Page 1 of 1
The energies are different for the same structure with two different optimization
Posted: Sat Dec 24, 2022 7:31 am
by jianhui_yang1
I run a structural optimization of magnetic materials. It stopped with reached required accuracy. Then I “mv CONTCAR POSCAR”, for the next structural optimization. However, the two results of energies are different. Br further, this phenomenon always exists in this magnetic system. How can I avoid this phenomenon?
The calculation files are in the attachment.
Re: The energies are different for the same structure with two different optimization
Posted: Tue Dec 27, 2022 3:51 pm
by alexey.tal
Hi,
From the provided files, I see that you start the second calculation from scratch (ISTART=0), which means that the charge density and wave functions from the previous calculation will not be used. Essentially, the second calculation starts optimizing the charge density from the superposition of atomic charge densities for the new structure (CONTCAR of the first calculation).
If you want to continue the geometry optimization you should start the second calculation from the converged charge density.