The former intention to have an Airbus-310 serve as the LH2 demonstrator has been
deferred in the meantime. Because of financial reasons, a 30 passenger regional airplane
of type Fairchild-Dornier 328 has been chosen now for modification into an LHa-driven
aircraft (Fig. 9-7). The advantages of this intermediate step are the availability of the
hydrogen engine, the lower investment in infrastructure, and the earlier introduction of a
series aircraft. Flight test of the DO-328 will start with one of the two engines converted
to hydrogen with the other one to follow after demonstrating reliable operation. The fuel
capacity is 420 kg (= 6 m3) of LHi stored in two cylindrical tanks underneath the wings
plus 1150 kg of kerosene for the second engine [88]. The ambitious goal is to demonstrate
its flight performance by the year 2002.
266
9.2.11. Biohydrogen Production System
With the support of NEDO, a project for biological hydrogen production has been
initiated in 1991 by the Research Institute of Innovative Technology for the Earth (RITE)
to run over 8 years [70]. R&D activities comprise the areas
• breeding improvement for photosynthetic microorganisms
genetic engineering and manipulation to enhance stability and improve efficiency,
• large-scale cultivation techniques,
• byproduct recovery techniques
acquisition of useful materials such as enzymes, bacterial protein, pigments, nucleic
acids,
• development of totally integrated system (see Fig. 9-8)
A computer model was used to analyze unintended impacts on climate, human health,
and environment for a hypothetical biohydrogen production system. The figures given in
Table 9-3 are based on a biohydrogen production capacity equivalent to 542 PJ/yr (w 17
GW). Emissions have been roughly estimated assuming a certain set of technical data for
a hypothetical production plant some place in Southern Europe or North Africa [67].
A biohydrogen system is presently being designed for a small town in Germany. The
demonstrating project whose realization is expected to start in 1996 includes the hydrogen
production by biomass gasification to be used in a fuel cell cogeneration power plant, as
a process gas in a metal industrial company, and in a later stage as a fuel for a fleet of
vehicles [100].
S P
H2 etc. *" f '"ti0n I Hydrogen
>- of hydrogen |
Organic Hz «tc.
acid
Photosythesis
Anaerobic process process Hz «tc.
(anaerobic bacteria) (photosynthetic
bacteria)
Waste water
| Organic
1 acid
Pretreatment
\ COz
Carbon dioxide
Photosynthesis
process
(microalgae)
Sea water Collection of
By-products
by-products
Fig. 9-8: Integrated system for hydrogen production from biomass, from [70]
267
Table 9-3: Area and freshwater demand and emissions of biohydrogen production
equivalent to 542 PJ/yr, from [67]
Equivalent to
8 % of the estimated annual
H2 production 542 PJ/yr energy consumption in
Germany in 2040
Area demand of production
2,431 km2 0.02 % of Sahara desert
facilities
Freshwater demand (90 % 10 % of River Nile during
1.7*109 m3/yr dry season
recycling)
ground water used for
Freshwater loss from 8
1.1*10 nrVyr agricultural irrigation in
production plant
Germany
Sewage sludge production in
Excess biomass production 3.7*106 t/yr of dry mass Germany
CO2: < 48.5* 106 t/yr
CH4: < 1.2*105 t/yr CO2: 7 % of emissions in
Production-related emission H2O: 20-75* 106 t/yr due to Germany in 1987
of greenhouse gases losses from production plant CH4: 3 % of emissions in
H2O: 33*10 6 t/yrduetoH 2 Germany
burning
9.3. LOCAL HYDROGEN SYSTEMS
Iceland possesses abundant renewable energy resources, but today's production is not
more than 4500 GWh for domestic consumption. Energy sources are hydro (500 MW),
geothermal (500 MW), corresponding to only 15 % and 2 %, respectively, of their estimated
potentials, plus the import of 630,000 t of refined oil products. Iceland has been producing
electrolytic hydrogen for 35 years at a rate of about 2000 t/yr consumed in fertilizer
production. Plans are currently being made to better exploit the Iceland natural energy
sources for hydrogen production (high-temperature electrolysis), export of electrolytic
hydrogen to the European market, and consumption in energy-intensive industries. One
alternative considered is the gasification of domestic peat (or imported coal) by means of
geothermal steam for methanol production and, combined with electrolytic hydrogen, and
subsequent hydrocarbon fuel synthesis [99]. Studies have been conducted where the use
of geothermal energy in the form of 250 - 270 °C steam is considered to support hydrogen
production in a Hot Elly type electrolysis and to operate turbines for the compression
steps during hydrogen liquefaction. LH2 production cost is estimated to be lower by 11
% compared with the conventional methods. Since the domestic fishing fleet is consuming
a significant part of the imported oil, consideration is given to H2-fueled ships for which
10 production plants along the coast line are foreseen with capacities ranging between 20
268
and 70 MW. On-board storage of the hydrogen is assumed as a metal hydride [22]. Within
the WEIT project soon to start, vans will be fueled with hydrogen from renewable energy
in Iceland (see section 9.5.2.2.).
A hydrogen project considered within the "Hydrogen Initiative Bavaria", WIBA,

and partly funded by the state government deals with Munich airport aiming at ground-
based traffic as H2-fueled cars. On-site pressure electrolysis is used for the generation of
hydrogen (32.5 Nm3/h) to operate three buses and one passenger car plus the respective
robot refueling station. In addition, LH2 deliveries are planned to be taken as a backup
fuel and to operate another passenger car including a robot refueling station. The project
was started in September 1997. Vehicle operation is intended to begin in fall 1998 [9].
Arcosanti is an experimental town in the desert of Arizona for 7000 people in the
final stage. It is based on the concept of "arcology", architecture -f- ecology, with efficient
circulation of people and resources, multi-use buildings, and solar energy. The American
Hydrogen Association has proposed an energy complex consisting of 40 genset dish units
to convert sunlight into electricity and to use excess electricity for hydrogen production,
and of a biomass conversion plant to convert organic matter and garbage into hydrogen
and carbon material [8].
A "Wind-Hydrogen-Energy Demonstration Plant" has been initiated in Argentina.

The initial primary power will be 1 MW, with half of it to be used for hydrogen production
in an alkaline electrolyzer at a rate of 100 Nm3/h. The storage is designed to be both as
a compressed gas and as a hydride. Hydrogen utilization is planned to be done in 10 kW
fuel cells, in an internal combustion engine, and in a catalytic burner. In a later stage,
hydrogen serves to fuel five cars and two buses [30].
In 1990, ENEA has started the Photovoltaic Hydrogen Cycle project, a small-scale
demonstration system including a KOH water electrolyzer, metal hydride storage for up to
18 Mm3 of hydrogen and a 3 kW SPFC [48].
In the "Clean Air Now (CAN) Solar Hydrogen Project" near Los Angeles, a stand-
alone electrolyzer is powered by photovoltaic cells with 48 kW total power. The produced
hydrogen stored as a compressed gas (up to 34 MPa) is used to fuel three Ford Ranger
trucks which are in commercial operation meeting the ULEV standard [89].
An energetically self-sufficient solar house has been investigated in Freiburg, Ger-

many. Pertinent components are a seasonal storage of solar energy by means of a hydro-
gen system with electrolyzer for HI generation, storage vessel, catalytic burner for high-
temperature heat generation, and a fuel cell for electricity generation. Important subject of
research is the central technique for optimizing the system efficiency [73].
Canadian water electrolysis technology has been successfully marketed. The El-250
technology with a 3 t/d capacity corresponding to approx. 7 MW as a part of a state-
of-the-art hydrogen liquefaction plant in Quebec is being demonstrated since 1987. A
similar plant with twice the capacity is being developed in Brazil [54]. Brazil has still a
huge potential of hydro power which is not used yet. Especially with respect to the hydro
potential available in the northeast of the country far from the industrialized areas in the
269
south, hydrogen is considered the energy carrier of choice with the option to export surplus
amounts as LH2 [35].
In the city of Karlsruhe, various companies are planning a Public Transportation

Project where city buses are powered by electrolytic hydrogen. The electrolyzer installed
close by the refueling station is at the same time part of the idea of a rational energy
utilization by a more economic operation of the electricity generating baseload power
plants and a more efficient utilization of primary energy.
A Renewable Hydrogen Transportation System project is currently developed for

the city of Palm Desert. Its goal is to operate a fleet of eight PEM fuel cell powered
vehicles, a refueling infrastructure including a photovoltaic hydrogen generation, and a
fuel cell service and diagnostic center [74].
An experimental program is carried out in Spain on solar hydrogen production using

photovoltaic cells with a peak power output of 8.5 kW and operating a 7 kW(e) alkaline
electrolyzer. R&D work is done on PAFC [51].
The Hydrogen Project of the Aare-Tessin AG (Atel) starting in 1989 is a demon-

stration project comprising the construction and operation of a 90 kW H2 production plant
(SPEWE) at a rate of 20 Nm3/h and the hydrogen consumption in 3 kW SPFCs [44].
A nickel hydride hydrogen rechargeable battery (NHHRB) which is similar to a

Ni-Cd battery has been developed in China for renewable electricity storage purposes. More
than 10 tons of hydride have been taken so far to produce 1.4 million sets of high-quality
NHHRBs [25]. The first NHHRB factory has opened in June 1997 with a production goal
of 30 million batteries and a 300 tons of battery materials consumption per year.
9.4. POWER GENERATION
Fuel cell research and development is being carried out for more than 35 years
in Japan, representing a typical example of the country's interest in demonstrating all
technology before commercialization, whether commercially practical or not [20]. During
the 1980s, major activity was on demonstrating the 1 MW class. Similar to the US ONSI
system, Japan has developed its own manufacture of PC25 type units. Over 50 PAFC power
plant units in the range of 50 - 500 kW are being operated in field tests since the beginning
of the 1990s. Some have reached an operational time of almost 40,000 h, which represents
already about the required stack lifetime for commercial application. Test operation of the
large-scale PAFC power plant units with 11 MW, 5 MW, and 1 MW were terminated in
1997. Total capacity of PAFC plants operating in March 1995 was 24.75 MW with more
than 1.1 million cumulative hours [20]. A PEFC development program has started in Japan
relatively late in 1992. The operation of units in the kW range has been demonstrated; the
next step will be a 30 kW unit. Focus is also laid upon fuel processing, e.g., of natural gas;
the completion of a respective 2-3 kW unit is expected for 2000. Within the WE-NET
project, a feasibility study for a 5 MW PEFC with an expected efficiency of more than 50
% fueled by LH2 is being conducted as well as for a 200 kW precursor unit [81]. Operation
tests of 100 kW MCFC stacks with external reforming have been completed to be the basis
for the construction of a 1 MW pilot plant (4 x 250 kW stacks) powered by natural gas
270
at 650 °C. Target efficiency is > 45 %, operation time > 5000 h. NEDO is supporting
SOFC projects including fundamental studies on material science and reaction kinetics and
on stack and system development with the goal of operating modules of several 10 kW
power with steam reforming of natural gas until 1999 [21].
Development of fuel cell technology has existed in the USA for about 30 years
starting with the US TARGET programm (1967 - 1976). Funding by DOE supported R&D
for stationary power generation with PAFC and MCFC. The technology has advanced to
near commercialization. In a Los Alamos project on "Fuel Cells for Stationary Power
Generation from Hydrogen", PEM fuel cell stack hardware has been developed and tested
in 1000 hours of continuous testing of a 100 cm2 call with a stable performance, and an
energy conversion efficiency of 60 % was achieved [50]. Combined cycles of advanced
fuel cell systems with gas turbines are high on the US priority list [21]. A 50 W methanol
fuel cell operated with a mixture of 3 % methanol and 97 % water is under development
in California using ruthenium as a catalyst on the membrane which allows the utilization
of up to 45 % of the potential energy in the methanol. Advantages are no need for the fuel
processor, humidification, thermal management system and the use of methanol as both a
fuel and coolant Prototype models have been operated intermittently for 3000 h and an
efficiency of 34 % was achieved [58].
The pilot project of a "Fuel Cell CHP Plant" is presently realized by the German
utility HEW in a residential area in Hamburg within the Euro-Quebec project. The practical
demonstration is done in two steps. Step 1 is the operation of a 200 kW(e) natural gas
fueled PAFC (ONSI) which is being operated since 1995 to meet district heat demand. In
step 2, the plant was completed by a 200 kW(e) hydrogen-fueled PAFC (ONSI). As the
first-in-the-world H2-based energy system in a municipal housing area, it has started its
operation in August 1997. The hydrogen fuel will be stored as LH2 in a 60 m3 tank to be
refilled every 2 - 3 weeks by a tank truck [79, 101]. The overall efficiency is 85 %. The
operation requires legal permission procedures which may set standards in Germany.
The Ansaldo company in Italy is studying H2 fueled PAFC power plants conceived
as 850 kW standard modules for applications in the chemical industries, e.g., chloralkali
plants, ammonia plants, petrochemical plants. The fuel consumption will be 750 Nm3/h of
hydrogen per MW electric power. Operation in the CHP mode would yield cogenerated
heat at 140 °C in the amount of 0.7 MW per MW electric power [14].
Within the frame of the project "Fuel Cells" established at the Research Center Julich
in cooperation with industrial companies, focus is laid upon R&D work for SOFC for
stationary applications and PEM-FC for mobile applications [68]. The main goals
are the improvement of system and understanding by solving basic problems of process
engineering as well as the environmental impact by analyzing the material streams. For
the SOFC, the so-called substrate concept has been developed which allows the production
of larger electrolyte sheets of up to 250 by 250 mm2 as a 20 /xm thick layer upon the
anode as a 2000 /zm thick support. The thin electrolyte allowed a lowering of the operation
temperature down to 800 °C (the goal is 700 °C). The construction and operation of more
than 400 single cells of 100 by 100 mm2 and more than 30 single cells of 250 by 250
mm2 were realized by the end of 1995. Operating temperatures were from 950 °C down
to 850 °C. The next step will be the operation of a 1 kW natural gas fueled SOFC unit
271
based on internal reforming with an operating temperature of 700 - 800 °C [75]. Assistance
is given by numerical modeling (micro model, stack model) and the establishment of the
necessary experimental data base. The PEM-FC activities at the FZJ include the overall
balancing of energy conversion chains for different vehicle propulsion concepts. Favorable
figures in terms of energy consumption and CC>2 emission were found to be achievable in
fuel cell cars with methanol as fuel. One goal is the design, construction, and operation
of a 25 kW stationary PEM-FC system including a compact methanol reformer, catalytic
burner, and gas purification.
Based on the MCFC, the German companies MTU and Daimler-Benz have developed
the "Hot Module" concept of a decentralized power plant, which can be fueled with natural
gas, coke furnace gas, or biomass gas. Efficiencies of up to 65 % are expected to be
achievable. A demonstration plant in Dorsten, Germany, with an electric power of 300 kW
working with pipeline natural gas is being connected to the electricity grid since August
1997 [37]. In addition, process heat at a temperature of 400 - 450 °C can be utilized; the
efficiency is about 50 %.
The German industry is pursuing the design of a natural gas fueled SOFC with a
self-supporting structure of an electrolyte of 100 fim thickness and an area of 50 by 50
mm2. Developmental steps of SOFC in the Siemens company have started with a 50 W
unit in 1993, passed 370 W, 1.8 kW, 10.8 kW, and eventually reached 20 kW in 1996.
The design specifications of a SOFC combined heat power plant (see Fig. 7-5) are an
efficiency of > 50 %, a fuel utilization of > 90 %, a thermal cyclability of > 100 cycles,
and an operational lifetime of > 40,000 hours. A 20 % pre-reforming of the natural gas
fuel is required. Its construction is expected within the next 2-3 years [40].
A self-supporting SOFC-based power plant system, HEXIS (Heat Exchanger Inte-

grated Stack), using circular planar elements has been developed by the Sulzer company
in Switzerland for natural gas as a fuel and a power of up to 7 kW [26]. Field tests have
started in summer 1997; commercialization is planned for the year 2002. The development
goal is the construction of small-size systems in the power range of 1 - 200 kW.
Goal of a joint Dutch/Danish project is the installation and operation of a 100 kW

SOFC power plant to be connected to the electricity grid in The Netherlands. Based on
the Westinghouse tubular design, the stack consists of 1152 fuel cell tubes which have a
length of 1.5 m and a diameter of 2.2 cm. Testing of this SOFC unit was successfully done
over 300 hours. It was put into operation at the end of 1997. The efficiency is expected
to approach 47 % at 100 kW(e). If in addition the waste heat of 150 kW could be utilized
in a gas turbine, the overall efficiency could be raised to 70 % [15].
The Norsk Hydro company is developing an environmentally benign concept of

electricity production by burning hydrogen in a steam turbine. Natural gas is taken as
primary energy source and converted in a steam reformer with the generated CO2 utilized
as injection gas in oil fields for oil recovery enhancement.
The cracking of propane is a subject of research at the Duisburg University,

Germany. Since the crack product carbon black is deposited on the catalyst surface resulting
in a decrease of the H2 production, it has to be removed in a purification step. A small-
272
scale arrangement of two chambers alternatingly used for cracking and purification has
been erected providing a continuous H2 flow of about 130 Nl/h (corresponding to approx.
0.5 kW) appropriate for being used in a fuel cell [19].
In a cooperative effort between the DLR and various other German companies, the
project of a hydrogen and oxygen fueled steam generator was pursued which allowed
the instantaneous provision of steam of any desired quality. Startup times of the steam
generator are in the order of 1 -2s with a combustion efficiency of > 99 %. In a first
step, a steam generator prototype was operated in 1991/92 in totally 285 combustion tests
with a power output range between 25 and 70 MW. The second step was the installation of
the demonstration facility for H2/O2 instant reserve, HYDROSS, consisting of a modified
H2/C>2 steam generator plus storage and supply system within a container [72].
Major tasks of the hydrogen energy program in the Russian Federation in which about
20 scientific organisations are involved include research on fuel cells and electrolyzers both
with solid polymer electrolyte, on H2 / C>2 steam generator (experiments with 20-100 kW
and 1 0 - 2 0 MW thermal power), and on catalytic combustion [83].
9.5. ACTIVITIES ON HYDROGEN AS A FUEL IN TRANSPORTATION
9.5.1. Aviation
An early rocket development in the USA was based on both nuclear and hydrogen
propulsion. A high-temperature graphite reactor was to vaporize and heat up liquid
hydrogen. A prototype operated successfully for a limited time. This technology was
said to be available if large-scale space propulsion For inter-planetary travel were needed
[86]. The first rocket that used a hydrogen / oxygen engine was a NASA Centaur launched
in 1963 with a thrust of 0.07 MN per engine. The Saturn-V rocket in the Apollo project
was equipped with a 2nd and 3rd stage H2 / O2 rocket engines which reached a thrust of 1
MN (with a lifetime of one hour under normal operating conditions). Due to its work on
the nuclear powered rocket using LH2 as a propellant (project ROVER), the Los Alamos
National Laboratory has gained since the 1960s comprehensive experience in large-scale
handling of gaseous and liquid H2 characterized by an excellent safety record [42, 41].
Also for the Space Shuttle main engine, an LH2-LOX fuel system was selected which
is installed in the External Tank. The LOX tank has a volume of 550 m3, the LH2 tank
has a volume of 1575 m3 both consisting ofaluminum with a 2.5 m foam insulation [80].
The three high-pressure engines for the Space Shuttle reach a thrust of 2.1 MN each [84].
NASA has designed various concepts of LH^-fueled aircraft distinguished mainly

by the tank configuration. The first successful in-flight test of an experimental hydrogen-
propelled aircraft, a twin-jet B57-Canberra with one engine modified for H2 operation, took
place in 1955/56. Also the unmanned U2 reconnaissance aircraft was fueled with LH2. In
1988, a four seater Grumman Cheetah converted to hydrogen with an LH2-fueled internal
combustion engine became the first and only, so far, airplane to take off, cruise, and land
by means of hydrogen power only [84].
273
Soiid Rocket Booster
Solid Rocket Booster
1st stage-
engine
LE-7
Fig. 9-9: Configuration of the Japanese H-II two-stage launch vehicle, from [7]
The X-30 National Aerospace Plane (NASP) was designed as a liquid hydrogen
propelled vehicle that could go into the orbit as well as travel over intercontinental ranges
at hypersonic speeds up to Mach 25 (= velocity needed to reach the orbit). Among other
technologies identified which were considered critical to its success was the air-breathing
supersonic combustion ramjet (= scramjet) propulsion system, where LH2 is injected into
the combustion chamber and ignited by hot compressed air. Slush hydrogen was considered
as a fuel in the higher Mach flights. What eventually did skyrocket was not the X-30, but
the estimated total cost of 17 billion US $ for flight testing. The high technical risk made
earlier supporters loose their interests, so proponents were unable to save the project from
termination which finally happened in 1995 [43]. New designs have since been pursued,
e.g., with the Mach 15 X-33 reentry vehicle, its smaller and lighter successor, the Mach
8 X-34, whose development has started in summer 1996, or the HYPER-X with its first
Mach 7 flight scheduled for 1999.
The National Space Development Agency of Japan, NASDA, has demonstrated

the first successful launching of the two-stage launch vehicle, H-II, with LH2 / LOX
propulsion system in 1994 (Fig. 9-9). Advanced technology was required for the LE-7
main engine. The LH2 and LOX cylindrical tanks have a diameter of approx. 4 m and a
length of 17 m/4 m and 4 m/2.3 m, respectively, for the first/second stage engine [7]. A
new and more powerful H2 rocket was successfully launched in November 1997. Since
1991, the use of slush hydrogen as a fuel for rockets or future aircraft is currently being
investigated. A concept study has been made on a civilian oriented reusable spaceplane,
SSTO (Single Stage To Orbit), Hope-X-Prototype24, based on an air-breathing propulsion
system and scramjet engine which allows a speed of > Mach 15. The plane of 94 m length
will have a 10 passenger capacity. The fuel is slush hydrogen; the oxidizer is liquid air [69].
The development of China's space industry was initiated in the 1950s. Successful
launching of the first domestically manufactured carrier rocket was in 1964. The Long
March No. 3 three-stage large carrier rocket is propelled by a liquid hydrogen fueled
combustion engine [5]. A small LH2 production plant is being operated near Beijing for
aerospace purposes.
24
The "Hope-X-Prototype" project was recently reported cancelled due to space research budget cuts m 1997 [17].
274
Experience in Germany with hydrogen as an aviation fuel has been gained already in
the early 20th century. Rigid airships containing huge hydrogen gas bags served for freight
and passenger transportation over more than 25 years until the "Hindenburg" accident 1937
in Lakehurst has immediately stopped these activities [84]. The German DLR operates the
P8 test facility for research and technology studies on high-pressure (up to 30 MPa) LOX
/ GH2 rocket engine fuel combustion as used in the ARIANE transportation system [71].
Basic research is conducted on chemical and physical processes during high-pressure H2-
C>2 combustion as well as material characterization. Experimental work is accompanied by
intensive computer modeling efforts to simulate transient and stationary flow phenomena.
In Munich, a "Hydrogen Technology Centre" is operated by the space division of Daimler-
Benz Aerospace (DASA) covering not only space-related hydrogen aspects. The activities
comprise conceptual studies as well as development, production, and testing of hydrogen
systems and components. Experimental work included NOX emission measurements on
an hydrogen combustion aero-engine.
Although many space rocket engine systems in Russia relied on kerosene as an

inexpensive fuel, the interest in H2 fueled scramjet engines started as early as 1954 with
the Mach 3 Burya project which was flight-tested in 1957. In 1991, a scramjet SSTO
was launched and operated over 130 seconds having probably achieved a speed of Mach 8
[13]. The Russian company ANTK-Tupolev is operating the "Flying Laboratory" Tu-155
since 1988, which is a hybrid version of the Tu-154 airplane. One of the three engines
can be fueled with either hydrogen or natural gas. Operating experience with LH2 is 10
hours. Efforts, however, are concentrating on LNG because of the enormous domestic
natural gas resources. Design development has begun in two directions: (i) a freight
aircraft TU-156LNG, which is a two-engine version of a TU-154M with a kerosene/LNG
hybrid propulsion and allows a payload of 14 t and a cruising range of 3600 km, and (ii)
a passenger aircraft TU-204LNG for a payload of 21 t and a cruising range of 4300 km.
Experience gained on cryogenic and engine component operation, infrastructure at airports,
training of personnel, and effects of emissions on the atmosphere is being introduced into
the joint German-Russian CRYOPLANE project [38].
The aviation company Aerospatiale in France has tested for several seconds the
operation of a slush hydrogen fueled scramjet engine. Ignition in the combustion chamber
called statoreactor (there are no moving parts) is to provide the thrust to reach Mach 6
speed [103].
9.5.2. Ground-based Vehicles
9.5.2.1. Urban and City Bus Projects
Clean energy systems have been realized such as the first "zero emission fuel (ZEV)
cell powered bus developed by Ballard Power Systems Inc., Canada. The company has
designed in phase 1 (1991 - 1993) a 20 passenger transit bus driven by 24 PEFC stacks
with 5 kW each. The bus was the world's first ZEV vehicle and has met its performance
targets. Phase 2 of Ballard's ZEV Fuel Cell Bus finished in 1995 was based on a 60
passenger commercial prototype bus with a power of 260 kW. In a subsequent phase 3,
small fleets of buses are to provide the basic data for a full commercial production [64].
Within this phase, the city of Chicago has started in early 1997 a 3-years bus fleet testing
275
program. A Ballard production of fuel cells for buses on a commercial base as a phase 4 is
expected to start after successful testing. Projects involving Canadian participation within
international programs include an urban bus transportation system using a mixture of 20
% HI and 80 % CH4 = "Hythane" as a fuel. This bus with four roof-mounted aluminum
tanks is a modified version of a natural gas driven bus. A demonstration phase with two
buses has started in 1994. Ballard has announced in 1997 the intention to also invest into
the manufacture of 250 kW stationary SPFC power plants.
Within the Euro-Quebec project, several city bus projects are being designed and
demonstrated. The world's first full-size city bus with an internal combustion engine
has been tested in Belgium since 1994. The H2 fuel is supplied from a 125 1 LH2 tank,
substituted by two 2001LH2 tanks in a later phase. A MAN city bus is the third generation
bus promoted within the frame of the Euro-Quebec project. It is operated with a dual-fuel
(LH2, gasoline) 170 kW internal combustion engine. The storage of 570 1 LH2 is in three
bottom-mounted tanks at 0.35 MPa. Boiloff loss is 3 % per day. Cruising range is 250
km for LH2 and 450 km for gasoline. It has been operated in Erlangen, FRG, starting in
April 1996, and is running since spring 1997 in Munich. The DE NORA and ANSALDO
LH2 Hybrid City Bus Demonstrator, sponsored within the Euro-Quebec project, is a Fiat
Iveco bus equipped with a PEFC fueled by gasified LH2 with 600 1 in three top-mounted
cylindrical tanks providing a constant power of 45 kW.
The first step towards an introduction of hydrogen as a fuel in Norway will be made
with the operation of four city buses using LH2 fuel in an internal combustion engine. The
600 1 tanks allow a cruising range of 250 km.
The Georgetown Fuel Cell Transit Bus Program aims at the development and
demonstration of a PAFC/battery propulsion system in urban transit buses [102]. The fuel
is methanol stored in a standard fuel tank and converted in an on-board reformer to hydrogen
to operate the fuel cell. A battery provides additional power necessary during acceleration
and fluctuating power demand, since the PAFC system is not suitable for rapid changes
in power level. The program encompassed the development of two 25 kW combined fuel
cell/battery systems in Phase I (completed in 1991), and in Phase n the fabrication of three
test bed buses, the first fuel cell powered vehicles to run on liquid fuel (the first delivered
in 1994). Vehicle testing and evaluation is subject of Phase IE. In the next step as a part
of a commercialization effort, two buses will be powered by a 110 kW PAFC plus a 60
kW battery; their completion is scheduled for 1998.
A city bus, NEBUS, powered by 10 PEFC stacks with 25 kW gross each has been
introduced in May 1997 as the third fuel cell prototype no-emission vehicle by Daimler-
Benz. Seven H2 tanks with 150 Nl at 30 MPa each allow cruising range of 250 km.
Development goal is its commercial production in 2004.
In the project Fuel Cell Bus, sponsored by the state of Bavaria, a consortium of
German industrial companies is planning to realize a zero-emission low-floor bus by the
year 1999 [18]. The basis is an MAN bus whose conventional diesel engine will be replaced
by four PEM fuel cell modules with a rated output of 120 kW (Fig. 9-10). The hydrogen
fuel will be stored as compressed gas at 25 MPa in nine pressure bottles on the roof,
sufficient for a 250 km cruising range.
276
SysternComponents CooSng System Power Electronics Hydrogen Storage System
FuelCefis E-Tractfon Motor
Fig. 9-10: Schematic of an MAN low-floor bus converted to a PEM fuel cell drive
within the German project "Fuel Cell Bus"
The Chula Vista Hydrogen Bus Project is to demonstrate zero-emission bus trans-
portation. The buses are equipped with Ballard PEM fuel cells and fueled by compressed
hydrogen with a range of 250 miles [24].
9.5.2.2. Cars
R&D activities on hydrogen energy in China have started in the 1970s. The project
of a gasoline / hydrogen hybrid automobile with hydride storage and internal combustion
engine was pursued in the early 1980s and demonstrated a 10 - 15 % energy saving and
a much lower pollution emission, however, it did not pass beyong research stage [25]. In
the fuel cell area, China has studied the hydrogen-air PEFC and the direct methanol fuel
cell. A double reactor for hydrogen production at a low-energy consumption level, 2.7 -
2.9 kWh per Mm3 hydrogen, is later to operate on a pilot plant scale [25]. Current research
is focusing also on the indirect methanol fuel cell for vehicles (cars and city buses).
A BMW 745i test vehicle designed in 1983/84 in cooperation with the DLR was
the first in Europe licensed for LH2 operation [85]. In a modified vehicle of the 735i type
fueled with liquid hydrogen, the tank has an inner volume of 93 1, the maximum pressure
allowed is 0.5 MPa. The cruising range is up to 300 km [90] (see also section 7.2.3.1.).
The refueling procedure tested within the SWB project (see section 9.2.6) requires less
than 20 minutes. The fourth generation test car (1995) operates at 80 kW with 130 1 LH2
tank providing a cruising range of 400 km. A CNG / gasoline driven hybrid BMW as an
intermediate step to LH2 has recently gone into series production.
The Musashi Institute of Technology in Japan has presented in 1975 its first test
vehicle fueled with cryogenic hydrogen. A more recent development, the Musashi No. 8
car presented in 1993, applies direct injection using a high-pressure LH2 pump. Musashi
No. 9 is a 4-ton refrigerator van which makes use of the low temperature of the LH2 to
keep the goods at a temperature of 0 °C. The tank capacity is 360 1 of LH2. The operation
of the truck has been successfully demonstrated [105].
277
Mazda has developed a rotary engine converted to hydrogen operation and employed
in two prototypes, HR-X1 and HR-X2, presented in 1991 and 1993, respectively. The HR-
X2 engine has a power of 96 kW fueled from a 280 kg metal hydride tank with a capacity
of 43 Nm3 of H2 allowing a cruising range of 230 km. Within the WE-NET project, the
Mazda Motor Corp. is testing the driving performance of a third hydrogen powered car
based on the MX-5 Miata design also with a metal hydride storage system and with a 400
kg heavier weight than the standard car. Its acceptance as a ZEV is expected [78].
In 1991, the American Academy of Science has presented the first fuel cell car. A Ford
Fiesta was modified to a hydrogen-fueled PEM-FC drive, the LASERCEL1 prototype. The
hydride storage with 2.3 kg of Ha capacity is refueled within 5 min and allows a cruising
range of 300 km [2].
From 1984 to 1988, Daimler-Benz has tested a fleet of 10 vehicles (5 passenger

cars running on a H2-gasoline mixture, 5 vans running on pure H2> in the period 1984 -
1988 based on low-temperature hydride storage. The total operation range was more than
750,000 km. In the project, the hydrogen powered vehicles were operated by "normal"
customers, the safety level was comparable with that of conventional cars.
In 1994, Daimler-Benz has introduced the New Electric Car, NECAR I, as the first
generation fuel cell vehicle, an MB 100 van in which 12 PEFC stacks with a total power
of 50 kW were deployed. A second generation, NECAR n, has been presented in 1996. It
is a 6-seat V-cIass van operated by two 25 kW Ballard PEFC stacks with 65 % efficiency
and fueled by compressed hydrogen in two 140 1 tanks under 25 MPa pressure. Cruising
range is 250 km. The mass-power ratio is 6 kg/kW compared with 21 kg/kW in NECAR
I [57]. The overall efficiency for this protype vehicle was measured to be 29 % so far
with a projected 40 - 45 % to be reached by 2003 [45]. In May 1998, Daimler-Benz has
successfully test-operated its latest fuel cell vehicle, NECAR 3, an A-class passenger car
powered by a 50 kW fuel cell system with an onboard conversion of methanol to hydrogen
by a miniaturized steam reforming system. A 40 1 methanol tank will allow a cruising
range of about 400 km [36]. In cooperation with the Canadian Ballard Power Systems,
Daimler is willing to invest into the world's first commercial fuel cell engine.
Within the project ELBA (Elektro-Auto mit Brennstoffzellen-Antrieb = Electric Car

with Fuel Cell Propulsion) at the Research Center Julich, the mobile application of fuel
cells is currently being demonstrated. A Fiat Ducato van with electric drive and lead battery
has been extended by a fuel cell system consisting of two PEFC with 2.5 kW power each,
a hydrogen pressure bottle to store 4 Nm3 at 20 MPa, 12 air pressure bottles to store 24.4
Nm3 of air (02) at 30 MPa, two venting systems for reaction heat removal with 750 Nm3/h
capacity. The fuel consumption is 4 Nm3/h. In the first test phase, the two 2.5 kW fuel cells
are utilized to charge the battery during operation. In the second phase, a more powerful
fuel cell (30 - 40 kW) could replace the battery. A comprehensive safety and risk analysis
for ELBA has been conducted which revealed no significant individual risk (10"6/yr) [31].
Within the WEIT (Hydrogen Energy Iceland Transfer) project, six delivery vans
powered by renewable gaseous hydrogen will soon be operated in Hamburg, Germany. The
project also includes the first ever publically accessible H2 refueling station; its construction
started in 1997. The H2 fuel will be provided by the gas industry in the first phase and
278
later imported from Iceland generated by renewable energies. The vans are conventional
cars converted to H2 propulsion. About 345 Nm3 of hydrogen in pressure bottles (20 MPa)
will allow a cruising range of 150 km [16, 62].
A project has started in 1996 to develop and install a 12 - 15 kW SPFC into a

Peugeot minivan equipped with a special tank to store hydrogen at 70 MPa [12]. In the
project FEVER (Fuel Cell Electric Vehicle for Efficiency and Range), Renault pursues a
car design with a 30 kW PEM fuel cell and a 120 1 LH2 tank to allow a 500 km cruising
range [59]. In both projects, the goal is to start test operation in 1997.
In 1996, Toyota has presented a fuel cell car, RAV4, which is fueled by methanol with
on-board conversion to hydrogen. The car employs a battery buffer storage for additional
power provision beyond 25 kW. Hydrogen is stored as a hydride in a rare earth based alloy
in three small tanks. The cruising range is estimated to be 50 % larger compared with an
ICE car running on methanol. Based on the same type of car, Toyota has unveiled the
RAV4L electric vehicle in September 1997 as its latest fuel cell car [60].
Production methods, utilization, and environmental impact of methanol or other

energy alcohols as synthetic liquid secondary energy carriers have been subjected to theo-
retical and experimental investigation at the Research Center Jiilich. The new production
methods allow a carbon conversion of > 90 % [63]. In the Center for Solar Energy and
Hydrogen Research in Stuttgart, FRG, methanol is being investigated to be produced in
a CC»2-neutraI manner. The hydrogen is taken from an electrolysis process while CC«2 is
recovered from the atmosphere by applying a KOH scrubbing solution to form a carbonate
which is then electrolytically decomposed by using sulfuric acid [95]. In particular, Japa-
nese studies treat the option to retain CO2 generated during the combustion of methanol, to
liquefy it and then to return it by maritime transportation back to the hydrogen / methanol
plants. Drawback is the additional technology needed connected with a 20 % efficiency
reduction [96].
The Chrysler company has designed a hydrogen fuel cell car, where in a multi-
step process hydrogen is extracted from gasoline to be used in an automotive PEFC [6].
A 50 kW gasoline processor coupled to an SPFC stack has been presented in 1997. A
proof-of-concept car is to be unveiled in 1999 [61]. The approach appears reasonable,
because it converts gasoline (or methanol or ethanol) by reducing drastically the emission
of pollutants and, at the same time, utilizes the existing gasoline distribution infrastructure.
It thus represents an intermediate step on the way to a renewable hydrogen base. The
single steps are:
• gasoline is heated until it is vaporized to a gas;
• a partial oxidation process is applied to the gasified gasoline for production of
hydrogen and carbon monoxide;
• steam is introduced to convert the CO in a water gas shift reaction into CO2 and
to produce more hydrogen;
• the residue of 1 % CO, which is still way too much for a PEFC, is converted to
CO2 by adding air and using a Pt catalyst. CO is reduced to the required level
of < 10 ppm;
• the clean hydrogen is then cooled in a heat exchanger down to the « 80 °C
operating temperature.
279
[I] ALDER, H.P., Hydrogen in Air Transportation, Feasibility Study for Zurich Airport,
Switzerland, Report EIR-Bericht No. 600 (1986).
[2] AMERICAN ACADEMY OF SCIENCE, Hydrogen Fuel Cell Vehicle
LASERCEL1™, American Academy of Science, Independence (1991).
[3] ANDREASSEN, K., et al., Norwegian Hydro Energy in Germany (NHEG), Int. J.
Hydrogen Energy 18 (1993) 325-336.
[4] ANON, Euro-Quebec Hydro-Hydrogen Pilot Project - Phase H Feasibility Study,
Final Report, Volumes I - HI, Government of Quebec and Commission of the
European Communities (1991).
[5] ANON, Space Industry, World Wide Web, http://www.chinaplus.com/db/html/
1123.html, China Plus (Hong Kong) Ltd (1997).
[6] ANON, Fuel Cell Technology, World Wide Web, http://www.media.chrysler.com/
wwwshow/ 2262.htm#genO, Chrysler Corporation (1997).
[7] ANON, Launch and Control Facilities for the H-H, World Wide Web,
http://yyy.tksc.nasda.go.jp/Home/Press/Press-e/p-h2_6_e.html, National Space
Development Agency of Japan, Tokyo (1997).
[8] ANON, The Arcosanti Project, http://www.getnet.com/charity/aha/arcosant.html,
American Hydrogen Association (1997).
[9] ANON, Spatenstich fur Wasserstoff am Flughafen Munchen, Wasserstoff-Gazette,
1 (1997) HI. Quarter, World Wide Web, http://www.hyweb.de/gazette/, Ludwig
Bolkow Systemtechnik, Ottobrunn (1997).
[10] ANON, SOFC-V Symposium in Aachen, Brennstoffzelle (1997) Issue 1, Deutsches
[II] ANON, Pumpen fur den Kryobetrieb, Newspaper article in Blick durch die Wirt-
schaft, Frankfurt, April 16, 1997.
[12] ANON, The Clean Fuels and Electric Vehicles Report 9 (1997) 132-133.
[13] ANON, Missile News: Special Report, http://www.cdiss.org/hyper2.htm, Center
for Defense and International Security Studies, Lancaster, UK (1998).
[14] ANON, Hydrogen Fed 850 kW Fuel Cells Module for Chemical Industry, World
Wide Web, http://www.ansaldo.it/CLC.module.htm, Ansaldo (1998).
[15] ANON, Brennstoffzellen-Kraftwerk, Newspaper article in Blick durch die Wirt-
schaft, Frankfurt, February 20, 1998.
[16] ANON, Hamburg beginnt mit Bau der weltweit ersten Wasserstofftankstelle fur
eine Flotte von sechs Lieferwagen, Brennstoffzelle (1998) Issue 1, Deutsches
[17] ANON, Ein Jahr der Entta'uschungen fiir Japan, Newspaper article in Blick durch
die Wirtschaft, Frankfurt, February 23, 1998.
[18] ANON, Fuel Cell Drive for Municipal Commercial Vehicles, World Wide Web,
http://www.hyweb.de/fcbus, Ludwig Bolkow Systemtechnik, Ottobrunn (1998).
[19] ANON, Katalytischer Propancracker zur Erzeugung von Wasserstoff fiir Membran-
brennstoffzellen, Gerhard Mercator Universitat Gesamthochschule Duisburg (1998).
[20] APPLEBY, A.J., Issues in Fuel Cell Commercialization, J. Power Sources 69 (1996)
153-176.
[21] ARENDS, G., NABffiLEK, H., TIETZ, F., SOFC V in Aachen Germany, European
Fuel Cell News 4 (1997) No. 3 3-4.
280
[22] ARNASON, B., SIGFUSSON, T.I., JONSSON, V.K., New Concepts in Hydrogen
Production in Iceland, Int. J. Hydrogen Energy 18 (1993) 915-919.
[23] BAM, Sicherheitsrelevante Ausrustung und Versorgung wasserstoffgetriebener Ver-
kehrsflugzeuge, Gefahren und SchutzmaBnahmen, BAM Report, Federal Institute
for Materials Testing and Research, Berlin (1994).
[24] BAMBERGER, B., Chula Vista Hydrogen Bus Project, World Wide Web,
http://www.eren.doe.gov/hydrogen/hychula3.htm, US Department of Energy
(1997).
[25] BAO, D., Some Practical Progress of Hydrogen Energy in China, (Proc. 7th
Canadian Hydrogen Workshop, 1995, Quebec City), MEHTA, S.K., BOSE, T.K.
(Ed.), Canadian Hydrogen Association (1995) 101-105.
[26] BARP, B., DJETHELM, R., Natural Gas Fuel cells for Residential Applications,
European Fuel Cell News 3 (1996) March 17-18.
[27] BARTHELS, H., et al., PHOEBUS Julich, An Autonomous Energy Supply System
Comprising Photovoltaics, Electrolytic Hydrogen, Fuel Cell, (11th World Hydrogen
Energy Conf., Stuttgart, FRG, 1996), VEZIROGLU, T.N., et al., Hydrogen Energy
Progress XI, International Association for Hydrogen Energy (1996) 1005-1015.
[28] BOECKH, M., Die Wasserstoff-Ara: Eine Vision verblaBt, Bild der Wissenschaft
34 (1997) No. 5 32-36.
[29] BOEING COMMERCIAL AIRPLANE COMPANY, An Exploratory Study to De-
termine the Integrated Technological Air Transportation System Ground Require-
ments of Liquid-Hydrogen-Fueled Subsonic Long-Haul Civil Air Transports, Re-
port NASA-CR-2699, National Aeronautics and Space Administration, Washington
(1976).
[30] BOLCICH, J.C., et al., Wind-Hydrogen Energy Demonstration Plant in Argentina,
(1996) 301-310.
[31] BONGARTZ, R., JAHN, W., MARX, J., Probabilistische Sicherheitsanalyse des
Brennstoffzellen-Fahrzeuges ELBA, Internal Report KFA-ISR-IB-5/96, Research
[32] BONHOFF, K., EU-Projekte, Brennstoffzelle (1998) Issue 1, Deutsches Brennstof-
fzellenforum e.V.
[33] CHIBA, M., ARAI, H., FUKUDA, K., WE-NET: Japanese Hydrogen Program, Int.
J. Hydrogen Energy 23 (1998) 159-165.
[34] CHTTOSE, K., et al., Progress on safety in the WE-NET Program Activtivities by
Mffl, (2nd Workshop on Safety, Menlo Park, 1997).
[35] DA SUVA, E.P., The Brazilian Potential for Hydrogen Form Energy Production
and Exportation: A Preliminary Study, (10th World Hydrogen Energy Conf., Cocoa
Beach, USA, 1994), BLOCK, D.L., VEZIROGLU, T.N., Hydrogen Energy Progress
X, International Association for Hydrogen Energy (1994) 315-321.
[36] DAIMLER-BENZ AG, NECAR 3 Wasserstoff-Erzeugung
wahrend der Fahrt, World Wide Web, http://www.daimler-
benz.com/repl.cgi?pagesrc=/research/specials/necar/necar_g.html, Daimler-Benz
AG, Stuttgart, (1997).
[37] DAIMLER-BENZ AG, High Tech Report 1997, Daimler-Benz AG, Stuttgart,
(1997).
281
[38] DASA, CRYOPLANE - Deutsch-Russisches Gemeinschaftsprojekt zum Einsatz
kryogener Treibstoffe in der zivilen Luftfahrt - Realisierbarkeitsstudie 1990/91/92,
Deutsche Aerospace Airbus GmbH, Hamburg (1992).
[40] DRENCKHAHN, SOFC Entwicklung bei SIEMENS, (Zweites Industrieseminar,
Julich, FRG, 1995), NABIELEK, H. (Ed.), Solid Oxide Fuel Cells, Research Center
Julich (1995) 25-39.
[41] EDESKUTY, F.J., Nuclear Propulsion, in: VANCE, R. W. (Ed.), Cryogenic
Technology, John Wiley & Sons Inc., New York (1963).
[42] EDESKUTY, F.J., Safety of Liquid Hydrogen in Air Transportation, Hydrogen in
Air Transportation, (Int. Symp., Stuttgart, 1979), DFVLR (1979) Paper 18.
[43] FAS, X-30 National Aerospace Plane (NASP), World Wide Web,
http://www.fas.org/irp/rnystery/nasp.htnrfl2, Federation of American Scien-
tists (1997).
[44] FISCHER, E., STUCKI, S., The Atel Demonstartion Plant for Hydrogen Production
and Use in Fuel Cells, (Proc. 3rd Techn. Workshop, Lugano, 1993), STRUCK, B.D.
(Ed.), Implementing Agreement for a Programme of Research and Development
on the Production and Utilization of Hydrogen, Annex DC: Hydrogen Production,
Document HUF-8, International Energy Agency (1993) 159-177.
[45] FRIEDRICH, J., PANIK, F., MOK, P., Mobile PEM Fuel Cells in the City of
Tomorrow, European Fuel Cell News 5 (1998) January 20-21.
[46] FUKIWAKE, M., Energy and Environmental Technology in Japan: The New
Sunshine Project, Hydrogen and Clean Energy (Int. Symp., Tokyo, 1995), NEDO
(1995) 47-54.
[47] FUKUDA, K., Japan's WE-NET Program, (Proc. 7th Canadian Hydrogen Works-
hop, 1995, Quebec City), MEHTA, S.K., BOSE, T.K. (Ed.), Canadian Hydrogen
Association (1995) 83-100.
[48] GALLI, S., Construction and Operation of a PV-Hydrogen Pilot Plant: Summary
of the Project, (Proc. 3rd Techn. Workshop, Lugano, 1993), STRUCK, B.D.
(Ed.), Implementing Agreement for a Programme of Research and Development
on the Production and Utilization of Hydrogen, Annex DC: Hydrogen Production,
Document HUF-8, International Energy Agency (1993) 151-157.
[49] GIACOMAZZI, G., GRETZ, J., Euro-Quebec Hydro-Hydrogen Project (EQHHPP):
A Challenge to Cryogenic Technology, Cryogenics 33 (1993) 161-111.
[50] GOTTESFELD, S., Hydrogen-Fueled PEM Fuel Cells for Stationary Power Gene-
ration, World Wide Web, http://www.eren.doe.gov/hydrogen/pem.htm, US Depart-
ment of Energy (1997).
[51] GRETZ, J., WURSTER, R., Hydrogen Activities in Europe, VDI Berichte No.
1201, VDI-Verlag, Dusseldorf (1995) 229-243.
[52] GREVEN, R.G., Hydrogen Production Projects of the Ontario Ministry of
Energy, Hydrogen Production, (2nd IEA Technical Workshop, Julich, FRG, 1991),
and Development on the Production of Hydrogen from Water, Document HUF-6,
Research Center Julich (1991) 11-18.
282
[53] GROTELÜSCHEN, M., "Struktur der Energiewirtschaft ist das hartnäckigste Pro-
blem", Newspaper article in VDI-Nachrichten, Düsseldorf, January 31, 1997.
[54] HAMMERLI, M., DOLENKO, A.J., BECK, N.R., The Canadian National Hydro-
gen R&D Program, (Proc. 7th Canadian Hydrogen Workshop, 1995, Quebec City),
MEHTA, S.K., BÖSE, T.K. (Ed.), Canadian Hydrogen Association (1995) 49-61.
[55] HIRAOKA, K., et al., Energy Analysis and CÜ2 Emission Evaluation of a Solar
Hydrogen Energy System for the Transportation System in Japan - I. Conceptual
Design of the System, Int. J. Hydrogen Energy 16 (1991) 631-638.
[56] HIRAOKA, K., et al., Solar Hydrogen-Methanol Energy System for Transportation
Sector in Japan, (10th World Hydrogen Energy Conf., Cocoa Beach, USA, 1994),
BLOCK, D.L., VEZIROGLU, T.N., Hydrogen Energy Progress X, International
[57] HOFFMANN, P., (Ed.), Hydrogen & Fuel Cells Letter 11 (1996) June.
[58] HOFFMANN, P., (Ed.), Hydrogen & Fuel Cells Letter 12 (1997) April.
[59] HOFFMANN, P., (Ed.), Hydrogen & Fuel Cells Letter 12 (1997) September.
[60] HOFFMANN, P., (Ed.), Hydrogen & Fuel Cells Letter 12 (1997) October.
[62] HOFFMANN, P., (Ed.), Hydrogen & Fuel Cells Letter 13 (1998) January.
[63] HÖHLEIN, B., Neue Energieträger für den Verkehr, Monographien des Forschungs-
zentrums Jülich, Vol 5, Research Center Jülich (1991).
[64] HOWARD, P.F., Ballard Zero-Emission Fuel Cell Engine, (Proc. 7th Canadian
Hydrogen Workshop, 1995, Quebec City), MEHTA, S.K., BÖSE, T.K. (Ed.),
Canadian Hydrogen Association (1995) 221-230.
[65] HOYT, J.G., Design Concept for LH2 Airport Facilities, Hydrogen in Air Trans-
portation, (Int. DGLR/DFVLR Symp., Stuttgart, 1979), Paper No. 22.
[66] HÜNNEKES, C, Stellenwert der Wasserstofftechnologie im neuen Energiefor-
schungsprogramm der Bundesregierung, Auf dem Weg zur Wasserstoffenergie -
Wie kommen wir weiter?, (BAM-Serninar, Berlin, 1995), Federal Institute for Ma-
terials Testing and Research, Berlin (1997).
[67] HÜSING, B., REIß, T., Perspectives and Limitations of Biological Hydrogen Pro-
duction, (11th World Hydrogen Energy Conf., Stuttgart, FRG, 1996), VEZIROGLU,
Energy (1996) 355-360.
[68] KABS, H., Das Jülicher Energieforschungsprogramm, Energieinnovation und
Forschung, (2nd Enquete with Virtual Exhibition, Graz, Austria, 1998),
Österreichischer Verband für Elektrotechnik (1998).
[69] KAHANER, O.K., SEMP Ship and Spaceplanes, World Wide Web,
http://www.mcc.com/projects/ilo/wais/atip/ship, Asian Technology Information
Program (1991).
[70] KOHNO, M., Development of Environmentally Friendly Technology for the Pro-
duction of Hydrogen, Hydrogen and Clean Energy (Int. Symp., Tokyo, 1995),
NEDO (1995) 243-249.
[71] KOSCHEL, W.W., HAIDN, O.J., KRÜLLE, G., P8 - The New French/German
Test Facility for H2/O2-High Pressure Rocket Engine Combustion Research, (llth
(1996) 1815-1825.
283
[72] KUSTERER, H., SCHERF, H., HYDROSS Einsatz von Wasserstoff zur Sofor-
treservebereitstellung, Wasserstoff als Energieträger, (Status Seminar, Würzburg,
1995), Projektträger Biologie, Energie, Ökologie, Research Center Jülich (1995)
95-128.
[73] LEDJEFF, K. (Ed.), Brennstoffzellen, C.F. Müller Verlag, Heidelberg (1995).
[74] LEHMAN, P., The Palm Desert Renewable Hydrogen Transportation System, 1996
US-DOE Hydrogen Program Review (Conf., Miami, 1996), Proc. Vol 2, US-
Department of Energy (1996) 737-746.
[75] MEISSNER, D., Grundlagen der Brennstoffzelle, in: HAKE, J.-F., et al. (Eds.),
3. Ferienkurs Energieforschung, Vol. 3200007, Research Center Jülich (1997)
211-227.
[76] MURASE, M., R&D Plans for WE-NET (World Energy Network), Hydrogen and
Clean Energy (Int. Symp., Tokyo, 1995), NEDO (1995) 55-64.
NET), 1994 Annual Summary Report on Results, NEDO-WE-NET-94, NEDO,
Tokyo (1995).
NET), 1995 Annual Summary Report on Results, NEDO-WE-NET-95, NEDO,
Tokyo (1996).
[79] NEWI, G., WIENBERG, G., CH4 / H2 Brennstoffzellen in Wohngebieten, VDI
Berichte No. 1201, VDI-Verlag, Düsseldorf (1995) 73-85.
[80] ODOM, J.B., Liquid Hydrogen/Oxygen External Tank Development for the Space
Shuttle, Hydrogen for Energy Distribution, (Symp., Chicago, 1978), Proc. 175-192.
[81] OKANO, K., R&D of Hydrogen Utilization Technologies: WE-NET Program and
Fuel Cell Activities in Japan, (llth World Hydrogen Energy Conf., Stuttgart, FRG,
[82] OUD, H.A., STEEB, H., The German-Saudi HYSOLAR Program, (llth World
Hydrogen Energy Conf., Stuttgart, FRG, 1996), VEZIROGLU, T.N., et al., Hy-
drogen Energy Progress XI, International Association for Hydrogen Energy (1996)
273-282.
[83] PEKHOTA, F.N., RUSANOV, V.D., MALYSHENKO, S.P., Russian Federal Hy-
drogen Energy Program, (llth World Hydrogen Energy Conf., Stuttgart, FRG,
York (1992).
[85] PESCHKA, W., Hydrogen: The Future Cryofuel in Internal Combustion Engines,
[86] PIGFORD, T.H., Historical Aspects of Nuclear Energy Utilization in the Half-
Century and Its Prospect Toward the 21st Century, J. Nucl. Sei. Tech. 33 (1996)
195-201.
[87] POHL, H.-W., Hydrogen in Future Commercial Aviation, Hydrogen and Clean
[88] POHL, H.-W., WILDNER, D., Hydrogen Demonstrator Aircraft, (11th World Hy-
drogen Energy Conf., Stuttgart, FRG, 1996), VEZIROGLU, T.N., et al., Hydrogen
284
Energy Progress XI, International Association for Hydrogen Energy (1996) 1779-
1786.
[89] PROVENZANO, J., SCOTT, P.B., ZWEIG, R., Demonstration of Fleet Trucks
Fueled with PV Hydrogen, (llth World Hydrogen Energy Conf., Stuttgart, FRG,
[90] REGAR, K.-N., FICKEL, C, PEHR, K., Der neue BMW 735i mit Wasserstoffan-
trieb, VDI Berichte 725, VDI-Verlag, Dusseldorf (1989) 187-196.
[91] ROSSMEISSL, N.P., WALTEMATH, L.A., The United States Department of
Energy Hydrogen Program, (llth World Hydrogen Energy Conf., Stuttgart, FRG,
1996), VEZTROGLU, T.N., et al., Hydrogen Energy Progress XI, International
[92] SCHMIDTCHEN, U., WURSIG, G., Lagerung und Seetransport groBer Mengen
flussigen Wasserstoffs am Beispiel des "Euro-Quebec Hydro-Hydrogen Pilot Pro-
ject" - Uberblick uber die in Deutschland anzuwendenden Gesetze, Verordnun-
gen und Technischen Regeln, Research Report 195, Federal Institute for Materials
Testing and Research, Berlin (1993).
[93] SCHRIBER, G., The Hydrogen Research Programme of the International Energy
Agency (IEA), (llth World Hydrogen Energy Conf., Stuttgart, FRG, 1996), VE-
ZIROGLU, T.N., et al., Hydrogen Energy Progress XI, International Association
[94] SEMKE, S., Energieforschung in Deutschland, in: HAKE, J.-F., et al. (Eds.), 2.
Part H, Research Center Julich (1996) 919-940.
[95] SPECHT, M., et al., Liquid Fuels from Hydrogen and Atmospheric CO2 - A
Climate-Neutral Energy Concept, (llth World Hydrogen Energy Conf., Stuttgart,
FRG, 1996), VEZIROGLU, T.N., et al., Hydrogen Energy Progress XI, Internatio-
nal Association for Hydrogen Energy (1996) 1311-1315.
[96] STUCK!, S., SCHUCAN, T., Speicherung und Transport von Wasserstoff in Form
175-194.
[97] SWB, Solar-Hydrogen, The Project in Neunburg vorm Wald, Solar-Wasserstoff-
Bayern GmbH (1996).
[98] SZYSZKA, A., Realization and Operation of SWB's Demoplant Assembling Major
Industrial-Scale Components of Solar Hydrogen Technology, Hydrogen and Clean
[99] VALFELLS, A., HUSSEINY, A.A., The Icelandic Option, Hydroelectricity, Ge-
othermal Heat, and Peat for Synfuel Manufacture, Alternative Energy Sources,
(Proc. 3rd. Int. Conf., Miami Beach, 1980), VEZIROGLU, T.N., (Ed.), Alter-
native Energy Sources HI, Vol. 5, Nuclear Energy/Synthetic Fuels, Hemisphere
Publishing Corporation (1983).
[100] WALCHER, G., GIRGES, S., WEINGARTNER, S., Hydrogen Project Bad
Bruckenau, (llth World Hydrogen Energy Conf., Stuttgart, FRG, 1996), VE-
[101] WEINMANN, O., Perspektiven einer Wasserstoffwirtschaft aus der Sicht eines
Energieversorgungsunternehmens, Auf dem Weg zur Wasserstoffenergie - Wie
285
kommen wir weiter?, (BAM-Seminar, Berlin, 1995), Federal Institute for Materials
Testing and Research, Berlin (1997) 41-46.
[102] WIMMER, R.R., Fuel Cell Transit Bus Testing & Development at Georgetown
University, (32nd Intersociety Energy Conversion Engineering Conf., Honolulu,
1997), Paper #97156.
[103] WINTER, L., Mit Wasserstoffbrei auf Uberschallkurs, Newspaper article in Die
Welt, Hamburg, December 20, 1994.
[104] WtJRSIG, G., PETERSEN, U., Early Stage Safety Analyses for Hydrogen Process
Systems, (llth World Hydrogen Energy Conf., Stuttgart, FRG, 1996), VEZIRO-
GLU, T.N., et al., Hydrogen Energy Progress XI, International Association for
[105] YAMANE, K., Developments in Liquid Hydrogen-Fuelled Vehicles in Ja-
pan, Musashi Institute of Technology, in: World Wide Web, http.Y/caddet-
ee.org/newsdesk/nw397_05.htm, Center for the Analysis and Dissemination of
Demonstrated Energy TechnologiesThe RA Energy Foundation (1997).
[106] ZIEMANN, J., et al., Low-NOx Combustors for Hydrogen Fueled Aero Engine,
286
Chapter 10
TRANSITION TO A FUTURE HYDROGEN ENERGY ECONOMY
10.1. TRANSITION FROM FOSSIL TO NON-FOSSIL ENERGY
Energy is fundamental to social and economic development. Three main features

characterize an energy economy:
(i) the choice of the appropriate primary energy,
(ii) the selection of the appropriate conversion technology, and
(iii) the need to find the proper methods of storage and transportation for the new type
of energy.
Increasing impacts on the environment with energy-related pollutants such as green-

house gases urgently require appropriate countermeasures. It is necessary to drastically
reduce the consumption of fossil fuels as the main source of anthropogenic CO2 emissions.
In a future hydrogen energy economy, the hydrogen will ultimately come from non-
fossil primary energies. Surplus energy from renewables with their limited energy density
and strong fluctuations could be consumed in the production of hydrogen as an energy
carrier. Hydrogen has a strong appeal from a pollution point of view and also from energy
security by reducing dependence on conventional fuels, although fossil fuels will provide a
long-term transitional resource. The main argument for a possible future hydrogen economy
besides the environmental issue is the possibility of seasonal storage and transport of electric
energy in large quantities and over long distances. A transitional production method for
hydrogen on the way to non-fossil H2 uses fossil raw material and energy for the reforming
reaction from non-fossil sources representing a resource-saving (30 - 40 %), low-emission
method.
The transition to a hydrogen/electricity energy economy is dependent on whether

the electric grid is capable of serving the peak demand for electricity. For Europe, it is
estimated that energy storage in the form of hydrogen is necessary, if the fluctuating portion
in electricity generation exceeds 20 - 30 % [18]. Starting on the basis of the existing
electricity grid, local and regional structures of renewable energies should be extended and
integrated. In the future, hydrogen will take over the jobs that electricity alone is merely
able to do: vehicle fuel, seasonal and strategic energy storage, raw material in the chemical
and iron industry. Utilities generate and supply electrolytic hydrogen as their third main
product besides electricity and heat. Electrolysis can be used at the same time to meet
demand fluctuations. Electricity and hydrogen are interconvertible using electrolysis and
fuel cells.
A restructuring of the existing energy system requires several decades and massive
capital investment meaning that only really striking arguments will be able to initiate the
change and it should start as early as possible. Motivation for the introduction of the
hydrogen economy is continuously increasing traffic in the future and the need for a local
reduction of emissions of high-density traffic in the cities. In the future, new zero-emission
concepts for aviation in the sensitive ecosystem of the atmosphere are required. Hydrogen
287
Source: lESVic
Fig. 10-1: Evolution of the energy system for land transportation, from [2]
technologies must undergo a significant technological improvement and cost reduction and
must advance beyond the proof-of-concept stage before significant investments will be
made in infrastructure changes. Technical and cost issues may prevent its widespread use
unless a framework of scientific principles, technical expertise, and analytical evaluation
and assessment is developed. With the present energy policy, there is no need for a rapid
introduction into the market, therefore efforts should be focused on the optimization of
existing technologies.
A conceivable intermediate step for the transition to a comprehensive hydrogen

economy is the concentration on a low-energy economy with minimum CC>2 emissions into
the atmosphere using local excess potentials of electricity, e.g., renewables hydro, wind or
cheap off-peak electricity in the day/night, working day/weekend, or winter/summer cycles.
Where the available potential of primary energy does not (yet) suit the demand structure,
i.e., far from energy consumption sites, storage systems must be used. Any additional
hydrogen required could, e.g., be imported from the solar belts or from fossil fuels with
solidification of CC>2 for disposal. The transition, however, i.e., the gradual reduction of
the primary fossil fuel and substitution by non-fossils, requires "investment" energy.
A significant portion of the energy for transportation and other applications can be
provided by hydrogen. The main problem in introducing hydrogen is not of a technical
288
nature, it is the large investment in new infrastructure. The change of an existing system
needs to be done in small steps. The time dependence of an emerging hydrogen economy
for the example of land transportation is visualized in Fig. 10-1. A first step could be taken
by the use of hythane (mixture of natural gas + hydrogen) thus minimizing infrastructure
cost. Hythane allows a gradual transition from natural gas to ultraclean hydrogen. Starling
with small proportions of Ha, the existing pipeline structure could be used. Continuous air
quality improvement can be realized by increasing the hydrogen content with a safe fall-
back position at every step along the way. "Smart" delivery systems could supply hydrogen
to fuel cells while delivering hythane to cookers. Hythane with only 5 % hydrogen shows a
disproportionately large reduction in the emission of pollutants. The first hythane evaluation
project is underway in Denver.
One of the main areas appropriate for conversion to hydrogen is the transportation
sector with immediate environmental benefits to urban areas. Commercially produced H2-
fueled cars based on either the internal combustion engine or on a fuel cell should be
available at a reasonable price, have good performance and adequate range, as well as
widespread availability of the fuel. Natural gas is expected to play a significant role as
preceding technology. Also methanol has often been cited as a good compromise in terms
of the desired characteristics of economics, environmental impact, supply logistics, and
vehicle technology. Its synthesis could be realized by renewable hydrogen plus carbon-
containing resources (e.g., biomass, COa) and its infrastructure would already be available.
The conversion of air transportation to hydrogen would be an example on a larger scale.
In the longer term, hydrogen as a motor fuel may have a role to play in reducing oil
consumption while addressing problems of urban air pollution and also global warming.
A vehicle using a fuel cell and electrolytic hydrogen produced from renewable or nuclear
energy would represent a true zero-emission fuel/vehicle system at any point from fuel
production to end use. The problem of infrastructure for LH2 vehicles is that 3-4 times
as many tank trucks required for distribution would be on the road. Very few changes to
consumer behavior are required to switch from gasoline or diesel to methanol.
Stationary off-grid applications such as gas turbine, fuel cell power plant, or internal
combustion engines operated in the CHP mode are appropriate for decentralized energy
provision. In particular high temperature fuel cells are applicable in households because
of their need for both electricity and heat.
The duration of the transition period will depend on the so-called harvest factor
of the new energy conversion plants, which is the ratio of the energy produced during
its lifetime to the energy invested for its construction. A harvest factor of 10 means that
a complete transition could be realized within a century; a harvest factor of less than
5, however, would prevent complete transition since the new system will not be self-
sustainable. According to estimates published in 1993, photovoltaic power plants have a
harvest factor of optimistically 1-3 with a forecast 3-6 range if the lifetime of the solar
cells can be extended to as much as 30 years. An optimistic nuclear energy scenario based
on the predicted figures of a PWR-1400 would result in a harvest factor of 14, provided a
40 y lifetime, a well functioning complete nuclear cycle, and, of course, public acceptance
can be ensured [15].
289
10.2. DESIGN OF A HYDROGEN ENERGY ECONOMY
The first concepts of a "hydrogen economy" were developed in the 1960s representing
integrated energy systems with the fuel hydrogen based on nuclear or solar primary energy.
The diagram in Fig. 10-2 depicts the first concrete concept of a future electricity supply
by solar hydrogen as was proposed in 1964 [8]: seawater is fed into pressure electrolyzers
powered by solar electricity, with the produced hydrogen and oxygen gases to be introduced
into a 5 MPa pipeline grid and with heavy water, D2O, as a byproduct. The gases would
be utilized in households and industries or for electricity production by means of magneto-
hydrodynamic generators (MHD) or fuel cells (FC).
Industnal electricity consumers

Chemical
industry
Transportation systems
—^~ Air
Steel
a i^ _
storage
tank
Local
electncity
consumers
industry MHD
-
large-scale
Refueling power plant
station
for cars
Heat for
Welding
household,
technology
industry
Fig. 10-2: Concept of a solar hydrogen economy from 1964, from [8]
290
In the concept of a prospective "Novel Horizontally Integrated Energy System
(NHIES)" with zero emission as outlined in [5], the principal problem is not seen in the
amount of fossil fuels consumed, but in the need to maintain cleanness during consumption,
i.e., the treatment of emissions. This concept is based on (see Fig. 10-3)
• decomposition of coal, oil, gas and cleaning
• stoichiometric relation
• synthesis and conversion.
Unlike the conventional vertically integrated system, primary energy inputs compri-
sing the entire spectrum, including nuclear, are decomposed and cleaned prior to combu-
stion resulting in streams of CO and H2, while C>2 is extracted from air. The demand
for final energy in the form of electricity, H2, CH4, and CHaOH is then met by taking
CO, H2, O2 out of the grid (the so-called 2nd grid) in the respective stoichiometric mix-
ture. It is a zero emission system in terms of NOX and SO2- A later elimination of
the CO grid represents the complete transition to a carbon-free hydrogen economy. For
flow and cost assessment, the MARNES linear programming model was used as an ana-
lytical tool, which includes the concept of stoichiometric allocation and the conduct of
streams of non-energetic materials such as NOX or CO2, and which predicts NHIES to be
macroeconomically cost-competitive with existing non-integrated systems [5].
The idea of a hydrogen economy as described in [11] is to transform the conventional

infrastructure of cities consisting of energy, water sewerage, electricity, garbage control,
and transportation to a hydrogen system based on the existing pipeline grids for town gas,
clean water, and sewerage. Hydrogen is used for the cogeneration of heat for pyrolysis of
biowastes and vaporization of the sewerage and of electricity for end use, e.g., charging
hybrid vehicles. Electrolysis of polluted water could be used for hydrogen production and
IDecomposition Stoichiometric Synthesis Final

& Purification Completion & Conversion Energy
N2 CO2
Air Separation 1 * * CO-turbine 1 Electricity I
0 <* )
— H2 -*-
Water Splitting
— 02 —*•
n\J. H...-J
2 ") co,
N0
,
Natural
Reforming | Methanol synthesis 1— fc Fuel/neat J—1
gas
0 «> )
Coal Refining
Crude oil
SOj
Ek"CtricJtyT Heat
L____ U ____. y ...J —————————————————————————————————————————————._____ H——————————————————————————————————————.
Uranium Nuclear power plant — Pu —fr- Nuclear power plant Electricity + heat 1
+ i
Fission products Fission products
Fig. 10-3: The "Novel Horizontally Integrated Energy System (NHIES)", from [5]
291
water recycling. The transformation is seen by [11] to be feasible with an estimated billion
dollars per 100,000 population and the technology required is "low tech with a few high
tech whistles and bells".
Various technical and economic assessments of hydrogen energy systems have sugge-
sted that hydrogen could become competitive for use in efficient, low-polluting COa-neutral
systems in the near future. The near-term option allowing for a smooth transition is to start
with new, small-scale local infrastructures. For H2-powered vehicles, the following scena-
rios are possible [12]:
• central Ha production by steam reforming and delivery by either trucks or gas
pipeline to refueling station
• withdrawal of excess H2 from industrial plants and delivery via gas pipeline to
refueling station
• decentralized Ha production by small-scale steam reforming or electrolysis units
on-site refueling station
A set of coherent programmatic pathways is applied in a project dealing with an

integrated analysis of how hydrogen-powered vehicles can enter the market in substantial
quantities. Each pathway investigated includes computer modeling comprising vehicle
technology, refueling infrastructure, market scenarios, government actions. The latter is
the key input of the models permitting the assessment of the effects of making government
investments [19].
A reasonable approach to a gradual introduction of hydrogen is small-scale residential

energy system modules comprising electrolyzer, metal hydride (or pressure vessel) as a
buffer system, and CHP fuel cell components, ideal for applications far away from the
grid. Roundtrip efficiencies of 30 % have already been demonstrated, 40 % are expected
to be possible in the near future [20].
In principle, the operation of a fleet of LHa powered vehicles, e.g. city buses, is more
appropriate than individual traffic. The infrastructure can be limited to a few refueling
stations. Also boiloff losses can be avoided in fleet vehicles; for a private car the estimated
loss seen over a whole year is between 30 and 60 % of the total fuel [16]. The daily
hydrogen demand for a typical refueling station is estimated to be 3000 - 54,000 Nm3/d
[12] possibly provided by small reformer units, e.g., working off an existing natural gas
pipeline.
The establishment of strategies is an indigenous task of policy. A whole variety

of measures are conceivable for tackling the problem of environmental damage caused by
the continuous contribution of the combustion of fossil fuels: direct legislative regulations
and rules, investments, tax system, research funding. Industry may contribute to utilizing
hydrogen technology by long-term investments, research activities and cooperation [3]. The
"ultra-low emission vehicle" (ULEV), the "super ultra-low emission vehicle" (SULEV), and
the "zero-emission vehicle" (ZEV) mandate in California25 makes this state appear a good
candidate for the introduction of hydrogen vehicles in a close cooperative effort between
23
According to the "Clean Air Act" (1990) in California, the percentage of total vehicles sold must be 2 % in 1998, 5 % in 2000, and
10 % in 2003.
292
fuel cell manufacturer, automobile industry, and academic groups. Legislative measures as
taken in the USA are described in [6].
10.3. CONCEPT OF A NUCLEAR HYDROGEN ECONOMY
Fig. 10-4 describes the possibility of a chemical industrial complex where the energy
required is provided by nuclear power based on small-sized modular HTGRs [7]. One
option (upper part of the figure) employs a direct cycle gas turbine HTGR for efficient
electricity production to be used in a water splitting process for hydrogen production,
which is eventually consumed in H2-driven turbines or fuel cell power plants. In the
other option (lower part of the figure), cogeneration of electricity and process heat/steam
is performed for the purpose of methanol and hydrogen production by means of steam
reforming of natural gas and/or thermochemical cycles.
The concept of a nuclear hydrogen economy as shown in Fig. 10-5 is based upon
hydrogen generation using nuclear power, the transport of energy stored in the hydrogen,
and its consumption. The production facility consists of an HTGR combined with a
thermochemical plant where the high-temperature heat is converted directly into chemical
Use of
j Electricity Electricity
I during Night
Electrolysis
electricity —— .J
Modular HTR
200 Hit Electricity 600 MWt

700 -950 r & Steam 280 MWe
co-generation power-generalion
Hydrogen
Reforming of Production by
Natural Gas Thermo-chem i ca I
Method
Production of
Methanol
Fig. 10-4: Flow chart of a chemical industrial complex based on modular HTGRs
as the energy source, from [7]
293
o|« Oxygen (Oa) from the atmosphere
Oxygen (C>2) into the atmosphere
Water
splitting
facility
Fig. 10-5: Hydrogen energy economy on the basis of nuclear power and water
splitting, from [17]
energy. Its basic product is hydrogen. Transport and distribution are provided by a pipeline
grid. Small-scale and large-scale storage tanks serve the purpose of short-term and long-
term energy demand balancing. Areas of consumption of hydrogen as an energy carrier
are given as raw material in the chemical and steel industry, as a fuel for vehicles, for
heating and cooking in households, and for electricity and long-distance heating provided
by the utilities.
Sources of renewable energies (wind, water, sun) are deemed sufficient to more than
replace all fossil fuel. Barriers to significant solar hydrogen utilization are the lack of high
efficiency photoconverters, high stability and long lifetime, and cost-effective materials and
systems. Environmental effects of energy use and the world's growing need for electricity
are reasons for a renewed emphasis on nuclear power, at least it could serve as a transition
facilitator. A transition period ranging from decades to more than a century is expected.
The aspect of energy independence represents an argument in favor of the transition to
hydrogen [13].
Nuclear energy may gain greatest importance in helping to get the hydrogen economy
started. On a mid-term basis, overcapacities in countries with a high percentage of nuclear
power (e.g., France) and reserve capacities, which exist in all larger distributed grids, could
be used in electrolysis plants to be immediately disconnected again if needed, e.g., at
peak load times. On a longer-term basis, nuclear power could be installed for hydrogen
production if economically competitive with fossil primary energies.
294
10.4. SOME INTERESTING ASSESSMENTS
The following contains some interesting ballpark estimates which necessarily arise
when discussing the world energy situation, in particular in connection with a future large-
scale hydrogen energy economy:
• Anthropogenic CC>2 emissions guide back into the past with respect to CC>2
concentration in the atmosphere when the formation of the fossil fuels started. The
speed at which the OO2 concentration is currently changed, is roughly estimated
to be 180,000 years per year of continued emissions [9].
About 2000 km2 of forest is necessary to bind the carbon dioxide emitted from a
300 MW coal power plant.
• In the European Community, the nuclear share of 33 % in electricity production
allows the saving of CC>2 emissions to the amount of 700 million t/yr. This
corresponds to the estimated CC>2 emissions of all passenger cars (more than 200
million) in Western Europe in the same time [1].
• The world hydrogen production today is 500*109 Nm3/yr with 97 % from fossil
fuels. If this amount were to be produced by electricity (4.5 kWh/Nm3 F^) from
power plants operated for 8000 h/yr, the installed power required would be 280
GW. This corresponds to more than 80 % of the currently available nuclear power
in the world.
The projection of a complete introduction of LH2 aviation in the future implies for
Europe the replacement of 45 million t/yr of kerosene (today: 30 million) by 16
million t/yr of hydrogen or 45,000 t/d; the present liquefaction capacity in Europe
is 20 t/d. The German share of LH2 aviation fuel is estimated to be 2.5 million
t/yr LH2 or 7,000 t/d with 60 % of this for Frankfurt airport alone. Liquefaction
plants are expected to be established in units with 60 t/d capacity corresponding
to the maximum available size today. Frankfurt airport is predicted to operate 65
of those liquefaction modules in the final stage.
The German figure of 2.5 million t/yr of LH2 translates into an electricity need
of 1.6*10n kWh/yr for H2 production and liquefaction (based on 4.5 kWh/yr
production and 10 kWh/kg liquefaction). This corresponds to 15 large LWRs or
250 modular HTGRs (German reference design) or to almost 3000 km2 of PV area
(PHOEBUS data).
• The Euro-Quebec project is designed to deliver 14,600 tons of liquid hydrogen per
year to Hamburg, FRG. One barge carrier of the reference design is able to transport
1065 tons of LH2, which translates into 13.7 shipments per year. The amount of
LH2 arriving in Hamburg corresponds to a power of 74 MW or 6.48*108 kWh per
year. A large-scale hydrogen economy would definitely require the employment of
larger tank ships such as the SWATH ship with a Transport Capacity of 8150 t of
LH2- Primary energy consumption in Germany in 1996 was 350 million TOE or
approx. 4.1*1012 kWh. The assumption that 1 % of this is provided by hydrogen
means 111 loaded LH2 SWATH ships per year arriving at Hamburg harbor.
In 1992, energy consumption in Germany was 2515 TWh with 2084 TWh used
in the form of fuel and 431 TWh used as electricity. If these amounts were
to be provided by electrolytic hydrogen and assuming efficiencies of 85 % for
water electrolysis, 90 % for transport and distribution, and 45 % for re-conversion
295
into electricity, the net electricity required would be 2724 TWh and 1252 TWh,
respectively, i.e., almost 4000 TWh per year. In relation to the H2 net heat of
combustion, this energy is contained in 1.4*1012 Nm3 or almost three times the
present world hydrogen production. Water consumption on the basis of a specific
value of 805 g of H2O/m3 of H2 is then 1.2*109 m3 (without the cooling water
needed, which for alkaline electrolysis is usually in the range of 25 - 30 % of
the water to be split) corresponding to about 0.5 % of the mean precipitation in
Germany [14].
• If car traffic in the city of Munich with 500,000 vehicles per day were converted to
hydrogen-fueled traffic and assuming a 20 km cruising range per day and 1 kWh/km
energy consumption, water vapor emissions would amount to about 1 million t/yr or
2 kg/(m2 yr) in the city area. This is about 0.1 % of the mean annual precipitation.
A 1000 MW water-cooled power plant emits about 14 million t/yr of water vapor
assuming 7000 operation hours [4].
• If all light-duty vehicles in the US in 2010 were hydrogen fuel cell vehicles and
if the hydrogen were produced by renewables, land use would be 3 % of the
total US land in the case of biomass, 2 % for wind and 0.1 % for photovoltaics.
Concerning the lower efficiency of ICE, land use would at least double for hydrogen
ICE vehicles compared with fuel cell vehicles [10].
• Assuming the whole fuel demand of Germany's traffic sector to be supplied by
methanol, about 220* 109 Nm3 of hydrogen is required, which is eleven times the
present annual H2 production in Germany.
10.5. FURTHER WORK
Beyond the further development and improvement of hydrogen technologies and cost
reduction as prerequisites for the introduction of H2 as an energy carrier on a larger scale,
it is the perceived safety risk that demands further efforts in parallel with the technological
development. More information about the safety aspects of hydrogen would be helpful in
reducing uncertainty ranges for more realistic safety and risk assessments of (nuclear) heat
utilization plants and in increasing confidence in a safe operation of hydrogen systems.
Better knowledge required for more detailed risk and safety assessments of the various H2
production, storage, and transportation processes is seen in different areas:
• investigation of the evolution of the H2 gas cloud and its mixing with and influence
on the local atmosphere in terms of spilled amounts and release rates conceivable
in large-scale accidents as well as the validation of respective calculation models;
• improvement of understanding of the turbulent processes in the flame zone, over-
pressures under various confinement conditions, pressure piling dependence on
geometric boundary conditions, thermal radiation; quantification of the various
parameters influencing flame acceleration and its influence on the initiation of a
detonation;
• research on the combined effects of mechanisms that contribute to boiloff in LH2
storage, in particular, moving storage systems (ortho-para conversion, heat leak,
sloshing, flashing, thermal stratification);
• evolution and development of better hydrogen release detection devices;
296
• examination of air in-Ieakage into LH2 (and SLPÏ2), solid air effects, and the
influence on combustion behavior, oxygen detection;
• continued atmospheric research on the impact of aviation on stratosphere and
enhanced water vapor emissions on troposphere;
• development of unified consistent standards for design, construction, operation,
maintenance and their international acceptance;
• human factors engineering for industrial plants in a systematic manner.
Concentrating on nuclear power in connection with hydrogen production, further

safety-related work is necessary on
• tritium and hydrogen transportation in the process heat system;
• process gas cloud explosion hazards inside/outside the reactor building, response
of containment structures, and substitute liquid energy carriers (energy alcohols)
on a natural gas and biomass basis;
• improvement of material and structural design by examination of long-term creep
behavior and high-temperature corrosion properties of reference materials and the
new superalloys including tests in a sulfur oxide atmosphere;
• passive cooling steam generator as an innovative safety concept which would not
necessarily require a reactor shutdown in the case of a disturbance in the process
heat system.
[1] ANON, Foratom: Kernenergie trägt zur Nachhaltigkeit bei, SVA-BuIletin, No. 14,
Schweizerische Vereinigung für Atomenergie, Bern (1997).
[2] BALLARD, G.E.H., Engines of Change, (llth World Hydrogen Energy Conf.,
[3] BEHREND, E., SCHMIDTCHEN, U., Perspektiven der Wasserstofftechnologie in
Deutschland, Auf dem Weg zur Wasserstoffenergie - Wie kommen wir weiter?,
Berlin (1997) 9-10.
[4] FRITSCH, M., Wasserstoff - Die Zukunftslösung? Umweltfreundlicher Energieträ-
ger für das 21. Jahrhundert, Bessere Umwelt (1994) No. l, Deutsche Aerospace,
Munich (1994).
[5] HÄFELE, W., MARTINSEN, D., WALBECK, M., The Concept of a Novel
Horizontally Integrated Energy System: Ways to Reduce CÛ2 Emissions, Automatic
Control, (10th World Congress, Munich, 1987), Preprints Vol. 5, International
Federation of Automatic Control (1987) 155-161.
[6] HÖHLEIN, B., Neue Energieträger für den Verkehr, Monographien des Forschungs-
zentrums Jülich, Vol 5, Research Center Jülich (1991).
[7] IDE, A., et al., The 3rd Report on KASEVIA NADA (Coast) Artificial Islands Study
- An Energy Island Part, Report in Preparation (1998).
[8] JUSTI, E., Leitungsmechanismus und Energieumwandlung in Festkörpern, Vanden-
hoeck & Ruprecht, Göttingen (1965).
297
[9] KLOPRIES, B., Die Wasserstoffwirtschaft aus heutiger Sicht - ein Resümee,
Wasserstoff als Energieträger, (Status Seminar, Würzburg, 1995), Projektträger
Biologie, Energie, Ökologie, Research Center Jülich (1995) 181-191.
[10] LIPMAN, I.E., DELUCCffl, M.A., Hydrogen-Fuelled Vehicles, Int. J. of Vehicle
Design 17 (1996) 562-589.
[11] MYERS, R., The Hydrogen Economy Stupid, World Wide Web,
http://www.igc.apc.org/raenergy/stupid.html, The RA Energy Foundation
(1997).
[12] OGDEN, J.M., Development of Infrastructure for Refueling Hydrogen Vehicles,
(1996) 1113-1122.
[13] ROTHSTEIN, J., Nuclear Energy, Environmental Problems and the Hydrogen
Energy Economy, Int. J. Hydrogen Energy 20 (1995) 275-281.
[14] SCHÄFER, H., Einführung in die Gesamtproblematik des Global Link, VDI Be-
richte No. 1129, VDI-Verlag, Düsseldorf (1994) 1-15.
[15] SEIFRITZ, W., Zwanzig Jahre Wasserstoffwirtschaft: Zur Ökobilanz beim
Übergang in eine dekarbonisierte Energiewirtschaft, Newspaper article in Neue
Zürcher Zeitung, Zürich, August 18, 1993.
[16] SOCHER, M., RJJEKEN, T., TA-Projekt "Risiken bei einem verstärkten Wasser-
stoffeinsatz", TAB Working Report No. 13, Büro für Technikfolgen-Abschätzung
[17] STRUCK, B.D., Wasserspaltung unter Einsatz von HTR-Wärme, KFA Annual
Report 1977/78, Research Center Jülich (1978).
organischer Verbindungen, VDI Berichte No. 1129, VDI-Verlag, Düsseldorf (1994)
175-194.
[19] THOMAS, C.E., Integrated Analysis of Transportation Demand Pathway Op-
tions for Hydrogen Production, Storage, and Distribution, World Wide Web,
http://www.eren.doe.gov/hydrogen/integrat.htm, US Department of Energy (1997).
[20] VANHANEN, J.P., LUND, P.D., TOLONEN, J.S., Electrolyzer-Metal Hydride-Fuel
Cell System for Seasonal Energy Storage, (llth World Hydrogen Energy Conf.,
298
Chapter 11
FINAL REMARKS
In principle, two different and independent areas have been covered in this report:
on the one side hydrogen as an energy carrier and its application in a future large-scale
hydrogen energy economy, and on the other side: nuclear power which represents an
important option for providing primary energy for hydrogen production. The combination
of both nuclear and hydrogen is not necessarily required but it represents a reasonable and a
feasible option to facilitate the emergence of a hydrogen economy. Both areas will help to
reduce global impact on the atmosphere. Hydrogen faces economic problems and possibly
public acceptance problems. Nuclear faces public acceptance problems.
Many research activities still need to be done in both areas in terms of an economic
breakthrough of hydrogen and its link to nuclear process heat. Since hydrogen is me-
rely used today as an energy carrier, research is ongoing in all subsections encompassing
technological development of components (storage, infrastructure, etc), safety-related inve-
stigations, and regulations. Nuclear power is used for hydrogen production by providing
electricity for use, e.g., in electrolysis and by providing process heat, e.g., for conversion
and upgrading fossil fuels. The former, electricity production, is a mature technology well
established worldwide. The latter, however, process heat provision, for which the HTGR
represents the ideal option, requires further research on plant components, material pro-
blems, as well as safety of the system permitting risk assessment for a whole industrial
nuclear process heat complex, in particular, with a view to a large-scale hydrogen economy.
The limited efforts on the further development of a new energy technology as has
been observed over the past decade is due to the fact that the price level for conventional
energies is currently relatively low with even falling tendency, also the energy reserve
situation is no concern, and no change is expected for the next years.
On the other hand, upheavals in the international political and economic scenery
- collapse of the socialistic world, economic rise of the Asian countries, beginning
crisis of the world economy - have occurred at the expenses of enhanced activities
in the protection of environment and climate. Furthermore an accelerated international
entanglement (globalization) has taken place which leaves national systems in a structural
and efficiency crisis. Within the tight international competition, energy costs have become
an increasingly more important issue reducing the chance for new technologies with benefits
in the long run. Environmental and climate aspects seem to lose societal attention lowering
the pressure on the political actors by a worried population; a climate change is too far away
to be perceived as a threat. The present goal of energy politics is the achievement of a high
level of congruence between supply security, economy, and environmental compatibility
meaning an optimized, not necessarily minimized, energy consumption.
After the Kyoto summit in 1997, the reduction of greenhouse gas emission is consi-
dered just one among several strategies. There is hope that by means of the new political
instruments (tradeable emission certificates, joint projects between industrial and develo-
ping countries, and others) of the climate protocoll, the political actors will proceed slowly,
299
but steadily to take internationally sound actions and direct the powerful transnational eco-
nomy and finance markets towards a low-emission economy and acquire acceptance and
the readiness of accepting the expenses by those who are concerned.
The keywords that characterize the strategy for a transition into the world's future
energy economy are: conservation, substitution, and diversification strategy, necessary
to reduce the burning of fossil fuels and to improve efficiencies, to replace them by
environmentally benign energy carriers, and to avoid new and too strong dependencies.
Of course, hydrogen as an energy carrier will be just one contender on the energy
market competing with others, e.g., electricity, and at the present stage facing economic
problems. The key issues, however, that are able to promote an innovative energy
technology based on hydrogen are its environmental benefit in a world with still increasing
demand for energy, its safety compared with other fuels, and, in the long run, its potential of
mitigating the jeopardy by a critical energy supply situation in future. Also nuclear power
represents but one contender among the non-fossil energy carriers, although providing
an environmentally benign primary energy source which could be used for hydrogen
production. Its major problem is its lack of public acceptance.
Various activities on a larger scale have been described in the report representing
important steps in the process of solving the world's energy problem. Highly ambitious
projects are currently being pursued with the "World Energy Network" in terms of hydrogen
as the future principal energy carrier and with the process heat concept of the HTTR reactor
in terms of nuclear as a primary energy provider.
The overall aim of all activities is an exchange of information among all interested
parties. The goal could be achieved by establishing international working groups to cover
the critical aspects:
• Introduction of "advanced systems", e.g., fuel cells in the conventional field and
HTGRs in the nuclear field;
• Definition of a sound safety and risk assessment methodology of non-nuclear
technologies including experiments on hydrogen safety;
• Feasibility of the combination of nuclear power plant / chemical plant;
• Analysis of the nuclear hydrogen production potential;
• Definition of fall-back positions with established systems.
Some of these activities are already underway, e.g., in an IAEA Co-ordinated Research
Program on Design and Evaluation of Heat Utilization Systems for the HTTR (CRP-4),
which is scheduled from 1994 to 1999. International support of these activities is essential.
300
Chapter 12
SUMMARY
In 1996, the world consumption of primary energy amounted to 8380 million TOE,
where 90 % originated from fossil fuels and the remainder from nuclear power and
renewable energies. Currently hydro (20 %) and nuclear (17 %) are the only significant
non-fossil contributors to electricity production. Burning hydrocarbons means liberation of
carbon oxide gases with undesirable side effects: CO2 is currently labeled as one of the
primary causes of the greenhouse effect. After the meetings of the world climate convention
in Toronto (1988) and Rio de Janeiro (1992), the third world summit in Kyoto (1997)
resulted in a commitment by all industrialized countries to reduce their overall emissions
of greenhouse gases by 5.2 % compared with 1990 and 1995 by the year 2008/2012,
respectively, an agreement which is for the first time obligatory according to international
law.
The second aspect of hydrocarbons is their limited supply on earth. The future energy
supply mix will depend on the magnitude of demand growth, changing objectives and social
priorities, and the technologies to meet them. A strong demand for more energy (plus 1.6
%/yr) is anticipated. A balanced and stable program without pursuing extreme policies has
the best prospects of achieving the lowest long-term social cost, therefore the future should
be faced with an array of options and flexible strategies.
Limited stocks of fossil energy resources and pending environmental issues are likely
to increase the significance of non-fossil energy forms - water, nuclear, solar, wind,
geothermal - in the long run. Nuclear power is considered by many to be the most
promising CCVfree energy technology which has helped to save a significant additional
CC>2 impact on the atmosphere. Principally used as base load power plants, nuclear off-
peak electricity could be used for hydrogen production. Nuclear power in the "Combined
Heat and Power" (CHP) mode can help raise energy conversion efficiencies. Hydrogen
as an additional secondary energy carrier for the non-electric market is the driving force
for its introduction as future energy carrier. The generation of hydrogen using nuclear
energy contributes to fundamental objectives of energy policy: energy security, saving of
resources, and protection of the environment. In a future low-emission energy economy, the
energy carrier hydrogen could become a strong contender using its advantages compared
with other alternatives.
NUCLEAR POWER AND ITS POTENTIAL FOR HYDROGEN GENERATION
More than half of the world primary energy consumption is used as hot water, steam
and space heat. Unlike fossil energy carriers, nuclear power is almost exclusively used for
commercial electricity production. Clearly nuclear heat production could play a major
and important role in the non-electric sector. Nuclear growth depends on the future
public perception of the comparative benefits, cost, and risks of alternatives. Based on
the comparative evaluation of tangible risks to public health, safety, and environment,
nuclear power appears to be a better choice than fossil fuels.
301
Potential of Nuclear Energy for Combined Heat and Power Production
With respect to a large-scale hydrogen production, nuclear power can play a significant
role if used as a provider of electricity in the electrolysis process or as a provider of high-
temperature heat in fossil fuel conversion or in thermochemical cycles. An introductory
option would be the use of cheap surplus or offpeak electricity. For the introduction of
nuclear process heat into the energy market, the estimation of present and future industrial
demand for process steam needs to be made first, as well as establishing how nuclear power
fits into the energy needs profile. Temperature requirements range from low level (e.g., hot
water, district heating, desalination) to a high level of up to 1000 °C (e.g., petrochemical
industry, water splitting).
With regard to the heat market, a light-water reactor (LWR) of the 1200 - 1400
MW(e) class has the potential to provide 10,000 GJ/h of low-temperature heat (« 320
°C) in the CHP mode. Since hot water and process steam cannot be transported directly
over long distances, nuclear power can be economically used only in areas with large
heat consumption density like chemical industrial complexes, or in long-distance heating
systems where some of the candidate chemical energy transmission systems can be operated
at lower temperatures.
The concept of the high-temperature gas-cooled reactor (HTGR) has the important
features of high-temperature heat, low-temperature heat, and electricity. Cogeneration
applications are estimated to allow efficiencies in the range of 80 - 90 %. Steam applications
include the generation of electricity in the steam cycle, the production of process heat in
cogeneration with electricity, and the production of district heat. Using energy generating
systems in the CHP mode, steam and electricity production in an HTGR or a conventional
plant can be accomplished consecutively by routing the steam first through a turbine and
then to the steam application system. Steam on a higher temperature level offers the chance
to open up new markets for process steam applications in the chemical and petrochemical
industries or for enhanced oil recovery. In heat applications, heat is delivered from the
helium coolant with an outlet temperature at the high end of 750 - 950 °C, for the conversion
of solid and liquid fossil fuels into "clean" gaseous and liquid fuels such as synthesis gas,
hydrogen, or methanol.
The HTGR allows the nuclear heat to be utilized for methane reforming and at
the same time for steam generation. According to the selected design, the shares of
synthetic gas and electricity production are variable over a wide range adjusted by the
steam generator inlet temperature. Unlike conventional fossil-fueled reformer tubes, the
helium-heated reforming plant connected to an HTGR has to meet the much more stringent
requirements of a "nuclear" component in terms of construction, quality assurance, and
scheduled checking. In process heat reactor designs with no intermediate heat exchangers,
the reforming tube placed in the primary circuit has the function of forming a radioactivity
barrier between the primary helium and the process gas. Another major difference between
nuclear and conventional steam reforming is the manner of heat input, which is convection
transport with the helium in the nuclear version and radiation in the conventional version.
The aim of reaching a heat flux density similar to that of the conventional method can
be achieve, -y employing a helium-heated counterflow heat exchanger. Helium under
pressure shows excellent heat transfer properties.
302
Since heat input for the coal gasification process is a significant cost factor, nuclear
process heat provided by an HTGR was considered as a substitute for conventional firing.
Another important advantage of nuclear coal gasification is its environmental effect since no
coal is fired to provide the reaction heat. The main problem, however, is the need to transfer
the heat via heat exchangers to the coal gasification process. In pilot facilities, both steam-
coal gasification and hydro-gasification have been demonstrated to work successfully under
"nuclear" conditions.
The process of hydrogen or methanol production by using a raw iron producing blast
furnace also for stack gas utilization, can be realized with a process heat HTGR. A concept
for the "non-integrated" employment of an HTGR in the iron and steel industry has been
proposed. The nuclear reactor would be sited at a location favorable for the reduction gas
generation from which the gas is delivered via a pipeline to the iron and steel industrial
sites. A process for the conversion of biomass into methanol has been proposed, where
wood is gasified by superheated steam. Electrolytic hydrogen is added to the product gas
and then converted to methanol or methane. Both heat and electricity could be provided
by an HTGR.
Also the high-temperature heat from HTGRs could be used in the water splitting pro-
cesses to convert heat energy directly into chemical energy. Three processes are appropriate
for application: high-temperature electrolysis, thermochemical cycle, and thermochemical-
electric hybrid cycle. The basic design for nuclear power for thermochemical-electric
hybrid processes is based on an HTGR to cogenerate low-temperature energy to be used
in the steam generator and high-temperature energy to be used in the steam reformer. The
heat required for this reaction is taken either from the steam generator or from the previous
exothermal reaction.
Chemical energy transmission systems could be most effectively utilized when inte-
grated with a parallel electric transmission system, taking the high-end temperature range
for driving the heat pipe and the low-end for generating electricity.
Nuclear Components
Nuclear reformer tube heating with a high-temperature reactor is performed with

helium, typically at 950 °C, as the heat source. A counterflow scheme allows the use of
internal return pipes for the product gas. Experience in construction and operation was
gained with the EVA-I and EVA-II facilities at the Research Center Julich. The perceived
disadvantages of a nuclear steam reformer with its comparatively low heat transfer and its
high system pressure can be overcome by design optimization to increase heat input into
the process gas and its conversion rate.
An intermediate heat exchanger (IHX) circuit separates the nuclear from the chemical
system. It thus minimizes contamination of the process heat exchanger and allows it to
be placed outside the reactor containment, prevents water and process gas ingress into
the nuclear core in case of a tube rupture, and reduces hydrogen permeation from the
process gas into the primary helium as well as tritium permeation in the reverse direction.
Drawbacks of an IHX are its technical problems due to the higher temperature compared
with the reformer, its additional electricity demand of 50 MW(e) in a 3000 MW(th) process
303
heat HTGR, and the fact that it presents an additional component which may fail. The
construction of the IHX for the Japanese HTTR was successfully finished and installed in
the reactor building in 1995.
SAFETY CONSIDERATIONS CONCERNING NUCLEAR POWER FOR

PROCESS HEAT AND/OR HYDROGEN PRODUCTION
Safety Design of Nuclear Heat Utilization Systems
After the severe accidents at the Three Mile Island and Chernobyl nuclear power
stations, new designs with improved safety features have been proposed focusing on more
intensive consideration of passive safety characteristics. A much more far-reaching demand
for the introduction of innovative nuclear power plants in the future is made on a design
such that fission product release is made impossible or, at least, restricted to the plant itself.
One possibility of assuring the restriction of radioactivity to the primary circuit is

the deployment of an intermediate loop as an additional safety barrier to the system. In
such an arrangement, all heat consuming components are physically separated from the
primary circuit by means of an intermediate heat exchanger. In the case of heat consuming
components to be arranged in the primary circuit, they have to act themselves as a safety
barrier.
The deposition of fission products on primary circuit surfaces and, in particular, on the
reformer tube walls causes difficulties during maintenance and catalyst refilling procedures.
Cesium and silver isotopes are of major concern. Numerous experimental and theoretical
efforts have been made to examine the plateout distribution of fission products in the primary
circuit both under normal operating and accident conditions. Experimental data obtained
from in-pile and out-of-pile deposition loops as well as from operating HTGRs were taken
to derive plateout parameters and to validate the corresponding calculation models.
In nuclear process heat plants, hydrogen and tritium exhibit high mobility at 950 °C
causing permeation through the walls of the heat exchanging components. Tritium produced
in the primary circuit could permeate into the product gas and cause a radiation problem to
the consumer. In the opposite direction, hydrogen from the secondary circuit could diffuse
into the primary circuit to cause corrosion of the fuel elements and of the graphite liner in
the core with the formation of methane. Possibilities of reducing the permeation streams
are the formation of oxide layers or, as active measures, an improved design of the gas
purification system or the construction of an intermediate circuit between the primary and
secondary circuit to serve as a sink for both hydrogen and tritium.
Tests in the EVA-n plant have shown that its operation at the same load is possible
even if up to 30 % of the reforming tubes were blocked, meaning that isolation of single
tubes would not disturb the operation thus increasing the plant's flexibility. Analytical
studies have been conducted at JAERI to investigate the steam generator as a passive heat
sink in case of a failure of the heat removal system of the reformer to ensure its coolability.
The system pressure in the gasifier is designed to be somewhat higher compared

with the pressure in the primary circuit, meaning that in case of a ruptured helium line no
radioactivity could escape from the primary system, but product gas and steam as well as
304
coal particles could enter the primary helium circuit. Leakage will be detected by humidity
sensors, helium lines shut off from the process gas circuit. The basic consequences of the
steam entrainment of a steam producing HTGR are equivalent to those of a water ingress
accident into the core.
Gas Explosion Research
Any process heat plant design implies piping through the containment to connect
the reactor vessel with the chemical plant. The fracture of a pipe could result in the
accumulation of a flammable gas mixture in the containment. Precautions must be taken
to minimize the risk of a fire or gas explosion such as avoidance of explosive gas
ingress, proper detection devices, sufficient safety distances, inerting, appropriate layout
of secondary coolant boundary, explosion-proofed wall, plant isolation valve.
The most important hydrogen source in an LWR core meltdown accident is the fuel
element cladding/water reaction at temperatures > 1000 °C. Steam and hydrogen will be
released simultaneously from the reactor vessel into the containment. The most likely mode
of H2 combustion in the containment is a deflagration as was observed in the TMI accident.
Intercell flow distribution and local steam condensation rates have a significant influence on
the local gas fractions in each compartment. A rapid condensation if possible may convert
an inert atmosphere into a detonatable mixture, which would threaten the containment
integrity and the survival of safety equipment. Assuming a 75 % cladding/water reaction,
the potential of DDT is practically present in all nuclear containments. A radiological
problem arises with the possible extensive formation of aerosols carrying Csl which
decomposes during H2 combustion and liberates volatile iodine.
Hydrogen-air-steam combustion experiments have been conducted to simulate local

detonations in a nuclear containment and to validate the respective numerical detonation co-
des. Experimental facilities used were the Battelle model containment and the containment
of the decommissioned HDR in Germany as well as one-compartment vessels at Hanford
and Nevada Test Site in the USA or the RUT facility in Russia. Many codes have been
developed to simulate the transport, convective mixing, and combustion of hydrogen in a
containment building. State-of-the-art are CFD models or lumped parameter models with
successful reproduction of, e.g., thermal stratification phenomena.
Research programs on gas explosions outside the nuclear containment have been
initiated in various countries focusing on specific aspects of the offsite explosion and the
consequences of flammable gas clouds, in particular, the impact of the pressure wave on
the nuclear containment building. Evaluation of the status of knowledge on hydrogen's
and other fuels' combustion behavior within the German LWR safety program has shown
that existing safety margins of modem nuclear containments guarantee a safe enclosure.
A variety of methods can be employed as hydrogen mitigation systems. Mixing de-

vices and venting systems can prevent locally high HI concentrations in the containment
atmosphere. Inerting, which is widely applied, limits the supply of oxygen. Recombiners
allow an immediate and continuous H2 removal with no power supply required. The deli-
berate ignition of lean mixtures appears simple and easy, but it represents some challenge
to building and equipment because of the difficulty of anticipating the consequences. In
305
any case, the strategy applied should depend on the actual state of the building during and
after the accident.
The PNP gas cloud safety program focused on the formation and explosion of clouds
of the process gases H2, CO, air and other representative combustible gases such as
acetylene, ethylene, propane, methane, and the damage caused by the blast wave to simple
structural elements (window pane, brick wall) after fast deflagration and detonation.
The release of product or feed gas into the environment and the formation of a
flammable gas cloud could result in heat radiation and blast wave impact from fire /
explosion. For the German PNP design, it was proposed to ensure that the gas be routed
through inerted channels away from the gasifier and flared at a safe spot. The safety concept
of the Japanese HTTR includes the appropriate blast wave resistance design of building
structures, gas leak detection device, fire extinguishing system, emergency shutoff valve,
safety distance which could be reduced by fireproof walls.
WORLD ACTIVITIES ON NUCLEAR POWER AND ITS

POTENTIAL USE FOR HYDROGEN PRODUCTION
Two decades of enormous economic growth in the People's Republic of China have
led to a significant need for energy and electricity, and the demand for more is greatly
increasing. Chinese primary energy consumption is dominated by coal, but nuclear power
is expected to gain more importance for base load electricity, for district heating, and for
the process heat industries. Presently there are 23 hydrogen production units in operation
with a total capacity of 582,600 Nm3/h.
The HTR-10, under construction at Tsinghua University since 1995, is designed

as a test reactor for multipurpose research. In a first phase, the HTR-10 is planned to
be operated in connection with a steam turbine which works in a cogeneration mode to
provide electricity and heat. In a second phase, the coolant outlet temperature will be raised
to 900 °C to operate the HTR-10 in a gas turbine cycle. A long-term test program for
nuclear process heat applications has been proposed in different phases including simulation
experiments for nuclear coal gasification on a laboratory scale and in a pilot facility in
connection with the HTR-10. A concept was proposed exploiting very heavy oil fields
with an estimated total amount of 90 million t in the north-east of China by the use of
nuclear process steam. HTGR designs with 1000 MW(th) were shown to be able to provide
steam with the required parameters to open up these resources.
Primary energy consumption in Germany totaled 340 million TOE in 1995 with
significant contributions from oil (40 %), coal (27 %), gas (20 %), and nuclear (10 %).
The consumption of primary energy has been stagnant since the 1980s due to energy
saving and industrial restructuring while gross internal production as well as electricity
consumption have still been growing. Growth of the latter, however, has been observed
to be getting slower.
The 46 MW(th) high-temperature gas-cooled AYR reactor in Julich as a first-of-its-

kind has been in operation for 21 years with a high availability demonstrating that nuclear
power can be converted into high-temperature heat. For more than 10 years, the reactor
was operated with an average outlet temperature of 950 °C. The reactor allowed successful
306
testing of a large amount of high quality HTGR fuel elements. The follow-up pebble
bed reactor THTR-300 has demonstrated the expected large-scale applicability of HTGR
technology. The present German HTGR reference design for electricity production is the
200 MW(th) HTR-MODUL with a large number of inherent safety features; a variant with
170 MW(th) has been designed for process heat applications.
The "Prototype Plant Nuclear Process Heat" (PNP) project was founded in 1972
with the goals of developing HTGRs for high gas outlet temperatures of 950 °C as a
source of process heat to test components for heat transfer to the process plant, and
to demonstrate processes and experimental facilities for coal gasification. The technical
feasibility of a nuclear process heat reactor for the refinement of coal has been established,
main components developed, and its basic licensing capability confirmed. Studies on
the technical feasibility and economic competitiveness of the processes for nuclear coal
refinement were completed in 1987.
The steam reforming process of methane was experimentally investigated in the

"Nuclear Long-Distance Energy" (NFE) project. Two different bundles of steam reformer
tubes were tested. Convective helium heating was verified for both designs. The complete
experimental chemical energy transmission system based on steam reforming of methane,
"EVA / ADAM", was operated for more than 10,000 hours including steady state conditions
at both full and partial load, transient procedures, and also special test situations such as
tube blockage. From 1976 to 1984, catalytic and non-catalytic steam-coal gasification was
established and successfully tested in a 1.2 MW semi-technical scale experimental facility.
The maximum capacity was 230 kg/h of hard coal. The process of hydro-gasification of
lignite as an alternative was successfully verified in a pilot plant with a throughput of 9.6
t/h of lignite corresponding to a power of 50 MW.
Starting in 1972, the project of a closed-cycle "High-Temperature Reactor with

Helium Turbine of Large Power", HHT, had the main goals of obtaining a higher efficiency
and reducing the number of components. But it was also considered to have a broad variety
of utilizations, in particular, in the lower process temperature range. Research activities
included a materials research program to investigate the lifetime of the turbine, hot gas
duct, and insulation materials in a helium atmosphere, the development and testing of the
hot gas duct and insulation, and the investigation of fission product behavior in the primary
helium circuit.
Research work has been done on the application of thermochemical (hybrid) cycle
processes for the production of hydrogen by proposing an HTGR for the cogeneration
of high-temperature energy for process heat and low-temperature energy for electricity
production.
Indonesia is an oil and gas producing country with an increasing domestic consump-
tion rate. In the 1970s, several giant natural gas fields such as Natuna were discovered
making gas a major export commodity. Minimizing the emission of carbon dioxide into
the atmosphere has suggested the nuclear option in exploiting the Natuna gas field. Se-
veral alternatives have been identified as being feasible and economically competitive,
distinguished by the desired fuel product and by the consumption of CC>2 which is an
unwanted component in the Natuna gas. A feasibility study has been initiated by the IAEA
307
to investigate the application of the HTGR for cogeneration purposes of CC>2 conversion,
desalination, and H.2 production in terms of economy and technology.
The significant growth of Japan's economy in the 1960s was based on cheap petroleum
imports which, however, ceased with the first oil crisis. The largest share in Japan's
electricity production of 907* 109 kWh (1993) is provided by nuclear power with 34 %;
others are natural gas, oil, hydro, and coal. Japan depends heavily on imported energy;
therefore intensive research efforts are being conducted on new energy carriers. The
economics of the many different designs for future reactors currently being pursued are
considered promising and are expected to form Japan's next generation of reactors.
Starting in 1991, the 30 MW(th) High-Temperature Engineering Test Reactor, HTTR,

is being constructed in Oarai and is expected to start its operation in summer 1998. The
helium-cooled reactor will have an outlet temperature of 850 °C in the first stage and
will be the first of its kind with an outlet temperature of 950 °C to be later connected
to a high-temperature process heat utilization system; its demonstration is planned for the
second core. Top candidate to be tested by JAERI in connection with the HTTR is steam
reforming of natural gas for hydrogen and methanol coproduction including carbon dioxide
recycling. The goal is the production of 1390 Nm3/h of hydrogen plus 1930 kg/h of
methanol, while the input is 950 kg/h of natural gas plus 10 MW process heat, which is
introduced via the He-He intermediate heat exchanger.
A bench-scale test facility for hydrogen production using the thermochernical iodine-
sulfur process has been established at JAERI. Its feasibility was successfully demonstrated
in a lab-scale apparatus. Further demonstration steps are the employment of a metallic
reactor for hydrogen production and a scale-up process to a 100 Nm3/h rate. The
bromine-calcium-iron or UT-3 process is considered in Japan to be superior to many other
thermochernical cycles. One aspect of the main ongoing research efforts is its combination
with an HTGR for hydrogen and electricity cogeneration. Another candidate for nuclear
process heat application is hydrogen production by means of high-temperature electrolysis
of steam. This has been tested by JAERI in a bench-scale arrangement of 12 tubular-type
cells at a temperature of 850 °C. The construction of a prototype system is planned with
a capacity of 10 and later 100 Nm3/h.
The FAPIG organized a working group to study how to use heat from HTGRs to
mitigate environmental and resource problems, to stabilize supply, and to promote effective
use of energy. Three types of heat utilization plants have been modeled, for each of which
system outline, basic plant layout, social effects, and technical issues were studied: a
cogeneration plant for a typical chemical plant with an estimated overall efficiency of 74
%, a coal gasification plant to produce 1700 t/d of product gas, and a hydrogen production
plant.
In the Russian Federation, about 20 different technologies have been studied in various
industrial sectors with the possible use of an HTGR including the production of hydrogen,
synthetic fuel, or methanol, and the reprocessing of oil. The conceptual design of a 200
MW(th) modular high-temperature gas-cooled reactor, VGM, has been developed and is
intended to be used for process heat applications. To assure a high level of safety, the
308
system is designed to have a helium-to-helium IHX and a helium-to-silicon oil process
heat exchanger (PHX).
Within the United States HTGR program, a reference conceptual design of an 1170
MW(th) HTGR plant has been developed based on a prismatic core for steam / electricity
cogeneration, which is readily adaptable to various applications such as a "Sensible Energy
Storage" system based on a molten salt circuit for thermal energy storage and transport, a
gas turbine cogeneration system, or a process heat system employing the chemical reforming
process with synthetic fuel or chemical feedstock as the principal output.
A direct-cycle, modular variant, MHTGR-GT, has been designed as a second gene-

ration meltdown-proof nuclear power plant combining the experience of both gas-cooled
reactor technology and industrial gas-turbine technology. The thermal reactor power is
450 MW with a projected efficiency of 46 %. A characteristic feature of the closed-loop
Brayton cycle is its potential, besides electricity generation, for efficiently utilizing waste
heat, e.g., for desalination, district heating, or steam generation.
CONVENTIONAL AND ADVANCED HYDROGEN PRODUCTION METHODS
Decomposition of Fossil Fuels
An easy and the most widely used method of hydrogen generation is steam reforming
of natural gas, which is clean and available in large amounts. During the process, natural
gas is mixed with steam to react at high temperatures, typically > 500 °C, and relatively
low pressures in the presence of a nickel catalyst. In the homogeneous water-gas shift
reaction, the equilibrium is moved towards more C02 in order to achieve a higher H2
yield and to preclude carbon deposition due to the Boudouard reaction. The yield of the
reforming process is a mixture of H2, CO, CO2, plus residual steam and methane. The
mixture of H2 and CO left after purification is called synthesis gas.
If the steam is completely or partly replaced by carbon dioxide, the H2 to CO

ratio is shifted towards more CO. This involves an environmental benefit, because two
greenhouse gases (CHU and CO2) are combined, resulting in a product gas which might be
more favorable for certain applications, e.g., methanol. However, currently no industrial
technology for the CO2 reforming of methane is established. The reasons for this are the
rapid deactivation of conventional catalysts if used without the presence of steam, and the
relatively high soot formation.
Partial oxidation is able to convert methane and other hydrocarbons, catalyzed or non-
catalyzed, at temperatures between 1100 and 1500 °C. Despite its lower efficiency compared
with steam reforming, partial oxidation provides, on the other hand, exothermicity and a
greater selectivity for synthesis gas production as well as advantages for certain applications
such as compactness, rapid startup or load change, lower overall cost. If steam is added
to the fuel and the oxidant, it is possible to heat balance the exothermal partial oxidation
reaction with the endothermal reforming reaction, meaning that no external heat source
(autothermal) is required.
An alternative to the steam reforming of methane is single-step thermocatalytic

decomposition or thermal cracking. Methane (and other hydrocarbons) decompose at
309
temperatures between 700 and 980 °C and if no air is present to form hydrogen and
carbon. Cracking was studied for various catalysts in both continuous flow and pulse
catalytic reactors. Variations in the operational parameters were found to produce CH4-H2
mixtures with a wide range of hydrogen concentrations, from 30 to 98 vol%.
Conventional steam reforming is a mature and well-established technology practiced

on an industrial scale and currently representing the least expensive H2 production method.
Today's reformer plants for hydrogen production vary in size, from typically around 100
Nm3/h capacity up to the largest currently existing with a maximum H2 production rate
of 140,000 Nm3/h. Recent reformer improvement developments have focused on high
efficiency, good load changing capability, and low cost through use of cheaper materials
and production methods. Hydrogen purity is up to 99.99 %. Small size and suitability
on a small scale makes them beneficial for integration into a fuel cell power plant. Fuel
cell manufacturers are developing advanced miniaturized designs for vehicle or household
applications which allow the reforming process to take place inside the fuel cell stack. A
future high-temperature fuel cell application could be the production of hydrogen, where
more Kb is generated in the internal reforming step than is converted into electricity.
The conversion of coal to both liquid and gaseous hydrocarbons has been commerci-
ally deployed worldwide. Gas production from coal is realized by means of a gasification
medium which reacts with the coal at temperatures > 800 °C. All organic constituents of
the coal will be converted with sufficiently long reaction times. If air or oxygen is injected
into the gasifier, part of the coal is burnt directly leading to an autothermal reaction.
For the steam-coal gasification process, steam of 600 - 1000 °C is needed to convert
the organic solid in the heterogeneous water gas reaction. The production of hydrogen
and carbon monoxide is optimal at high temperatures and low pressures. The heat of the
hot gas is recuperated for use in the production of high-pressure steam. Coal gasification
with oxygen and steam is conventional chemical engineering and is employed to generate
about 15 % of the world hydrogen production and primarily used for ammonia synthesis
in the fertilizer industry. Different types of autothermal coal gasification used on a large
scale are represented by the Lurgi, Winkler, Koppers-Totzek processes or modifications
of them. There is also the option of implementing the steam-coal gasification process
in an Integrated Gasification Combined Cycle (IGCC) power system, which is currently
considered the cleanest and most efficient (34 —> 38 %) coal-fueled technique so far.
In hydro-gasification, coal is converted to synthetic natural gas in a fluidized bed at

temperatures of around 800 °C in an exothermal reaction. Parallel reactions are the steam
reforming and the water-gas shift reactions. In order to obtain a high conversion rate of
coal, the methane fraction in the product gas of the reforming step (which will then be the
feed gas for the hydro-gasification) should be no higher than 5 %, which requires a low-
temperature separation step. The advantage of hydro-gasification compared with steam-coal
gasification is its 200 K lower pre-heating temperature which reduces potential corrosive
attack. A major drawback, however, is the relatively large amount of residual coke of up
to 40 %. Standardized reformer equipment for power industries is available.
Other sources of hydrogen are provided by the conversion of oil or other feedstocks
such as municipal solid waste and biomass, decomposition of hydrogen sulfide, or recovery
from refinery offgases.
310
Splitting of Water
The electrolysis of water is usually applied if cheap electricity is available or high-

purity hydrogen is required. The electrolytic decomposition of water requires electric energy
which is dependent on the reaction enthalpy and the temperature. Typical cell voltages are
about 1.8 to 2.0 V, which translate into efficiencies of 60 - 70 %. The electrolysis of steam
at higher temperatures (900 - 1000 °C) reduces the electric energy required and lowers the
activation barriers at the electrolyte surfaces resulting in higher efficiencies.
Alkaline water electrolysis is the only large-scale method to be commercialized, where

the Ha is actually a byproduct of chlorine production. Solid polymer electrolyte water
electrolysis is considered to be a promising method because it can operate at high current
densities and has the advantages of cell compactness, simplicity in design and operation
as well as lack of corrosion problems. At the high end, 800 - 1000 °C, zirconia-based
ceramic membranes are used, which show a high conductivity for oxygen ions at operating
temperatures. The electricity demand at 1000 °C is 30 % lower compared with that of room
temperature. The high-temperature technology is able to create so-called regenerative cells
which can work either in the electrolyzer mode or in the fuel cell mode.
In the thermochemical method, the splitting process is subdivided into different

partial reactions, each running on a lower temperature level. A large number of possible
cycles has been tested and checked in terms of appropriate reaction temperatures and
velocities and in the economic respect. Major problems arise due to the large material
flows and the introduction of impurities. The main problem is to achieve satisfactory
overall energy efficiencies which are no higher than around 40 % and, in connection with
severe materials and equipment design difficulties, do not allow large-scale thermochemical
process applications so far. Systems investigated include metal-halide, bromine-calcium-
iron (UT-3), sulfuric acid methanol process, and iron-chlorine processes. One of the
thermochemical cycles with the most considerable potential is the iodine-sulfur (IS) process
developed by General Atomics, which has demonstrated its feasibility for many years and
is now being pursued at JAERI.
Thermochemical hybrid cycles offer the possibility of running low-temperature reac-

tions on electricity. The expectations for realization of hybrid processes are similar to those
for purely thermochemical processes. Various hybrid processes are energetically possible,
but not always practicable. Important criteria are the minimum voltage for the electrolysis
step, readability, efficiency. The sulfuric acid hybrid or Westinghouse process is expected
to reduce material streams as compared with the IS process. The electric energy demand
here is a factor of 7 (in reality 3 - 4 ) lower than in the electrolytic water splitting process.
Technological improvements are still possible.
Photolytic processes use light energy to promote the water splitting reaction. It is of
great interest because solar energy can provide an unlimited energy source for hydrogen
production. In photoelectrochemical systems, inorganic materials are being used for direct
conversion of (sun) light into hydrogen at a semiconductor electrode. In photobiological
systems, hydrogen is produced as a result of the way certain biological organisms use
311
sunlight. Some photosynthetic algae and bacteria absorb energy from sunlight and use it
for water splitting in the process of generating plant tissue or creating heavier hydrocarbons
or oxygen for assimilation. The commercialization of the photosynthetic processes requires
a drastic increase of the conversion efficiency as well as an improvement and upgrading of
the microorganisms. Anaerobic bacteria could be used for pre-treatment of waste water to
degrade organic substances that the photosynthetic bacteria are not able to handle.
If hydrogen production processes are compared, the interdependence of efficiency,

capital investment and value of the byproduct has to be taken into account. Reforming of
natural gas and partial oxidation of heavy oils are presently the least expensive and most
frequently applied methods of hydrogen production. Electrolytic hydrogen has a small, but
stable market; it is preferred if there is a fluctuating demand, a need for small amounts
or the requirement for high-purity hydrogen. Since coal is abundantly available in the
world, R&D efforts should concentrate on coal-based processes; the advanced processes of
methane cracking and coal cracking (HYDROCARB) have a huge economic potential.
Liquefaction of Hydrogen
The minimum (theoretical) work required for the liquefaction of hydrogen is 3.92
kWh/kg; a realistic value is about 10 kWh/kg or approx. 30 % of the H2 net heat of
combustion. In contrast to the gases in air, the production and liquefaction of hydrogen
is conducted in two separate energy-intensive steps. Conventional processes used are the
Linde-Hampson process and the Claude process. The magnetic refrigeration method looks
promising, because it offers a compact cooling system with an extremely long lifetime,
lower capital investment and it promises higher efficiencies. The liquefaction plants in
the world presently produce liquid hydrogen (LHi) with a total capacity of 265 t/d but
not more than 19 t/d in Europe. Based on current experience, the construction of plants
with a capacity of up to 250 t/d is expected to be feasible, the goal in Japan's WE-NET
project is even 600 t/d.
STORAGE AND TRANSPORT OF HYDROGEN
The commercial provision of energy worldwide has turned into a very complex
structure, not limited by national boundaries. Hydrogen storage and transport technologies
have to be developed to fit these structures. Relevant issues and R&D activities are
summarized below.
Storage of Hydrogen
Gaseous hydrogen at ambient temperature can be compressed in high pressure tanks

which are easy to handle. Pressures range from < 1 MPa to around 30 MPa. Pressure
vessels must be periodically tested and inspected to ensure their safety. A classical storage
type is no-vessel underground mass storage used in seasonal peak consumption periods.
Geological formations with porous permeable layers underneath a dense layer are chosen,
which can be either an aquifer or a depleted natural gas resource or a specifically formed
cavern in a salt mine. Typical ratios of the capacities of a 100 km high-pressure pipeline
section to cavern storage to porous storage are 70 to 1300 to 20,000.
312
Liquid hydrogen must be stored in a superinsulated tank and even then depending on
the insulation quality and the surface-to-volume ratio a certain fraction is an unavoidable
boiloff to keep the rest cold. A supporting structure and interconnections between the
fluid container and the outer vessel have to assure that the inner shell can freely undergo
contraction / expansion. A multilayer insulation and a vacuum jacket inside help minimize
the transport of heat into the cryogen. The largest LH2 tank constructed so far is the
NASA 3407 m3 vacuum perlite-insulated spherical storage tank used in the US space
shuttle program. The tank has a boiloff rate of 0.03 % per day. Improved concepts of
insulations have been developed for future large-size LH2 storage tanks with a volume of
about 24,400 m3. Projected tanks for large-scale LH2 storage within the WE-NET project
have a storage capacity of 50,000 m3. Target boiloff is 0.1 % per day.
The safety of LH2 storage tanks has been investigated in extensive experiments.
Passenger car tanks have been exposed to deformation and perforation, as well as fire and
burst testing to check the performance in conceivable traffic accidents. Tests with a 61 m3
container representative of maritime transportation were conducted, the first with an LH2
tank of that size, to investigate the thermodynamic behavior of the LH2 in the tank.
Metal hydride is an intermetallic compound that traps hydrogen at moderate pressures.

It stores large amounts of hydrogen in an exothermal diffusional process. Some drawbacks
are its cost and its heavy weight, and its requirement for very pure hydrogen. Metal hydride
advantages include convenience, compactness, stable storage, and intrinsic safety.
New classes of hydrogen storage are nanocrystalline metal hydrides, which are
polycrystals with a typical grain size < 20 nm, or high-strength glass microspheres with 25 -
500 fim in diameter and with 1 fim wall thickness, which are loaded in a high pressure bath
of hydrogen. After cooling the hydrogen is trapped. Engineered microspheres could make
storage and transport of large amounts of hydrogen much more efficient. A qualitative step
forward appears to be the discovery of graphite fibers which were reported to store 30 1
of H2 per gram; it would allow a cruising range of 8000 km for a H2 fueled car. The
tremendous storage capability, however, is still unexplained and needs further verification.
Transport of Hydrogen
In most countries, pipeline systems for natural gas serve both transportation and
distribution purposes. The transport of hydrogen in the same energy quantity would require
a 3.3 times higher compression because of its lower heat of combustion per unit volume.
Transport of gaseous hydrogen in a pipeline system is a technology which has long been
applied. More than 750 km of hydrogen pipelines (1990) have been put in place worldwide.
Transportation cost is expected to be in the same order as natural gas for regional distribution
and to be higher for large-scale transportation.
Vehicle-bound distribution systems for both gaseous and liquid hydrogen encompass
trucks, rail cars and ships. A tank ship transport of LH2 is reasonable if transported
over long distances. The comparatively low volumetric energy density of LE^, however,
requires very large tank ships and a much higher fraction of the hydrogen transported is
needed for the transport itself.
313
Chemical Energy Transmission Systems
The key components of a chemical energy transmission system or a chemical heat

pipe are a primary energy source to provide heat, an input catalyst reactor where in an
endothermal reaction energy is stored in newly created products, and an output catalyst
converter, where the reverse exothermal chemical reaction helps to extract the stored heat
for further consumption.
The energy transport system by steam reforming / methanation, also known as

EVA/ADAM, was theoretically and experimentally investigated on a representative scale
and demonstrated its technical performance and efficiency as a thermochemical pipeline
system. A complete EVA/ADAM system was successfully operated in Germany under
the realistic conditions of an industrial application with a transported power of 5.4 MW.
Hydrogen evolution from cyclohexanes at the demand site and the reverse reaction at
the supply site offers the advantages of safe and commonly used transport at ambient
temperature in chemical tank ships or trucks and mild reaction conditions. The advantage
of ammonia dissociation is the complete absence of any side reactions. The fact that
ammonia condenses at around room temperature and separates from the reactant mixture
allows storage of feedstock for both endothermal and exothermal reaction in the same
vessel.
Chemical heat pump systems are important if the utilization of low-quality heats («
80 °C), e.g., solar or geothermal or nuclear heat, waste heat in factories, is considered for
heat storage and temperature upgrading.
APPLICATIONS OF HYDROGEN
Hydrogen has many and versatile uses: It plays a significant role as a chemical
feedstock for industrial and petrochemical processes, e.g., in the conversion of coal into
clean synthetic fuels or in the generation of process heat. Furthermore it possesses a huge
potential as a direct fuel for producing mechanical energy, heat, or electricity for a future
large-scale use.
Hydrogen as Raw Material
About 40 % of the hydrogen produced worldwide is consumed in ammonia synthesis.

Feed gas for the steam reformer is methane or gasoline. Adding air in a secondary reformer
leads to partial oxidation of the residual methane and of the CO. After separation of the
CO2, the product gas is a mixture of nitrogen and hydrogen whose ratio (desired 1:3) is
adjusted by the operating conditions. NHs production on an industrial scale allows a daily
output of 1000 - 2000 t consuming 80,000 - 160,000 Nm3/h of H2.
In the conventional method for the generation of methanol from synthesis gas, a
mixture of CO, CO2, and H2 is compressed and introduced into a fixed-bed catalytic
reactor. The reactions are exothermal and volume-reducing, thus low temperatures and
high overpressures are desirable. A catalyst is required to maximize methanol output.
Methanol production generates a surplus of hydrogen which can, by adding CO2, be utilized
to increase the methanol yield. Methanol is basically used in the chemical industry as
314
intermediate product. It is gaining further attention as secondary energy carrier with less
emission, e.g., as a vehicle fuel.
Steady utilization of hydrogen in refineries commenced more than 50 years ago. It is

mainly used for hydro-cracking and hydro-treating in upgrading processes. In the catalytic
hydrogenation process, hydrogen is used to convert coal to gasoline and to remove sulfur,
oxygen, and nitrogen. Rigid specifications concerning the contents of the catalyst poison
CO have to be met. A purity of 98 % hydrogen is desirable. The Fischer-Tropsch synthesis
is the catalytic hydrogenation of CO to form hydrocarbons to be used as raw material in the
petrochemical industry. The high purity of vaporized LH2 is utilized in the semiconductor
industry.
Hydrogen as a Fuel
Liquid hydrogen was early recognized to be an important rocket and aircraft fuel. Air-
breathing engines with supersonic combustion, scramjet, offer an excellent thrust potential.
With its high combustion heat per mass unit, fast ignition and high heat sink capacity,
hydrogen is a perfect candidate, in particular, for supersonic applications. In addition, it
avoids all pollutants of fossil fuels except for NOX. The main penalty is the storage volume
of the hydrogen tanks which is larger than kerosene by a factor of about four. Hydrogen
in space transportation is being used by the USA, Russia, Europe, China, Japan, and India.
Internal combustion engines that run on hydrogen are very efficient, because they
take advantage of the fast-burn and far-lean combustion characteristics of hydrogen. Since
the 1970s, many automobile companies have designed and operated prototypes of cars or
trucks. It was shown that the most practicable solution for onboard H2 storage is liquid
hydrogen.
Catalytic combustion of hydrogen exhibits many advantages in comparison with flame

combustion. It occurs at low temperatures, is safe and leads to a very high conversion of
the burning gas (99.9 %). The NOX formation as present in conventional combustion at
temperatures of about 1700 K is almost completely suppressed.
The fuel cell is an energy conversion device that works in reverse of electrolysis. In
the cell, the energy of a fuel is directly converted in an electrochemical reaction into D.C.
electricity with an approx. 60 - 70 % efficiency. The principal advantages of a fuel cell in
comparison with the indirect energy conversion (which goes via heat first) are its cleanness
and its high efficiency, which is nearly constant independent of size and operating range.
Thus configurations of power plants could range from kilowatts to hundreds of megawatts.
Heat from the electrochemical reaction is available for cogeneration application.
The low-temperature polymer electrolyte and alkaline fuel cells are an efficient,
compact, robust, and quiet method of generating electricity. They have the highest power
density of all cell types, good load change behavior and good efficiency at partial load,
thus being ideally appropriate for mobile applications. The indirect methanol fuel cell
operates on liquid or gaseous methanol combined with a reforming process and offers a
good storage capability in mobile systems. Most practical experience has been gained in
the operation of phosporic acid fuel cells (PAFC). The molten carbonate fuel cell (MCFC)
can be operated with hydrogen, methanol, methane or coal gas with external or with partial
315
or full internal reformation. The solid oxide fuel cell (SOFC) is attractive because of
its very high temperature which allows fast chemical reactions, but on the other hand
implies stringent requirements for materials development and construction. SOFC converts
gaseous hydrocarbons either directly or after internal reforming with a very low emission
level. Stationary applications of fuel cells embrace large-scale central power generation,
distributed generation, and cogeneration. The potential of fuel cell power plants is strongly
related to the global demand for new power and heat capacity. A cogeneration system with
waste heat recovery can achieve efficiencies of up to 80 %. PAFC, MCFC, and SOFC are
currently the strongest contenders for commercial stationary power generation.
Hydrogen is an ideal fuel for gas turbines. Due to its rapid mixing with air, a smaller
combustion chamber is sufficient and the efficiency is higher compared with conventional
fuels. Gas turbines modified for liquid hydrogen operation yield an up to 10 % higher
thermal efficiency and output compared with fossil-fueled turbines. For systems with
advanced heat exchange, efficiencies of more than 50 % are estimated to be achievable. The
combustion is stable, largely free of pollutants, and favorable for lifetime and maintenance.
Of disadvantage is NOX production at high flame temperatures. The H2-O2 steam generator
as a novel power plant component is able to provide instantaneously spinning reserve
capacity upon demand. Combustion heat is used to vaporize feed water where the steam
is routed to the intermediate pressure steam turbine.
SAFETY RISKS OF A LARGE-SCALE HYDROGEN APPLICATION
Hydrogen gas is colorless, odorless, tasteless, non-toxic, and undetectable for human
senses. If released in a confined area, hydrogen can cause suffocation by dilution of
the oxygen content. Mixtures of hydrogen with air are flammable over a wide range of
concentrations. Explosion of the mixture can occur with very low impact. The hydrogen
flame is nonluminous, comparatively hot, but hardly radiates any heat. The investigation
of the combustion behavior of hydrogen connected with the pressure development under
accident conditions is the basis of all safety-related considerations.
The broad introduction of hydrogen as an energy carrier requires a concomitant and

prognostic evaluation of its safety and risk. The inadvertant release of hydrogen caused
by a leak or spillage and its mixing with air creates explosion hazards making mandatory
the assessment of conceivable risks and its hazard potential after an inadvertant release.
Systematic prognosis procedures are applied in probabilistic safety and risk analyses. The
tools for the deterministic analysis are calculation models to simulate all the steps of an
accident scenario as realistically as possible. Model validation is based on the evaluation
of experiments and postexamination of previous accidents.
NASA experience with hydrogen began in the 1950s, when hydrogen was considered
the principal liquid rocket fuel. Safety engineering soon started to address hydrogen hazards
and to develop procedures for the safe operation of equipment and facilities. The large-scale
industrial application of hydrogen has led to a safety concept with a scale of safety measures
aimed at minimizing risks by preventive safety measures and by mitigation measures.
Hydrogen has long been recognized to have a deleterious effect on metals by changing
their physical properties. Depending on the type of appearance of this phenomenon, named
316
embrittlement, a distinction is made between hydrogen reaction embrittlement, internal
hydrogen embrittlement, and environmental hydrogen embrittlement. The effect is strongly
connected with locally high hydrogen concentrations which can be caused by stress-
enhanced diffusion rates to lattice defects and reaction sites to initiate cracks. Despite
of the fact that some metals can become brittle due to hydrogen or hydrogen compounds,
there is a good safety record which is explained by possibly conservative assumptions for
the design stress and a scheduled retesting of the vessels.
The evaluation of reports and statistics of accidents with hydrogen has revealed a
number of accidents in the order of a few cases per year. The Hindenburg accident in 1937
at Lakehurst is very familiar to the public. However, the airship did not explode, it tragically
caught fire due to a highly combustible skin coating used at that time. The accident with
the Challenger Space Shuttle was presumably caused by so-called "phantom" fires resulting
from undetected small cryogenic hydrogen leakages from the external tank and their early
ignition by the anomalous plume from the rocket booster. The spectacular accident in
Hanau in 1991 demonstrated the consequences to be expected from the catastrophic failure
of a tank with pressurized hydrogen. The explosion resulted from a fatigue rupture near
a longitudinal welding seam.
Although hydrogen has a reputation as a particularly dangerous fuel, the overall

statement was made that the technology of handling hydrogen as a high-pressure gas or as
a liquid is successful and safe, primarily based on the existence of well-trained personnel
and extreme concern about system reliability. Measures for accident prevention are the
appropriate choice of materials, coatings, adsorption layers, the deployment of stainless
steel liners in composite vessels and piping, and proper welding procedures.
Liquid Hydrogen Pool Spreading and Atmospheric Dispersion of Hydrogen Gas Clouds
Starting from large-scale liquid fuel (LNG, LPG) maritime transportation and appli-
cation in the 1970s, interest increased in the assessment of the consequences of cryogen
spill accidents. If hydrogen is released in its liquid form, it can accumulate and form a
cryogenic pool on the ground and immediately starts to vaporize. A cutoff of the mass
input results in a breakup of the pool from the central release point creating an inner pool
front. The ring-shaped pool then recedes from both sides until it has completely died away.
LH2 vaporization is determined by heat transport into the pool. Of all contributions, heat
transfer by conduction from the "hot" ground plays the dominant role. The most detailed
investigation of LH2 pool spreading so far was conducted by the Research Center Jiilich
during a series of LH2 spill experiments in 1994, where the LH2 pool size on water and
on a solid ground was measured as a function of time and location.
State-of-the-art modeling of cryogenic pool spreading and vaporization is based on the

solution of the so-called shallow-layer differential equations. This allows the tracking of the
dynamic behavior of pool growth and recession and calculates pool height and velocity as a
function of time and location. Respective calculation models with successful performance
have been implemented in the British GASP code and in the German LAUV code.
The accidental release of (gaseous or liquid) hydrogen leads to the evolution and
dispersion of a gas cloud whose shape is influenced by type and rate of release and by
317
the atmospheric conditions as well as topography. The dilution with ambient air makes
the hydrogen cloud pass through the whole broad flammable range from high to low
concentrations. Condensation of the always available humidity in the ambient air, which
makes the hydrogen cloud visible, reduces the cloud density and liberates latent heat,
both effects supporting the positive buoyancy. Release types are distinguished according to
vaporization of a liquid hydrogen pool on the ground, where the hydrogen escapes upwards,
away from the site of release, a jet stream released through a small opening in a pressurized
system, where the cryogen is widely sprayed in a multi-phase mixture, and the catastrophic
failure of a pressure vessel with the explosion-like expansion of the stored gas.
Little experimental work has been conducted in the past to study the atmospheric
dispersion behavior of a hydrogen gas cloud. The NASA series of seven LH2 spill tests
demonstrated that the evolving H2 cloud was existent for a couple of 100 m downwind,
especially if the ground was able to cool sufficiently. Also the completely different
dispersion behavior of the "light" hydrogen and the "heavy" propane was shown.
Only a few theoretical efforts have been made to simulate the dispersion of hydrogen
gas since there is scanty experimental data. The specification of the source term is of
particular importance for the accurate modeling of gas cloud evolution and dispersion.
State-of-the-art modeling of the transient behavior of gases with positive or negative
buoyancy in the atmosphere is provided by computer fluid dynamics (CFD) models which
simulate complex flow processes by solving the Navier-Stokes equations. Besides their
need for immense computer time and storage capacity, which can only be realized by the
most advanced computer generation, these models require a detailed input of initial and
boundary conditions for the calculation grid. In the two-equation k-e turbulence model,
special partial differential equations are solved to describe the transport of turbulence as
well as its generation and dissipation. Of all the approaches, the k-e model offers the highest
relative independence of empirical relations. K-e modeling has been realized in a variety
of computer codes. Representatives of this category of codes are the German BASSIM,
the JRC-Ispra code ADREA-HF, or the commercially available British CFX code.
Combustion Behavior of Hydrogen
If liquid hydrogen is released in ambient air with a high humidity content, the
flammable fraction of the gas cloud is generally enclosed in the visible cloud, while under
conditions of a dry atmosphere, flammable fractions could also exist outside the visible
cloud. The ignition of a gas cloud always requires an ignition source. Under normal
circumstances, a spreading cloud is always ignited at its perimeter. According to accident
statistics, the most probable reasons for the ignition of hydrogen gas clouds are autoignition
on hot surfaces, an electric spark, or an electric discharge. But only a small portion of
the energy of a hydrogen cloud is expected to be liberated in an open air explosion; it is
estimated to be in the range of 0.1 - 10 %, mostly < 1 %.
The "Unconfined Vapor Cloud Explosion" is the explosive type of deflagration where
about 60 % of the combustion energy is expressed as kinetic energy in the form of
pressure waves. Heat transport by conduction and diffusion from the reaction zone into the
preheating zone enables the propagation of the flame. A near-ground deflagration results
in a hemispherical fireball. Depleting fuel terminates the radial spreading and the buoyant
318
forces allow for a liftoff of the fireball resulting in the typical mushroom shape. A "Boiling
Liquid Expanding Vapor Explosion", BLEVE, characterizes the failure mechanism of a
liquid gas container where the major part of the contents is released by flash vaporization
and explodes while the rest forms a pool which burns down. This type of explosion is
caused if a storage tank is exposed to a heat source from outside.
In contrast to a deflagration, the chemical reaction in a detonation is faster by several

orders of magnitude where the reaction front propagates at a supersonic speed of several
km/s. The shock wave passing through the flammable mixture results in its instantaneous
compression, heating, and spontaneous ignition; the release of the chemical energy keeps
the shock wave going. The damage potential is much stronger than in a deflagration.
The shock front possesses a 3D cellular structure characteristic of the fuel-air mixture.
The decisive criterion for flame acceleration up to the deflagration-to-detonation transition
(DDT) is the state of turbulence in the gas cloud. The explosion of an unconfined hydrogen
gas cloud in the presence of a weak ignition source was never observed to have turned into
a detonation. The ignition energy required for a direct detonation can only be provided by
either an intensive turbulent hot gas jet or by an explosive.
Numerous large-scale and small-scale test facilities have been used to investigate
flame acceleration and DDT. The most dangerous configurations were, as expected, those
with (partial) confinement and obstruction. Large-scale tests were conducted in the USA
in the FLAME apparatus. In Germany, a series of balloon tests of different geometries
with mixtures of hydrogen and other hydrocarbon fuels with air at volumes between 7.5
and 2100 m3 were conducted with ignition by an explosive to cause a detonation. In the
Russian RUT facility, one of the key results was the observation of a 12.5 vol% H2 lower
detonability limit, demonstrating that larger-scale facilities seem to extend the detonability
range of a flammable gas mixture.
Although the mechanisms of DDT and flame acceleration are qualitatively understood,
there are still difficulties in fully describing them on a quantitative basis. The most advanced
approaches are the CFD models, which are similar to the dispersion models, based on
the Navier-Stokes equations with a k-e approach to deal with turbulence. They need,
however, to be adapted to high-speed flows of explosions and short time steps characteristic
of chemical reactions. Numerical simulation methods are applied to study turbulent
combustion, flame acceleration processes, and (re-initiation of) detonations. Today's
modeling and simulation show a good agreement with a variety of fast combustion
phenomena. Examples of codes are, e.g., the Dutch REAGAS or a 2D development by the
German DLR or again the BASSIM and CFX codes.
In a deflagration, the volume expansion of the burnt products reacts like a piston
pushing the unburnt gas mixture aside. The static pressure loading in slow deflagration
processes is conservatively described by the adiabatic, isochoric, complete combustion
pressure representing an upper bound in a closed volume. Pressure and density decrease in
the wave direction. In contrast, a detonation generates a compression wave. Its pressure-
time behavior is characterized by a distinct rapid increase and a subsequent exponential
decay. In the far distance, deflagration and detonation show about the same pressure shape.
In a first approximation, it is possible to deduce the strength of the destroying pressure
waves from the evaluation of damage. Charts with characteristic destruction lines give
319
peak overpressure of a damage-equivalent pressure wave as a function of the distance.
"Experimental" data were derived from explosives and nuclear weapon explosions, which
are, however, only partially applicable to a deflagration. A simplification of the load-
damage analysis has led to the development of the "Pressure-Impulse" methodology as a
computer simulation technique for quick and easy predictions.
Safety distances can be fixed according to physically defined criteria, e.g., the dose
of thermal radiation or peak overpressure, to have reached a certain threshold value. A
basic prerequisite is knowledge of the source term which is dependent on leak size and
thermodynamic conditions of the leaking material. Different relations illustrate the variation
in conservativism of different institutions that generate safety criteria. The safety distances
are distinguished for people and for less demanding equipment, e.g., adjacent storage tanks,
working buildings, or distinguished with respect to fireballs, shrapnel, structural response
or physiological effects.
Environmental Issues
Water vapor is the most climate-relevant atmospheric gas contributing a portion of

65 % to the overall greenhouse effect. The (anthropogenic) effect of water vapor released
into the atmosphere within a large-scale hydrogen technology is disputed among climate
researchers. With respect to the greenhouse effect caused by hydrogen fueled aircraft, it
can be stated that for mid-latitude atmospheric conditions at an altitude below 10 km, the
contribution of water vapor is negligible. Above 10 km, the emitted water vapor behaves
like a greenhouse gas exceeding the effect of kerosene at altitudes > 12 km. The formation
of contrails is avoided at lower cruising altitudes, however, at the expense of higher fuel
consumption.
Nitrogen oxides (NOX) are generated in combustion processes as a function of

temperature, pressure, and time. Emissions are small compared with conventional fuels.
The generation and emission of NOX peaks at excess air ratios of about 1, but is steeply
reduced by applying a lean H2, low-temperature burning process.
CURRENT ACTIVITIES ON HYDROGEN IN THE WORLD
A variety of hydrogen energy applications of a demonstration character have been

started or are projected worldwide encompassing autonomous or partial power systems on
different scales as well as vehicle research projects. Some examples of large-scale H2
systems are presented in the following.
The most ambitious hydrogen energy R&D program in the world is currently being
conducted in Japan where private industry is demonstrating rigorous investment efforts
exceeding those levels found in the USA and European countries. The "International
Clean Energy Network Using Hydrogen Conversion", WE-NET, pursued under the New
Sunshine project, has the ultimate goal of establishing a worldwide energy system based
on (liquid) hydrogen. WE-NET is projected to run over 28 years, 1993 - 2020. Phase
I (1993 - 1998) deals with surveys on key technologies and elemental research. Main
targets are an efficient membrane electrolysis, large-scale liquefaction and storage, LH2
ship transportation, and the development of a hydrogen combustion turbine.
320
The "Euro-Quebec Hydro-Hydrogen Pilot Project", the concept of a clean energy
system based on hydrogen, was presented in 1990. The idea of this project is to take 100
MW electricity from hydropower in Canada for electrolytic hydrogen production. Liquid
hydrogen is then shipped by barge carriers to Hamburg, where it is distributed to a variety of
consumers. The Euro-Quebec project is designed as a local hydrogen energy economy. Its
prime goal is the demonstration of the possibilities of safe transport, storage, and distribution
of (liquid) hydrogen. Since 1992 with the start of phase HI, pre-approval activities and
a demonstration program have been started, the most important include various city bus
projects, development and testing of advanced LH2 mantime transport containers, emission
testing on sectors of LH2 adapted combustors for aircraft engines, power plants in CHP
mode, and last but not least, a study on rules and regulations, safety, and acceptability
of hydrogen.
A solar hydrogen demonstration plant project on an industrial scale has been imple-
mented since 1985 by Solar-Wasserstoff-Bayern in Germany and is scheduled to run until
1999. It aims at testing important subsystems of a future hydrogen energy economy on
a long-term basis. Components tested are photovoltaic fields, advanced-type electrolyzers
of the 100 kW category, AFC and PAFC power plants, gas-fired heating boilers, an LH2
refueling station, and recently also a PEFC powered fork lift truck. Despite the fact that
not all of the systems installed worked satisfactorily, the overall impression is positive and
the results are promising. A new project was started in 1997 in Germany by the "Hydrogen
Initiative Bavaria", WIBA, to demonstrate H2-fueled ground-based traffic at Munich airport.
A Photovoltaic Hydrogen Fuel Cell Demonstration facility, PHOEBUS, has been

erected and is currently operated at the Research Center Julich encompassing all components
for a decentralized autonomous, all-year electricity supply from solar energy using hydrogen
as the energy storage medium. The characteristic components of PHOEBUS are 312 m2
photovoltaic modules, a pressure electrolyzer, an alkaline fuel cell, a battery for short-term
storage of electricity, pressure vessels for long-term storage of hydrogen and oxygen, as well
as an energy management system. A third of the PV electricity is used to produce hydrogen
in an electrolyzer to be later consumed in a fuel cell. The overall energy consumption is
assumed to be 17.4 MWh/yr corresponding to an efficiency of 59 %.
In a joint German-Russian cooperation, the "Cryoplane" project started in 1990 to

investigate the feasibility of an aircraft propelled by LNG or LH2- Topics covered by
the project are scenarios for the transition to an alternative fuel, infrastructure, aircraft
design, fuel system layout, engine modifications, and ecological issues. Precursor phases of
commercial LH2 aviation will be the development of the appropriate infrastructure and the
operation of a demonstrator. The Cryoplane project is currently in a phase of developing and
testing the fundamental technologies of single components such as the fuel tank, cryogenic
pump and hydrogen combustor. A regional airplane of Domier-328 type has been selected
to serve as a demonstrator and is projected to fly by the year 2000.
TRANSITION TO A FUTURE HYDROGEN ENERGY ECONOMY
In a future hydrogen energy economy, the hydrogen will ultimately come from non-
fossil fuel, although fossil fuels will provide a long-term transitional resource. Sources of
renewable energies (wind, water, sun) are deemed sufficient to more than replace all fossil
321
fuel. In the concept of a nuclear hydrogen economy, the production facility would consist
of an HTGR combined with a thermochemical plant where the high-temperature heat is
directly converted into chemical energy. The aspect of energy independence represents an
argument in favor of promoting the transition to hydrogen.
Nuclear energy may gain its greatest importance in helping to get the hydrogen
economy started and to facilitate the transition. On a mid-term basis, overcapacities in
countries with a high percentage of nuclear power (e.g., France) and reserve capacities,
which exist in all large distributed grids, could be used in electrolysis plants to be
immediately disconnected again if needed, e.g., at peak load times. On a longer-term basis,
nuclear power could be installed for hydrogen production if economically competitive with
fossil primary energies.
The duration of the transition period will depend on the so-called harvest factor of the
new energy conversion plants, which is the ratio of the energy produced during its lifetime
to the energy invested for its construction. A harvest factor of 10 means that a complete
transition could be realized within a century; a harvest factor of less than 5, however,
would even prevent complete transition since the new system will not be self-sustainable.
Photovoltaic power plants, as estimated in 1993, have harvest factors of optimistically 1
- 3 with a forecast 3-6 range if the lifetime of the solar cells can be extended to as
much as 30 years. An optimistic nuclear energy scenario could result in a harvest factor
of 14, provided there is a well functioning complete nuclear cycle, and, of course, public
acceptance. A transition period ranging from decades to over a century is expected.
A conceivable intermediate step for the transition to a comprehensive hydrogen

economy is the concentration on a low-energy economy with minimum CO2 emissions
into the atmosphere using local excess potentials of electricity, e.g., renewables such as
hydro, and wind or cheap off-peak electricity in the day/night, working day/weekend, or
winter/summer cycles.
A change of the existing system for the introduction of hydrogen as an energy carrier
for use in efficient, low-polluting systems in the near future needs to be done in small
steps. Examples in the traffic area are city bus projects or LH2 aviation in the longer run.
Furthermore small-scale energy storage systems comprising electrolyzer, metal hydride,
and fuel cell components, ideal for applications far away from the grid will be a reasonable
approach. Natural gas is expected to play a significant role as the preceding technology.
The problem is not the amount of fossil fuels consumed, it is the need to maintain
cleanness during consumption, i.e., the treatment of emissions. Hafele has envisioned the
concept of a "New Horizontally Integrated Energy System", NHIES, with zero emission,
underlying the ideas of decomposition of coal, oil, gas and cleaning, stoichiometric relation,
and synthesis and conversion. The NHIES has to compete with the presently existing
vertically integrated energy systems. The "ultra-low emission vehicle" (ULEV) and the
"zero-emission vehicle" (ZEV) mandate in California makes this state appear to be a good
candidate for the introduction of hydrogen vehicles in a close cooperative effort between
fuel cell manufacturer, automobile industry, and academic groups.
322
The establishment of strategies is an indigenous task of policy. A variety of
measures is conceivable for tackling the problem of environmental damage caused by
the continuous contribution of the combustion of fossil fuels: direct legislative regulations
and rules, investments, tax system, research funding. Industry may contribute by long-term
investments, research activities and cooperation for utilizing hydrogen technology.
FINAL REMARKS
In principle, two different and independent areas have been covered in this report:
on the one side hydrogen as an energy carrier and its application in a future large-scale
hydrogen energy economy, and on the other side nuclear power which represents an
important option for providing primary energy for hydrogen production. The combination
of both nuclear and hydrogen is not necessarily required but it represents a feasible option
to facilitate the transition. Both areas will help reduce global impact on the atmosphere.
Hydrogen faces economic problems and possibly public acceptance difficulties. Nuclear
power faces public acceptance problems.
The limited efforts on the further development of a new promising energy technology
as has been observed over the past decade are due to currently low conventional energy
prices with even falling tendency, also the energy reserve situation is no concern, and no
change is expected for the next years. Current goals of energy policy include the achieve-
ment of a high level of congruence between supply security, economy, and environmental
compatibility meaning an optimized, not necessarily minimized, energy consumption.
After the Kyoto summit in 1997, the reduction of greenhouse gas emission is
considered just one among several strategies. There is hope that by means of the new policy
instruments (tradeable emission certificates, joint projects between industrial and developing
countries, and others) of the climate protocol, the policy actors will proceed slowly, but
steadily to take internationally sound actions and direct the powerful transnational economy
and finance markets towards a low-emission economy and acquire acceptance and the
readiness of accepting the expenses by those who are concerned.
The keywords that characterize the strategy for a transition into the world's future
energy economy are: conservation, substitution, and diversification strategy, necessary
to reduce the burning of fossil fuels and to improve efficiencies, to replace them by
environmentally benign energy carriers, and to avoid new and too strong dependencies.
Of course, hydrogen as an energy carrier will be just one contender on the energy
market competing with others, e.g., electricity, and is currently facing economic problems.
The key issues, however, that are able to promote an innovative energy technology based on
hydrogen are its environmental benefit in a world with still increasing demand for energy,
its safety compared with other fuels, and, in the long run, its potential of mitigating the
jeopardy by a critical energy supply situation in future. Also nuclear power represents but
one contender among the non-fossil energy carriers, although providing an environmentally
benign primary energy source which could be used for hydrogen production. Its major
problem is its lack of public acceptance.
323
Many research activities still need to be done in both areas aiming at an economic
breakthrough of hydrogen and its link to nuclear process heat. The necessary concomitant
efforts concern safety considerations for a large-scale hydrogen economy. This could
be promoted by enhanced information exchange among all interested parties, e.g., by
establishing international working groups to cover critical aspects, including:
• Introduction of "advanced systems", e.g., fuel cells in the conventional field and
HTGRs in the nuclear field;
• Definition of a sound safety and risk assessment methodology of non-nuclear
technologies including experiments on hydrogen safety;
• Feasibility of the combination nuclear power plant / chemical plant;
• Analysis of the "nuclear" hydrogen production potential;
• Definition of fall-back position with established systems.
This would be useful to promote confidence in the technologies and implementation

of steps towards a hydrogen economy.
324
Appendix A
THERMOCHEMICAL CYCLES
FOR HYDROGEN PRODUCTION
A.l. PRINCIPLE
Since a direct thermolysis of water, which requires temperatures of > 2500 °C, is not
practicable under normal circumstances, the splitting process is subdivided into different
partial reactions, either one running on a lower temperature level. The principle is given
by the following cycle:
2AB + 2H 2 O -* 2 AH + 2 BOH + heat at T!

2 BOH -> 2 B + 1/2 O2 + H2O - heat at T2
2 AH -» 2 A + H2 - heat at T3
2A + 2 B - » 2 A B + heat at T 4
After the first cycle (vanadium - chlorine) was proposed in 1964, some 2000 - 3000
potential thermochemical cycles have been tested and checked in terms of appropriate
reaction temperatures and velocities and in economic respect. Thermochemical cycles
are Camot cycle-limited meaning that high temperatures could improve the conversion
efficiency. Major problems arise by the large material flows, by the introduction of
impurities, and by the potential creation of toxic and environmentally unacceptable species.
The main goal is to achieve satisfactory overall energy efficiencies which are currently not
higher than around 40 % and, in connection with separation steps and severe materials
and equipment design difficulties, do not allow, so far, large-scale thermochemical process
applications [11]. Studies are focusing on reaction kinetics, thermodynamics, reactant
separation, material stability, flow sheeting, and cost analysis.
Thermochemical multistage cycles are usually classified in families according to the

chemicals involved. Cycles can consist of up to 8 equations, usually 3 - 6 , with up to 5
elements others than H2 and O2 involved [5].
A.2. THERMOCHEMICAL CYCLES
A.2.1. Oxide System
The most simple water splitting process is the two-step cycle with metal oxides
(MeO) as the redox system
(1) 3MeO + H2O -> Me3O4 + H2

(2) Me3O4 -» 3 MeO + 1/2 O2 - heat
where Me = Mn, Fe, Co
325
The High-Temperature Solar Chemistry Section of the PSI in Switzerland has realized
a sunlight-concentrating device based on the metal iron (Me=Fe) with a power of 60 kW
to provide the required temperatures of 2200 °C (for step (2)). Efficiency is expected to
be 20 % [13].
In other cases of water splitting reactions, the metal reacts exothermally with water
at a temperature < 800 K [13]
x Me + y H2O «——> MexOy + y H2 + heat
For the electrothermal dissociation of metal oxides into their elements, high tempera-
tures are needed which can be lowered by using chemical reducing agents. The combination
with CH4 as a reducing agent in a reforming process results in the coproduction of metals
and synthesis gas:
MexOy + y CH4 <——> x Me + y (2 H2 + CO)
Due to its molar ratio, the synthesis gas mixture is particularly suitable for methanol
production. The iron system, Me - Fe, has been extensively studied in laboratory electric
furnaces [13]. The other two systems have a very low hydrogen yield only [9].
A.2.2. Halide Family Processes
In the metal-halide system, the hydrogen producing reaction is written as [9]
3 MeX2 + 4 H2O -» Me3O4 + 6 HX + H2

where Me = Mn, Fe
X = Cl, Br, I
The Mn system is insignificant due to negligible yield of hydrogen. For the oxygen
producing reaction, three main reaction paths have been proposed. In one of them, the
oxidizing reaction of H2O is conducted in two steps through the addition of an oxide phase
which reacts to give the corresponding halide [9]:
Fe3O4 + 8 HX -> 3 FeX2 + 4 H2O + X2

AO + X2 -» AX2 + 1/2 O2
AX2 + H2O -> AO + 2 HX
where A = Mn, Ca
X = Cl, Br
The bromine-calcium-iron process based on the combination A = Ca and X - Br

is also named "UT-3" (University of Tokyo-3) process:
326
(1) CaBr2 + H2O -» CaO + 2 HBr 700 - 750 °C
(2) CaO + Br2 -» CaBr2 + 1/2 O2 500 - 600 °C
(3) Fe3O4 + 8 HBr -» 3 FeBr2 + 4 H2O + Br2 200 - 300 °C
(4) 3FeBr2 + 4H 2 O -> Fe3O4 + 6 HBr + H2 550 - 600 °C
Stages (1) and (3) produce reactants for the stages (2) and (4). All reactions in the
UT-3 cycle are solid-gas reactions with the net result of a steam decomposition into H2
and O2 [16]. Therefore it is easy to separate gas from solid products. The process has
four reactors connected in series in a loop. After one cycle, the reactors are switched and
the direction of the cycle is reversed. Solid reactants are given in the form of spherical
pellets. Heat exchangers are used to regulate reaction temperatures. Conversion rates for
some reactions are small meaning that gas flow and thus reactor size is relatively large.
Another promising thermochemical cycle is based on the "iron-chlorine compound"

(or "Mark 9") [4, 8]:
(la) 3 FeCl2 + 3 H2O -» 3 FeO + 6 HC1 1100 °C

(Ib) 3FeO + H2O -» Fe3O4 + H2 650 °C
(2) Fe3O4 + 8 HC1 -* FeCl2 + 2 FeCl3 + 4 H2O 100 - 300 °C
(3) 2FeCl3 -» 2FeCl2 + Cl2 420 °C
(4) C12 + H2O -» 2 HC12 + 1/2 O2 - heat 700 - 1000 °C
investigated in Germany, Italy, Japan, and the USA. Its theoretical efficiency is 44 %,
in practice, however, no higher than 30 % due to difficulties given by material problems,
the coupling to the process heat source, and the separation of the gaseous phases. Subjects
of investigation at the Research Center Julich were the "modified AGNES" process with
the reaction (4), the reverse Deacon process, conducted in two steps:
(4a) C12 -i- 2 Mg(OH)Cl -» 2 MgCl2 + H2O + 1/2 O2

(4b) 2MgCl 2 + 2H 2 O -» 2 HC1 + 2 Mg(OH)Cl
A.2.3, Sulfur Family Processes
Three surviving thermochemical cycles are based on the sulfur family processes which
all have in common the thermal decomposition of sulfuric acid at high temperatures. A
typical representant is the four steps process:
(1) 3/2 SO2 + H2O -» H2SO4 + 1/2 S

(2) H2SO4 -» H2O + SO2 -I- 1/2 O2 - heat 850 °C
(3) 3/2 S + H2O -» H2S + 1/2 SO2
(4) H2S -» H2 + S - heat
327
The most considerable potential has the iodine-sulfur (IS) process which has been
first investigated at General Atomics, San Diego, USA, for many years demonstrating its
feasibility and evaluating first chemical engineering flow sheets. The IS process as pursued
at JAERI consists of the following three steps:
(1) I2 + SO2 + 2 H2O <—> 2 HIX + H2SO4 < 100 °C

(2) 2 HI <—» H2 + I2 700 °C
(3) H2S04 <—> H20 + S02 + 1/2 02 850 °C
comprising the exothermal water consuming Bunsen reaction, the endothermal decom-
position of hydrogen iodide in an electrodialysis cell as the hydrogen producing reaction,
and the thermal decomposition of sulfuric acid as the oxygen producing reaction [12]. The
decomposition of H2SO4 and HI was found to be an expending procedure causing severe
corrosion problems. According to analytical studies, an efficiency of 40 - 50 % under
optimal operating conditions can be expected [10]. General Atomics gives a figure of 47
% [15]. The IS process was originally designed to be coupled with a high-temperature
reactor or a fusion reactor and more recently also with a solar thermal plant The cycle has
been verified on a laboratory scale (1 - 10 Ml of H2 per hour) in Japan and is intended to
be later connected to HTTR process heat for demonstration (see also chapter 8).
The sulfuric acid-methanol process is given by
H2SO4 -+ H2O + SO2 + 1/2 O2 - heat 700 - 1000 °C

CH4 + H2O -> CO + 3 H2 700 - 1000 °C
CO + 2 H2 -» CH3OH 300 °C
CH3OH + H2O + SO2 -> CH4 + H2SO4 w 100 °C
with the goals to design and optimize the processing steps and to demonstrate its
feasibility on an industrial scale [2].
In the "sulfate" process by
4 FeSO4 -» 2 Fe2O3 + 3 SO2 + SO3 +1/2 O2 HT

Fe2O3 + 3 SO2 + SO3 + H2O -> 4 FeSO4 + H2 LT
the high-temperature (HT) step is easily selected at > 980 K, whereas the low-
temperature (LT) step results in H2S and FeS2 formation rather than H2. This brought
up the idea to substitute the LT step by an electrolysis step.
A.3. THERMOCHEMICAL HYBRID CYCLES
The thermochemical hybrid process is a combined cycle process with both ther-
mochemical and electrolytic reactions of water splitting. The hybrid process offers the
328
possibility to run low-temperature reactions on electricity. The expectations for realization
of hybrid processes are similar to those for purely thermochemical processes [3]. Various
hybrid processes are energetically possible, however, not always feasible. Important criteria
are the minimum voltage for the electrolysis step, realizability, efficiency.
A.3.1. Sulfuric Acid Bromine Hybrid Process
The sulfuric acid-bromine hybrid process (or "Mark 13") which has been tested at
the JRC Ispra, Italy, consists of the reactions:
(la) H2SO4 «—» H2O + SO3 700 - 1000 °C

(Ib) SO3 *—» SO2 + 1/2 O2 700 - 1000 °C
(2) SO2 + Br2 + 2 H2O <—> 2 HBr + H2SO4 20 - 100 °C
(3) 2 HBr «—> H2 + Br2 - electricity 80 - 200 °C
with reaction (3) as the electrochemical step. This cycle using only liquid or gaseous
species operated successfully in 1978 in a laboratory scale plant with a hydrogen production
rate of 100 1/h over 150 h. An efficient way was found to remove SO2 by reaction with
bromine which was the basis for creating a new flue gas desulfurization process [14]. The
efficiency is given as 37 % [15].
A new innovative system is being developed to apply solar energy for the step of
HBr production. Concentrated solar radiation is absorbed by bromine vapor to produce
bromine atoms which then react with steam at temperatures between 600 and 1000 °C to
form HBr and oxygen. A 1 kW solar system was constructed and tested. HBr conversion
rates were observed to be 70 % of the theoretical maximum [6].
A.3.2. Sulfuric Acid Hybrid Process
The sulfuric acid hybrid cycle (Westinghouse process) is given by the reaction
equations
(la) H2SO4 -* H2O + SO3 > 450 °C

(Ib) SO3 «—> SO2 + 1/2 O2 > 800 °C
(2) 2 H2O + SO2 «—> H2SO4 + H2 - electricity 80 °C
Optimum efficiency is reached with a 65 % concentration of sulfuric acid in water.

The weak point is again the decomposition of the sulfuric acid introducing significant
corrosion problems. The cycle efficiency is 40 % which has the potential to be increased
to 46 % if the electrolysis is conducted in several stages.
The cycle, developed by Westinghouse in 1975, has been experimentally investigated

at the Research Center Julich in close cooperation with JRC Ispra. It has eventually led to a
three-compartment electrolysis cell with a hydrogen production rate of 10 Nl/h successfully
operated in a 600 h test run. Cell conditions were 80 °C at 1.5 MPa [15]. Tube materials
329
selected have shown an excellent corrosion resistance. The possibility to use direct sunlight
for the 863 decomposition step was also demonstrated.
The two-step sulfuric acid hybrid process is using in the thermal (« 900 °C) oxygen-
producing step the decomposition of sulfuric acid and sulfonic acid in the electrochemical
hydrogen-producing step:
H2SO4 <——f H2O + SO2 + 1/2 O2

H2O + H2SO3 *—» H2SO4 + H2
The hybrid process with just two steps is expected to reduce material streams as
compared with the above described IS process. The electric energy demand here is a factor
of 7 (in reality 3 - 4 ) lower than in the electrolytic water splitting process.
AJ3. Hydrocarbon Hybrid Process
Hydrocarbon hybrid processes are promising thermochemical cycles. The hydrogen

producing step is typically done by methane splitting connected with the formation of CO.
In the methane-methanol-methana! process, the major steps are methane splitting,

methanol synthesis, methanal synthesis, and methanal electrolysis, for which the chemical
reactions are [1]:
(1) H2O + CH4 -» CO + 3 H2 - 250 kJ/mol 700 - 950 °C

(2) 2H2 + CO -> CH3OH + 90 kJ/mol 250 - 300 °C
(3) CH3OH -> CH2O + H2 - 83 kJ/mol 400 - 500 °C
(4) CH2O + H2 ->• CH4 + 1/2 O2 - 0.4 V 50-300 °C
The main energy input is in the steam reforming step (1), which is a well developed
and frequently applied technology as is the step (2) of methanol synthesis. Secondary
products such as methane and higher hydrocarbons created during the methanol synthesis
are considered not significant. The cycle is characterized by a selectively working catalyst
and a high reaction rate. The methanal (= formaldehyd) is generally applied on a large-scale
in an autothermal operation at about 600 °C by combining exothermal oxidation (methanol
+ oxygen —>• methanal + water) with endothermal dehydration (methanol —> methanal +
water). The electric energy required in the methanal electrolysis in step (4) is about a third
of that needed in the direct electrolysis of water. In contrast to the previous steps, this
process step (4) is not well known and needs further development work. Advantages of
the cycle are that all reaction partners are in either liquid or gaseous state and neither one
is exotic nor poisonous. Recuperation, i.e. the reuse of released heat energy, is significant
for an economic process. Anticipating design improvements, the above process is expected
to achieve an efficiency of 51 % [1].
The cycle processes "methane-methanol" and "methanal-methanol" based on only

three and two, respectively, main reactions require a larger electricity input. The process
330
"sulfonic acid-sulfuric acid" consumes the least electricity, but is probably plagued by
severe material problems. Other processes such as "aceton-propane-propylen" and "chinon-
hydrochinon" need more instrumental effort [1].
The methane production, reactions (3) and (4), could also be made by means of
methanol plus iodine
2 CH3OH + I2 -» 2 CH3I + H2O + 1/2 O2

2 CH3I + H20 -» CH3OH + I2
with a conversion rate achieved in experiments of up to 50 % of methane [15].
Production of methanol is made in the following hybrid cycle
(1) CO2 + 3 CH4 + 2 H2O <——» 4 CO + 8 H2

(2) 4CO + 8H2 <——> 4CH3OH
(3) 3 CH3OH -»• 3 CEU + 3/2 O2 - electricity
(4) CO2 + 2 H2O -» CH3OH + 3/2 O2
The carbon source CO2 could be taken from the waste gas of coal power plants.
This cycle represents a good candidate for the production of a liquid energy carrier with
nuclear heat as well [15].
A combined procedure for the co-production of hydrogen and electricity from water
splitting and coal has been proposed with the following reaction system
(1) C + 2 H2SO4 <— » CO2 + H2O + 2 SO2 - heat 250 - 300 °C

(2) 2 SO2 + 4 H2O <——> 2 H2 + 2 H2SO4
(3) C + 2 H2O -» C02 + 2 H2
The fuel coal, represented by "C", is oxidized with chemically bound oxygen. The
heated combustion gases are routed to a gas turbine to run the electrolysis step. The SO2
and H2O are applied in a Westinghouse cycle. An optimum efficiency is seen at 57 % [15].
A.3.4. Metal-Metal Hydride Hybrid Process
In the metal-metal hydride system, the following reactions occur with "Me" repre-
senting an alkali metal [15]
(1) H2O + 2 Me -» 2 MeH + 1/2 02 - electricity 250 - 400 °C

(2) 2 MeH -> 2 Me + H2 - heat 700 - 900 °C
where Me = Li, Na, K or mixtures
331
Oxygen gas is developed while the hydrogen is bound into the metal. A few tests
were conducted with the lithium hybrid process, which has been developed at the RWTH
Aachen, Germany. Out of some 40 candidate materials for the cell membrane, TiNi alloys
were shown to be corrosion resistant for more than 500 h under electrolysis conditions.
A thermodynamic analysis indicates an efficiency of 45 %, if hydrogen at a pressure of
4 MPa is provided [15].
Using experience with sodium from the breeder technology, the RWTH Aachen has
investigated the sodium cycle, Me=Na, in 1988/89 with a promising > 50 % overall
efficiency. The high-temperature step was operated continuously on a laboratory scale.
Cathodic membrane material problems were encountered in the electrochemical step [7].
The Me/MeH hybrid cycle offers an interesting possibility of hydrogen production in

connection with a high-temperature gas-cooled reactor.
REFERENCES TO APPENDIX A
[I] BARNERT, H., Anmerkungen zur thermochemischen Produktion von Wasserstoff

aus Wasser mittels Hochtemperaturreaktor-Warme, Report Jiil-1660, Research Cen-
ter Jiilich (1980).
[2] BARNERT, H., Wasserstoff aus Wasser + HTR-Warme, KFA Annual Report 1974,
Research Center Jiilich (1975) 25-30.
[3] BECKURTS, K.-H., DIETRICH, G., Projekt Fernwarme Versorgung fur Millionen-
Stadte, Bild der Wissenschaft 13 (1976) January 64-70.
[4] FRIE, W., Thermochemische Kreisprozesse zur Wasserstoff-Erzeugung, Wasser-
stoff als Energietrager, (Status Seminar, Wurzburg, 1995), Projekttrager Biologic,
Energie, Okologie, Research Center Jiilich (1995) 47-60.
[5] FUNK, J.E., Hydrogen Production, Tutorial during the 10th World Hydrogen Energy
Conf., Cocoa Beach, USA, June 19, 1994.
[6] HEATON, H., Production of HBr from Bromine and Steam for Off-Peak, World
Wide Web, http://www.eren.doe.gov/hydrogen/hbr.htm, US Department of Energy
(1997).
[7] HERZOG, F., Development of the Metal/Metal Hydride Process, in: EEA, Imple-
menting Agreement for a Programme of Research and Development on the Produc-
tion of Hydrogen from Water, Annex IX: Hydrogen Production, Technical Report,
Document HUF-4, Research Center Julich (1990) 99-133.
[8] KNOCHE, K.F., Stand der Arbeiten zur Wasserstoff-Erzeugung mit nuklearer
ProzeBwSrme, Chem.-Ing.-Tech. 49 (1977) 238-242.
[9] LUNDBERG, M., ROSEN, E., Model Calculations on Some Thermochemical Wa-
ter Splitting Processes, Hydrogen Production, (2nd IEA Technical Workshop, Julich,
FRG, 1991), STRUCK, B.D. (Ed.), Implementing Agreement for a Programme of
Research and Development on the Production of Hydrogen from Water, Document
[10] ONUKI, K., et ah, Studies on Water Splitting Hydrogen Production, (3rd JAERJ
Symp., Oarai, 1996), Proc. JAERI-Conf 96-010, Japan Atomic Energy Research
Institute (1996) 462-474.
[II] PESCHKA, W., Liquid Hydrogen: Fuel of the Future, Springer-Verlag Wien New
York (1992).
332
[12] SHIMIZU, S., ONUKI, K., NAKAJIMA, H., Bench-Scale Studies of the Iodine-
Sulfur Process, Hydrogen Production, (2nd IEA Technical Workshop, Jiilich, FRG,
HUF-6, Research Center Jiilich (1991) 49-62.
[13] STEINFELD, A., et al., Solar-Processed Metals as Clean Energy Carriers and
Water-Splitters, (llth World Hydrogen Energy Conf., Stuttgart, FRG, 1996), VE-
[14] VAN VELZEN, D., Desulphurization and Denoxing of Waste Gases Producing
Hydrogen as a By-Product, Hydrogen Production, (2nd IEA Technical Workshop,
Julich, FRG, 1991), STRUCK, B.D. (Ed.), Implementing Agreement for a Pro-
gramme of Research and Development on the Production of Hydrogen from Water,
Document HUF-6, Research Center Julich (1991) 99-111.
[15] WEIRICH, W., et al., Thermochemical Processes for Water Splitting - Status and
Outlook, Nucl. Eng. Des. 78 (1984) 285-291.
[16] YOSHIDA, K., Hydrogen Production by UT-3 Thermochemical Water Decomposi-
tion Cycle, Hydrogen and Clean Energy (Int. Symp., Tokyo, 1995), NEDO (1995)
39-46.
NEXT PAGE(S)
left BLANK
333
Appendix B
CHEMICAL ENERGY
TRANSMISSION SYSTEMS
B.I. STEAM REFORMING SYSTEM (EVA/ADAM)
Within the project "Nukleare Fernenergie" (NFE) (- "Nuclear Long-Distance

Energy"), the energy transport system by natural gas steam reforming / methana-
tion was theoretically and experimentally investigated at the Research Center Julich in
Germany. Its goal was to demonstrate the operation of a complete helium-heated system
for steam reforming of methane on a representative scale and to demonstrate the technical
performance and efficiency of a thermochemical pipeline system [5]. Fig. B-l shows
a schematic of this long-distance energy transportation system. The heat source was an
electric heater simulating HTGR conditions.
In 1972, the first experimental setup, the so-called EVA ("Einzelspaltrohr-Versuchs-

Anlage" - "Single Tube Test Facility") was established to account for the endothermal
reforming reactions:
CH4 + H2O <—» CO + 3 H2 - 206 U/mol

CH4 + 2 H2O <—> CO2 + 4 H2 - 165 kJ/mol
The tube with a variable length of 10 - 15 m and a variable inner diameter of 100
- 150 mm was entered at the bottom by helium of 950 °C which flowed upwards in an
annulus at the outside providing heat to the reformer tube inside. The feed gas methane
was heated up in a counter current flow and reached a temperature of 825 °C at the tube
bottom. The resulting product gas was then cooled down by flowing upwards again through
return pipes inside the reformer tube (internal pigtails). The temperature profiles in the EVA
system are shown in Fig. B-2, left hand side.
The construction of a follow-up reformer test stand, EVA-II, was started in 1980.
EVA-n consisted of a bundle of 30 reforming tubes based on the baffle plate design with
an inner/outer diameter of 100/120 mm and a length of 11.4 m. Helium throughput was 4
kg/s. The other components of this helium loop besides the 10 MW electric heater were
a hot gas duct, steam generator, and helium circulator. Methane input was 0.6 kg/s. The
product gas was generated at a rate of 4400 kg/h.
The reverse process of methanation was realized in 1979 in the test facility ADAM
("Anlage mit drei adiabaten Methanisierungsreaktoren" = "Facility with three adiabatic
methanation reactors"). A three-stage methanation reaction including a dilution of the
product gas by feedback of already retransformed gas was chosen to mitigate the large
heat release, to minimize soot formation, and to maximize CH4 output. The feed gas was
synthesis gas at a rate of 600 Nm3/h in ADAM-I and 9600 Nm3/h in ADAM-II [6].
335
EVA / ADAM Energy Transmission System
EVA ADAM
(methane reforming) (methanation)
L
Long-distance pipeline for H2, CO, CO2
L
Long-distance pipeline for CH4
Fig. B-l: Schematic of long-distance energy transportation in the EVA/ADAM

facility, from [1]
Temperatures I Product Temperatures

Process in EVA | gas in ADAM
as (Energy uptake) _I_ (Energy release)
9 Product
/gas
Reaction t
Helium •»•••
I 1 .Methanation reactor
Reformer Helium
tube
>
'=;:•- Catalyst
Reacting
process
gas
Return
pipe
Returning
product
/ gas
Helium 400 600 800 "C1000 200 400 °C 700

Temperature ——* Temperature ——*
Fig. B-2: Temperature behavior in the methane reformer EVA (left) and in the
methanation system ADAM (right), from [1]
The two significant features of the ADAM plant were the complete reconversion
to methane, thus releasing the complete reaction enthalpy again and the realization of a
650 °C gas mixture after the first methanation step which allowed for the generation of
336
superheated high-pressure steam for further use (here added to the heating system of the
Research Center Julich). The temperature profiles in the ADAM system are shown in Fig.
B-2, right hand side [1].
The complete EVA-II / ADAM-II system was operated in Julich for a total time of
10,150 hours. It has demonstrated for the first time the chemical cycle process under the
realistic conditions of industrial application with a transported power of 300 kW (EVA /
ADAM) and of 5.4 MW (EVA-n / ADAM-II), respectively. The test facilities, however,
were not operated for the sake of maximizing the energy efficiency, they were rather used to
examine the components' behavior under various load conditions including fast transients
[1, 5, 6].
The CH4 - H2O system needs a second return pipe from ADAM to EVA for the
liquid water, if a cleanup of the water and the loss of unconverted hydrocarbons are to
be avoided [4].
The EVA / ADAM system has been established, besides in Germany, also in the
USA and the Russian Federation [9].
B.2. CO2 REFORMING SYSTEM
The CO2 reforming process of methane
CH4 + CO2 <—» 2 CO + 2 H2 - 247 kJ/mol
has been studied as a chemical heat pipe system since 1980 as an alternative in order
to overcome some of the potential difficulties. In contrast to the CH4 - H2O system, a
third pipe is not required since it is not necessary to clean the water. Thus an open cycle
system could be adopted [4].
At the Weizmann Institute in Israel, the closed-cycle CO2 reforming process of natural
gas has been adapted and modified to solar energy as the primary heat input. The use of
gaseous CO2 eases the daily startup / shutdown cycles. A lab-scale facility with 5 - 1 0
kW power was successfully operated. A scale-up plant designed for 480 kW heat input
to the reforming reaction and 250 kW heat output in the methanation step is being tested
since 1994. Catalysts are rhodium and ruthenium. So far 150 cycles have been completed
[3]. The solar power absorbed was 97 kW, a methane conversion rate of 70 % maximum
was achieved. A computer code has been developed for loop design and optimization
purposes [9].
B.3. METHANOL SYSTEM
The production of energy alcohol on the basis of biomass plus CO2-free primary
energy features a CO2-neutral system since the CO2 release upon combustion of methanol
has been formerly extracted from the atmosphere for the biomass production.
337
The methanol-carbon dioxide system is described by
CH3OH + H2O «——» CO2 + 3 H2 - 131 kJ/mol
The heat released during the methanol synthesis corresponds to not more than 15 %
of the energy stored in the hydrogen. If, however, this heat can be fed into the reforming
process, a high storage efficiency could be gained. No additional CO2 is emitted, if
atmospheric CO2 is used as the carbon source.
The nuclear variant of using CH3OH as the energy carrier taking the HTGR as the
primary source for the required process heat is being described in section 4.3.2.3. Congenial
systems are the methanol-carbon monoxide system
CH3OH <——> CO + 2 H2
and the methanol-methyl formate (= formic acid salt) system:
2 CH3OH + H20 <—— » HCOOCH3 + 2 H2
B.4. METHYLCYCLOHEXANE - TOLUENE SYSTEM
Among the many candidates of hydrogen transportation or seasonal storage methods,

the H2 evolution from cyclohexanes at the demand site and the reverse reaction, the
hydrogenauon of benzene, at the supply site offers the advantages of safe and commonly
used transport at ambient temperature in chemical tank ships or trucks and mild reaction
conditions. Cyclohexanes have a hydrogen density and content comparable with metal
hydrides [8]. Methylcyclohexane is the choice of preference, since unlike benzene, toluene
is not carcinogenic.
The methylcyclohexane (MCH) cycle is described by:
C7H8 + 3 H2 - 215 kJ/mol
The hydrogen storage capacity of MCH is 6.2 wt%. Dehydrogenation at 1 MPa is

an endothermal catalytic reforming step with a high yield of toluene (98 %) at elevated
temperatures (> 400 °C). A critical point is the initial conditioning and regeneration of the
catalyst The process conditions of the hydrogenation of the toluene are 2 MPa at 220 °C
achieving a yield of 99 % [2]. A comparison of the MCH vector with the LH2 vector has
been made within the frame of the Euro-Quebec project. Characteristics of the LH2 and
the MCH vectors are listed in Table B-l.
338
Table B-l: Characteristics of the two vectors investigated in the EQHHPP, from [10]
Liquid Hydrogen Methylcyclohexane / Toluene

No hydrocarbons involved during Hydrocarbons involved due to
production and utilization conversion limitations
Low environmental impact upon High environmental impact upon
inadvertant release inadvertant release
Use as liquid or gaseous fuel Use as gaseous fuel only
Sophisticated transport technology not Conventional transport in chemical tank
yet commercially available ships
Limited storage period Unlimited storage period
Part of liquefaction energy can be
Energy needed for dehydrogenation
recovered
Long-term storage of energy by means of MCH is reasonable in cases where the

energy structure results in a large cost difference of seasonal electricity production (e.g.,
hydro, solar).
B.5. NH3 SYSTEM
The ammonia dissociation reaction as a closed-cycle liquid hydride energy storage

system has been investigated primarily by Australian researchers in connection with solar
primary energy [7]:
2NH 3 N2 + 3 H2 - 59 kJ/mol
The advantage of the above process is the complete absence of any side reactions.
The relatively low exothermal reaction rates can be raised, if higher pressures are applied.
The fact that ammonia condenses at around room temperature and separates from the
reactant mixture allows storage of feedstock for both endothermal and exothermal reaction
in the same vessel. Experience with a pilot plant operating at pressures up to 16 MPa and
temperatures up to 720 °C indicate a 1 kW closed-loop system to be feasible [7].
A hydrogen storage system using formates seems to be promising. The reversible

catalytic reaction with sodium formate to decompose to bicarbonate and hydrogen is
NaCOOH + H20 NaHCO3 + H2
The hydrogen capacity is 2.9 wt%. Its advantages are cheap raw materials and the
potential of storing H2 in solid form [11].
339
REFERENCES TO APPENDIX B
[I] BOLTENDAHL, U., HARTH, R., Warmetransport auf kaltem Wege, Bild der
[3] EPSTEIN, M, Recent Test Results of CO2 Reforming of Methane in a Closed
Loop, (3rd JAERI Symp., Oarai, 1996), Proc. JAERI-Conf 96-010, Japan Atomic
Energy Research Institute (1996) 228.
[4] HANNEMAN, R.E., VAKIL, H., WENTORF, R.H.Jr., Closed Loop Chemical
[5] HARTH, R., JANSING, W., TEUBNER, H., Experience Gained from the EVA H
[6] HOHLEIN, B., et al., Methane from Synthesis Gas and Operation of High-
Temperature Methanation, Nucl. Eng. Des. 78 (1984) 241-250.
[7] LOVEGROVE, K.M., High Pressure Ammonia Dissociation Experiments for Solar
Energy Transport and Storage, Int. J. Energy Research 20 (1996) 965-978.
[8] SAITO, Y., Chemical Heat Pump and Hydrogen Transportation, Hydrogen and
Clean Energy ant. Symp., Tokyo, 1995), NEDO (1995) 123-130.
896-904.
[10] WURSIG, G.-M., Shipping Liquid Hydrogen, MER (1991) December 10-14.
[II] ZAIDMAN, B., WIENER, H., SASSON, Y., Formate Salts as Chemical Carriers in
Hydrogen Storage and Transportation, Int. J. Hydrogen Energy 11 (1986) 341-347.
340
Appendix C
CHEMICAL HEAT PUMP SYSTEMS
C.I. 2-PROPANOL - ACETONE SYSTEM
A key process is the dehydrogenation catalysis of 2-propanoI under mild reaction con-
ditions converting low-quality heat into chemical energy and recovering high-temperature
heat or hydrogen gas at the energy demand site.
(CH3)2CHOH <—> (CH3)2CO + H2 - 55 U/mol
The 2-propanol / acetone / hydrogen chemical heat pump is operated in three steps:
1. 2-propanol dehydrogenation
is an endothermal reaction at low temperatures (« 80 °C) assisted by a catalyst.
2. Acetone hydrogenation
is the reverse exothermal reaction at higher temperatures (150 - 200 °C).
3. Separation
of the 2-propanol from acetone and hydrogen by condensation in a fractional
distillation column at ambient temperatures.
High thermal efficiencies can be achieved with properly designed catalysts. A pilot
plant for this chemical heat pump system was constructed in Oarai, Japan, and successfully
tested in 1994. A pilot plant driven by solar energy was realized at the Florida Solar Energy
Center demonstrating an average yield of 180 ml/h of hydrogen (over several days).
C.2. CYCLOHEXANE - BENZENE SYSTEM
The reaction couple consists of cyclohexane dehydrogenation at 200 °C under normal

atmospheric pressure and benzene hydrogenation at 350 °C.
C6H12 <—> C6H6 + 3 H2 - 206 kJ/mol
This system can be realized in a tube wall type reactor.
The use of a membrane reactor with a hydrogen-permeable palladium alloy membrane

inside a catalyst-packed bed allowed a shifting of the chemical equilibrium to efficiently
drive the dehydrogenation process at lower temperatures, e.g., waste heat. It thus serves
as a chemical heat pump system.
C.3. HYDRIDE SYSTEM
A closed-loop heat pump system based on hydrides, a metal hydride super heat
pump (MHSHP) system, has been investigated at JAERI with the purpose not only to
341
amplify nuclear waste heat, but also to mass store hydrogen for transportation purposes. The
L reactor contains an MHX alloy where hydrogen is stored. A medium-temperature heat
source liberates the hydrogen. After transfer to the H reactor, the hydrogen is absorbed in
the Ca alloy and upon reaction liberates high-temperature heat which can be decoupled from
the system by a helium heat exchanger. Candidate MHX materials are currently subject to
further research. In a test facility, heat conversion from 500 to 900 °C was achieved. The
MHSHP system is considered to be connected to the UT-3 hydrogen production cycle.
C.4. SPONGE IRON
The use of sponge iron resulting from the direct iron ore reduction as a hydrogen sto-
rage system has been found to be practicable although with the penalties of a heavy weight,
its solid state as a fuel, and not too high an amount of energy to be stored. Applications
might be seen in heavy transportation systems, e.g., fuel cell powered locomotives.
342
ABBREVIATIONS
ADAM Facility with Three Adiabatic Methanation Reactors

(Anlage mit drei adiabaten Methanisierungsreaktoren)
AFC Alkaline Fuel Cell
AICC Adiabatic Isochoric Complete Combustion
ALWR Advanced Light-Water Reactor
BMC Battelle Model Containment
BWR Boiling Water Reactor
GETS Chemical Energy Transmission System
CFD Computer Fluid Dynamics
CHP Combined Heat and Power
CNG Compressed Natural Gas
DDT Deflagration-to-Detonation Transition
DLR Deutsche Forschungsanstalt fur Luft- und Raumfahrt
DMFC Direct Methanol Fuel Cell
EC European Community
EPR European Pressurized Water Reactor
EPRI Electric Power Research Institute
EQHHPP Euro-Quebec Hydro-Hydrogen Pilot Project
ET External Tank
EVA Single Tube Test Facility (Einzelspaltrohr-Versuchsanlage)
FAPIG First Atomic Power Industry Group
FZJ Forschungszentrum Jiilich GmbH
FZK Forschungszentrum Karlsruhe GmbH
GRS Gesellschaft fur Anlagen- und Reaktorsicherheit mbH
HDR Superheated Steam Reactor (HeiBdampfreaktor)
HHT High-Temperature Reactor with Helium Turbine of Large
Power (Hochtemperaturreaktor mit Heliumturbine groBer
Lei stung)
HTGR High-Temperature Gas-Cooled Reactor
HTTR High-Temperature Engineering Test Reactor
IAEA International Atomic Energy Agency
ICE Internal Combustion Engine
ICT Institut fur Chemische Technologic der Fraunhofer Gesellschaft
ffiA International Energy Agency
IGCC Integrated Gasification Combined Cycle
343
IGFC Integrated Gasification Fuel Cell
IHX Intermediate Heat Exchanger
IMFC Indirect Methanol Fuel Cell
INET Institute for Nuclear Energy Technology, Beijing
IS Iodine - Sulfur (Thermochemical Cycle)
JAERI Japan Atomic Energy Research Institute
JCT Japan - China - Taiwan
KVK Component Experimental Loop
(Komponenten-Versuchskreislauf)
LH2 Liquid Hydrogen
LMFBR Liquid Metal Fast Breeder Reactor
LNG Liquid Natural Gas
LOX Liquid Oxygen
LPG Liquid Petroleum Gas
LWR Light Water Reactor
MCFC Molten Carbonate Fuel Cell
MCH Methylcyclohexane
MHR Modular Helium Reactor
MHTGR Modular High-Temperature Gas-Cooled Reactor
NASA National Aeronautics and Space Administration
NASDA National Space Development Agency of Japan
NBP Normal Boiling Point
NFE Nuclear Long-Distance Energy (Nukleare Fernenergie)
NHIES Novel Horizontally Integrated Energy System
NOX Nitrogen Oxides
NRC US Nuclear Regulatory Commission
NTS Nevada Test Site
OECD Organization for Economic Cooperation and Development
PAFC Phosphoric Acid Fuel Cell
PEFC Polymer Electrolyte Fuel Cell
PEM-FC Proton Exchange Membrane Fuel Cell
PHWR Pressurized Heavy Water Reactor
PNP Prototype Plant Nuclear Process Heat
(Prototypanlage Nukleare ProzeBwarme)
PRA Probabilistic Risk Assessment
PSA Probabilistic Safety Assessment
PV Photovoltaics
344
PWC Pressurized Water Cooler
PWR Pressurized Water Reactor
REH 2 Hydrogen from Renewable Energies
RT Room Temperature
RUT Rocket Shock Tube
RWTH Rheinisch-Westfalische Technische Hochschule, Aachen
SLH2 Slush Liquid Hydrogen
SNG Synthetic Natural Gas
SOFC Solid Oxide Fuel Cell
SPEWE Solid Polymer Electrolyte Water Electrolysis
SPFC Solid Polymer Fuel Cell
STP Standard Temperature and Pressure
SULEV Super Ultra-Low Emission Vehicle
SWATH Small Waterplane Area Twin Hull
TMI Three Mile Island
TNO Netherlands Organization for Applied Scientific Research
TNT Trinitrotoluene
TOE Tons of Oil Equivalent
ULEV Ultra-Low Emission Vehicle
WE-NET International Clean Energy Network Using Hydrogen
Conversion (World Energy Network)
ZEV Zero Emission Vehicle
345
CONTRIBUTORS TO DRAFTING AND REVIEW
Breitung, W. Research Center Karlsruhe, Germany

Chirivella, J. Ergo-Tech Systems, Tujunga, USA
Chitose, K. Mitsubishi Heavy Industries, Yokohama, Japan
Daish, N. Cambridge Environmental Research Consultants, Cambridge,
United Kingdom
Edeskuty, F. J. Los Alamos National Laboratory, Los Alamos, USA
Hicken, E. F. Research Center Julich, Germany
Ide, A. Fuji Electric, Kawasaki, Japan
Katscher, W. Research Center Julich, Germany
Kluyskens, D. Hydro-Quebec, Montreal, Canada
Meissner, D. Research Center Julich, Germany
Mergel, J. Research Center Julich, Germany
Nabielek, H. Research Center Julich, Germany
Nishikawa, Y. Institute of Applied Energy, Tokyo, Japan
Pehr, K. BMW, Munich, Germany
Schmidtchen, U. Federal Institute for Materials Research and Testing, Berlin,
Germany
Statharas, J. NCSR Demokritos, Attiki, Greece
Sun, Y. L. Institute for Nuclear Energy Technology, Beijing, People's
Republic of China
Wursig, G.-M. Germanischer Lloyd, Hamburg, Germany
347

Hydrogenfeul

Uploaded by

Copyright:

Available Formats

Hydrogenfeul

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Hydrogenfeul

Uploaded by

Copyright:

Available Formats

IAEA-TECDOC-1085

INTERNATIONAL ATOMIC ENERGY AGENCY /A\

Orders should be accompanied by prepayment of Austrian Schillings 100,-

HYDROGEN AS AN ENERGY CARRIER AND ITS

Hydrogen is a strong emerging candidate with a significant role as a clean, envi-

CHAPTER 4. ENERGY SITUATION AND ACTIVITIES ON NUCLEAR POWER

1.1. WORLD ENERGY SITUATION

As described in a study by the European Commission, global energy supply is

1.2. GLOBAL ENVIRONMENTAL IMPACT

The United Nations Convention on Climate Change has prompted initiatives by

Three conferences of parties to the United Nations Framework Convention on

1.3. WHY NUCLEAR POWER?

• on-site generation of energy-intensive products

Nuclear power is considered by many to be the most promising CO2-free energy

1.4. WHY HYDROGEN?

NUCLEAR POWER AND ITS POTENTIAL

2.1. SIGNIFICANCE OF PROCESS HEAT FOR THE INDUSTRY

0 200 400 600 800 1000 1200 1600

0 200 400 600 800 1000 1200 1600

Ratio of steam Ratio of

Steam conditions Steam Power

2.2. NUCLEAR POWER PLANTS

conventional plant or HTGR

320 °C, 2 MPa

Product Production Reaction Temperature [°C]

Hydrogen - Water electrolysis 1 ->, 1

Methanol - Methanol synthesis I | (Exothermal)

'-,- "v j LWR

The procedure of using electricity in the electrolysis of water to produce hydrogen

2.2.2. High-Temperature Gas-Cooled Reactor

2.2.2.1. Features of HTGR-Based Chemical Processing

Physical requirements for process heat HTGRs are

Chemical industrial plants could be optimized in terms of product spectrum and

2.2.2.2. Chemical Processes for Hydrogen Production

Fig. 2-7: Dependence of hydrogen and electricity production on helium temperature

Temperature split-up [°C] 950...800...250 950...700...250 950...600...250

66 % compared with the conventional processes [35]. Cogenerated electricity is applicable

2.2.2.2.1. Reforming Process

For the industrial processes of ammonia synthesis, hydrocracking, hydrogenation, a

Unlike conventional fossil-fueled reformer tubes, the helium-heated reforming plant

The combination of all three steps appears economically attractive. A balance

For the hydro-gasification process (Fig. 2-9), an intermediate circuit is deemed

Coupling hydro-gasification of lignite with a methanol synthesis process is possible

Gas Residual coke

2.2.2.2.4. Biomass Conversion

+ 2 H2O + 5.66 H2O -» 6 CH3OH + 2.68 O2

Additional requirements to an energy system as to be environmentally benign and

2.2.2.2.5. Water Splitting

There is a great economic potential of nuclear thermochemical cycles, i.e., the

Electricity Process heat

Battery G H2S04 Battery H

23.1. Nuclear Steam Reformer

Reformer tube heating with a high-temperature nuclear reactor is performed with

Feed Cos Inlet

Product Gos Outlet

625 870 I 950 39/ 40

Parameter Conventional Plant Nuclear Plant

To overcome the perceived disadvantages of a nuclear steam reformer with its

Helium side heal transfer enhancement

o AICC: « VO. STEAM