State-Averaged Quantum Algorithms for Multiconfigurational Surface Chemistry: A Benchmark on Rh@TiO2(110)

TL;DR

This paper benchmarks quantum algorithms for complex surface chemistry problems, demonstrating adaptive ans"atze can achieve near-reference accuracy efficiently in multistate, strongly correlated systems.

Contribution

It provides a controlled benchmark for quantum algorithms in multistate surface chemistry, comparing SA-fUCCSD and SA-ADAPT approaches on a challenging Rh@TiO2 system.

Findings

SA-fUCCSD improves with circuit depth but needs many parameters.

SA-ADAPT achieves near-CASSCF accuracy with fewer operators.

Modified operator selection accelerates convergence.

Abstract

Accurate modeling of surface catalytic processes often requires methods capable of describing strong correlation, charge transfer, and multiple closely lying electronic states. While density functional theory remains widely used, its limitations for localized electronic states motivate the use of wavefunction-based approaches and, more recently, quantum computing algorithms. However, the performance of quantum ans\"atze in chemically motivated, multistate settings remains largely unexplored. Here, we benchmark state-averaged factorized unitary coupled cluster with singles and doubles (SA-fUCCSD) and the adaptive, problem-tailored ansatz (SA-ADAPT) using an embedded cluster model of NO adsorption on Rh-doped TiO2(110). The system exhibits pronounced multiconfigurational character and multiple state crossings, providing a stringent test. State-averaged CASSCF serves as a reference, and the quantum ans\"atze are evaluated as solvers for the corresponding CASCI problem within a fixed orbital basis. We find that SA-fUCCSD improves with increasing circuit depth but requires many parameters and shows sensitivity to initialization. In contrast, SA-ADAPT achieves near-CASSCF accuracy with significantly fewer operators. A modified operator selection scheme, incorporating multiple operators per iteration, substantially accelerates convergence. Our results demonstrate the efficiency of adaptive ans\"atze for multistate problems and establish a controlled benchmark for quantum algorithms in chemically motivated systems beyond minimal models.