Quantum Computer-Enhanced Surface Reaction Simulations for Battery Materials through Sample-Based Quantum Diagonalization and Local Embedding

Quantum chemistry calculations are essential for predicting electronic structures, however, exactly solving the many-body Schrödinger equation is impractical due to electron interactions which are hard to capture with Density Functional Theory. Quantum computing presents a viable alternative, especially for localized electrons in surface reactions. In previous work, we have introduced a quantum embedding approach in which the majority of (core) electronic orbitals are treated with classical computing approaches. The most relevant ones are selected into an active space that is processed on quantum hardware. [1]

Here, we apply an embedding method to the Oxygen reduction reaction at Lithium battery electrode surfaces. Using the Local Unitary Cluster Jastrow ansatz and Sample-based Quantum Diagonalization (SQD), we evaluate efficient circuits in near-term quantum devices. SQD then allows Quantum Selected Configuration Interaction to be performed on classical computers, informed by quantum-computed electron configurations.

Our calculations show improvements over results obtained with Coupled Cluster Single Doubles, which are validated against Complete Active Space Configuration Interaction, giving insights into Li-O2 surface reactions and highlighting quantum computing’s potential in materials science.

[1] Gujarati, T.P. et al. Quantum computation of reactions on surfaces using local embedding. npj Quantum Inf 9, 88 (2023). https://doi.org/10.1038/s41534-023-00753-1