New quantum chemistry method to unlock secrets of advanced materials

A new computational approach developed at the University of Chicago promises to shed light on some of the world’s most puzzling materials—from high-temperature superconductors to solar cell semiconductors—by uniting two long-divided scientific perspectives.

“For decades, chemists and physicists have used very different lenses to look at materials. What we’ve done now is create a rigorous way to bring those perspectives together,” said senior author Laura Gagliardi, Richard and Kathy Leventhal Professor in the Department of Chemistry and the Pritzker School of Molecular Engineering. “This gives us a new toolkit to understand and eventually design materials with extraordinary properties.”

When it comes to solids, physicists usually think in terms of broad, repeating band structures, while chemists focus on the local behavior of electrons in specific molecules or fragments. But many important materials—such as organic semiconductors, metal–organic frameworks, and strongly correlated oxides—don’t fit neatly into either picture. In these materials, electrons are often thought of as hopping between repeating fragments rather than being distributed across the material. 

“Accurately describing electrons on individual fragments is possible, but then you lose the global picture of how charges move across a material,” explained Daniel King, co-first author of the paper. “Our approach squares that circle: you model the local fragments, but also capture how electrons hop between them.”

The new method builds on a framework called the Localized Active Space (LAS) approach, originally developed by Research Assistant Professor Matthew Hermes. By extending it to periodic solids, the team has created a hybrid method that merges local quantum chemistry with global band theory.

To prove the method’s power, the researchers applied it to several challenging test cases. Hydrogen chains, for instance, have long been difficult to model: classic density-function theory methods misclassify these systems as metals, while more accurate approaches dictate that they should behave like insulators. The new LAS approach was able to correctly show how the electrons within hydrogen chains give it insulator properties. 

In another example, the team used LAS to simulate a p–n junction, the fundamental component of solar cells and computer chips. The method revealed how charges separate and move across the function when light hits them—a process that could be hard to capture before. 

“As a proof of principle, this is step one,” said Bhavnesh Jangid, a fourth-year graduate student in the Gagliardi Group and co-first author. “We showed that our method captures the right physics at high accuracy. There are now other advanced methods we’d like to integrate into the approach to keep improving it.”

The researchers envision their method as both a tool for understanding existing materials and, eventually, for designing new ones. “All materials are quantum mechanical at heart,” said King. “This is an elegant step toward really seeing how quantum mechanics drives the properties we use in everyday life.”

The new work, published in Nature Communications, was supported in part by Q-NEXT, a U.S. Department of Energy National Quantum Information Science Research Center that brings together the world’s leading quantum information researchers. The LAS method is available open-source from the Gagliardi Group and the team said they are continuing to refine it to make it accessible and easy to use for other researchers investigating quantum transport properties.