Nandini Ananth and Peter McMahon of Cornell University are supported by an EAGER award from the Chemical Theory, Models and Computational Methods program in the Division of Chemistry to develop new methods for using quantum computers to solve dynamics in quantum-simulation problems. Computer simulations that fully solve the fundamental quantum mechanical equations that describe the way electrons, atoms, and molecules interact and move in time hold the key to predicting and characterizing chemical reactions. It has long been appreciated that this task requires quantum computers which, unlike classical computers, can, in principle, solve the most complex of quantum mechanical equations. With the advent of near-term (first generation) quantum computers, substantial effort is underway to develop new algorithms to solve chemical problems on these devices. While there has been some success in methods to characterize the static energies of molecular systems, dynamic algorithms capable of being run on current hardware are lacking. Ananth and McMahon are developing a hybrid algorithm that combines both classical and quantum computers to perform quantum dynamics simulations, where the quantum computer need only run short programs (circuits) that are suitable for execution on near-term machines. This research has immediate broader impacts in the quantum-computing industry in the United States and may open the door to an array of new applications and scientific research. The development and testing of methods from this project will be transferred to industry practice through the open sharing of code and results. Students trained in this research will join the next generation workforce in quantum information systems.
These hybrid quantum-classical scheme may enable, for the first time, quantum dynamic simulations on a quantum computer, moving beyond the current algorithms that are focused on primarily static properties like spectra. Conventional wisdom is that dynamics beyond trivially short timescales will require fault-tolerant quantum computers because standard long-time dynamics simulations require the sequential computation of a large number of short-time propagation steps leading to significant error accumulation. Ananth and McMahon are developing a hybrid quantum-classical algorithm for long-time dynamic simulations that overcomes this limitation. Specifically, they are using the path-integral framework to represent a longtime dynamics simulation as a series of independent short-time propagation steps that can be performed using near-term quantum computers. The hybrid nature of the algorithm arises from combining efficient classical Monte Carlo methods to sample important real-time paths in configuration space weighted by quantum transition probabilities. This research may demonstrate the feasibility of condensed phase quantum dynamic simulations and enable finite temperature simulations of quantum systems on near-term quantum computers. This effort may significantly expand the range of chemical problems that can be studied to include reaction rate calculations and characterization of mechanisms and non-linear spectra.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
