Author Topic: xyz input and output files with VNL  (Read 3852 times)

0 Members and 1 Guest are viewing this topic.

Offline quantumtoday

  • Heavy QuantumATK user
  • ***
  • Posts: 29
  • Country: fr
  • Reputation: 0
    • View Profile
xyz input and output files with VNL
« on: October 10, 2016, 17:16 »
Hi,

I would like to ask about problems of xyz input and output files using with VNL:

- If I download a xyz file from somewhere (for example, http://materia.fisica.unimi.it/manini/dida/structures.html) and open by VNL, I do not see the periodic propertiers of the structure and can not repeat unit cells using "bulk tools" in Builderor and also can not view BZ.
- Similarly, If I create a periodic structure in VNL using the existing database, and then export to a xyz file. Then I import this xyz file in Vesta or Xcrysden, I can not repeat unit cells.

So do you know why I did have this problem? How can I fix it?

Many thanks.

Sincerely,
Tran
« Last Edit: October 10, 2016, 17:28 by quantumtoday »

Offline Anders Blom

  • QuantumATK Staff
  • Supreme QuantumATK Wizard
  • *****
  • Posts: 5418
  • Country: dk
  • Reputation: 89
    • View Profile
    • QuantumATK at Synopsys
Re: xyz input and output files with VNL
« Reply #1 on: October 10, 2016, 19:47 »
That's because the XYZ file format is designed for molecules, not periodic structures.
Some people have added their own extra lines to XYZ files to include the cell, however the files from the site you mention do not do that.

Offline quantumtoday

  • Heavy QuantumATK user
  • ***
  • Posts: 29
  • Country: fr
  • Reputation: 0
    • View Profile
Re: xyz input and output files with VNL
« Reply #2 on: October 11, 2016, 12:31 »
Thanks Anders Blom for the reply.

I have just experienced that the CIF format can keep periodic properties and work in both VNL and Vesta. But Xcrysden does not support this format (I remember that).

There is still a concern, how do we edit a xyz file to present a periodic structure? (since many xyz files available out there)


Offline Petr Khomyakov

  • QuantumATK Staff
  • Supreme QuantumATK Wizard
  • *****
  • Posts: 1290
  • Country: dk
  • Reputation: 25
    • View Profile
Re: xyz input and output files with VNL
« Reply #3 on: October 11, 2016, 14:22 »
When you import a XYZ file into VNL, you import it as a molecule configuration, i.e., there is not periodic box (periodic boundary conditions) assigned to it.

What you can do is to transform this "molecule" structure to bulk structure (i.e., you assign periodic boundary conditions to the structure with some default lattice parameters), using molecule to bulk plugin in the Builder, see 'Convert to bulk' at http://docs.quantumwise.com/tutorials/builder_manual/builder_plugins/builder_plugins.html.

Using Bulk Tools, you can then change the default lattice type and lattice parameters to the actual lattice type and lattice parameters of your bulk structure.   

After that, you must be able to use the Builder Tools, e.g., to repeat the structure.

Offline quantumtoday

  • Heavy QuantumATK user
  • ***
  • Posts: 29
  • Country: fr
  • Reputation: 0
    • View Profile
Re: xyz input and output files with VNL
« Reply #4 on: October 11, 2016, 15:14 »
Thanks Petr. Your suggestion is a good solution indeed.

Just sometimes after we convert to a bulk configuration, the builder detects incorrectly the crystal symmetry Info (wrong group space, for instance) , so we might need to correct manually before opening again and doing Fit Cell and Repeat.

Offline Anders Blom

  • QuantumATK Staff
  • Supreme QuantumATK Wizard
  • *****
  • Posts: 5418
  • Country: dk
  • Reputation: 89
    • View Profile
    • QuantumATK at Synopsys
Re: xyz input and output files with VNL
« Reply #5 on: October 12, 2016, 01:10 »
There is still a concern, how do we edit a xyz file to present a periodic structure? (since many xyz files available out there)

This is a bit like saying, how do we edit a JPG file to also contain text :-) You can't, because it's not part of the standard.

Of course, you can make your own file format "xyz with cell vectors", and in fact I did that myself, but it's not a standard, so no other software might be able to read the file. And, coming back to your original files, they actually do not contain the cell, so there is no way VNL could read information that's not in the file.