How long (approximately) would it take to learn DFT calculations from scratch?

It depends on what you are considering "learning" it, let's define two cases:

1) You want to be able to employ one or two softwares (like ORCA Quantum Chemistry) with some degree of understanding and being able to tweak parameters and understand results: It might take from a couple weeks to months, depending or your learning process and on the subject you are researching.

2) You want to master the concepts to a level in which you can develop techniques, processes and softwares: It's something that would take years of learning and experience gathering.

Your case alligns as the first one, so, I'd say that in a couple months of hard work you are going to be able to understand what you are doing. Of course, being really good at it and being able to solve complex tasks is a set of skills that takes a lot of experience to get.

Dear Mr. Fachini,

Thank you very much for your answer.

Yes, I would like to start in the way you described as the case (1). I will probably choose the Quantum Espresso program because it is free for anyone and I will be able to install it on my personal home computer.

First I would like to investigate the band structure of various heterostructures containing SnTe or SnInTe layers with a monoclinic distortion of the unit cell (I would like to use for the calculations the real unit cell parameters which I have measured).

Then I would like to start inventing my own materials, checking (theoretically) their band structure for potential topological or thermoelectric properties. If interesting results come out, suggesting that, for example, a given material would be a Weyl/Dirac semimetal or a good thermoelectric, I might ask someone to grow such a sample to test it experimentally and/or maybe I will strive to publish it. These are my main goals, at least for the near future.

With best regards,

Adrian Sulich

Adrian Sulich, I'd say it's possible to learn how to apply it in a short time and start getting good results in some months, for sure.

The main problem with DFT on periodic conditions, like the one you described, is that they are usually quite expensive in terms of computational resources. If you try to run it on "publication" theory level/accuracy you shall face the need of alocating resources in high performance computational facilities. But to keep the optimism, I think that the learning step is surely possible on personal computers.

Also, I'd consider approaching someone experienced with the systems you are working on because said person would surely know tips on how to make the calculations less expensive without sacrificing too much on quality.

Good luck on your endeavor, you'll probably find the experience very interesting and instigating!

Dear Mr. Fachini,

thank you very much for your response.

Best regards,

Adrian Sulich