This thread is locked.Only browsing is available.
Top Page > Browsing
Date: 2015/10/21 22:43
Name: Yi Wang   <>

Yi Wang
School of Materials Science and Engineering
Nanjing University of Sci. & Tech.
Page: [1]

Re: student ( No.1 )
Date: 2015/10/21 22:52
Name: Yi Wang  <>

Sorry, I misunderstood the forum form. Is it possible the administrator help me change the title?

My question is "Is it possible to use openmx to simulate vacancy formation in bulk metals"

I have built a geomentry of bulk tungsten of 3*3*3 supercell of the conventional cell of bcc tungsten, and delete one atom in the center. I thought this should represent the vacancy formation under constant volume.

However, when I perform the simulation, I find the atomic forces get bigger and bigger, and the neighboring atoms around the vacancy do not move toward the vacancy, but move away from it.

The same situation also happens if I set a pseudo-atom at the vacancy position, and set the basis-set E for it.

So I want to ask is it actually possible to simulate vacancy in bulk metals and if the answer is positive how could I properly simulate the vacancy in bulk metal?
Re: student ( No.2 )
Date: 2015/10/26 19:18
Name: Artem Pulkin


Can it be that you upscaled your system (by mistake)?

Re: student ( No.3 )
Date: 2015/11/14 16:05
Name: Yi Wang  <>

Dear Artem Pulkin,

Thank you and sorry for this late reply. I have rechecked the input file and I found it might be caused by the k-point setting of my input.

I may have set a too small k point setting for the vacancy formation calculation, and causing the bulk lattice constant too large compared with a properly calculated one.

I rerun the model with a new input from scratch after serious check, and it now runs fine.

It was all my fault that I failed too check all parameters.

Many thanks for the contributors and the community, the OpenMX method is great for DFT calculations.

Page: [1]