Clean steels reveal how hydrogen cracks steel – and why slower is not safer

Hydrogen is the smallest molecule there is – and that is exactly what makes it awkward for steel. Hydrogen atoms slip into the metal lattice, gather at internal flaws and, under stress, open up cracks that can bring a pipeline or a pressure vessel down long before its nominal lifetime. This hydrogen-induced cracking has been studied for decades, yet one of its key variables has remained hard to isolate: the role of non-metallic inclusions, the tiny oxide or sulphide particles that every industrial steel contains. A new study has now taken an unusually clean route to the answer.

Why “clean steels” make the experiment possible

Commercial steels contain a mix of inclusion types, so it is almost impossible to say which of them is doing the damage. The research team, led by Gautam Sudha, used secondary steelmaking to prepare so-called clean steels that contain essentially only one kind of inclusion. Deoxidising the melt with aluminium produced steels with only alumina (Al₂O₃) inclusions; deoxidising with silicon produced steels with only silica (SiO₂) inclusions. With that, two otherwise comparable materials could be tested head to head.

Alumina traps far more hydrogen than silica

Using the classical Devanathan–Stachurski permeation technique, in which hydrogen is driven electrochemically through a thin steel membrane and detected on the other side, the researchers compared how fast hydrogen moves through each steel and how much of it gets held up on the way. The result, published in the International Journal of Hydrogen Energy, was clear: the alumina inclusions trapped considerably more hydrogen than the silica inclusions. The alumina steels therefore showed a lower apparent hydrogen diffusion coefficient – hydrogen takes longer to get through.

The counter-intuitive lesson: slower is not safer

At first glance, a steel that lets hydrogen pass more slowly sounds like the better choice for a hydrogen pipeline. The study shows why that intuition can be dangerous. The hydrogen that is held back does not disappear; it accumulates at the inclusions. Once enough of it has gathered there, it can nucleate a crack and drive its growth until the component eventually fails. In the words of the authors, a lower diffusion coefficient does not necessarily mean a safer steel. What matters is not only how quickly hydrogen travels, but where it ends up – and whether those sites are the ones that start cracks.

This has practical consequences for how steels are assessed. A single permeation number, read on its own, can point in the wrong direction. The type, size and distribution of inclusions – and therefore the deoxidation route chosen in the steelworks – are part of the hydrogen-compatibility story and need to be evaluated alongside the diffusion data.

The Power-to-X perspective: materials are infrastructure

Debates about hydrogen usually revolve around electrolysers, prices and demand. But a Power-to-X economy also has to be built – out of steel. Pipelines, buffer storage, compressor stations, filling stations and the vessels in which hydrogen is converted into eMethane, eMethanol or eAmmonia all depend on steels that tolerate hydrogen reliably for decades. In Europe, existing gas pipelines are being repurposed for hydrogen, and Switzerland is weighing how it will connect to the European Hydrogen Backbone. In every one of these cases, the question of which steel, and which steelmaking route, is suitable decides cost, safety and the speed at which the infrastructure can be rolled out.

Findings like this one therefore matter well beyond the metallurgy lab. They help explain why some steels that look fine on paper fail in service, and they give steelmakers a lever – inclusion control – to produce materials that are fit for a hydrogen future. For Switzerland’s defossilisation, where hydrogen and its derivatives will have to be transported and stored at scale, robust, well-understood materials are as much a prerequisite as renewable electricity and recycled CO₂.

Source: TechXplore, “Clean steels help reveal unique mechanisms of hydrogen induced cracking” (4 September 2026); Gautam Sudha et al., “Revealing the unique role of Al₂O₃ and SiO₂ inclusions on hydrogen diffusion and trapping behaviour in steel”, International Journal of Hydrogen Energy 246 (2026) 155581, DOI 10.1016/j.ijhydene.2026.155581.

🤖 AI transparency: This text and its visual were created using AI, reviewed and approved by a human.