Simulating Electron Transfer on Noisy Quantum Computers: A Scalable Approach to Open Quantum Systems

Simulating large electronic networks with vibrational environments remains a fundamental challenge due to the long lifetimes of electronic-vibrational (vibronic) excitations on the picosecond scale. Quantum computers are a promising platform to simulate the dynamics of open quantum systems aided by intrinsic hardware-noise, with successful demonstrations of models with two electronic sites. We simulated a microscopic model of electron-transfer (ET) with a single donor and up to nine acceptor sites on a superconducting processor of IBM, using a model-specific error mitigation scheme. Our results using up to 20 qubits reveal a probability of ET that is well aligned with classical calculations where electronic and vibronic transfer resonances can be identified at the expected driving forces. We conducted 10 independent experiments per system size on different days, accounting for hourly fluctuations in error rates. We find that the most important ingredient for large-scale simulations is a large number of available qubits connected by high-fidelity gates, with coherence times above the threshold set by the target open system. Because the vibronic mechanism of electron transfer is entanglement-driven, our simulation is a natural application-based benchmark, in which the hardware capacity to produce and sustain entanglement is quantified by the maximum system size for which the hardware produces accurate results.