A Fermionic Variable Quantum Eigensolver with Ultracold Strontium

University Of California-San Diego, La Jolla CA

Investigators

Abstract

The chemical and physical properties of molecules are determined by their electronic structure. Computing the latter starts with finding the molecular ground state because this knowledge helps compute many properties, such as their dipole moment and charge density. Since classical algorithms are computationally expensive, a tremendous academic and commercial effort has made advances in using quantum computers as an efficient future alternative. These efforts are based on the quantum version of classical bits known as qubits. However, electrons in a molecule are not qubits but rather fermions, and translating fermions into qubits is resource-intensive in computation and storage. This project aims to complete the construction of a proof-of-principle quantum computer based on fermions to determine the ground state of molecular systems. The goal is to set the groundwork for a fermionic computer that is better than classical supercomputers at finding the electronic structure of molecules, hopefully setting the stage for quantum chemistry computations that will impact industry and medicine. Along with these efforts, this project will have an educational impact by training graduate and undergraduate students in quantum science and technology. The project impacts will broaden by creating a Young Quantum Physicist Program to bring middle and high-school students, with an effort to achieve a diverse group representing society, to our labs and introduce them to the rapidly developing quantum technology. This proposal aims to leverage the latest advances in building qubit-based quantum computers using neutral atoms in optical tweezers to assemble and benchmark a new one based on fermions. This quantum computer will implement a hybrid classical-quantum algorithm called Variational Quantum Eigensolver using fermions. This implementation will find the ground state of small molecules and develop new tools and approaches to apply the algorithm to larger molecules. Obtaining this molecular ground state is the first step in calculating many properties that determine the molecules’ physics and chemistry. Starting from a degenerate fermi gas of strontium atoms, we will load a register array of 20 by 20 optical tweezers with atoms in their electronic and motional ground states. Using a second tweezer that traps particles in the excited clock state, but not those in their ground state, and another tweezer tuned on the clock transition, we will implement an effective tunneling gate through shuttling and optical pulses on this transition. A fourth tweezer will drive a Rydberg blockade on pairs of tweezers interacting in the VQE circuit. Our goal is to achieve state-of-the-art performance in all of these operations. With this toolbox, we will find the ground state of small molecules. Finally, we will apply the intuition gained in optimizing the classical-quantum algorithm for small molecules, as well as in circuit depth and measurement, to engineer an integrated algorithm for larger molecules —a simultaneously hardware-efficient and chemically-inspired VQE— possibly resorting to quantum machine learning. Most importantly, the goal at the end of the project is to establish a roadmap for scalability and achieving optimizations of fermionic quantum computers based on neutral atoms beyond the capacity of classical computers. 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.