Simulating CO₂ adsorption on metal-organic frameworks using NISQ hardware and cuQuantum

Metal-organic frameworks (MOFs), porous 3D surfaces consisting of a metallic core and organic ligands, are a leading class of materials for solid adsorbents for carbon capture and sequestration (CCS). A key challenge in their practical realization for this purpose is the huge number of possible structures and configurations, along with highly-variable adsorption performance for CCS. While traditional condensed matter methods, like density functional theory and molecular dynamics, can provide insights and predictions of the viability of particular MOFs for CCS, they are strongly correlated systems, making them a prime candidate for more accurate quantum chemistry simulations on a quantum computer. Though noisy intermediate-scale quantum (NISQ) era QPUs do not have the qubit counts to simulate large systems, smaller, proof-of-concept experiments can be performed to start to understand what advantages quantum computers may provide. We here present some results from the variational quantum eigensolver (VQE) to estimate the ground-state energy of isolated MOF components to characterize their performance in CO₂ capture, using both NISQ era hardware and state-vector and tensor network simulations using NVIDIA's cuQuantum simulation toolkit.