This thread is locked.Only browsing is available.
Top Page > Browsing
esp and scale factors
Date: 2022/07/24 23:08
Name: Mehdi Vejdanihemmat   <m.vejdanih@basu.ac.ir>

Dear Developer,

1) How do we know what values are appropriate for scale factors in "esp"? or is it OK to let the program esp use its own values with command "./esp System.Name"?

2) Is the program esp appropriate for a slab with for example 15 atoms?

Best Regards,
Mehdi Vejdanihemmat
メンテ
Page: [1]

Re: esp and scale factors ( No.1 )
Date: 2022/08/16 18:19
Name: T. Ozaki

Hi,

Starting from the example shown in the manual, you can change the values and see how the result are changed.
The too large and too small values tend to give a result being far from our physical intuition.

As explained in http://www.openmx-square.org/tech_notes/tech9-1_0.pdf , the scale factors are used to find sampling points
for the fitting so that the hartree potential at the sampling points can be reproduced by the sum of effective point charge model
as much as possible. Apparently, the ESP method cannot be applied to bulks. Also, applying the ESP method for large molecules may
get in a trouble, since the atoms located in deeper inside are far from the sampling points, which may lead to a numerical
uncertainty in fitting.

So, the application might be limited to relatively small molecules.

Regards,

TO
メンテ
Re: esp and scale factors ( No.2 )
Date: 2022/07/26 15:42
Name: Mehdi Vejdanihemmat  <m.vejdanih@gmail.com>

Dear Dr. Ozaki,

Thank you very much for your useful reply.

Best regards,
Mehdi Vejdanihemmat
メンテ

Page: [1]