UC Merced Leads National Effort to Unlock Quantum Secrets of Twisty Molecules

A team of scientists led by UC Merced is embarking on a project to understand how the twisted shapes of specific molecules can influence the spin of electrons - a phenomenon that could revolutionize solar energy, electronics and quantum computing.

The research, funded by the U.S. Department of Energy's Scientific Discovery through Advanced Computing (SciDAC) program, focuses on a mysterious effect known as chirality-induced spin selectivity, or CISS.

In simple terms, CISS explains how chiral molecules - those that have a helical shape and show a "handedness," such as left- or right-handed gloves - can serve as filters for electron spin, a quantum property akin to a tiny magnetic orientation.

Chiral molecules appear identical but are arranged differently in three dimensions, so they can't be superimposed on their mirror images, much like our left and right hands. Often, their helical shape is responsible for their handedness. One well-known example of a chiral molecule is thalidomide, a drug whose two mirror-image forms had drastically different effects: One was therapeutic, while the other caused severe birth defects.

"Understanding how chiral molecules control electron spin could open the door to new technologies in energy harvesting and spintronics," said Department of Chemistry and Biochemistry Chair Professor Christine Isborn, the project's principal investigator. "But the physics behind this effect is still not fully understood."

The multi-institutional team includes experts in chemistry, physics, engineering and applied mathematics from UC Merced, the University of Michigan, Rutgers University- Newark and Lawrence Livermore National Laboratory. Together, they aim to develop powerful new computational tools to simulate the movement and interaction of electrons and atomic nuclei in real time.

A Quantum Puzzle with Big Implications

CISS has captured the attention of scientists worldwide because of its potential applications. It could improve solar panels by converting sunlight into electricity more efficiently, enable new types of sensors and memory devices and even help separate mirror-image molecules in pharmaceuticals.

Despite its promise, CISS remains poorly understood. Existing computer models struggle to replicate the strength of the effect seen in experiments. That's where the UC Merced-led team comes in.

"We're developing scalable, high-accuracy simulations that go beyond current methods," said applied mathematics Professor Harish Bhat, who, along with chemistry Professor Henrik Larsson, is a co-investigator. "Our goal is to model the quantum dynamics of electrons and nuclei in chiral systems with unprecedented detail."

Three-Pronged Approach to Discovery

The project is organized into three main research thrusts:

1. Quasi-Exact Modeling: Using advanced wavefunction methods that attempt to solve the Schrödinger Equation, the team will simulate electron behavior in small chiral molecules with near-perfect accuracy. These simulations will serve as benchmarks for testing more scalable approaches. Larsson and Paul Zimmerman, a chemistry professor at the University of Michigan, lead this thrust.
2. Machine Learning Meets Quantum Physics: By analyzing data from the high-accuracy simulations, researchers will train machine learning models to improve the performance of time-dependent density functional theory (TDDFT), a widely used method in quantum chemistry. This will help capture the complex spin dynamics in larger systems. Bhat heads up this effort.
3. Exascale Computing: The team will harness the power of supercomputers - capable of performing billions of calculations per second - to simulate electron and nuclear motion in realistic materials. These simulations will help scientists understand how environmental factors such as temperature and molecular vibrations influence CISS.

"We're combining the best of physics, chemistry, and computer science to tackle a problem that's both fundamental and practical," said Vikram Gavini, an engineering professor at the University of Michigan and thrust lead for exascale computing.

A National Collaboration with Global Impact

The project is supported by an $8 million grant from the DOE's Basic Energy Sciences and Advanced Scientific Computing Research programs. This is the first UC Merced-led SciDAC project and the only project selected this year to be led by a university. It leverages cutting-edge mathematical software and computing facilities at Lawrence Livermore National Laboratory, including the El Capitan supercomputer, one of the world's fastest.

Applied mathematicians and computer scientists from the Modular Finite Element Methods team at Lawrence Livermore, part of the SciDAC FASTMath Institute, will collaborate with scientists on the research project to accelerate understanding of the fundamental physics of CISS.

"This is a true team effort," Isborn said. "We're bringing together experts in quantum dynamics, machine learning, and high-performance computing to solve a problem that could reshape how we think about electron spin, energy and information."

The researchers plan to make their software and data publicly available, enabling other scientists to build on their work. They also aim to host workshops and webinars to share their findings with the broader scientific community.

Training the Next Generation

Beyond the science, the project has a strong educational component. Graduate students and postdoctoral researchers will gain hands-on experience in cutting-edge computational methods, preparing them for careers in academia, industry and national labs.

"We're not just solving a scientific puzzle, we're training the next generation of researchers," Larsson said.

Looking Ahead

If successful, the project could lead to new materials and devices that use electron spin in novel ways, from ultra-efficient solar cells to quantum computers that store information in spin states.

"CISS is one of the most intriguing effects in modern chemistry and physics," said Neepa Maitra, a physics professor at Rutgers University-Newark. "By understanding it at a fundamental level, we deepen our understanding of the correlated quantum motion of electronic charge, spin, and their coupling to nuclei, and could unlock new possibilities for technology."

The research is expected to span four years, with major milestones in algorithm development, simulation accuracy and experimental validation.

"This is a bold and ambitious project," said Isborn. "But with the right tools and the right team, we believe we can crack the code of CISS."

Lorena Anderson