Accelerating the discovery of materials with LUMI to advance clean energy and zero-emission vehicles

In this blog series, we will delve into the collaborative efforts of seven international research collaboration projects between Finland–Japan and Finland–Colorado that use the LUMI supercomputer to address global challenges and top-level research topics in different fields. Join us as we interview the project leads and hear how these collaborations came to life and how they utilise LUMI for cutting-edge research in their field!

Professor Antti Karttunen and Dr. Kevin Conley from the School of Chemical Engineering at Aalto University, Finland, use LUMI supercomputer in their research, which accelerates the discovery of materials for clean energy and zero-emission vehicles.

– Our collaboration project is called Heat transport in semiconducting energy materials (HeatTra). We use high-performance computing to accelerate the discovery of materials for clean energy and zero-emission vehicles. We will focus on the controlled heat transport in complex and imperfect energy materials, Karttunen tells.

The results from the modelling on LUMI will be useful for colleagues conducting experimental research.

– The modelling will help us to understand heat transport in materials such as high-entropy thermoelectric alloys that could be used for direct heat-to-electricity conversion. Experimental research groups working on thermoelectric materials can use the findings from our project to produce novel thermoelectric materials for improving the energy efficiency of several different industries, he continues.

New partnerships from Colorado

The collaboration is carried out with researchers from the group of Professor Eric Toberer of the Physics Department of the Colorado School of Mines, USA.

– Professor Toberer’s group seeks to discover and design new energy materials through collaborative, interdisciplinary research. Their investigations often require the union of solid-state chemistry, materials science, and condensed matter physics.

The opportunity to create a new partnership came through the Finland–Colorado research collaboration call opened by CSC.

– Our collaboration has started in the framework of the HeatTra project. When CSC opened the Finland-Colorado collaborative call, we realized that this could be a great possibility to start a collaboration with the Toberer group, which is leading a expert in the field of thermoelectric energy conversion materials. At Aalto, we had previous experience in high-performance computing and simulation of thermoelectric materials. Prof. Toberer was immediately interested in meeting over Zoom to discuss potential collaboration. We formulated the theme for our collaboration together, wrote the proposal, and eventually received the funding, Karttunen explains.

Image: Professor Antti Karttunen, Aalto University, Finland. Image by Jaakko Kahilaniemi.

LUMI plays a key role in the project

The LUMI supercomputer and its computing capacity create opportunities for research on a much larger scale and over a longer period of time than before.

– The capabilities of LUMI play a key role in the HeatTra project. We will study heat transport in energy complex, disordered energy materials in much larger size scale and longer time scale than has been possible before.

The project utilizes the LUMI infrastructure in a versatile way.

– We will combine Density Functional Theory calculations, deep learning-based interatomic potentials, and molecular dynamics simulations. This combination of methods requires versatile hardware with high performance both in the CPU and GPU realms. We generate data on LUMI-CPU using conventional quantum chemical methods, train deep learning models on LUMI-GPU, and utilize the models in calculations that benefit from both GPU and CPU power. We will build know-how on state-of-the-art machine-learning-based interatomic potentials and their utilization in energy materials research, Karttunen explains.

The new workflows developed by the project will be beneficial for future research.

– The high-performance workflows developed during the project can also be applied in many other materials research projects later.

Read also the previous parts of the blog series:

Developing large computer model ensembles with LUMI to simulate ice flows in the Antarctic

LUMI powers the study of light scattering in space

Authors: Maari Alanko, Elisa Halonen and Pihla Kauranen, CSC – IT Center for Science