TURBOMOLE Users Forum

TURBOMOLE Modules => Molecular Properties, Wavefunction Analysis, and Interfaces to Visualization Tools => Topic started by: antti_karttunen on March 05, 2008, 06:10:32 PM

Title: Cube format and ECPs
Post by: antti_karttunen on March 05, 2008, 06:10:32 PM

Hello,

I noticed a small issue when plotting molecular orbitals using the new cube format (a very useful addition, by the way!). If a system contains atoms that use an ECP, dscf and ridft do not take the ECP electrons into account when writing the cube header. For example, gold (79 electrons, 60-electron ECP) is written as "19" in the header. Due to this, visualization programs interpret the gold atoms as potassium atoms (19 electrons). This is not a big issue, as it can be easily remedied with a simple sed command wrapped in a shell script. Maybe it can be corrected in a future release...

Title: Re: Cube format and ECPs
Post by: christof.haettig on March 06, 2008, 10:02:54 AM

Can send this: turbomole@cosmologic.de ?

Do you have such a script which can provide in this forum for other users?

Regards,
 Christof

Title: Re: Cube format and ECPs
Post by: antti_karttunen on March 06, 2008, 11:40:17 AM

Hello,

I'll also report the issue to turbomole@cosmologic.de. The script I was talking about is available at
http://cc.joensuu.fi/~ankarttu/tm/tmcube.tar.gz
I have not tested it extensively, but it worked for all my cube files. It also includes another feature that is useful for those who visualize the cubes with GaussView.