Computing band gaps of periodic materials via sample-based quantum diagonalization

Abstract: State-of-the-art techniques in computational materials science often fail to capture many-body interactions in periodic systems. One particular application is the electronic structure problem, where established approaches such as Density Functional Theory are applied to predict experimental results. In this contribution, we present a sample-based quantum diagonalization (SQD) workflow for simulating the electronic ground state of periodic materials, and for predicting their band gaps. We obtain a fermionic Hamiltonian for crystalline solids by first projecting the electronic density obtained by density functional theory onto a localized basis set and then adding interaction parameters obtained by a self-consistent field theory. We then build an approximation of the system's ground-state wavefunction with the local unitary cluster Jastrow (LUCJ) ansatz. By sampling the quantum circuit and classically diagonalizing the Hamiltonian in the resulting configuration subspace, we predict the band gap energy of representative materials and obtain agreement with experimental results. The predicted bandgaps were obtained from calculations using 38 and 46 qubits, performed on a superconducting quantum processor, with minimal error-mitigation overhead. Our self-consistent approach offers a scalable pathway for utility-scale quantum simulation of materials.