QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: Dhirendra on May 13, 2016, 08:02

Title: Ghost atoms, relaxation, workfunction
Post by: Dhirendra on May 13, 2016, 08:02

Hi,

I am getting the problems in relaxation when I include the ghost atoms. so,

How accurate the surface relaxation without ghost atoms?
And also how accurate the Workfunction calculations without ghost atoms?

Also, without ghost atoms, is it expected that with the basis set the workfunction of a metal changes significantly?

Title: Re: Ghost atoms, relaxation, workfunction
Post by: Jess Wellendorff on May 13, 2016, 09:04

For work function calculations, the ghost atoms are most important for the actual  work function calculations, less important for the geometry optimization. In general, including ghost atoms is crucial to get an accurate work function. Hard to say what the basis set dependence is, most likely system dependent...

Title: Re: Ghost atoms, relaxation, workfunction
Post by: Dhirendra on May 15, 2016, 10:19

Thanks Jess,
If I want to use an alternate way of modifying the basis, can you suggest me which parameter to look for?

Title: Re: Ghost atoms, relaxation, workfunction
Post by: Dhirendra on May 18, 2016, 11:37

Gentle Reminder.

Title: Re: Ghost atoms, relaxation, workfunction
Post by: Anders Blom on May 18, 2016, 11:44

The alternative to using ghost atoms is to extend the range of the basis functions.

Look for the keyword "radial_cutoff_radius" in the basis set definition (if you select Script detail = Show defaults in the Script Generator, these will be written to the script).