Description

---

This article’s main topic is that of their new computational method, G4(MP2)-6X. Essentially what this new method does is use BMK/6-31+G(2df,p) for geometry optimization, apply six additional scaling factors for the correlation energy components compared with G4(MP2), and then have a sort of correction term as reoptimized HLC parameters. This paper is interesting to me, because at this point I’ve read a few different papers on these types of new methods, and I have to wonder, how well do this parameters actually capture the real world, or is it more like DFT where it essentially guesses at the right answer? The idea is that these parameters are meant to grasp the actual underlying physics of the problem.

---