Collaborative Research: MPS/CHE-EPSRC: Magic as a Quantum Descriptor for Chemistry

Researchers from Indiana University and Virginia Tech, along with their collaborators at King's College London in the United Kingdom, are working to establish a quantum property known as magic as a new descriptor for the electronic structure of molecules. Predicting molecular behavior on classical computers is a central challenge in chemistry, and knowing which molecules are hard to simulate helps identify where future quantum computers may provide an advantage. For decades, entanglement has been the primary way to gauge this quantum complexity, yet recent advances in quantum information science reveal that a distinct property called magic can better capture what makes a system difficult to simulate. The research team will develop rigorous measures of magic for molecules, create computational methods that exploit molecules with low magic, and test whether magic tracks familiar chemical ideas such as bond breaking and radical character. Their discoveries could give chemists a new organizing principle for electron correlation, bonding, and reactivity, and could clarify when chemical problems are best addressed by classical or quantum computation. The project will train postdoctoral researchers and graduate students in an interdisciplinary area spanning chemistry, physics, and quantum information science, and will release open-source software and tutorials that make these tools available to the community. This award is made under the NSF-UKRI lead agency opportunity. This project defines magic as nonstabilizerness, a measure of how far a quantum state lies from the set of stabilizer states that classical computers can simulate efficiently. The work is organized into three integrated aims. The first aim will develop chemically meaningful measures of magic for molecular systems, including non-local magic, which isolates intrinsically quantum correlations from artifacts of local basis choices, and task-oriented measures, which quantify the combined complexity of a state and an operator when predicting specific observables. The second aim will advance complementary simulation methods that exploit low magic structure, including stabilizer state expansions in the Schrodinger picture, Pauli propagation in the Heisenberg picture, and hybrid schemes that combine the two. The third aim will apply these tools to chemistry, mapping how magic evolves along reaction coordinates, comparing magic with established concepts such as static correlation and radical character, and characterizing how magic is distributed across molecular fragments. Together, these efforts will test whether magic serves as a meaningful and computationally useful descriptor of electronic structure across diverse chemical systems. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.