Update: my peer reviewed article in Energy Science and Engineering (Aug 2024) summarizes some of the points in this piece.
https://scijournals.onlinelibrary.wiley.com/doi/10.1002/ese3.1861
A recent study https://www.dvgw.de/medien/dvgw/forschung/berichte/g202006-sywesth2-steel-dvgw.pdf
carried out by Open Grid Europe GmbH with the assistance of the University of Stuttgart, paid for by DVGW (Deutscher Verein des Gas- und Wasserfaches- the German Association for Gas and Water) did rather careful, extensive and thorough testing of a wide and characteristic variety of pipeline steels in hydrogen atmospheres of various pressures.
The report draws a shocking conclusion that has been parroted on high by the #hopium dealers at Hydrogen Europe and various other pro-hydrogen lobby groups:
“Hence, all pipeline steel grades investigated in this project are fundamentally suitable for hydrogen transmission.”
Well that’s it- case closed then! All gas transmission pipelines are fundamentally suitable to transmit pure hydrogen! The fossil gas distribution industry is saved! The “sunk cost” of all that infrastructure is rescued! And all those worry-warts like myself who were pointing out the hazards of such a conversion were just wrong!
While I’m totally happy to find out when I’m wrong, so I can change my opinion to be consistent with the measured facts, I’m afraid that in this case, the answer is rather more complex than just “Paul Martin is wrong- gas pipelines are safe for use with hydrogen”.
TL&DR Summary: extensive materials testing in this study proves that molecular hydrogen does cause pipeline materials to fatigue crack faster (up to 30 times faster than they would in natural gas) and to lose as much as 1/2 their fracture toughness (making them more likely to break). But if you reduce the design pressure of the pipeline substantially- to 1/2 to 1/3 of its original design pressure- the gas industry would consider that “safe enough” under the rules intended for designing new hydrogen pipelines. That would of course drop the capacity of the existing gas pipeline by a lot, requiring that either the lower capacity be accepted or the line be “twinned” or replaced if it were switched to hydrogen. And a host of other problems per my previous article on this topic, are also unresolved.
What Was Studied
Modern gas transmission pipelines are generally made of low alloy, high yield strength carbon steels typified by API 5L grades X42 through X100. The study examined steels commonly used in pipeline service in Germany, ranging from mild steels of low yield strength such as historical grade St35 (35,000 psi yield), through API 5L X80 (80,000 psi yield strength), including some steels used in the manufacture of pipeline components such as valve bodies. In many cases, specimens were prepared in such a way that the bulk material of the pipeline, a typical weld deposit and the heat affected zone of the parent metal were all tested. Thorough, careful work.
The specimens were tested in a cyclic (fatigue) testing apparatus which could be filled with hydrogen atmospheres of varying pressures. The major factors examined were fatigue crack growth rate and fracture toughness, because these parameters are known, not merely suspected, to be affected in a detrimental way in these steels by the presence of hydrogen.
What They Found
To hopefully nobody’s surprise, the testing found that the presence of hydrogen does greatly accelerate fatigue crack growth, and significantly negatively affects fracture toughness in the tested steels.
Specifically, they were able to build a good model of the fatigue cracking behaviour of these materials. They found, to quote p. 169 of the study:
- At lower stress intensities and hydrogen pressure, crack growth is comparable with crack growth in air or natural gas
- At higher hydrogen pressures, crack growth very rapidly approaches the behaviour at a partial pressure of H2 = 100 bar (~ 1500 psi) , even at lower stress intensities
- The position of the transitional area from “slow” crack growth to H2-typical rapid crack growth (my emphasis) depends on the hydrogen pressure, although it cannot be predicted exactly
They also found that fracture toughness Kic was negatively affected by the presence of hydrogen. Fracture toughness was, as expected, reduced even in low yield strength steels like St35, even when small amounts of hydrogen were added. Fracture toughness was strongly reduced in higher yield strength steels such as L485 (a common modern pipeline steel used in Germany). Even 0.2 atm H2 dropped fracture toughness greatly, and fracture toughness continued to drop steeply as pH2 was increased.
Hmm…so how did they draw the conclusion that these steels are “fundamentally suitable for hydrogen transmission”?
By comparison against the requirements of the hydrogen pipeline design/fabrication code/standard, ASME B31.12.
The study found that the crack growth rate was consistent with the assumptions used in the hydrogen design de-rating method used in B31.12. They also found that in all the steels tested at pH2 = 100 bar, the minimum required Kic value of 55 MPa/m^½ was exceeded.
The TL&DR conclusion here is as follows: yes, hydrogen causes pipeline steels to fatigue crack faster and to lose fracture toughness to a considerable extent, relative to the same steels used in air or natural gas. But that’s okay…because it doesn’t crack faster or lose more fracture resistance than expected in a design code used for dedicated hydrogen pipelines.
A design code that fossil gas pipelines are not designed and fabricated to, by the way!
What Does This Mean? Hydrogen’s Impact on Pipeline Design Pressure
Transmission pipelines are designed, fabricated and inspected in accordance with codes and standards which vary from nation to nation. The common standards in use in the USA, which serve as a reference standard in many other nations, are ASME B31.8 for fossil gas and other fuel pipelines, and ASME B31.12 for bespoke hydrogen pipelines. While the latter do exist (some 3000 km of dedicated hydrogen pipelines in the USA alone), the former are much more extensive (some 3,000,000 km of them in the USA). And if you a) own such a pipeline or b) depend on it to supply the gas distribution network you own, and c) know that without hydrogen, you’ll be out of business post decarbonization, you will be very motivated to conclude that you can re-use your gas pipeline to carry hydrogen in the future. Hmm, sounds like a bit of a potential conflict of interest, no?
In both ASME standards, the design pressure of the pipeline is determined via a modification of Barlow’s hoop stress equation, involving the specified minimum yield strength of the piping (S), the pipe nominal wall thickness (t), pipe nominal outer diameter (D), a longitudinal joint factor (E), a temperature de-rating factor T, and a design safety factor F, which depends on service class/severity and location. For hydrogen per B31.12, a new factor Hf, a “material performance factor” is applied to effectively de-rate carbon steel pipeline material design pressure to an extent rendering it (arguably) safe for use with hydrogen:
P = 2 S t/D F E T Hf
These helpful tables excerpted from ASME B31.8 and B31.12 were borrowed from Wang, B. et al, I.J. Hydrogen Energy, 43 (2018) 16141-14153
Design factor F, used in both codes, varies between 0.8 and 0.4 in ASME B31.8 based on “location class”, which is based on factors including proximity to occupied buildings.
B31.12 for hydrogen has two design factor tables: one for new, purpose-built hydrogen pipelines, with F values matching those in B31.8 for fossil gas (option B), and one for re-use of pipelines not originally designed to B31.12, which uses a lower (more conservative) table of F values ranging from 0.5 to 0.4 (option A). The latter, option A, would apply to any fossil gas pipeline repurposed to carry hydrogen.
For many existing gas pipelines, repurposing the line to carry hydrogen would require de-rating of the design pressure from the current level which is often 72% or 80% of specified minimum stress, to perhaps 40-50%.
For hydrogen piping, the material de-rating factor Hf ranges from 1 for low yield stress piping materials used at low pressures, to 0.542 for high tensile, high yield strength materials operating at high system design pressures. No such material de-rating factor is required in ASME B31.8 for the design of fossil gas pipelines.
In the extreme case, a pipeline designed and fabricated for fossil gas per ASME B31.8 in a low criticality (class 1 division 1) location far away from occupied buildings, made of a high yield strength steel, would have its design factor reduced from 0.8 to 0.5, and an Hf applied of 0.542. The result would be a reduction in design pressure to 34% of the original value, i.e. a reduction of almost three-fold.
A reduction in design pressure represents a very significant reduction in pipeline energy carrying capacity and would require either “twinning” of the line with new pipe, or replacement with new pipe.
So: Can We Use Existing Gas Transmission Pipelines for Pure Hydrogen?
The answer is much more complicated than a simple yes or no!
Can they be re-used? Maybe- but the pipe material isn’t the only issue. There are many others, covered in my paper here:
(which I will shortly update with this new information in relation to piping materials- that’s why I love LinkedIn as a publishing medium, because it makes updates easy!)
Can they be re-used at their existing design pressure and hence at their existing energy carrying capacity? The answer to that is almost certainly NO. At bare minimum, de-rating of the design pressure would be required, likely to a significant extent. This would necessitate either twinning the line with new pipe to carry the same amount of energy, replacing the existing pipe, or accepting the reduced capacity.
Will they blow up and kill people if used for hydrogen? Well…they will crack much faster, even at reduced stress, and will be much more likely to break, than if they carried fossil gas without hydrogen in it. Gas pipelines are often operated at a pressure which varies with respect to time, cycling frequently, whereas dedicated hydrogen pipelines tend to be run at more constant pressures, resulting in less rapid fatigue. But if the design criteria of a code (B31.12) not used in the design and construction and testing of the original pipe are retroactively applied to the existing pipeline, the industry might consider that to be “safe enough”. The DVGW testing demonstrates that the design assumptions used in the hydrogen pipeline design code to set its “hydrogen design de-rating factor” are met, in metallurgical terms.
Let’s just say, that’s far from a ringing endorsement for the concept. If I were a regulatory body in charge of ensuring that gas utilities keep their pipelines safe, I’d be paying very close attention to any pipeline being re-purposed for hydrogen. The gas industry itself is in at very least a potential conflict of interest in regard to this matter, and the regulatory bodies will need to step up and ensure that if any pipeline is converted to carry hydrogen- even hydrogen blends- that this is done in a way that is truly safe.