Author Topic: Potential for interface  (Read 4545 times)

0 Members and 1 Guest are viewing this topic.

Offline berna

  • Heavy QuantumATK user
  • ***
  • Posts: 56
  • Country: tr
  • Reputation: 0
    • View Profile
Potential for interface
« on: December 19, 2016, 00:41 »
If there was not any potential for interface what should be done?

Offline zh

  • Supreme QuantumATK Wizard
  • *****
  • Posts: 1141
  • Reputation: 24
    • View Profile
Re: Potential for interface
« Reply #1 on: December 19, 2016, 05:03 »
Your question is too short. Please let your question understandable to others, if you ask help.

Offline berna

  • Heavy QuantumATK user
  • ***
  • Posts: 56
  • Country: tr
  • Reputation: 0
    • View Profile
Re: Potential for interface
« Reply #2 on: December 19, 2016, 11:24 »
I wrote pages full before than nothing. If I don't want help, why do I spend my time here! I think my question is quite understandable. Because there is only one place using potential (Tersoff, ReaxFF, Pedone). But my question is in long form:
I want to create ZnO(10-10)-Au(111) surface.  "calculate distance using potential" doesn't active in shift-surface which is a part of the  interface widget,

Offline Daniele Stradi

  • Supreme QuantumATK Wizard
  • *****
  • Posts: 286
  • Country: dk
  • Reputation: 3
    • View Profile
Re: Potential for interface
« Reply #3 on: December 19, 2016, 12:04 »
Dear Berna,

thank you for sharing the details of your issue with the ATK community.

Classical potentials such as those you want to used are not only used in the Interface Builder, which I guess from your question is the VNL plugin you are actually using. They are actually mostly used in ATK as a calculator and as such they can be selected in the Script Generator.

Unfortunately, if the shift-surface button is not active, it means that there is no classical potential in ATK available   
for your desired combination of elements (in this case, Zn, O, Au). If you have your own potential, you can actually implement it as specified in: http://docs.quantumwise.com/tutorials/combining_potentials/combining_potentials.html.

Another, probably easier option is to shift the surface manually and set the distance using chemical intuition: for most system, bonding distances are in the range 2.0-3.5 Angstrom, so setting a distance in this range will probably provide a reasonable starting guess. Then, you can optimize the structure of the interface using DFT, which is also implemented in ATK. Although I am not an expert, I am aware that the ZnO-Au interface has been studied a lot, so I think you can also check the literature to obtain reasonable estimates for the correct binding distance.

Regards,
Daniele.






Offline leszek.nowakowski

  • New QuantumATK user
  • *
  • Posts: 1
  • Country: pl
  • Reputation: 0
    • View Profile
Re: Potential for interface
« Reply #4 on: October 24, 2024, 14:44 »
Hello, I'm aware it's been 8 years almost since the topic was created, but I could not find the solution for a very similar question. I'm also building an interface composed of elements, which doesn't have build-in potential in QATK (cerium), so automatic calculation of shifting is impossible. The procedure of adding own potential ( as desrcibed by Daniele ) http://docs.quantumwise.com/tutorials/combining_potentials/combining_potentials.html is not valid for calculations of shifting interfaces, since it can't be send straight to the editor so I can't really see the source code, and finding it by myself is kind of tricky. So i have a two questions: -> is there a possibility to add own potential somehow (maybe save it in a file and add to a set of prebuild potentials in QuantumATK), and then this potential can be used in Shift Surfaces window; -> if there is not such an option, I can build the script myself by Iteration, as described in https://www.youtube.com/watch?v=6tL77iZDJsY&ab_channel=SynopsysQuantumATK . So I have to move the one of the materials from interface (fortunately atoms building the materials are tagged "Left Interface" and "Right Interface") along x, y, z (this can be actually quite easy with moveAtoms() function apllied to tagged atoms) and also rotate them along z-axis ( I don't know how to rotate atoms). So iterating through a set of move_x, move_y, move_z and rotate_z parameters and calculating the total energy with a set potential will do the job, but I'm struggling to do that. Could You help me please? EDIT: I managed to set up Iterative calculation. I have a variable from generator called move_x, which should move the atoms of one part of the interface along x-axis. I order to do that, I set up script in Custom block : configuration.shiftAtoms([move_x, 0,0]), but it moves all of the atoms in configuration. How to select with this script only these atoms which are tagged "Interface_Left_0" ??? EDIT: I found it myself in the documentation. You can use
Code
configuration.shiftAtoms([move_x, 0, 0]*Angstrom, configuration.indicesFromTags(tags="Left Interface 0"))
Code
configuration.indicesFromTags(tags="Left Interface 0")
picks the indices of atoms with specific tag, and this indices can be sent to shiftAtoms.
« Last Edit: October 25, 2024, 15:27 by leszek.nowakowski »