HPC Simulations Uncover How Hydrogen Moves Through Metals
- Rafal Tekreeti
- 2 days ago
- 3 min read
By: Abdelrahman Hussein, Margo Cauwels, Lisa Claeys, Tom Depover, Kim Verbeken
Hydrogen is expected to play a crucial role as an energy carrier in the green energy transition. This vision of a hydrogen-powered future—often referred to as the hydrogen economy—requires a large infrastructure network for production, storage, and distribution, much of which relies on metallic materials.
However, when hydrogen comes into contact with metals, it can penetrate the surface and diffuse into the bulk, eventually accumulating at internal defects such as interfaces, dislocations, and grain boundaries. This accumulation can lead to the formation of microcracks, which may grow and cause sudden and catastrophic failures—a phenomenon known as hydrogen embrittlement.
To prevent this, we must understand how hydrogen moves inside metals and how much can be absorbed under various conditions. But this is a major challenge: hydrogen is the smallest atom in the periodic table, which gives it very high diffusivity, and it is notoriously difficult to track with existing experimental techniques.
To address this, we developed a new class of fully kinetic, full-field models. These models simulate hydrogen transport at high spatial and temporal resolution across realistic metallic microstructures, using advanced computing resources. By applying these simulations as digital twins (see Figure 1), we were able to interpret and predict the results of key experimental techniques such as thermal desorption spectroscopy (TDS).
The simulation results in Figure 2a show a remarkable match with experiments. Traditionally, the two peaks observed in TDS would be interpreted using the concept of "trapping" — with each peak attributed to a different detrapping activation energy.
However, our full-field simulations tell a different story. As shown in Figure 2b, the desorption behavior in a partially charged sample is not governed by distinct trap energies. Instead, most hydrogen near the edges desorbs directly, producing the main peak. At the same time, a portion of hydrogen diffuses inward toward the center of the sample, then diffuses back out, giving rise to the second peak or tail. We refer to this mechanism as “inward diffusion.”
This insight has major implications. It suggests that hydrogen transport in metals may behave very differently from what the classical trap-based framework predicts. With this new understanding, there's a pressing need to rethink how we analyze TDS data — and I believe the future lies in leveraging digital twins and machine learning to extract physical parameters from experimental data.
The role of VSC in this research has been absolutely instrumental. The simulations involved models with over 3.4 million elements and 1.7 million degrees of freedom, requiring memory and computational power far beyond what was available on my local machine. I also had to run multiple large-scale simulations, which would have taken months without access to high-performance computing.
Because this modeling framework is entirely new, I developed an in-house finite element code in C++ called PHIMATS, which is available as open-source on GitHub. I received valuable technical support for compiling and linking the code with PETSc on the HPC infrastructure.
Beyond the infrastructure, the technical training was equally valuable. Courses like Introduction to HPC by Ewald Pawels and MPI Programming by Geert Jan Bex, as well as additional material on Geert’s YouTube channel, provided much of the learning materials I needed to use these resources effectively. Without this combination of hardware and technical support, this research simply would not have been possible.
🔗 Read the full scientific publication in ScienceDirect: here
🔍 Your Research Matters — Let’s Share It!
Have you used VSC’s computing power in your research? Did our infrastructure support your simulations, data analysis, or workflow?
We’d love to hear about it!
Take part in our #ShareYourSuccess campaign and show how VSC helped move your research forward. Whether it’s a publication, a project highlight, or a visual from your work — your story can inspire others.
🖥️ Be featured on our website and social media. Show the impact of your work. Help grow our research community
📬 Submit your story: https://www.vscentrum.be/sys
