Description

CCSD(T) is the gold standard of computational methods. Even though CCSD(T) can successfully predict many energetic properties, large basis sets are needed which would increase the computational cost, and there are some failures. Composite methods, or model chemistries, combine modest levels of theory with larger basis sets and more robust theories with smaller basis sets to approximate the results obtained at a much higher level of theory. Correlation Consistent Composite Approach (ccCA) includes extrapolation of energies to the CBS limit and the addition of core–valence electron interactions. The success of composite methods rests on the assumption that important electronic effects can be calculated and used as additive terms to a reference energy, which is the ‘‘additivity assumption”. Two of the extrapolation schemes have been used to determine atomic and molecular energies to the CBS limit. Relativistic effects have been considered. For the enthalpies of formation, CCSD(T) results in a mean absolute deviation (MAD) of 0.87 kcal mol^-1, while ccCA results in an MAD of 0.80 kcal mol^-1. The computational time of ccCA is 20 times faster than CCSD(T). At the end the authors answered the question in the title. CCSD(T) scales as O (nocc3 nvirt4) instead of O(nocc3+nvirt4). The bottleneck of ccCA represents 72% of the total time instead of 62%. A common question arising from this paper is, what is the point of developing a composite method if you can only apply it to molecules under very specific conditions?

Keywords

Composite methods; ccCA; CCSD(T); Complete basis set; Extrapolation