QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: Hesam on December 20, 2013, 20:05

Title: Random Alloy
Post by: Hesam on December 20, 2013, 20:05

Hello

I was wondering if I can create random alloys like InGaAs in ATK?

Thanks

Title: Re: Random Alloy
Post by: zh on December 21, 2013, 02:21

At the current stage, it is not possible, because the methods such as virtual crystal approximation and coherent-potential-approximation are not implemented in ATK.

Title: Re: Random Alloy
Post by: Hesam on December 21, 2013, 04:35

Thanks but what about totally random alloy model? I mean putting the atoms randomly in case of alloy (i.e. putting Indium or Gallium randomly in In(x)Ga(1-x)As).

Title: Re: Random Alloy
Post by: Nordland on December 21, 2013, 19:10

Yes this can be done however the challenge is to figure out how In(x)Ga(1-x)As looks like.

Title: Re: Random Alloy
Post by: Hesam on December 22, 2013, 03:23

Thanks, could you explain how?

There is an option to replace a selected atom in a crystal to any atom. If there be a way to loop over all cations and based on a random number select the cation would be Indium or Gallium (comparing random number to x), then we can create In(x)Ga(1-x)As easily.

Title: Re: Random Alloy
Post by: zh on December 23, 2013, 01:18

Even in such way, it is difficult to make sure that the created structures make sense. The random alloys do not have an ordered structure. To have statistical sampling, you may have to calculate huge candidate structures for  given x and supercell size.  You may refer to "http://prb.aps.org/abstract/PRB/v82/i2/e024203". In a word, it is unfeasible to create a reliable structure of random alloy by replacing the atoms randomly and simply.

Title: Re: Random Alloy
Post by: Shinji Usui on December 24, 2013, 02:48

Dear Hesam

If you want to create a completely random structure of alloy, Packmol (http://www.ime.unicamp.br/~martinez/packmol/) will be a good tool to try.

BR
Shinji