Multiwfn forum

Multiwfn official website: Multiwfn forum in Chinese:

You are not logged in.

#1 2018-10-16 12:21:00

Registered: 2017-09-11
Posts: 29

Comparison on several methods of weak interactions

Comparisons on several methods of weak interactions
By I_was_a_baby

Abstract: I do some computations for comparing the several methods about describing weak interactions. It can be seen that, if the method of CCSD(T) can be treated as Golden standard, the ability of various methods to describe weak interactions is uneven, where the double hybrid functional, B2PLYP, is the most excellent. The M06-2X, a suitable functional even without DFT-D3 correction, however, make me sad that deviation to the Golden standard is beyond my imagination.



#2 2018-10-16 19:33:43

Tian Lu (Multiwfn developer)
From: Beijing
Registered: 2017-09-11
Posts: 1,639

Re: Comparison on several methods of weak interactions

SCS-MP2 is never suitable for weak interaction evaluation, it only improves thermochemistry accuracy of MP2. The costs of SCS-MP2 and MP2 are exactly identical.

MP3 is just a rubbish, it performs poorly in all kinds of studies, in many cases the result is even worse than MP2. MP2.5 has been substantially demonstrated to be robust for evaluating weak interaction energy.

Original version of B2PLYP is no longer recommended, B2PLYP-D3(BJ) should be used instead for all situations. It is just a coincidence that B2PLYP performs very good in this instance.

In this instance, M06-2X-D3 shows its usual performance, the error is completely acceptable.


Board footer

Powered by FluxBB