New paper titled "Exhaustive search for optimal molecular geometries using imaginary-time evolution on a quantum computer" was published in npj Quantum Information by Taichi Kosugi, Hirofumi Nishi, and Yu-ichiro Matsushita of Quemix.
This study proposes a method for geometry optimization of molecules for the first-quantized eigensolver using a non-variational quantum algorithm called PITE®️ (Probabilistic Imaginary Time Evolution). In this method, the nuclei within a molecule are regarded as classical point charges, while the electrons are treated as quantum particles. The electronic states and candidate molecular structures are encoded as a superposition of many-qubits states, and the minimum energy of the energy surface is determined by creating a histogram from repeated measurements. Demonstrations were conducted to illustrate that the circuit depth per step scales polynomially with the increase in the number of electrons, with the potential for further reduction with additional qubits. Furthermore, based on resource estimation, it was indicated that an overall quantum advantage in terms of molecule size and candidate number may be exhibited by the approach when a good initial prediction is employed. The proposed method was also validated through numerical simulations. Additionally, a scheme adapted for variational calculations that prioritize the reduction of circuit depth, especially for noisy intermediate-scale quantum (NISQ) devices, was explored. Classical systems composed solely of charged particles were considered as a special case within this scheme.
This new, efficient scheme represents a significant step towards achieving large-scale quantum chemistry calculations on quantum computers.
Quemix has strong confidence in the potential of PITE®️ to make substantial contributions to the industrial sector. If you are a company or an academic institution interested in collaborative research utilizing PITE®, please feel free to contact us here: https://www.quemix.com/en/contact
