Ground states of strongly-correlated materials on quantum computers using ab initio downfolding

Antonios M. Alvertis,Abid Khan,Norm M. Tubman

2024-09-19

Abstract:The accurate first-principles description of strongly-correlated materials is an important and challenging problem in condensed matter physics. Ab initio downfolding has emerged as a way of deriving accurate many-body Hamiltonians including strong correlations, representing a subspace of interest of a material, using density functional theory calculations as a starting point. However, the solution of these material-specific models can scale exponentially on classical computers, constituting a challenge. Here we propose that utilizing quantum computers for obtaining the properties of downfolded Hamiltonians yields an accurate description of the ground state properties of strongly-correlated systems, while circumventing the exponential scaling problem. We demonstrate this for diverse strongly-correlated materials by combining ab initio downfolding and variational quantum eigensolvers, correctly predicting the antiferromagnetic state of one-dimensional cuprate $\text{Ca}_2\text{CuO}_3$, the excitonic ground state of monolayer $\text{WTe}_2$, and the charge-ordered state of correlated metal $\text{SrVO}_3$. By utilizing a classical tensor network implementation of variational quantum eigensolvers we are able to simulate large models with up to $54$ qubits and encompassing up to four bands in the correlated subspace, which is indicative of the complexity that our framework can address.

Quantum Physics,Strongly Correlated Electrons

What problem does this paper attempt to address?

The paper attempts to address the problem of accurately describing the ground state properties of strongly correlated materials using quantum computers. Specifically, the authors propose a method that combines ab initio downfolding with the Variational Quantum Eigensolver (VQE) to overcome the exponential scaling issues encountered by classical computers when solving specific models of these materials. Through this approach, they are able to accurately predict the ground state properties of various strongly correlated materials, such as the antiferromagnetic state of the 1-dimensional cuprate Ca₂CuO₃, the excitonic ground state of the 2-dimensional material WTe₂, and the charge-ordered state of the correlated metal SrVO₃. Additionally, by implementing VQE with classical tensor networks, the authors demonstrate that the framework can simulate large-scale models involving up to 54 qubits and covering four energy bands in the correlated subspace, indicating the method's capability to handle complex problems. In summary, this study showcases the significant potential of current and future quantum computers in simulating the properties of strongly correlated materials.