View Article Online / Journal Homepage / Table of Contents for this issue

Volume 18 | Number 20 | 2008

www.rsc.org/materials Volume 18 | Number 20 | 28 May 2008 | Pages 2285–2392
Published on 03 April 2008. Downloaded by KIER on 5/20/2022 9:24:30 AM.

Journal of Materials Chemistry

Hydrogen Storage and Generation
‘Journal of Materials Chemistry
book of choice’
Why not take advantage of free book chapters from the RSC?
Through our ‘Journal of Materials Chemistry book of choice’ scheme
www.rsc.org/materials Volume 17 | Number 26 | 14 July 2007 | Pages 2669–2760
Journal of Materials Chemistry will regularly highlight a book
from the RSC eBook Collection relevant to your research interests.
Read the latest chapter today by visiting the Journal of Materials
Chemistry website.

The RSC eBook Collection offers:

 Over 900 new and existing books
ISSN 0959-9428
 Fully searchable
PAPER
Yi Xie et al. FEATURE ARTICLE

 Unlimited access
A facile solution-phase deposition Vasilios Georgakilas et al.
approach to porous selenium Decorating carbon nanotubes with
materials metal or semiconductor nanoparticles 0959-9428(2007)17:26;1-Y

Why not take a look today? Go online to find out more!

www.rsc.org/materials
Pages 2285–2392

ISSN 0959-9428
Theme issue: Hydrogen Storage and Generation
Registered Charity Number 207890
FEATURE ARTICLE FEATURE ARTICLE
Tejs Vegge et al. Roel van der Krol et al.
Ammonia for hydrogen storage: Solar hydrogen production with
challenges and opportunities nanostructured metal oxides 0959-9428(2008)18:20;1-#
 FEATURE ARTICLE www.rsc.org/materials | Journal of Materials Chemistry

Ammonia for hydrogen storage: challenges and opportunities†

Asbjørn Klerke,a Claus Hviid Christensen,a Jens K. Nørskovb and Tejs Vegge*c
Received 2nd January 2008, Accepted 13th March 2008
First published as an Advance Article on the web 3rd April 2008
DOI: 10.1039/b720020j

The possibility of using ammonia as a hydrogen carrier is discussed. Compared to other hydrogen
storage materials, ammonia has the advantages of a high hydrogen density, a well-developed technology
for synthesis and distribution, and easy catalytic decomposition. Compared to hydrocarbons and
alcohols, it has the advantage that there is no CO2 emission at the end user. The drawbacks are mainly
the toxicity of liquid ammonia and the problems related to trace amounts of ammonia in the hydrogen
Published on 03 April 2008. Downloaded by KIER on 5/20/2022 9:24:30 AM.

after decomposition. Storage of ammonia in metal ammine salts is discussed, and it is shown that this
maintains the high volumetric hydrogen density while alleviating the problems of handling the
ammonia. Some of the remaining challenges for research in ammonia as a hydrogen carrier are outlined.

Introduction highest potential hydrogen storage capacity. Meeting this target

alone is a significant challenge, which is accentuated by the fact
There are significant challenges related to the production, that up to half the storage capacity by weight is often lost in the
distribution and storage of hydrogen as a significant energy system integration.3 The search for ultra-high capacity materials
carrier of the future.1 The requirements of the transportation has led to the synthesis and characterization of more and more
sector have received particular attention, as outlined in the exotic and reactive compounds: complex metal hydrides like
FreedomCAR 2015 system targets:2 high reversible storage NaAlH4,4,5 LiAlH46,7 and AlH3,8 metal organic frameworks
capacity (3 kWh kg
1 or 9 wt.% hydrogen and 2.7 kWh L
1 or (MOFs),9 borohydrides like LiBH410,11 and Mg(BH4)2,12 desta-
0.081 (kg H2) L
1 system), fast kinetics (0.02 (g H2) s
1 kW
1), bilized borohydrides,13 ammonia borane14 and amide/imide
good reversibility (1500 cycles), ambient operating temperature systems.15 All display high potential storage capacities, but fall
(
40 to 60  C), low cost ($2 kWh
1 or $67 (kg H2)
1), high short on a number of the other targets. Currently, complex metal
hydrogen purity (<1 ppm CO) and safety (applicable standards). hydrides do not reach the target for reversible hydrogen storage
Among these required properties, research efforts have capacity, MOFs require operational temperatures around liquid
primarily been focused on finding materials that feature the nitrogen (
196  C), systems based on borohydrides are
practically irreversible and suffer from relatively slow kinetics,16
a
Center for Sustainable and Green Chemistry, Department of Chemistry, production and regeneration of ammonia borane is very costly,17
Building 206, Technical University of Denmark, DK-2800 Kgs. Lyngby, and the amide/imide systems have low reversible capacity.
Denmark
b
Center for Atomic-scale Materials Design, Department of Physics,
Among hydrogen’s main competitors as an energy carrier,
Building 310, Technical University of Denmark, DK-2800 Kgs. Lyngby, methane/natural gas and methanol offer clear advantages in
Denmark terms of energy capacity and distribution infrastructure, but
c
Materials Research Department, Risø National Laboratory for suffer from their intrinsic carbon content causing end-user CO2
Sustainable Energy, Building 228, Technical University of Denmark,
DK-4000 Roskilde, Denmark. E-mail: [email protected]
emissions, since onboard capture is not feasible. Furthermore,
† This paper is part of a Journal of Materials Chemistry theme issue on the IEA (International Energy Agency) recently projected that
hydrogen storage and generation. Guest editor: John Irvine. a gap between production and demand of natural gas may occur

Asbjørn Klerke earned his M.Sc. Tejs Vegge is a senior scientist

from the Technical University of and leader of the Energy
Denmark in 2006. He is Storage and Conversion group
currently pursuing a PhD degree at Risø National Laboratory for
at the Center for Sustainable Sustainable Energy, Technical
and Green Chemistry, under the University of Denmark. His
supervision of Prof. C. H. group focuses on characteriza-
Christensen. The focus of the tion and design of novel mate-
PhD project is hydrogen from rials and catalysts for hydrogen
ammonia stored in metal and ammonia storage, using
ammine complexes, with special integrated density functional
focus on the catalytic ammonia theory calculations and
decomposition over transition advanced neutron and X-ray
Asbjørn Klerke metal surfaces. Tejs Vegge scattering techniques.

2304 | J. Mater. Chem., 2008, 18, 2304–2310 This journal is ª The Royal Society of Chemistry 2008
 as soon as 2010–2012;18 effectively rendering this solution short production and transportation are well developed and already in
term and environmentally non-sustainable. widespread use on an industrial scale.26 The synthesis of
Aside from hydrogen, ammonia provides the only carbon-free ammonia in the Haber–Bosch process is among the largest and
chemical energy carrier solution for the transportation sector. best studied chemical processes in the world.27 The annual
Ammonia is not a greenhouse gas (GHG), and in light of the production of ammonia (mainly for agricultural use) exceeds 120
enormous global climate challenges with respect to GHG out- million tons.28 The synthesis reaction is performed under
lined, e.g., in the Kyoto protocol, combined with a high hydrogen 200–350 bar pressure over a multi-promoted iron catalyst oper-
density, NH3 could be an interesting alternative to hydrogen. ated at temperatures from 300 to 550  C in a reactor with two to
Ammonia is primarily produced from natural gas, but a large four catalyst beds giving around 15% conversion to ammonia.
number of coal-based ammonia plants are currently being put into The unconverted synthesis gas is recycled. The hydrogen for the
production;19 an aspect which is particularly important for coun- synthesis process is primarily produced from methane by steam
tries like China, with a rapidly expanding need for transportation, reforming and water gas shift reactions.29 Although ammonia
limited oil and gas resources, but enormous coal reserves.20 A synthesis is an exothermic reaction (Fig. 1), the practical
Published on 03 April 2008. Downloaded by KIER on 5/20/2022 9:24:30 AM.

highly developed ammonia infrastructure is already in existence, production of ammonia from hydrogen and nitrogen incurs
and with a centralized, coal-based production of ammonia with a small energy loss of 1.5 GJ t
1 compared to the 28.4 GJ t
1
simultaneous sequestration of the produced CO2, ammonia could energy stored in the ammonia. In comparison, the production of
offer not only a short term solution to needs of the transportation hydrogen has an energy cost of 8.0 GJ t
1.30 Hydrogen produc-
sector, but a long-term, zero-carbon-emission solution with a pro- tion is necessary for all proposed hydrogen-based energy carriers,
jected lifespan of 200 years.21 Still, ammonia is often disregarded as but in the case of ammonia, this can also come from other
a potential fuel in the transportation sector, primarily because of its sources than methane. The current price of ammonia is in the
too high toxicity,22 but recent research on reversible NH3 storage in range of $250–300 t
1, the equivalent of $1.4–1.7 (kg H2)
1.31 This
metal ammines (see Fig. 1), e.g. Mg(NH3)6Cl223 and price is mainly governed by the price of methane that constitutes
Ca(NH3)8Cl2,24 illustrates how a toxicity lower than that of gaso- above 75% of the running cost of ammonia production.
line can be obtained, while maintaining a practical, reversible Ammonia production is a prime candidate for CO2 seques-
storage capacity of up to 10 wt.% hydrogen.23,24 tration, because CO2 is being removed from the synthesis gas by
For use in polymer electrolyte membrane fuel cells (PEMFC), scrubbing to avoid deactivation of the ammonia synthesis
ammonia must first be decomposed into nitrogen and hydrogen, catalyst. Furthermore, ammonia plants are typically located
but NH3 can be decomposed catalytically at temperatures above close to natural gas reservoirs, where the captured CO2 might be
300  C (Fig. 1) with present-day catalysts. This is, however, an pumped back.30
area of active research and the development of superior decom- The hydrogen for ammonia synthesis can also be produced
position catalysts remains a worthwhile challenge for an from coal by gasification. Coal is interesting for hydrogen
ammonia-powered transportation sector. production for several reasons; first of all, coal is cheap
In the following, we review the status, challenges, and perspec- compared to other hydrocarbon energy sources,20 secondly, the
tives of ammonia stored in metal ammines as a carrier material in the coal reserves are large enough to last for at least 200 years,20 and
transportation sector, and place the results in the context of state-of- finally, coal can be efficiently converted into hydrogen by
the-art in hydrogen storage materials and existing transport fuels. gasification with a high efficiency.32–34 Thus, even though an
intense search for alternatives to coal is continuously on-going,
Ammonia synthesis the end result could be that we will need to exhaust all the
available coal reserves.
The use of ammonia as an energy carrier has several advan-
tages.25 One of these is that most of the technologies for
Ammonia safety
The risk of explosion is an essential safety concern in the
transportation sector, which can block the potential use of any
energy carrier. High capacity carriers are generally very reactive,
but significant differences can be observed in the flammability
and explosion limits (see Table 1).
From Table 1, it is clear that hydrogen is very flammable, as
are natural gas/methane, methanol, and gasoline vapors, whereas
ammonia is not flammable in air. With respect to explosion
limits, both hydrogen and ammonia require significantly higher
concentrations than gasoline vapor and natural gas.
The toxicity of ammonia is clearly also an important safety
aspect. Comparing the toxicity of liquid ammonia to that of
gasoline and methanol, liquid ammonia is approximately three
orders of magnitude higher in ‘‘apparent toxicity’’, i.e. the vapor
pressure relative to the toxicity, at room temperature.37 The
Fig. 1 The life-cycle of hydrogen stored as ammonia in metal ammines, apparent toxicities takes both the IDLH (Immediately
M(NH3)yX2. Dangerous to Life or Health) concentration and the vapor

This journal is ª The Royal Society of Chemistry 2008 J. Mater. Chem., 2008, 18, 2304–2310 | 2305
 Table 1 Vapor pressures of different energy carriers, their relative toxicities, flammability and explosion limits in air29,34–36

Vapor pressure, Apparent toxicity Flammability Explosion

pi, at 293 K/bar IDLHa (ppm) pi,293K/IDLH limits (in air) limits (in air)

Gasoline 0.047 750 63 1.4–7.6% 1.1–3.3%

Methanol 0.13 6000 21.6 6–36% 5.5–44%
Hydrogen — — — 4–74% 18.3–59%
Natural gas — — — 5.3–15% 5.7–14%
Liquid ammonia 8 300 2.7  104 — 16–25%
Mg(NH3)6Cl2 1.4  10
3 300 4.65 — —
a
NIOSH: Immediately Dangerous to Life or Health (IDLH) concentration.

pressure into account, since substances with a low vapor pressure methanol or methane, but these fuels need to be reformed and
Published on 03 April 2008. Downloaded by KIER on 5/20/2022 9:24:30 AM.

can have a low IDLH concentration and still rate safer than this gives a possibility of carbon poisoning of the anode.46,47 The
a substance with higher IDLH concentration but higher vapor carbon oxide-free ammonia can be fed directly into a high
pressure. temperature SOFC and the endothermic ammonia decomposi-
For ammonia stored in metal ammines, e.g. Mg(NH3)6Cl2, the tion reaction will help cool the SOFC, which combined with an
apparent toxicity falls below those of both gasoline and entropy gain means that ammonia as a direct fuel increases the
methanol; in comparison, the promising and intensely studied fuel cell performance at high temperatures.45
borohydrides12 are known to display partial release of highly In low temperature fuel cell systems, such as polymer elec-
flammable and toxic diborane during hydrogen desorption.38,39 trolyte membrane (PEMFC) and alkaline fuel cells (AFC),
Safety issues remain a main challenge for using ammonia in the ammonia needs to be decomposed to hydrogen and nitrogen
transportation sector, but the considerations outlined above are before it is fed to the fuel cell. In some applications, it might be
in line with a recent EU-funded report, which concluded that desirable to remove the inert nitrogen to supply a concentrated
‘‘the use of ammonia as a transport fuel wouldn’t cause more hydrogen stream. This can be done, e.g., by use of a hydrogen-
risks than currently used fuels’’.40 selective membrane system.48,49 The ammonia decomposition is
done catalytically over a suitable catalyst at temperatures from
Storage and transportation of ammonia ca. 300 to 520  C (see below). The ammonia decomposition
reaction is equilibrium-limited, so unconverted ammonia must
The infrastructure for handling and transporting bulk ammonia be taken properly into account when the fuel cell system is
is already well established. Liquid ammonia is being transported designed. If the system operates with an AFC, ammonia is
around the world in ships, pipelines, trains and trucks.25,41,42 The tolerated and can be recycled to the decomposition unit.50 For
transportation of liquid ammonia in closed systems to decen- proton-conducting fuel cells such as PEMFC and PAFC
tralized sites for production and regeneration of metal ammine (phosphoric acid fuel cell), ammonia is problematic since the
salts can minimize the overall cost and be scaled up while acidic electrolyte reacts with ammonia and thereby deactivates.
maintaining safety, but some additional steps need to be taken to This means that for continuous operation, the ammonia
minimize the risks involved.40 For automotive applications, the concentration must be below 1 ppm.51
end user should therefore not come in contact with liquid
ammonia, only the metal ammines produced and regenerated at Ammonia decomposition
a nearby location.
As outlined above, the toxicity and vapor pressure of liquid The ammonia decomposition reaction has traditionally been
ammonia make it undesirable for direct use in mobile applica- studied to gain more information about the ammonia synthesis
tions, mainly because of the potential risk of accidents where catalyst.52 This is slowly changing as ammonia is gradually
ammonia is released and because of the technical challenges of becoming more widely accepted as one of the more promising
securing the end user against contact with liquid ammonia during hydrogen carriers.53 At present, the commercially used catalyst
refilling and while performing periodical maintenance.25,42 for ammonia decomposition is nickel on alumina,54 which is
The storage of ammonia is currently done in two different mechanically strong and heat resistant, but further development
ways: for large quantities up to 50 000 t, ammonia is stored at 1 and optimization are needed.
bar and
33  C in insulated tanks. The temperature is kept down During the last ten years, different research groups have
by slow vaporization, and the ammonia vapor is continually developed and optimized catalysts specifically for use in ammonia
compressed back to a liquid. For small tanks, below 1500 t, decomposition for production of COx-free hydrogen.55–57 There
ammonia is stored under pressure in stainless steel spheres.29 now exists a general consensus that caesium-promoted ruthenium
supported on graphite is the best known catalyst.58,59 Several
studies have shown that barium is also a useful promoter for
Ammonia and fuel cells
ammonia decomposition on ruthenium surfaces.60
The direct use of ammonia in high temperature solid oxide fuel The currently best promoted ruthenium catalysts are suffi-
cells (SOFC) has been shown to work with a performance similar ciently active at temperatures from about 300  C. In Fig. 2, two
to pure hydrogen43,44—in fact, even surpassing it at temperatures different studies of the decomposition rate of ammonia are
above 200  C.45 The SOFC can also run on other fuels like shown. The first is a study by Raróg-Pilecka et al.,58 which shows

2306 | J. Mater. Chem., 2008, 18, 2304–2310 This journal is ª The Royal Society of Chemistry 2008
Published on 03 April 2008. Downloaded by KIER on 5/20/2022 9:24:30 AM.

Fig. 2 Dependence of the ammonia decomposition rate on xNH3 over the Ru/C catalyst at 400  C (left).58 The temperature dependence of the ammonia
decomposition rate for Ru/HSAG in micro-fabricated reactors61 (right).

the decomposition rate in a conventional reactor at various number of metal salts are known to form stable metal ammines,
ammonia concentrations and at a fixed temperature of 400  C. and these are generally well described in the classical chemistry
The second study shows the rate of ammonia decomposition as literature.64–66
a function of temperature, at a constant inlet gas concentration The solid storage of ammonia solves the safety issues of
of 20% ammonia in argon and for a constant flow rate. These driving with ammonia in liquid form under pressure. At the same
experiments were performed in micro-fabricated reactors, which time, the volumetric hydrogen density is high for the metal
feature excellent heat transport to the endothermic reaction.61 ammines compared to that of liquid ammonia and metal
The challenge with low temperature ammonia decomposition hydrides as shown in Fig. 3. Fig. 3 also compares the mass of the
is that the equilibrium conversion is dependent on the tempera- different hydrogen storage materials, and among the solid
ture and the reaction is endothermic, i.e. the concentration of materials, the metal ammines stand out by being lighter than the
ammonia at equilibrium increases with decreasing temperature. metal hydrides, but heavier than the liquids and gases, when the
At 425  C and 1 bar pressure, the equilibrium conversion of weight of the storage tank is not taken into consideration. The
ammonia is 98–99%, which makes unconverted ammonia current developments in high pressure hydrogen storage move
a significant challenge for the overall system efficiency if the towards lighter cylinders of composite materials but still with
temperature is lowered with a conventional reactor design. a high weight compared to the stored amount of gas, and the
Furthermore, the decomposition rate decreases significantly safety issues of driving and refueling to pressures between 35–70
when equilibrium is approached according to the general MPa need to be carefully considered. Liquid hydrogen is difficult
formula r ¼ r+(1
ß), where b describes the approach to equi- to handle on a small scale, mainly because of the very low
librium as a factor between 0 and 1.62 To avoid the equilibrium temperature of
252  C and the evaporation loss of 2–3% per
limitation, removal of the hydrogen through a membrane is an day. The amount of insulation also adds to the mass and volume
option,63 which would provide a pure hydrogen feed, free of of the overall storage system.67 Liquid ammonia needs to be
ammonia and nitrogen, and it will help to increase the conversion transported in pressurized tanks, and for the safety issues
by shifting the equilibrium further towards complete conversion. mentioned above, it appears unlikely to obtain the required
public acceptance. This leaves the solid storage of hydrogen and
ammonia as the best alternatives for mobile use.
Solid storage of ammonia
From the literature values, suitable metal ammine salts can be
The idea of using ammonia as a hydrogen carrier has been selected for a given application. So far, the best studied material
promoted by the further development of safe storage of ammonia is Mg(NH3)6Cl2, which was selected due to the low vapor
in solid form by binding it in metal ammine complexes.23 A large pressure of 2 mbar at room temperature (see Table 1) and the

Fig. 3 Mass and volume of 10 kg hydrogen stored reversibly by 8 different methods, based on the best obtained reversible densities reported in the
literature without considering the space or weight of the container.29,68–73

This journal is ª The Royal Society of Chemistry 2008 J. Mater. Chem., 2008, 18, 2304–2310 | 2307
 high gravimetric (9.19 wt.%) and volumetric (109 (g H2) L
1)
hydrogen density available from a compacted tablet.23 Recently,
research on Ca(NH3)8Cl8 has also shown promise for mobile
applications, since the hydrogen density is as high as 9.78 wt.%
and the release of ammonia is achieved at lower temperatures
than that of Mg(NH3)6Cl2, thereby reducing the energy needed
to desorb the ammonia, but also the stability at 60  C.2
The lower desorption temperature of Ca(NH3)8Cl8 results in
a higher ammonia vapor pressure at room temperature (0.7 bar),
but this is still an order of magnitude lower than that of liquid
ammonia.24 The number of metal ammines investigated for use
as hydrogen storage materials is still quite small, but due to their
similar chemical nature, the general findings can be used as
Published on 03 April 2008. Downloaded by KIER on 5/20/2022 9:24:30 AM.

a guideline for the behavior of other metal ammines. This has

been tested by detailed modeling using density functional theory
(DFT) calculations, which showed good agreement between
theoretical calculations and experimental data.74 DFT calcula- Fig. 4 Temperature programmed desorption of ammonia from
Ca(NH3)8Cl2 and Mg(NH3)6Cl2 measured with a heating of 1  C min
1
tions are able to reproduce the experimentally observed trend in
(adapted from ref. 24).
desorption enthalpies for various ammines at all decomposition
steps,24 making such calculations a valuable tool in the design
and prediction of novel metal ammine compounds with specific calculations, which show that metal ammine crystals can release
stabilities, e.g. by varying the metal or halide components in the ammonia by a change of the lattice constant and a slight internal
ammine salt. rearrangement; this can be interpreted as the individual
Release of ammonia from metal ammines has been investi- crystallites gradually shrinking to form pores.75
gated in several papers53,75,76 in the efforts to map out their Solid state storage of ammonia has the additional advantage
chemical and physical properties. Generally, metal ammines that the storage density can be considered constant at tempera-
reversibly absorb and desorb ammonia as shown in Fig. 1, where tures around 0  C. This is not the case for liquid ammonia as
the desired absorption and desorption enthalpies are also shown in Fig. 5, where the vapor pressure and volumetric
indicated. When the enthalpy of desorption of the first ammonia hydrogen density is plotted as function of the temperature in
is below 40 kJ mol
1, the vapor pressure at room temperature is a normal operating interval.
close to 1 bar, making absorption of ammonia in the salt difficult.
If the desorption enthalpy of the last ammonia is above 70 kJ
mol
1, the temperature needed for desorption is too high for
System integration
practical applications—with the possible exception of the SOFC, For practical use of ammonia and metal ammines as hydrogen
where sufficient heat will typically be easily available. carries on a larger scale, efficient system integration must be
The desorption kinetics of metal ammines has mainly been achieved to minimize the energy loss. This means that the waste
studied by temperature programmed desorption (see Fig. 4). heat from the fuel cell and the catalyzed ammonia decomposition
These experiments show that generally, the metal ammines have process must be used efficiently to release ammonia from the
very similar properties and can be formed into compact tablets metal ammine and to heat the ammonia feed gas.79 The safety
essentially without any voids that would lead to a loss of
volumetric storage capacity. Even in this compact form (>95% of
bulk density), they maintain their excellent properties with
respect to fast ammonia desorption kinetics.74 Generally, the
desorption characteristics can be estimated directly from the
desorption enthalpies and the fact that the metal ammines have
very similar entropies.77 A main limitation of ammonia desorp-
tion is heat transfer, since the activation energy for ammonia
desorption is relatively low and because the metal ammines are
poor heat conductors; hence the required heat is mainly
transported by the ammonia gas.78 Another reason for the fast
desorption is the intrinsic formation of a nanoporous, sponge-
like skeletal structure from the compact metal ammines during
ammonia desorption, which has recently been documented in
a detailed in situ small-angle X-ray scattering (SAXS) study.76
This is an important aspect, since compact tablets of metal
ammines maintain their physical shape even after desorption of
all the ammonia, corresponding to a loss of approximately half Fig. 5 The temperature dependence of the vapor pressure (solid lines)
the mass and three quarters of the occupied volume. The and volumetric hydrogen densities (dashed lines) for NH3 (l),
formation of nanopores has also been investigated by DFT Ca(NH3)8Cl2 and Mg(NH3)6Cl2.

2308 | J. Mater. Chem., 2008, 18, 2304–2310 This journal is ª The Royal Society of Chemistry 2008
 issues concerning liquid ammonia also make it important to hydrogen storage materials. Desorption of ammonia from metal
design a system where the end user does not come into contact ammines can easily be controlled and the desorption temperature
with liquid ammonia during refueling. This can be done by can be selected to fit a given application. The use of metal
handling the regeneration of the metal ammine salts at decen- ammines is currently being commercialized for automotive
tralized locations, where the heat of formation can also be reused. applications in connection with SCR catalysts to reduce the
If the metal ammine system is integrated with a high tempera- amount of NOx produced by diesel cars or trucks.82
ture SOFC, the waste heat from the fuel cell is more than sufficient There are several remaining challenges involved with an
to supply the heat required for ammonia release and ammonia ammonia mediated hydrogen economy. A technology must be
decomposition at the anode. As previously mentioned, ammonia developed to remove CO2 from the production system, but this is
has the advantage over reformed methane or methanol in a SOFC a general challenge for the idea of a carbon-free energy system.
that it is carbon-free and that no reformer is needed.80 The residual Alternatively, it would be very interesting if one could produce
ammonia and unused hydrogen can be sent directly to a (catalytic) ammonia electrochemically in a process analogous to the one
burner as is common practice with most operating SOFCs, and certain enzymes, the nitrogenases, use in nature.83 This would
Published on 03 April 2008. Downloaded by KIER on 5/20/2022 9:24:30 AM.

the ammonia will contribute to lowering NOx emissions.81 allow decentralized, sustainable ammonia production on the
Systems using ammonia together with low temperature fuel basis of photo-voltaic cells or wind power. Optimization of the
cells using proton conducting membranes are more complicated use of the energy stored in ammonia is also needed. This is
for various reasons. The heat from the fuel cell is available at low possible, either from better ammonia decomposition catalysts or
to intermediate temperatures, which is sufficient for complete from better direct ammonia fuel cells. Other alternatives could be
desorption of ammonia from Ca(NH3)8Cl2, but not for e.g. electrolysis of ammonia to hydrogen in an alkaline solution.84
Mg(NH3)6Cl2. However, any required additional heat can come The benefit of electrolysis is clean hydrogen production at low
from the ammonia decomposition reactor that will operate at temperature. Current fuel cell research is focusing on bridging
temperatures from 300  C, which is the minimum temperature if the temperature gap between SOFCs and PEMFCs. That could
the ammonia residue should be below 2%. The challenge is that give very efficient direct ammonia fuel cells with running
the decomposition is an endothermic process requiring 46 kJ temperatures from 400–600  C.85 In terms of storing ammonia in
(mol NH3)
1. The heat for ammonia decomposition needs to be metal ammines, there appear to be many unexplored opportu-
supplied by electrical heating or by burning hydrogen or nities for designing suitable salts that feature a desired desorption
ammonia. The remaining ammonia from the decomposition temperature, as well as in mixtures with other hydrogen storage
needs to be removed to keep the ammonia concentrations <1 materials.86 Finally, the establishment of fully integrated systems
ppm to avoid poisoning of the fuel cell. This can be done is required to achieve the necessary operating experience with all
efficiently by leading the gas through an acid media which can aspects of these systems.
reduce the ammonia concentration down to ppb levels. More
elegantly, the unconverted ammonia could also be removed by Acknowledgements
absorption into MgCl2 or CaCl2 or other even more reactive
metal salts, to (re)form the metal ammine. The Center for Sustainable and Green Chemistry is sponsored by
The possibilities are plentiful and very little research effort has the Danish National Research Foundation and the Center for
so far gone into this technology compared to other related Atomic-scale Materials Design is supported by the Lundbeck
subjects, such as CO poisoning in PEMFC or DMFC. Foundation. The authors acknowledge the European Commis-
sion DG Research (contracts SES6-2006-51827/NESSHy and
MRTN-CT-2006-032474/HYDROGEN) and the Danish
Outlook and perspective Council for Strategic Research (# 2104-05-0016).
The utilization of ammonia as a potential hydrogen carrier is still
receiving limited attention compared to e.g. the complex metal References
hydrides and other direct hydrogen storage methods, even though 1 G. Crabtree, M. Dresselhaus and M. Buchanan, Phys. Today, 2004,
the ammonia technology seems to be significantly closer to market. 57(12), 39.
The research devoted to the use of carbon-based hydrogen carriers 2 http://www1.eere.energy.gov/hydrogenandfuelcells/pdfs/freedomcar_
as light hydrocarbons and methanol has also been substantial, targets_explanations.pdf.
3 D. Mosher, X. Tang and S. Arsenault, High Density Hydrogen
although the challenges regarding CO2 for the end user remaining Storage System Demonstration Using NaAlH4 Based Complex
unsolved. In this perspective, ammonia stored in metal ammines Hydrides, FY 2006 Annual Progress Report, DoE Hydrogen
appears as an attractive alternative which can solve many of the Program, 2006, pp. 281–284.
4 D. Sun, S. S. Srinivasan, G. Chen and C. M. Jensen,
main targets setup for a hydrogen storage material, i.e. fast Rehydrogenation, J. Alloys Compd., 2004, 373(1–2), 265–269.
kinetics, high hydrogen storage capability, high availability and 5 T. Vegge, Phys. Chem. Chem. Phys., 2006, 8, 4853–4861.
low cost. The less attractive sides to ammonia as a hydrogen carrier 6 G. Sandrock, J. Reilly, J. Greatz, W.-M. Zhou, J. Johnson and
include the current methane-based production of ammonia J. Wegrzyn, J. Alloys Compd., 2006, 421, 185–189.
7 A. Andreasen, T. Vegge and A. S. Pedersen, J. Solid State Chem.,
without carbon sequestration, and the toxicity of the liquid. 2005, 178, 3672–3678.
Recent research into solid state storage of ammonia in metal 8 Sandrock, J. Reilly, J. Greatz, W.-M. Zhou, J. Johnson and
ammines has solved many of the problems regarding end user J. Wegrzyn, Appl. Phys. A, 2005, A80, 687–690.
9 H. Furukawa, M. A. Miller and O. M. Yaghi, J. Mater. Chem., 2007,
safety, since the main risk of exposure to liquid ammonia has
17, 3197.
been avoided. The metal ammines also have higher volumetric 10 A. Züttel, P. Wenger, S. Rentsch, P. Sudan, P. Mauron and
hydrogen densities than liquid ammonia and other proposed C. Ernmenegger, J. Power Sources, 2003, 118, 1.

This journal is ª The Royal Society of Chemistry 2008 J. Mater. Chem., 2008, 18, 2304–2310 | 2309
 11 Z. qodziana and T. Vegge, Phys. Rev. Lett., 2004, 93, 145501. 51 R. Halseid, P. J. S. Vie and R. Tunold, J. Power Sources, 2006, 154,
12 K. Ch1opek, C. Frommen, A. Leon, O. Zabara and M. Fichtner, 343–350.
J. Mater. Chem., 2007, 17, 3496. 52 J. B. Hansen, in Ammonia, Catalysis and Manufacture, ed. A. Nielsen,
13 J. J. Vajo, S. L. Skeith and F. Mertens, J. Phys. Chem. B, 2005, 109, 3719. Springer, Heidelberg, 1995, pp. 149–198.
14 A. Feaver, S. Sepehri, P. Shamberger, A. Stowe, T. Autrey and 53 C. H. Christensen, T. Johannessen, R. Z. Sørensen and J. K. Nørskov,
G. Cao, J. Phys. Chem. B, 2007, 111, 7469–7472. Catal. Today, 2006, 111, 140–144.
15 W. Lohstroh and M. Fichtner, J. Alloys Compd., 2007, 332, 446–447. 54 http://www.nikki-chem.co.jp/eng/products/pro3_2.html.
16 S. Orimo, Y. Nakamori, G. Kitahara, K. Miwa, N. Ohba, S. Towata 55 A. S. Chellappa, C. M. Fisher and W. J. Thomson, Appl. Catal., A,
and A. Zuttel, J. Alloys Compd., 2005, 447, 404–406. 2002, 227, 231–240.
17 P. V. Ramachandran and P. D. Gagare, Inorg. Chem., 2007, 46, 56 T. V. Choudhary, C. Sivadinarayana and D. W. Goodman,
7810–7817. Catal. Lett., 2001, 72, 197–201.
18 http://www.iea.org. 57 J. C. Ganley, E. G. Seebauer and R. I. Masel, J. Power Sources, 2004,
19 J. F. Clausen and C. A. Zee, TRW-Systems and Energy, EPA contract 137, 53–61.
68-02-2635. 58 W. Raróg-Pilecka, D. Szmigeel, Z. Kowalczyk, S. Jodis and
20 International Energy Agency (IEA), Key World Energy Statistics, 2007. J. Zielinski, J. Catal., 2003, 218, 465–469.
21 BP Statistical Review of World Energy, June 2007. 59 S.-F. Yin, B. Q. Xu, X. P. Zhou and C. T. Au, Appl. Catal., A, 2004,
22 http://www.cdc.gov/niosh/. 277, 1–9.
Published on 03 April 2008. Downloaded by KIER on 5/20/2022 9:24:30 AM.

23 C. H. Christensen, R. Z. Sørensen, T. Johannessen, U. Quaade, 60 R. Z. Sørensen, L. J. E. Nielsen, S. Jensen, O. Hansen, T. Johannesen,

K. Honkala, T. D. Elmøe, R. Køhler and J. K. Nørskov, J. Mater. U. Quaade and C. H. Christensen, Catal. Commun., 2005, 6, 229–232.
Chem., 2005, 15, 4106–4108. 61 R. Z. Sørensen, A. Klerke, U. Quaade, S. Jensen, O. Hansen and
24 R. Z. Sørensen, J. S. Hummelshøj, A. Klerke, J. B. Reeves, T. Vegge, C. H. Christensen, Catal. Lett., 2006, 112, 77–81.
J. K. Nørskov and C. H. Christensen, Indirect, reversible high-density 62 A. Boisen, S. Dahl, J. K. Nørskov and C. H. Christensen, J. Catal.,
hydrogen storage in compact metal ammine salts, J. Am. Chem. Soc., 2005, 230, 309–312.
submitted. 63 http://www.powerandenergy.com/hy_separator.shtml.
25 P. J. Feibelman, Phys. Today, 2005, 58, 13–14. 64 F. Ephraim, Ber. Dtsch. Chem. Ges., 1912, 45, 1322–1331.
26 L. Green, Int. J. Hydrogen Energy, 1982, 7, 355–359. 65 F. Ephraim, Z. Phys. Chem., 1913, 81, 513–542.
27 R. Schlögl, Angew. Chem., Int. Ed., 2003, 42, 2004–2008. 66 W. Biltz, Z. Anorg. Allg. Chem., 1923, 130, 93–139.
28 International Fertilizer Industry Association, World Ammonia (NH3) 67 A. Zûttel, Mater. Today, 2003, 6(September), 24–33.
Statistics by Region, 2006, http://www.fertilizer.org/ifa/statistics.asp. 68 D. Mosher, X. Tang and S. Arsenault, High Density Hydrogen
29 M. Appl, Ullmann’s Encyclopedia of Industrial Chemistry: Ammonia, Storage System Demonstration Using NaAlH4 Based Complex
Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 2007. Hydrides, FY 2006 Annual Progress Report, DoE Hydrogen
30 I. Dybkjaer, in Ammonia, Catalysis and Manufacture, ed. A. Nielsen, Program, 2006, pp. 281–284.
Springer, Heidelberg, 1995, pp. 199–308. 69 X. Liu, Y. Zhu and L. Li, Int. J. Hydrogen Energy, 2007, 32,
31 ICIS pricing Chemical Pricing Reports, Ammonia, http:// 2450–2454.
www.icispricing.com. 70 K. Nomura, S. Fujiwara, H. Hayakawa, E. Akiba, Y. Ishido and
32 J. Wu, Y. Fang, H. Peng and Y. Wang, Fuel Process. Technol., 2004, S. Ono, J. Less-Common Met., 1991, 169, 9–17.
86, 261–266. 71 E. Suissa, I. Jacob and Z. Hadari, J. Less-Common Met., 1984, 104,
33 W. de Jong, Ö. Ünal, J. Andries, K. R. G. Hein and H. Spliethoff, 287–295.
Biomass Bioenerg., 2003, 25, 59–83. 72 M. A. El-Osairy, I. A. El-Osery, A. M. Metwally and M. A. Hassan,
34 P. Häussinger, R. Löhmuller and A. M. Watson, Ullmann’s Int. J. Hydrogen Energy, 1993, 18(6), 517–524.
Encyclopedia of Industrial Chemistry: Hydrogen, Wiley-VCH Verlag 73 K. J. Laidler; and J. H. Meiser Physical Chemistry, 3rd edn,
GmbH & Co. KGaA, Weinheim, 2000. Houghton Mifflin Company, New York, 1999.
35 A. English, J. Rovner, J. Brown and S. Davies, Kirk–Othmer 74 T. Vegge, R. Z. Sørensen, A. Klerke, J. S. Hummelshøj,
Encyclopedia of Chemical Technology: Methanol, John Wiley & T. Johannessen, J. K. Nørskov and C. H. Christensen, Indirect
Sons, Inc., New York, 2005. hydrogen storage in metal ammines, in Solid State Hydrogen
36 E. Fiedler, G. Grossmann, D. B. Kersebohm, G. Weiss and C. Witte, Storage: Materials and Chemistry, ed. G. Walker, Woodhead
Ullmann’s Encyclopedia of Industrial Chemistry: Methanol, Publishing, Cambridge, UK, 2008, in press.
Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 2000. 75 J. S. Hummelshøj, R. Z. Sørensen, M. Y. Kustova, T. Johannesen,
37 http://www.cdc.gov/niosh/idlh/intridl4.html. J. K. Nørskov and C. H. Christensen, J. Am. Chem. Soc., 2006,
38 J. Kostka, W. Lohstroh, M. Fichtner and H. Hahn, J. Phys. Chem. C, 128, 16–17.
2007, 111, 14026. 76 H. S. Jacobsen, H. A. Hansen, J. W. Andreasen, Q. Shi,
39 E. Jeon and Y. W. Cho, J. Alloys Compd., 2006, 422, 273. A. Andreasen, R. Feidenhans’l, M. M. Nielsen, K. Ståhl and
40 N. J. Duijm, F. Markert and J. L. Paulsen, Safety assessment of T. Vegge, Chem. Phys. Lett., 2007, 441, 255–260.
ammonia as a transport fuel, Risø-R-1504(EN), 2005, http:// 77 E. Lepinasse and B. Spinner, Rev. Int. Froid, 1994, 17, 309–321.
www.risoe.dk/rispubl/SYS/syspdf/ris-r-1504.pdf. 78 T. D. Elmøe, R. Z. Sørensen, U. Quaade, C. H. Christensen,
41 T. Eggmann, Kirk-Othmer Encyclopedia of Chemical Technology: J. K. Nørskov and T. Johannessen, Chem. Eng. Sci., 2006, 61,
Ammonia, John Wiley & Sons, Inc., New York, 2001. 2618–2625.
42 G. Thomas and G. Parks, Potential Roles of Ammonia in a Hydrogen 79 T. Johannessen and R. Z. Sørensen, TCE, 2005, 773, 30–33.
Economy, U.S. Department of Energy, 2006. 80 N. Maffei, L. Pelletier, J. P. Charland and A. McFarlan, Fuel Cells,
43 G. G. M. Fournier, I. W. Cumming and K. Hellgardt, J. Power 2007, 4, 323–328.
Sources, 2006, 162, 198–206. 81 E. Fontell, M. Jussila, J. B. Hansen, J. Pålsson, T. Kivisaari and
44 G. Meng, C. Jiang, J. Ma, Q. Ma and X. Liu, J. Power Sources, 2007, J. U. Nielsen, Wärtsilä – Haldor Topsoe SOFC Test System,
173, 189. presented at SOFC-IX, 2005, Quebec, Canada, May 15–20, http://
45 C. H. Christensen, presented at International Symposium on Materials www.topsoefuelcell.com/tofc/PDF/Wartsila%20-%20Haldor%20
Issues in a Hydrogen Economy, Richmond, Virginia, 2007. Topsoe%20SOFC%20Test%20System.pdf.
46 C. M. Finnerty, N. J. Coe, R. H. Cunningham and R. M. Ormerod, 82 http://www.amminex.net/.
Catal. Today, 1998, 46, 137–145. 83 T. H. Rod, A. Logadottir and J. K. Nørskov, J. Chem. Phys., 2000,
47 M. Itome and A. E. Nelson, Catal. Lett., 2006, 106, 21–27. 112, 5343.
48 R. Metkemeijer and P. Achard, Int. J. Hydrogen Energy, 1994, 19, 84 F. Vitse, M. Cooper and G. G. Botte, J. Power Sources, 2005, 142,
535–542. 18–26.
49 R. Metkemeijer and P. Achard, J. Power Sources, 1994, 49, 271–282. 85 N. Maffei, L. Pelletier, J. P. Charland and A. McFarlan, J. Power
50 K. Kordesch, V. Hacker, J. Gsellmann, M. Cifrain, G. Faleschini, Sources, 2005, 140, 264–267.
P. Enzinger, R. Fankhauser, M. Ortner, M. Muhr and 86 P. A. Chater, W. I. F. David, S. R. Johnson, P. P. Edwards and
R. R. Aronson, J. Power Sources, 2000, 86, 162–165. P. A. Anderson, Chem. Commun., 2006, 2439.

2310 | J. Mater. Chem., 2008, 18, 2304–2310 This journal is ª The Royal Society of Chemistry 2008