Benchmarking distinguishable cluster methods to platinum standard CCSDT(Q) non covalent interaction energies in the A24 dataset

TL;DR

This paper evaluates low-cost approximation methods, DC-CCSDT and SVD-DC-CCSDT, for their accuracy in reproducing high-level CCSDT(Q) energies in the A24 dataset, demonstrating improved fidelity and computational efficiency.

Contribution

It introduces and assesses the performance of DC-CCSDT and SVD-DC-CCSDT methods as efficient alternatives to high-cost coupled cluster calculations.

Findings

SVD-DC-CCSDT outperforms CCSDT and CCSD(T) in accuracy.

SVD-DC-CCSDT is a low-cost, high-fidelity method for large basis sets.

The methods enable calculation of complex energies previously intractable.

Abstract

Recent disagreement between state-of-the-art quantum chemical methods, coupled cluster with single, double and perturbative triples excitations and fixed-node diffusion Monte Carlo, calls for systematic examination of possible sources of error within both methodological approaches. Coupled cluster theory is systematically improvable toward the exact solution of the Schr\"odinger equation, however very quickly is limited by the computational cost of the calculation. Therefore, it has become imperative to develop low-cost methods that are able to reproduce CC results, beyond the CCSD(T) level of theory. Here, the DC-CCSDT and SVD-DC-CCSDT methods are examined for their fidelity to the CCSDT(Q) correlation interaction energies for the A24 dataset and are shown to outperform CCSDT and CCSD(T). Furthermore, with (T)-based corrections of the SVD approximation the SVD-DC-CCSDT method becomes an accurate and relatively low-cost tool for calculation of previously intractable post-CCSD(T) energies in atomic orbital basis sets of unprecedented size.