Page 1 of 1

Negative optical absorption spectrum

Posted: Fri May 28, 2010 8:26 am
by sdwang
Dear all:
I have run a 2D SiH which unit cell has 4Si atoms and 2H atoms, and the cell is hexagon. After I run scf and nscf, I run yambo to calculate RPA with LF effect, but the OAS has some negative value in some energy window,what is the problem?
My procedures and inputfile are:
SCF:
&control
title='sih'
calculation='scf',
restart_mode='from_scratch',
prefix='sih'
pseudo_dir = '/home/sdwang/espresso-4.1.2/pseudo',
outdir='/home/sdwang/xx/'
tprnfor=.true.
tstress=.true.
/
&system
ibrav = 4, celldm(1)=7.3129,celldm(3)=5.9428, nat=6, ntyp=2,nosym = .false.
ecutwfc = 20.0,force_symmorphic=.true.,
/
&electrons
diago_thr_init = 1.0e-6
diago_full_acc = .true.
/
ATOMIC_SPECIES
Si 28.086 Si.pz-vbc.UPF
H 1.0 H.pz-vbc.UPF
ATOMIC_POSITIONS {crystal}
Si 0.018854666 0.989439465 0.069572870
Si 0.685506252 0.322743807 0.202897090
Si 0.352171736 0.656074748 0.234737189
Si 0.685520109 0.322770386 0.101412873
H 0.018859178 0.989446202 0.003955405
H 0.352168058 0.656065391 0.300354573
K_POINTS (automatic)
6 6 1 0 0 0
NSCF:
&control
title='sih'
calculation='nscf',
restart_mode='from_scratch',
prefix='sih'
pseudo_dir = '/home/sdwang/espresso-4.1.2/pseudo',
outdir='/home/sdwang/xx/'
wf_collect= .true.
/
&system
ibrav = 4, celldm(1)=7.3129,celldm(3)=5.9428, nat=6, ntyp=2,nbnd=80,nosym = .false.
ecutwfc = 20.0,force_symmorphic=.true.,
/
&electrons
diago_thr_init = 1.0e-6
diago_full_acc = .true.
/
ATOMIC_SPECIES
Si 28.086 Si.pz-vbc.UPF
H 1.0 H.pz-vbc.UPF
ATOMIC_POSITIONS {crystal}
Si 0.018854666 0.989439465 0.069572870
Si 0.685506252 0.322743807 0.202897090
Si 0.352171736 0.656074748 0.234737189
Si 0.685520109 0.322770386 0.101412873
H 0.018859178 0.989446202 0.003955405
H 0.352168058 0.656065391 0.300354573
K_POINTS (automatic)
6 6 1 0 0 0

Then I run
>p2y -N -S
>yambo -N:
<---> [01] Job Setup
<---> [02] Input variables setup
<---> [02.01] Unit cells
<---> [02.02] Symmetries
<---> [02.03] RL shells
<15s> Shells finder |####################| [100%] 15s(E) 15s(X)
<15s> [02.04] K-grid lattice
<15s> [02.05] Energies [ev] & Occupations
<15s> [03] Transferred momenta grid
<15s> X indexes |####################| [100%] --(E) --(X)
<15s> SE indexes |####################| [100%] --(E) --(X)
<15s> [04] Game Over & Game summary
>yambo -i
setup # [R INI] Initialization
MaxGvecs= 2000
>yambo -N
>yambo -o c
optics # [R OPT] Optics
chi # [R CHI] Dyson equation for Chi.
% QpntsRXd
1 | 1 | # [Xd] Transferred momenta
%
% BndsRnXd
1 | 80 | # [Xd] Polarization function bands
%
NGsBlkXd= 155 RL # [Xd] Response block size
% EnRngeXd
0.00000 | 10.00000 | eV # [Xd] Energy range
%
% DmRngeXd
0.10000 | 0.10000 | eV # [Xd] Damping range
%
ETStpsXd= 100 # [Xd] Total Energy steps
% LongDrXd
And for "NGsBlkXd" , I have test different value from 20 to 155, but the OAP is still has negative value.Why?
But when I set NGsBlkXd=1, there is no LF effect,the OAP seems OK.
When include local field effect, what is the value of "NGsBlkXd" I shoul set??
Thanks!

Re: Negative optical absorption spectrum

Posted: Fri May 28, 2010 9:22 am
by Daniele Varsano
Dear Shudong,
When include local field effect, what is the value of "NGsBlkXd" I shoul set??
This, of course, depend on your system, and as other variables have to be carefully checked
until you have a converged results. About your negative spectra, it is not easy to see what's
happening. You should post your report and output. Anyway, in which direction are you calculating
your spectra:
% LongDrXd
the input you reported is truncated, and if it is really like that it is wrong as the direction it
is not set.

Cheers,

Daniele

Re: Negative optical absorption spectrum

Posted: Fri May 28, 2010 9:58 am
by sdwang
Daniele Varsano wrote:Dear Shudong,
When include local field effect, what is the value of "NGsBlkXd" I shoul set??
This, of course, depend on your system, and as other variables have to be carefully checked
until you have a converged results. About your negative spectra, it is not easy to see what's
happening. You should post your report and output. Anyway, in which direction are you calculating
your spectra:
% LongDrXd
the input you reported is truncated, and if it is really like that it is wrong as the direction it
is not set.

Cheers,

Daniele

I am sorry.The LongDrXd=1.000000 | 0.000000 | 0.000000 | # [Xd] [cc] Electric Field
%
I post my report and output in attahment.

Thanks!