Argonne supercomputer advances energy storage research

The lack of large-scale energy storage hampers many renewable energy sources, such as sunlight-dependent solar power and unpredictable wind power. Researchers at Lawrence Livermore National Laboratory (LLNL) are working to change that, leveraging a grant on Argonne National Laboratory’s Theta supercomputer to better understand the ion transport dynamics that lie at the heart of batteries and other energy storage systems.

Essentially, the problem is that simulations of energy storage materials are too perfect: or rather, the materials they simulate are too perfect. Imperfections found in real-world materials are essential for ion transport, but are generally not well captured by simulations of these materials.

“We want to understand these frontier regions, so we’ve injected a lot of high-performance computing power into them,” said Brandon Wood, deputy director of LLNL’s Future Energy Applications Laboratory. The HPC power in question: the Theta system from Argonne, via an endowment from the INCITE program of the Ministry of Energy. Theta is an HPE system that delivers 6.92 petaflops Linpack, and it ranked 70th on the most recent Top500 list.

Researchers use Theta to run ensemble simulations of a wide range of material imperfections. Wood said Theta allows one to look beyond individual problem formations and into a “range of combinations to see the big picture”. These atomistic simulations incorporated the minute dynamics of ion mobility, and the results of these ultra-precise simulations were then integrated into a larger model to understand the interactions between different types of interfaces.

This multi-scale approach, the researchers explained, was crucial to obtaining meaningful results. Using HPC to study the control variables, the researchers found that there were strong interdependencies between changes in atomic structure and the arrangement of interfaces, and that both variables, along with the operating temperature of the material, had effects on ion transport.

“We need to understand how all of these different factors interact, and there’s an added layer of complexity because the relationship itself depends on operating conditions such as temperature or pressure,” Wood said. “This relationship between operating conditions, atomic-scale material properties, and meso-scale material properties is central to what we are trying to achieve.”

The research team is interested in applying their knowledge to improving solid-state batteries and hydrogen storage technologies.

“It gives guidance not just on what I would ideally want, but how to do it,” Wood said. “If you change your chemistry in a certain way and change the processing temperature in another way, it will result in improved performance or a better trade-off between the various relevant factors in these materials.”

To learn more about this research, read the report in ASCR Discovery here.

Previous UHS Board of Trustees Approves Clinical Research Course
Next Top 10 Places to Get a Personal Loan Online