.py file from geometry optimisation

[Para_13]

Using VNL 10.8 I created a script that I did a geometry optimisation on and had a look at the trajectory file in the viewer window. Is there a way to get a script from this .nc file to use in DFT calculations? I attempted to open it using word pad and copy the cooridinates of the final relaxed configuration but I think there must be an easier way.

[Para_13]

Sorry, to clarify, I'm trying to get a python script of the most energetically favourable system after the geometry optimisation has been performed. Is this possible?

[Anders Blom]

Certainly. The NC file should, if all went well, contain two configurations, the original one and the optimized one. If the script was produced by VNL, most likely the original one has Id=gID000 and the optimized one gID001.

If you select the NC file in the VNL file browser, these two objects should be visible in the Result Browser (the top-right panel). If you now drag the configuration (not the NC file, but the configuration object) to the Builder, you can save it as a Python file. You can also drag it directly to the Editor.

[Para_13]

Thanks.

Building on what I was doing with the diamond lattice, I'm now trying to set up a H-terminated surface using VNL 10.8.2 but when I attempt the geometry optimisation the entire program crashes. I was wondering if you have heard of this happening before?
 
Also, I have attatched a copy of the script I was using. This is before I have attempted the geometry optimisation and I've added the two C and two H atoms at the surface. I apologise for the way the script is presented but that's the way it saves and I have no idea why.
 
Could this be because of a bug in the program?

[Anders Blom]

This is not a large system, so I would be quite surprised if it cannot run. I tested it with 1x1x1 k-points on my 10.8.2 and although it's a bit slow to converge it runs fine.

(I didn't run a geometry optimization, just one step. For more assistance you will need to show the actual script you run, that failed. This is generally the case if asking a question about a script that fails; there's no way to test what that problem is without access to the script.)