Hello everybody!
As VASP is a difficult program to use and one can face various problems - from difficulties completing calculations, to reliability in the results - I would like to have a library of solved examples to cross examine my calculations or figure out solutions. The issue that I face now is DFT+U convergence problem but it won't be the last. I already read June Gunn Lee's "Computational Materials Science AN INTRODUCTION". It offers a good theoretical background and some solved examples but I would like something dedicated to the latter. Does somebody have something to suggest?
Dear Giannis,
Maybe you can take a look this book Computational Materials Science: From Ab Initio to Monte Carlo Methods https://amzn.to/3hPgw2X
and Density Functional Theory: A Practical Introduction
https://amzn.to/3djI2lH
I hope it will help you
Hi Giannis,
I don't think there is a complete book with solved examples. You can find all possible calculations which can be done using Vasp in the vaspwiki. Corresponding to each INCAR tags, there are lots of examples which will help you to go through and understand the basic concept.
Thank you.
Hi Romakanta,
Vaspwiki is very helpful for many cases but there are keywords that are not fully analyzed and explained especially for a beginner.
Giannis Karamanis,
The best recommendation I have is to use this e-book from John Kitchin's research group: https://github.com/jkitchin/dft-book
The Git page is a repository of example scripts that generate some results. They provide a link to a PDF version with all of the example scripts and output. They cover lots of examples in VASP, including a couple of pages on DFT+U, which might be of interest.
Note that they have used Atomic Simulation Environment (ASE) to generate the VASP inputs and process the results. If you're running VASP without any wrapper such as ASE, you should still be able to use a corresponding setup and reproduce the results.
By the way, I have no idea of the details of your convergence problem, but if you're having issues with DFT+U, make sure you look into the LMAXMIX tag for VASP. From VASPwiki:
"L(S)DA+U calculations require in many cases an increase of LMAXMIX to 4 for d-electrons (or 6 for f-elements) in order to obtain fast convergence to the groundstate."
The first (and only) time I used +U, I had severe convergence issues for Fe3O4 surfaces. Setting LMAXMIX = 4 (since the +U correction is applied to the Fe d orbital) fixed it. Back when I was working on the problem, there was nothing in the DFT+U section of the VASP manual that indicated LMAXMIX should be changed.
I think the book from David Sholl is a good one for the beginner of DFT. For using VASP, VASP wiki and the VASP discussion forum dedicated to its members are the best.
For DFT I would recommend that you can keep a hard copy of any practical book as your quick reference to go. I have the one written by Dr. David Sholl and Dr. Janice A. Steckel.
- Badges
- Science topic
- Similar topics
- Computer Science
- Program