Localized Quantum Chemistry on Quantum Computers.

Area of Science:

• Quantum computing
• Computational chemistry
• Quantum algorithms

Background:

• Quantum chemistry calculations for large, strongly correlated systems face exponential scaling challenges.
• Current quantum algorithms require resources beyond the capabilities of existing quantum devices.

Purpose of the Study:

• To present a novel quantum algorithm, local active space unitary coupled cluster (LAS-UCC), for calculating ground-state energies of large chemical systems.
• To reduce the computational cost and resource requirements for quantum chemistry simulations.

Main Methods:

• Combining localization of multireference wave functions with quantum phase estimation (QPE).
• Utilizing variational unitary coupled cluster singles and doubles (UCCSD) ansatz within the LAS-UCC framework.
• Demonstrating accuracy through molecular dissociation and bond-breaking simulations.

Main Results:

• LAS-UCC exhibits linear scaling with system size for specific geometries, offering a polynomial reduction in gate count compared to QPE.
• Achieved higher accuracy than variational quantum eigensolver (VQE) with UCCSD and classical local active space self-consistent field (SCF) methods.
• Resource estimates provided for systems up to 20 H2 molecules.

Conclusions:

• LAS-UCC presents a significant advancement in quantum computational chemistry, enabling more efficient and accurate calculations of strongly correlated systems.
• The algorithm's scalability and accuracy pave the way for tackling larger and more complex chemical problems on future quantum computers.