A traditional Japanese kagome woven basket, showing the trihexagonal lattice pattern, also seen in the atomic structure of kagome metals. Image/A. Gude, Flickr.
Kagome basket weaving is a centuries-old tradition in Japan. Artisans craft strands of bamboo into an open-weave lattice of hexagons separated by triangles, which are shaped into delicate baskets and other objects.
The unique tessellated structure also gives its name to a special class of materials — kagome metals — an exciting category of quantum materials that have an atomic structure with unusual superconducting properties. These materials could be of significant interest for applications in high-performance future electronics.
The scientific community has faced a significant challenge in understanding what drives the complex behaviors, particularly the superconductivity of kagome materials. New research led by Zhenglu Li, assistant professor of materials science at the USC Viterbi School of Engineering, uses a computational approach to unlock the mystery of kagome superconductors, offering unique insights into the way electrons interact with the lattice dynamics.
The Li Group’s work has been published in leading physics journal Physical Review Letters.
Superconductivity is a holy grail phenomenon in materials development for advanced electronics. It’s the process whereby materials can conduct electricity without any resistance when cooled to low temperatures.
Imagine, for example, wires that could carry “five times as much electricity” as current cables, or magnetically levitating trains and future fusion power plants.
Kagome metals based on the element vanadium are intriguing because their superconductivity coexists with other quantum phases, such as charge-density waves – a phenomenon where the electrons in a material reorder into density waves that break the intrinsic lattice symmetry. This has led to intensive debates within the scientific community on the mechanisms causing these behaviors in the materials.
Detail of the trihexagonal kagome lattice that forms the basis of the atomic structure of kagome superconductors. Image/ Wikimedia Commons.
“The kagome metals were found to have very rich phases of electrons. The material can be metallic and superconducting – but if you distort the structure a little bit, it can also host charge-density waves simultaneously,” Li said.
“People are interested in this kagome metal (CsV3Sb5 in this work) because the rich phases indicate complex interactions and provide an excellent platform for controlling these phases,” he said.
Li said that despite a vast scope of experimental studies, the community had not reached a consensus on what caused the material to exhibit many of these unique phases, including superconductivity. The Li Group rose to this challenge, developing a highly accurate new methodology to compute the coupling between electrons and atomic lattice vibrations, termed electron-phonon coupling, a critical interaction in materials that dominates conventional superconductivity, electrical and thermal conductivity in transistors, and optical absorption efficiency in solar cell materials.
“My group has developed highly accurate ‘many-body’ approaches that capture detailed information to help us understand materials that host interacting electrons and atoms. We use this approach to compute one of the key interactions in the materials — the electron-phonon interaction,” Li said. “This work examines whether this electron-phonon coupling can drive the superconductivity observed experimentally, as well as some other experimental low-energy excitation features in the material, such as photoemission kinks — one behavior of interacting systems, manifested as sudden changes in the energy-momentum relations of electrons, typically indicating coupling between electrons and other excitations.”
Li’s team concluded that the electron-phonon coupling is strong enough to cause the material’s superconductivity. The method used a “first principles” approach, meaning that it did not have adjustable parameters; instead, it relied on basic laws of quantum mechanics with general approximations, which also predict results that can be directly compared with experiments.
“We got nearly perfect agreement. First of all, our calculations demonstrate that the photoemission kinks (features in the dispersion relations) originate from electron-phonon coupling — it agrees perfectly with the experimental spectrum,” Li said. “Second, using these electron-phonon coupling ingredients that we validated against the experiment, we proceed to calculate if the strength is enough to mediate superconductivity, and indeed it does. Not only that — it also gives the comparable spectral signatures of superconductivity with the experiments.”
Harnessing the world’s most powerful supercomputers for the future of quantum materials
Assistant professor in the Mork Family Department of Chemical Engineering and Materials Science Zhenglu Li
Li and his team’s work on kagome superconductors has been supported by a U.S. Department of Energy INCITE (Innovative and Novel Computational Impact on Theory and Experiment) Award for 2024 – 2025, giving the team invaluable access to two of the world’s largest supercomputers — Aurora at Argonne National Laboratory and Frontier at Oak Ridge National Laboratory. The Li Group has been allocated a million GPU (graphics processing units) node hours over two years on these machines to further develop and apply their computational approaches to study quantum materials.
“These two machines are the first in the whole world to ever reach what we call exaflops scale,” Li said. An exaflop is a measure of performance for a supercomputer that can calculate at least 1018 or one quintillion floating-point operations per second.
“Taking Frontier as an example, it has over 70,000 GPUs, and we can access the whole machine if we need to do such a large-scale single calculation,” Li said.
“This work, in particular, is not possible without this scale of computational power. This is rooted in the high accuracy in our quantum many-body methods, such that the computational cost is extremely large,” Li said. “This award is critical to this work and other research projects in my group.”
Advancing computational materials research at USC
The Li Group’s work is an essential step toward creating more sustainable quantum materials for high-performance computing — a key goal of USC’s Frontiers of Computing, a $1 billion-plus initiative supporting ethical research in quantum computing, AI, robotics and more. The insights gained from this work could be key to unlocking the next generation of advanced energy-efficient devices that will shape our world for years to come. This research is also supported by the Ershaghi Center for Energy Transition at the USC Viterbi School of Engineering.
“I think this is a good showcase in general of the computational capability of USC,” Li said.
Published on April 8th, 2025
Last updated on April 8th, 2025
