This thread is locked.Only browsing is available.
Top Page > Browsing
why the change in result for the transport calculation for negf-8zgnr-0.3 when done with openmx3.5
Date: 2016/03/07 17:03
Name: Pavan choudary   <choudarykvsp@gmail.com>

as i could not install openmx3.7 as it is not compaitable with my system i tried openmx3.5 for negf calculations and when i ran the example file and plotted the graph for negf-8zgnr-0.3 there was a difference when compared with that one of 3.7 in the manual

my question is why the difference ?

i haven't changed any parameters


メンテ
Page: [1]

Re: why the change in result for the transport calculation for negf-8zgnr-0.3 when done with openmx3.5 ( No.1 )
Date: 2016/03/15 13:22
Name: T.Ozaki

Hi,

It seems that your SCF calculation is trapped to a local minimum.
For the case, if you set the following keyword as

NEGF.SCF.Iter.Band 10

you may be able to reproduce the result shown in the manual.

Regards,

TO
メンテ
Re: why the change in result for the transport calculation for negf-8zgnr-0.3 when done with openmx3.5 ( No.2 )
Date: 2016/04/08 15:45
Name: Pavan choudary  <choudarykvsp@gmail.com>

do we need to use the keyword for rest of the calculations that we want to perform?


Thank you,

regards

kvsp

メンテ
Re: why the change in result for the transport calculation for negf-8zgnr-0.3 when done with openmx3.5 ( No.3 )
Date: 2016/04/08 18:29
Name: T. Ozaki

Hi,

Especially for the NEGF calculations, converged electron density tends to depend on
the SCF history. So, it would be better to perform the SCF calculation in a rather
conservative way with the keyword.

Regards,

TO
メンテ

Page: [1]