Alter stoichiometry of existing compounds/recalculate

Science Talk
1

Hello, Thanks for this rich and helpful website.
I intend to explore two to three ternary compounds Al2InN3 (mp-1246874) and also AlIn2N3 (and possibly AlInN2).
My questions are:

  1. Is it possible to build my own large slabs of these compounds and have some automated-via-the-present-website DFT calculation (geometry optimized, total energy, DOS) of them. Say VASP calculations?

  2. Is it possible to alter the stoichiometry of existing compounds. For ex., Al2InN3 (mp-1246874) exists in the data basis, but is it possible to change it to say Al(2-x)In(1+x)N3 and re-evaluate (geometry, energy)?

Thanks in advance!

2

Hi @Gueorgui_Gueorguiev,

  1. We’re not a compute on-demand service. We do have the MPComplete service, which lets users upload new structures, that we’ll compute as the resources are available. That being said, we purposely limit what gets computed to well structured primitive cells with as few atoms as possible. All the tools we use for high-throughput DFT are open-source, so you’re more than welcome to run the calculations yourself.

  2. You can definitely make partially disordered structures and then use one of many ordering algorithms to enumerate orderings and compute their properties. These are available either directly or via interfaces in pymatgen. There are several theoretical frameworks to describe the properties of an alloy or partially disordered system based on the properties on the ensemble of ordered structure properties.

3

Hi Shyamd,

Many thanks for your answer few days ago. I have some follow-up questions.

  1. Regarding the first point of our previous conversation, I intend to try the MPComplete service, sticking to primitive cells with size and symmetry that already have been calculated and exist in your data basis (such as Al2InN3) but with altered stoichiometry ratio(s). Would a row of such variations over existing compounds be eligible for calculation?

  2. When you mention that one could make partially disordered structures and then use one of many ordering algorithms to enumerate orderings and compute their properties
    directly or via pymatgen, could you be more specific? I am not familiar with pymatgen, how easy/intuitive is to use it?

  3. Regarding the “Interface Reactions” app. This app certainly provides diversity of interesting possibilities. I read the documentation, but I still have some questions: Is there any universal definition/expression for the chemical potential variable “myu” as used? Is “myu” always the chemical potential (as usually defined in chemical physics) of the open element? Can I use “Interface Reactions” app to examine interface of two solids, e.g., InN and AlN, exposed to N2 (if yes, how I can determine the “myu” say at known temp and pressure)? What if open element is gas phase atomic Indium? Should I take the chemical potential of In? How can I account for the temp.?

Thanks again for your kind help.
Gueorgui