Multiwfn forum
Multiwfn official website: http://sobereva.com/multiwfn. Multiwfn forum in Chinese: http://bbs.keinsci.com/wfn
You are not logged in.
- Topics: Active | Unanswered
#1 2021-07-18 19:57:26
- splinter
- Member
- Registered: 2020-12-15
- Posts: 8
Obtaining Real space functions at different molecular orientations
I know that Multiwfn can compute a real space function (like electron density and related functions) on a 3D grid at the standard molecular orientation provided in the wfx/fchk file (for Gaussian). However, I would like to obtain the grid data at multiple molecular orientations (which could be obtained by rotating the molecule). A brute force way to do that would be to generate wfx files at different molecular orientations and then process them using the Multiwfn software. However, I was wondering if there is an inbuilt functionality in Multiwfn to obtain the grid data at different orientations using only one electronic structure computation. I would be grateful for any help.
Offline
#3 2021-08-05 18:25:28
- splinter
- Member
- Registered: 2020-12-15
- Posts: 8
Re: Obtaining Real space functions at different molecular orientations
Thank you for your reply, and apologies for my late response.
I was thinking of a way to resolve my issue within the framework of Multiwfn. To do that, I was trying to understand the source code of Multiwfn. Broadly speaking, the real space functions subroutine takes the cartesian coordinates (of the grid) as input and returns the value of the function at the provided grid point.
The energies and properties of a molecule don't change upon its rotation. The same is true for a real space function associated with a molecule, as the MO energies don't vary with molecular rotation. Unfortunately, I was under the wrong impression that the optimized MO coefficients also don't change with molecular rotation. But after comparing wfx files of a molecule with different orientations, the MO energies were the same (as they should be); however, the MO coefficients were quite different for the two molecules. The reason behind the difference in MO coefficients seems to be the algorithm to perform SCF. I was wondering if some mathematical operation could be applied to the MO coefficients to obtain the coefficients for the same molecule in a different orientation.
I would be grateful if you could make any remarks on my thoughts or provide feedback if I am thinking in the right direction.
Offline
#4 2021-08-05 23:11:28
Re: Obtaining Real space functions at different molecular orientations
Orbital shape and orbital energy are independent of molecular orientation, however, the expansion coefficients are commonly dependent of orientation. This point is easy to understand: Norm of a vector is independent of orientation, while its X,Y,Z components are dependent of orientation.
It is not quite easy to write your own code to rotate coefficients. If you are using Gaussian, after performing single point at orientation A, then when you calculate single point of orientation B, you can use guess=read keyword to read wavefunction in .chk file of orientation A, the SCF convergence should be reached after only one SCF cycle, so the cost is negligible.
Offline