Treating strong correlation with internally corrected coupled cluster methods: Is distinguishable cluster approach systematically improvable?
Varun Rishi and Edward F. Valeev
Department of Chemistry, Virginia Tech, Blacksburg, VA
Abstract
Treatment of strong correlation is an open problem in quantum chemistry. Methods that do not catastrophically fail when transiting from weak to strong correlation regime are still being researched upon. As we have found a solution to the weak correlation (or the dynamic correlation) problem in coupled cluster (CC) methods, the big question to address is if single-reference CC methods can be a starting point in our search for methods that work in strong correlation. One such method is the distinguishable cluster approach (DCD, DCSD) which has been formulated as a modified version of CCSD equations and can be viewed as one of the internally corrected CC methods as defined by Paldus et al. We probe if such an approximation can be systematically improved i.e. can we add the effect higher excitations and approach the full CC limit. We focus on identifying terms in the triples amplitude equations that can be removed apriori which would otherwise hinder the convergence of coupled cluster solution to the right limit for the bonddissociation and other instances of strong correlation in quantum chemistry.
References
1 V. Rishi and E. F. Valeev (to be submitted).
2 V. Rishi, A. Perera and R. J. Bartlett, J. Chem. Phys. 144, 124117 (2016).
3 D. Kats and F. R. Manby, J. Chem. Phys. 139, 021102 (2013).
4 P. Piecuch and J. Paldus, Int. J. Quantum Chem. 40, 9 (1991). 5 J. Paldus, J.Math. Chem. 55 (2), 477 (2017).
6 I.W. Bulik, T.M. Henderson, and G.E. Scuseria, J. Chem. Theor. Comput. 11, 3171 (2015).