QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: nidhi pandey on February 26, 2020, 12:05

Title: Geometry optimization
Post by: nidhi pandey on February 26, 2020, 12:05

I am trying to build a device for molecular spintronics using nickel and benzene dithiolate. The geometry optimization is almost taking 48 hours on 8 cores and still not converging. Can anybody help? thanks in advance.

Title: Re: Geometry optimization
Post by: hsuya on February 27, 2020, 06:30

Can you share the script?

Title: Re: Geometry optimization
Post by: nidhi pandey on February 27, 2020, 07:38

This is the script.

Title: Re: Geometry optimization
Post by: hsuya on February 27, 2020, 07:56

Firstly, start with less number of k-points and density mesh cut-off values and then move ahead from there.
Also I think you should relax the structure before making the device.
And try using MPI parallel processes if you haven't already.