QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: Dhirendra on February 2, 2016, 13:57

Title: Ge GGA+U
Post by: Dhirendra on February 2, 2016, 13:57

Hi,
I am trying to obtain the results published here doi:10.1016/j.commatsci.2015.10.023 for GGA+U calculations on Ge.
I do not obtain the convergence for the negative values of Up.  I have attached the .py file. Am I missing something?

Title: Re: Ge GGA+U
Post by: Dhirendra on February 2, 2016, 13:59

The file has 3.7(debugging), with -3.7 convergence issues arise.

Title: Re: Ge GGA+U
Post by: Jess Wellendorff on February 3, 2016, 08:39

Tried your script, the calculation runs through just fine. What is the problem?

Title: Re: Ge GGA+U
Post by: Dhirendra on February 3, 2016, 08:44

I was debugging with what Up values converge, and in that the file I had shared has a Up=3.7. If you change it to -3.7 as suggested in the referred paper,  convergence issues arise. Actually any negative value (less than -1 or -0.5) would do the same.

Title: Re: Ge GGA+U
Post by: Anders Blom on February 3, 2016, 12:10

Check your file and compare to the attached one which runs fine and gives very good band structure.