Accurate thermochemistry from explicitly correlated distinguishable cluster approximation.

Area of Science:

• Quantum chemistry
• Computational chemistry
• Electronic structure theory

Background:

• The distinguishable-cluster approximation is a powerful tool for electronic structure calculations.
• Explicitly correlated methods, like F12-type approaches, incorporate interelectronic distances to improve convergence.
• Coupled-cluster theory with singles and doubles (CCSD) is a standard but computationally demanding method.

Purpose of the Study:

• To develop and benchmark an explicitly correlated version of the distinguishable-cluster approximation.
• To assess the performance of this new method for closed and open-shell systems.
• To evaluate its applicability to systems with single- and multireference character.

Main Methods:

• Implementation of F12-type explicitly correlated techniques within the distinguishable-cluster framework.
• Benchmarking against coupled-cluster theory with singles and doubles (CCSD).
• Application to systems exhibiting both single- and multireference electronic characteristics.

Main Results:

• The explicitly correlated distinguishable-cluster approximation shows significant improvements over CCSD.
• The method is applicable in a black-box manner across different electronic system types.
• Obtained optimized geometries are comparable to or better than CCSD(T) (coupled-cluster singles and doubles with perturbative triples) quality.

Conclusions:

• Explicitly correlated distinguishable-cluster theory provides a robust and accurate approach for electronic structure.
• This method offers a practical alternative to traditional coupled-cluster methods, especially for challenging systems.
• The black-box nature and high accuracy make it suitable for broad applications in computational chemistry.