This thread is locked.Only browsing is available.
Top Page > Browsing
NEGF convergence
Date: 2015/10/02 18:28
Name: Artem Pulkin

Hi,

I am trying to converge quite a tricky system. It is a slab of Bi2Te3 inside a bulk of magnetic Co. The most successful configuration I found so far is

scf.Init.Mixing.Weight 0.003
scf.Min.Mixing.Weight 0.001
scf.Max.Mixing.Weight 0.005
scf.Mixing.History 100
scf.Mixing.StartPulay 200

The problem is that the Pulay mixing still does not work, after 200th iteration it disturbs the charge density out of equilibrium. The total charge near 200th iteration looks like this:

total= 894.38344 ideal(neutral)= 894.00000
total= 894.38339 ideal(neutral)= 894.00000
total= 894.38335 ideal(neutral)= 894.00000
total= 894.38334 ideal(neutral)= 894.00000
total= 869.92569 ideal(neutral)= 894.00000
total= 587.09154 ideal(neutral)= 894.00000
total= 499.45050 ideal(neutral)= 894.00000

If I try to increase the mixing weights I also have problems with simple mixing during first 200 iterations. What else can I do for this system? Is it even possible to use Pulay mixing here?

Regards,

Artem
ƒƒ“ƒe
Page: [1]