From optimal basis sets to accurate potentials: elemental carbon clusters at the focal point
J. Li,a F.N.N. Pansini,b J. Providência,c C.M.R. Rochad and A.J.C. Varandas*a,b
aSchool of Physics and Physical Engineering, Qufu Normal University, 273165 Qufu, China
bDepartamento de Física, Universidade Federal do Espíto Santo, Vitória, 29075-910, Brazil
cDepartment of Physics, University of Coimbra, 3004-516 Coimbra, Portugal
dDepartamento de Química, and Centro de Química, Universidade de Coimbra 3004-535 Coimbra, Portugal
E-mail: varandas@uc.pt
After a brief remark on the Born-Oppenheimer approximation, 1 we report on two topics. The first is extrapolation of electronic structure calculations to the complete one-electron basis set limit (CBS). Starting with a survey of recent work, 2-4 we address a recent proposal of optimal basis sets for correlated electronic structure calculations. Optimal double-zeta and triple-zeta basis sets of correlation consistent type are suggested for directly CBS extrapolating the correlation energy, 5 whose accurate determination is a major bottleneck in quantum chemistry. Method-specific in the sense of performing best for the chosen level of theory, the new basis sets are shown to perform accurately, with absolute correlation energies from (oVdZ,oVtZ) extrapolations outperforming by three- to fivefold those obtained using traditional ansatzes of similar flexibility: thence, of (VtZ,VqZ) or even better quality. Their performance for other properties of interest to chemists are also illustrated. The second topic is elemental carbon clusters. Starting with C2, C3 and C4, we first survey accurate ab initio-based global potential energy surfaces for such species, which illustrate intricate topological features there embedded. 6.7 Because global fits to ab initio points are of overwhelming difficulty for large cluster sizes, one wonders about the predictive capability of double many-body expansion (DMBE) theory when the development is truncated at the first few-body terms, namely at four-body level. 8 Despite some remarkable predictions for the energetics of the linear global minima and thermochemical/structural properties, further work is required ab initio and modelwise to assess the difficulties and virtues of such an approach. Prospective remarks conclude the talk.
1A.J.C. Varandas, J. Providência, and J.P. Providência Eur. Phys. J. D 69, 114 (2015); work in progress.
2A.J.C. Varandas, M.M. González, L.A.M. Cabrera and J.M.G. Vega Chem. Eur. J. 23, 9122 (2017).
3A.J.C. Varandas, Ann. Rev. Phys. Chem. 69, 177 (2018).
4A.J.C. Varandas, Phys. Chem. Chem. Phys. 20, 22084 (2018).
5A,J.C. Varandas and F.N.N. Pansini J. Chem. Phys. (submitted).
6A.J.C. Varandas and C.M.R. Rocha, Philos Trans A Math Phys Eng Sci 376, 20170145 (2018).
7 C.M.M. Rocha and A.J.C. Varandas Phys. Chem. Chem. Phys., 20, 10319 (2018).
8 C.M.M. Rocha, J. Li and A.J.C. Varandas J. Phys. Chem. A. (accepted).