Quantum-Accelerated DFT+DMFT for Correlated Subspaces in Hemoglobin 

Density functional theory (DFT) often fails to describe strong on-site correlations in transition-metal and bio-inorganic systems. Dynamical mean-field theory (DMFT) addresses this limitation, but its cost is dominated by the impurity solver. We propose and validate a hybrid HPC–quantum workflow for DFT+DMFT where high-performance DFT generates low-energy models and a quantum impurity solver, the Quantum Sampled Configuration Interaction (QSCI), stochastically probes the impurity Hamiltonian to reconstruct its spectral distribution.  

As a preparatory step, we benchmark the approach on a correlated d-manifold model of NiO to assess circuit design, sampling accuracy, and convergence. Building on this groundwork, we apply the hybrid algorithm to hemoglobin, performing full-molecule DFT on HPC to identify a correlated Fe-centered heme subspace that is embedded and solved on a quantum device via QSCI while keeping the surrounding protein environment at the DFT level. We discuss algorithmic validation, computational scaling, and convergence characteristics, outlining a practical route toward hybrid HPC–quantum simulations of correlated molecular systems.