<![CDATA[Multiwfn forum / Regarding using formchk to convert chk file of ONIOM task]]> - //www.umsyar.com/wfnbbs/viewtopic.php?id=607 Thu, 13 Jan 2022 06:54:49 +0000 FluxBB <![CDATA[Regarding using formchk to convert chk file of ONIOM task]]> //www.umsyar.com/wfnbbs/viewtopic.php?pid=2182#p2182 It is frequently asked how to use Multiwfn to perform wavefunction for fch file produeced by ONIOM task of Gaussian. In this post I would like to describe this point.

By default, the fch file converted from chk file of ONIOM task via formchk doesn't record wavefunction information, that means it is impossible to use Multiwfn to perform wavefunction for it.

Actually, the orbitals calculated by different levels in ONIOM are recorded in different fields of chk file, -XY option of formchk should be explicitly used to decide which part will be extracted, the meaning of X and Y is shown below

X: L, M, H corresponds to Low, Medium, High computational levels, respectively
Y: L, M, S corresponds to Large, Medium, Small regions, respectively.

In the situation of ONIOM2 (i.e. two levels are involved in ONIOM), there are three cases:

Using high computational level to calculate "model": -MS
Using low computational level to calculate "model": -LS
Using low computational level to calculate "real": -LM
in this context, "real" refers to the whole system, and "model" refers to high layer atoms.

Therefore, for an ONIOM(QM:MM) task, if you want to use Multiwfn to conduct wavefunction analysis for the high layer calculated by QM, you should use "formchk -MS xxx.chk" to convert xxx.chk to xxx.fch.

I found the above method works for Linux version, but for Windows version the aforementioned option of formchk doesn't work properly, at least for G16 B.01.

]]>
Thu, 13 Jan 2022 06:54:49 +0000 //www.umsyar.com/wfnbbs/viewtopic.php?pid=2182#p2182