D4.9: Benchmarking Report

This document is a continuation of deliverable D4.3 of the NEASQC project and is concerned with benchmarking the applications for quantum computing researched and implemented within the quantum chemistry work package of the project. In D4.3 we reported assessments contributed from three of the partners within the work package: AstraZeneca presented a benchmarking study for chemical reactivity with our previously implemented Pre-Born-Oppenheimer quantum computing computational package [1], where the showcase was the reaction of water hydrogenation. HQS Quantum Simulations demonstrated a performance analysis of their implemented variational algorithms [1], comparing a Variational Quantum Eigensolver (VQE) implementing UCCSD, an Adaptive Variational Quantum Eigensolver (AdaptVQE) using UCCSD, a variational Hamiltonian ansatz (VHA), and a variational imaginary time evolution (VITE) with two-local ansatz [2–5]. TotalEnergies estimated the feasibility to execute families of Hardware Efficient (HE) quantum computing ansatzes and quantum Unitary Coupled Cluster (qUCC) ansatzes on (noisy) near-term quantum computers, applied to a calculation of the ground state energy of strained benzene [6, 7]. In this report, that constitutes deliverable D4.9 of the NEASQC project, we extend on these contributions: In Chap. 2, AstraZeneca extended their investigations to the interaction of CO2 with a hydrogen radical using the Nuclear-Electronic Orbital (NEO) Quantum Computing Framework [8, 9], with a specific focus on simulating the reduction of CO2. In Chap. 3, HQS Quantum Simulations focuses here on the evaluation of methods to improve quantum measurements, one utilizing shadow measurement in the context of spectroscopy [10], one based on enhanced sampling using Bayesian statistics [11], and one based on projecting the result to fulfill so-called n-representability constraints which may be violated in a noisy quantum computation [12]. This is reported on in more detail in deliverable D4.7, with software implementations available on the NEASQC GitHub [1]. In Chap. 4, TotalEnergies continued their studies regarding benzene to a calculation of a system of graphene and a CO2 molecule using a VQE approach, which could be used in the context of understanding recapture processes of carbon dioxide. Similar to above, this study is reported on in more detail in deliverable D4.8, with related software found on the NEASQC GitHub [13]. Finally, in Chap. 5, ICHEC gives an overview of the software tool KraChem which was developed within the NEASQC project. The tool is designed to automatically fragment molecular systems into smaller subsystems, allowing breaking down large problems into subproblems, making them digestible for NISQ computations. This automatic subspace generation was reported on earlier in detail in deliverable D4.6, and is available on the NEASQC GitHub as well [14].