This thread is locked.Only browsing is available.
Top Page > Browsing
ESP
Date: 2010/11/24 00:48
Name: Gabriel Greene   <gabriel.greene@tyndall.ie>

To whom it may concern,

I am attempting to use the electro-static potential fitting method to calculate the atomic potentials between an Fe-C interface (for the purposes of schottky barrier analysis).

However, the program complains about an unknown van der waals radius and asks me to set the van der waals radius for each atom. Is there a keyword in the input file I can use to specify this?

******************************************************************
******************************************************************
esp: effective charges by a ESP fitting method
Copyright (C), 2004, Taisuke Ozaki
This is free software, and you are welcome to
redistribute it under the constitution of the GNU-GPL.
******************************************************************
******************************************************************

Constraint: charge
Scale factors for vdw radius 1.40000 2.00000
unknown van der Waal radius of atom 26
Please set your value for van der Waal radius of atom 26

I should also note that atom 26 corresponds to an iron atom.

Thanks very much for your help,
Regards,
Gabriel Greene
メンテ
Page: [1]

Re: ESP ( No.1 )
Date: 2010/11/27 03:36
Name: T.Ozaki

Hi,

It is possible to define the vdW radius in a function 'set_vdw()' in the code 'esp'.

But I do not recommend to use EPS for such a case, since ESP is expected to work
for small molecules. For large molecules and bulks, the ESP method inherently
possesses large uncertainty in determining the ESP charge especially for atoms being
far from the sampling points which are used for fitting the electrostatic potential.
In case of large molecules and bulks, such a situation easily happens for atoms which
are located in the inner region. Although several extensions of ESP for bulks have been
recently developed by other groups, the current version of OpenMX has no functionality
for the extension. Instead, the Mulliken and Voronoi analysis can be a better choice
in OpenMX at this moment.

Regards,

TO
メンテ
Re: ESP ( No.2 )
Date: 2010/11/26 23:37
Name: Gabriel Greene  <gabriel.greene@tyndall.ie>


Thanks very much for the response and helpful suggestions.

Best regards,
Gabriel Greene
メンテ

Page: [1]