Gap energy Eg

Posted on : September 27, 2015 (Sun) 21:17:28

by Mariem

Dear;

I manually determine the value of Eg from Origine software but
the value is not specific.
How ca I get accurate value of Eg?

best regards
Mariem

[Re:01] Gap energy Eg

Posted on : September 28, 2015 (Mon) 19:43:54

by Hitoshi

Hello Mariem,

Would you please let me know how to determine the energy gap by using the Origine software?
As far as I know, AkaiKKR does not output the value of energy gap.
But I possibly help you to read the value from the result.

Hitoshi

[Re:02] Gap energy Eg

Posted on : September 29, 2015 (Tue) 05:03:47

by MARIEM

Dear Hitoshi;

Manually, I determine the energy gap by making a zoom of the graph (DOS). Then I used the pointer to get better precision values of energy. Then calculate Eg by subtraction.

The result is different when I adjust a zoom of image (large graph) and when I make the graph smaller.

That why I asked you about the accurate way to get Eg.

best regards
Mariem

[Re:03] Gap energy Eg

Posted on : September 29, 2015 (Tue) 15:30:30

by Hitoshi

Hello Mariem,

Okey, I understood.

* After the go calculation, you can reduce ewidth for subsequent dos or spc calculations.
* msex in source/specx.f determines the energy resolution of dos and spc calculations. You can increase it.
* spc is better than dos in order to find E_CBM and E_VBM.

Following is an example.
Please use the second latest version (May 22, 2015) or older, do not use the latest (August 26, 2015).
GaAs is a direct band gap semiconductor, which has CBM and VBM at the Gamma point.

First, do go calculation to get potential.

c----------------------GaAs----------------------------------
go data/gaas
c------------------------------------------------------------
c brvtyp a c/a b/a alpha beta gamma
fcc 10.684 , , , , , ,
c------------------------------------------------------------
c edelt ewidth reltyp sdftyp magtyp record
0.001 1.5 sra mjwasa nmag 2nd
c------------------------------------------------------------
c outtyp bzqlty maxitr pmix
update 4 200 0.035
c------------------------------------------------------------
c ntyp
4
c------------------------------------------------------------
c type ncmp rmt field mxl anclr conc
Ga 1 1 0.0 2 31 100
As 1 1 0.0 2 33 100
Vc1 1 1 0.0 0 0 100
Vc2 1 1 0.0 0 0 100
c------------------------------------------------------------
c natm
4
c------------------------------------------------------------
c atmicx atmtyp
0 0 0 Ga
0.25 0.25 0.25 As
0.5 0.5 0.5 Vc1
0.75 0.75 0.75 Vc2
c------------------------------------------------------------

For spc (or dos) calculation, edit source/specx.f to increase msex (say msex=401).
And also, edit source/cemesr.f from "data ref/0.75d0/" to "data ref/0.5d0/".
Then do make.

After that, please prepare the input file for spc calculation.
As mentioned above, both of CBM and VBM is located at the Gamma point.
Decrease the ewidth.

c----------------------GaAs----------------------------------
spc data/gaas
c------------------------------------------------------------
c brvtyp a c/a b/a alpha beta gamma
fcc 10.684 , , , , , ,
c------------------------------------------------------------
c edelt ewidth reltyp sdftyp magtyp record
0.001 0.2 sra mjwasa nmag 2nd
c------------------------------------------------------------
c outtyp bzqlty maxitr pmix
update 4 200 0.035
c------------------------------------------------------------
c ntyp
4
c------------------------------------------------------------
c type ncmp rmt field mxl anclr conc
Ga 1 1 0.0 2 31 100
As 1 1 0.0 2 33 100
Vc1 1 1 0.0 0 0 100
Vc2 1 1 0.0 0 0 100
c------------------------------------------------------------
c natm
4
c------------------------------------------------------------
c atmicx atmtyp
0 0 0 Ga
0.25 0.25 0.25 As
0.5 0.5 0.5 Vc1
0.75 0.75 0.75 Vc2
c------------------------------------------------------------
c kvector
0.0 0.0 0.0
c------------------------------------------------------------

After the spc calculation, you can find data/gaas_up.spc. The second column is energy and the third is the Bloch-spectrum function (BSF).
Plot the BSF as function of energy, you may find two peakes: one is VBM and the other is CBM.
In order to find a peak center, fit to Lorentzian (or Gaussian or something else, I do not know which is the best.).





Exactlly speaking, the BSF should be delta-function-like peak without any energy uncertainty for ordered system.
This broadening comes from numerical reason (refer to page 29-30 of the following URL), and it may be prevented by decreasing the edelt.
ttp://www.slideshare.net/cms_initiative/cmsi-38020198

Hitoshi