Roads2HyCom Hydrogen and Fuel Cell Wiki

http://www.ika.rwth-aachen.de/r2h
Effects of Hydrogen on Materials
Document Tracking ID 5015
Newest revision:
http://www.ika.rwth-aachen.de/r2h/Effects_of_Hydrogen_on_Materials
This revision:
http://www.ika.rwth-aachen.de/r2h/index.php?title=Effects_of_Hydrogen_on_Materials&oldid=5015
Click here to read our disclaimer
Print date: Wed, 13 Aug 2014 04:15:31 +0000
About Roads2HyCom
Roads2HyCom is a project supported by the European Commission's Framework Six program. Its purpose is to assess and monitor
hydrogen and fuel cell technologies for stationary and mobile energy applications.This is done by considering what the technology is capable
of, relative to current and future hydrogen infrastructures and energy resources, and the needs of communities that may be early adopters of
the technology.By doing this, the project will support the Commission and stakeholders in planning future research activities. Project main
website: http://www.roads2hy.com
HyLights, Roads2HyCom and the Hydrogen and Fuel Cells Technology Platform (HFP)
The European Commission is supporting the Coordination Action "HyLights" and the Integrated Project "Roads2HyCom" in the field of
Hydrogen and Fuel Cells. The two projects support the Commission in the monitoring and coordination of ongoing activities of the HFP, and
provide input to the HFP for the planning and preparation of future research and demonstration activities within an integrated EU strategy.
The two projects are complementary and are working in close coordination. HyLights focuses on the preparation of the large scale
demonstration for transport applications, while Roads2Hycom focuses on identifying opportunities for research activities relative to the needs
of industrial stakeholders and Hydrogen Communities that could contribute to the early adoption of hydrogen as a universal energy vector.
Further information on HyLights is available on the project web-site at http://www.hylights.org.
Contents
1 Degradation mechanisms
1.1 Degradation of
steels

1.2 Polymer
Degradation

1.3 Degradation of
Copper

1.4 Other
Degradation
Mechanisms


2 Permeation and Diffusion
2.1 Permeation
through Polymers

2.2 Diffusion
through Steels

2.3 Diffusion
through Copper

2.4 Leakage from
Connections


Degradation mechanisms
Before achieving total hydrogen economy, it is very likely that at least in the transition time existing
infrastructure is used to transport hydrogen. The discussion beneath therefore also discusses the materials
used in the already present natural gas infrastructure. In the natural gas infrastructure, the main ducts are
made out of steel, while locally polymer materials are used, and finally, copper is the material of preference
in household applications. The possible effect of gaseous hydrogen on each of these three types of materials
is discussed below.
In general, in the discussion below, no specific steel or material grades are recommended. The reason being:
the current document, once prepared, will not be updated. New experience in using alloys for application in
hydrogen is, on the other hand, constantly being generated. This means that if a certain grade would be
recommended here, in the mean time better grades can have been developped. Any material grade mentioned
below should therefore be seen as an example only. But upon applying the knowledge in real, practical
systems one must always check for the recommendations in the currently existing guidelines, which are
regularly updated. Examples of such updated guidelines are the NASA Safety standard for hydrogen and
hydrogen systems, and the EIGA Gaseous Hydrogen stations document.
Effects of Hydrogen on Materials
1/6
Degradation of steels
Several degradation mechanisms are known for hydrogen originating from the gas phase and hydrogen
formed by electrochemical loading. The discussion below is largely based on the papers by Hirth et al..
Hydrogen embrittlement
Hydrogen can make steels brittle at ambient temperatures. This is possible if the hydrogen molecules can
dissociate on the steel surface, and if subsequently the hydrogen atoms formed can diffuse into the bulk of
the steel. The mechanism of hydrogen embrittlement is not clear yet. On one hand, hydrogen is suspected to
decrease the atomic bonds in the iron lattice (decohesion theory), on the other hand it is suspected to cause an
increased local plastic deformation. The consequence then may be a lowered fracture toughness and crack
growth resistance, thus leading to a higher probability of fracture. The measure of embrittlement depends on
the hydrogen concentration, the type of steel, and the microstructure of the steel.
Most literature, however, states that the influence of hydrogen at normal surrounding conditions (-20 to +50
C) is minimal. This apparent contradiction is due to the fact that the usual oxide on a steel surface prevents
the hydrogen from dissociating and diffusing into the bulk of the steel. This is not necessarily due to the
oxide skin being a physical barrier for the hydrogen molecule, but more due to the blocking of the catalytic
activity of a clean (non-oxidated) metal surface.
There are also experiments with direct evidence for embrittlement occuring in the case of a clean surface. In
practice, a clean metal surface can be present when through a mechanical load local plastic deformation
occurs. This is possible, if in the presence of crack-shaped defect, the yield limit of the material is locally
exceeded as a consequence of (low frequency) changing loads of sufficient size. Research is going on on this
topic in the European NaturalHy project, with numerous experiments on crack growth under industrial
conditions (performed at TNO). A very thorough review of the theories behind, and the effects of hydrogen
embrittlement is presented in Deliverable 6 of that project. Changing loads can amongst others arise through
a change in hydrogen pressure, a change in temperature or by a physical load. Apart from the NaturalHy
project, also much research had been performed on this subject in the HySafe project.
Another problem arises when ultrapure (say, better than 4.0 i.e., >99,99 vol% purity) hydrogen is being used.
This high purity is sometimes required to prevent poisoning of fuel cells. But, as a side effect, it may also
cause a deterioration of the oxide scale on the steel, thus allowing the hydrogen to diffuse into steel. Less
pure hydrogen can contain enough traces of oxygen, carbon monoxide or water to maintain a good oxide
scale, but with ultrapure hydrogen this is not necessarily the case. Not all trace impurities help in preventing
hydrogen embrittlement, hydrogen sulfide for one increases the attack.
Hydrogen embrittlement is divided into three categories, hydrogen environment embrittlement (HEE),
hydrogen stress cracking (HSC) and loss in tensile ductility. These are somewhat related, these categories are
not mutually exclusive. HEE is usually associated with gaseous hydrogen, while HSC is associated with
hydrogen form other sources. The attack is found in (under at least some conditions) in amongst others steels,
Ni-base alloys, metastable stainless steels, titanium alloys, carbon and low alloy steels.
In conclusion, if there are mechanical loads and/or ultrapure hydrogen present, then hydrogen can enter steel
at room temperature. This in turn causes weakening of the steel, which makes it more likely to fail. Care
Effects of Hydrogen on Materials
2/6
should therefore be taken to choose an appropriate material with the right dimensions, and to avoid
mechanical loads. From a practical point of view, it is thus far not possible to assess whether a material is
brittle or not by means of non-destructive testing. However, cracks formed due to hydrogen embrittlement
can be detected using the standard non-destructive testing techniques, like for example acoustic emission.
Blistering
In electrolysis conditions, atomic hydrogen can enter the steel, and then locally recombine at a defect. This
then causes a local very high pressure of di-hydrogen, ultimately causing the formation of a blister. This kind
of degradation is most frequently observed when the alloy is charged with hydrogen under electrolytical
conditions.
Hydrogen attack
At high temperatures, gaseous hydrogen can attack carbon and low alloy steels. Hydrogen is thought to react
with carbon in the steel to form methane. This results in decarburisation and the formation of fissures, thus
leading to premature failure. This kind of degradation mechanism can be mitigated by the use of steels with
stronger carbide formers.
Other mechanisms
Other degradation mechanisms relevant for this review include microperforation by high pressure hydrogen
(found in compressors), and hydride formation (found for several elements including titanium).
Polymer Degradation
Research towards the fysical interaction between hydrogen and polymer materials is not known. In the 70s
and 80s Brown has performed research towards the influence of helium, nitrogen, argon, oxygen, and water
on the mechanical behaviour of some polymer materials (PE, PP, PMMA, PS, PET). The findings of this
study are that some effects are noticeable till about 100 C above the boiling point of each gas. The boiling
point of hydrogen is -253 C. So, based on the research by Brown it can be expected that hydrogen at normal
surrounding conditions will have no fysical interaction with polymer materials to the extent that it will cause
a measurable deterioration in properties.
Chemical attack by hydrogen is also unlikely. Unsaturated hydrocarbons can be hydrogenated in a catalytics
process, in which the unsaturated bond is converted into a saturated one. However, the typical materials used
for hydrogen transport (PE, PVC) do not contain carbon carbon double bonds that can be saturated. There is
no further literature known in which there is reported increased polymer material degradation due to
hydrogen.
Conclusion: increased degradation due to hydrogen in unlikely, but hard evidence from literature is missing.
Effects of Hydrogen on Materials
3/6
Degradation of Copper
Hydrogen embrittlement of technically pure copper is only known to occur if the alloy contains oxygen (in
the form of copper oxides). This form of degradation is not very likely because ducts of copper are usually
produced from oxygen-free copper. Other forms of hydrogen embrittlement are not known. Further more,
copper is usually only applied at low pressures and low temperatures, thus reducing the risk of material
degradation.
Other Degradation Mechanisms
In many hydrogen storage tanks, a combination of a carbon fiber reinforcement and a hydrogen impermeable
liner is used. Depending on the application, the liner may be metallic or made from a polymer.
In case the liner is metallic, a problem can arise due to the difference in thermal expansion coefficient of the
metal and the carbon fiber. This is, because a change in the temperature of the tank (due to filling and
emptying, or due to radiation from the sun or outside weather conditions) will cause the materials to expand
and shrink repeatedly. If the liner and the reinforcement do not epxand at the same rate, the liner may
separate from the reinforcement. This results in a weakening of the storage tank and may cause a leak in due
time.
In the case the liner is made from a polymer material, the difference in thermal expansion coefficient is less,
but the barrier against hydrogen permeation is also lower (as discussed below).
Permeation and Diffusion
Permeation through Polymers
Due to the small size of hydrogen molecules, diffusion of molecular hydrogen through polymer materials is
relatively fast. This means that, when tubes from ordinary polymer materials are used in confined,
non-ventilated spaces, hydrogen can slowly seep through the wall of the tube and accumulate in the
surrounding space. This is a risk that is especially clear for insulated and/or buried tubes and inside closed
rooms and closets, where the hydrogen can build up. In a later stage it can then ignite or even explode.
A typical permeation rate for one kilometer of transport tube with an outer diameter of 110 mm, wall
thickness of 10 mm, made out of HDPE, filled with 10 bar hydrogen is 70 liters per 24 hours. The higher the
hydrogen pressure inside the tubing and the thinner the wall thickness, the higher the permeation rate will be
(see also Appendix 1). In comparison, the permeation values for natural gas are 10-50 times lower than for
hydrogen.
Effects of Hydrogen on Materials
4/6
Of course by using different materials or coatings the diffusivity barrier across the tube can made larger.
However, the natural gas pipeline system, which is currently the most likely candidate to be used for
hydrogen transport, is already in place, and the materials used in the end sections are often polymers (like
PE100 or HDPE, PVC-CPE, and MDPE). However, the tubing inside houses is usually made from copper.
As for the case of mobile applications of hydrogen, often carbon fiber reinforced hydrogen storage tanks are
used. These tanks contain a liner as a diffusion barrier to the hydrogen. This liner can be both metallic or
polymeric in form. As described above, thermal cycling of such liner/reinforcement combinations can cause
premature failure of such storage tanks.
In addition to the problem of diffusion through polymer tube walls, there is the the problem of sealing.
Connections between different pieces of tubing (which be polymer material, but can also be steel) will need
to have better seals than the ones conventionally used for natural gas. Other wise the hydrogen will leak out
through the seal, yielding similar risks as for the case of hydrogen diffusion through the polymer tube wall.
Conclusion: some permeation of molecular hydrogen will take place when using polymer materials. The
safety risk will strongly depend on the possibility that hydrogen can accumulate, i.e., by being trapped in
closed spaces.
Diffusion through Steels
Diffusion of hydrogen through steels is in general not a significant problem, since hydrogen diffusion
through steels is much slower than through polymer materials. Problems are only expected to occur at
possible defects, connections and flanges, where defective seals can facilitate hydrogen leakage. It is
important to note that, unlike with polymer materials, molecular hydrogen first has to split up into hydrogen
atoms before it can diffuse into a steel.
Diffusion through Copper
Diffusion through copper is not a significant problem, since copper is only used at low pressure and the
diffusion rate constant of hydorgen in copper is small. Problems are only expected to occur at possible
defects, connections and flanges, where defective seals can facilitate hydrogen leakage.
Leakage from Connections
The general recommendation to prevent leakage from connections in tubes or pipes, is that all connections
should be welded. The use of flanged or threaded connections should be kept to a minimum, since they can
act as a source for hydrogen leakage. The use of compression fittings is not recommended. For more detailed
Effects of Hydrogen on Materials
5/6
information, please refer to the EIGA guidelines.
Safety and Security Assessment
Safety and Security Assessment | Hydrogen Safety Measures | Hydrogen Accidents | Hydrogen Regulations
Hydrogen and Fuel Cell Technologies
Socio-economic Assessment | Safety and Security Assessment | Technology Assessment | Mapping of R&D
Activities | Hydrogen and Fuel Cell Technology Watch
Effects of Hydrogen on Materials
6/6