Scalable Ground State Energy Estimation

Estimating molecular ground-state energies is a central application of quantum computing, requiring both the preparation of accurate quantum states and efficient energy readout. While adiabatic state preparation (ASP) offers convergence guarantees, it has not yet been experimentally demonstrated beyond very small system sizes. On the theoretical side, recent advances have enabled ASP implementations that avoid Trotter errors.

Here, we apply ASP to prepare the ground state of a six-qubit encoding of the H3+ molecule and extract its energy using an efficient variant of iterative quantum phase estimation. Our results improve upon the classical Hartree–Fock energy, despite execution on a noisy trapped-ion quantum computer. To enhance performance, we employ circuit-optimization techniques to reduce gate depth and systematically analyze the impact of different noise sources. We find that the algorithm is resilient to both coherent and incoherent noise but remains sensitive to leakage errors. These insights highlight the importance of targeting leakage suppression in future hardware development.