QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: marmotte on June 10, 2014, 10:15

Title: Born effective charges
Post by: marmotte on June 10, 2014, 10:15
Dear all,

Should the calculated values fulfill the acoustic sum rule \sum_s Z_{s,ij}^*=0 ?

I'm finding a huge number like 6 ? is it a problem of relaxation or strained structure ? I did calculation from the experimental structure.

Thank you
Title: Re: Born effective charges
Post by: Troels Markussen on June 10, 2014, 16:30
Hi,

The acoustic sum rule should ideally be fulfilled. It should be OK to use the experimental structure. I would suggest that you try to decrease the displacement of the atoms (delta_z in the tutorial http://quantumwise.com/documents/tutorials/latest/Polarization/index.html/chap.born.html), if you didn't already try that.

Troels
Title: Re: Born effective charges
Post by: Kim_W on October 16, 2017, 10:46
How to get accurate value of born effective charges for 2D materials (e.g. graphene, single layer black phosphorus etc.).  The result was changed with different vacuum spacing.
Title: Re: Born effective charges
Post by: Ulrik G. Vej-Hansen on October 19, 2017, 20:57
Please provide the python script(s) you have used and a more detailed explanation of what you have tried and how it differed from your expectations.
Title: Re: Born effective charges
Post by: Kim_W on October 20, 2017, 09:57
When I changed the value of c in GaN.nc script, the value of BEC was also changed slightly. Besided,  the result of my calculation is different from the previous reports of 3.08-3.23.
Title: Re: Born effective charges
Post by: Ulrik G. Vej-Hansen on October 20, 2017, 11:19
Thanks, I will look into it and get back to you.
Title: Re: Born effective charges
Post by: Ulrik G. Vej-Hansen on October 23, 2017, 13:43
Unfortunately, our current implementation is only guaranteed to work for orthogonal cells, as stated in this tutorial on polarization: https://docs.quantumwise.com/tutorials/polarization/polarization.html

I am sorry for any inconvenience this has caused, and I will make sure that we update the manual page to make this clearer.
Title: Re: Born effective charges
Post by: Kim_W on October 25, 2017, 09:22
Thank you a lot. Then, I try to turn the lattice into orthogonal cell use surface(cleave) tool, however, the result is also different from the previous reports. Can you help me?
Title: Re: Born effective charges
Post by: Ulrik G. Vej-Hansen on October 25, 2017, 13:16
I will have a look at the scripts and get back to you.
Title: Re: Born effective charges
Post by: Ulrik G. Vej-Hansen on October 26, 2017, 13:18
Could you provide the source for the reference numbers you quote, please?
Title: Re: Born effective charges
Post by: Kim_W on October 27, 2017, 04:14
[1] Computational discovery of single-layer III-V materials
[2] GaN: From three- to two-dimensional single-layer crystal and its multilayer van der Waals solids
Title: Re: Born effective charges
Post by: Ulrik G. Vej-Hansen on October 27, 2017, 10:08
Please provide the direct reference or the DOI, so I can find them. The title alone is not adequate for academic papers.
Title: Re: Born effective charges
Post by: Kim_W on October 30, 2017, 04:31
 DOI: 10.1103/PhysRevB.87.165415
 DOI: 10.1103/PhysRevB.93.085431
Title: Re: Born effective charges
Post by: Ulrik G. Vej-Hansen on October 30, 2017, 16:55
Thanks, I will take a look at them.

On closer inspection of the script, I have realized that you have simply copied the one from the tutorial directly, without modifying it. Please go back and read through the tutorial in detail, so you understand how the script is created. This should allow you to modify it to suit the 2D system you are interested in.
Title: Re: Born effective charges
Post by: Kim_W on November 1, 2017, 03:17
The question is: How to set the length of lattice vector in z-direction when calculate the BEC for 2D materials?
Title: Re: Born effective charges
Post by: Petr Khomyakov on November 1, 2017, 11:01
It is not a problem of setting the lattice vector in the z(c)-direction. Your setting for that might be good enough.

The problem is that you have to first get an idea of what is the physical meaning of Born charges for 3D and 2D systems, as well as their definition, then go through the BEC_r.py script and modify it according to the definition for 2D systems.
Title: Re: Born effective charges
Post by: Ulrik G. Vej-Hansen on November 1, 2017, 14:49
I can add that I get a value of 3.1 when I use the approach outlined in the tutorial, with the modifications mentioned by Petr.
Title: Re: Born effective charges
Post by: Kim_W on November 7, 2017, 03:25
Can you send relevant script to my email?   [email protected]
Title: Re: Born effective charges
Post by: Ulrik G. Vej-Hansen on November 7, 2017, 13:14
I will post it here, so others can also benefit from it.