Problem in understanding GW+BSE corrected optical response

[barnali]

Respected Daniele Varsano,
Thank you. I have reduced the damping range and now my first peak of epsilon 2 appear exactly at direct band gap.

Sincerely
Barnali
Ph.D student (CSIR SRF)
Department of physics,
Assam University, India

[haseebphysics1]

Dear Daniele,

Pardon me if my question seems silly to you! I want to know why the bandgap (whether direct or indirect) energies are printed in two columns in the report file of Yambo? i.e; what is the meaning of the third column in the following lines:

[X]Indirect Gaps [ev]: 2.322622 2.634496
[X]Direct Gaps [ev]: 2.323948 3.652211

And what you would call this material? Direct or indirect bandgap semiconductor? Since there is very minute difference between the above two values.


Thanks,

[Daniele Varsano]

Dear Haseeb,
the second numbers have not big importance, it just tells you the max value of energy difference between the bands responsible for the gap.
Here the direct and indirect gaps are pretty similar, to probably it is hard to say, what you can deduce anyway is that most probably valence (conduction) band has two similar maxima (minima) at different k points. The best thing to do is to plot the bandstructure.

Best,
Daniele

[haseebphysics1]

Dear Daniele,

I want to ask a question relating to BSE.

I have done all the ground-state calculation with a small gaussian smearing value in Quantum ESPRESSO (degauss =1.0e-02). And these smearing seems to be picked by Yambo as well, as you can see here:

Code: Select all

 [02.05] Energies [ev] & Occupations
  ===================================

  [X]Fermi Level        [ev]:  10.11708
  [X]VBM / CBm          [ev]: -1.151160  1.172789
  [X]Electronic Temp. [ev K]: 0.2587E-1  300.2   
  [X]Bosonic    Temp. [ev K]: 0.2586E-1  300.1   
  [X]Finite Temperature mode: yes
  [X]El. density      [cm-3]: 0.614E+24
  [X]States summary         : Full        Metallic    Empty
                                 0001-0160               0161-0360
  [X]Indirect Gaps      [ev]: 2.323948  2.635822
  [X]Direct Gaps        [ev]: 2.323948  3.652211
  X BZ K-points :  72

Now, my question is:

1: What is the difference between the electronic and bosonic temp as mentioned in the above report file? How it is related to the Q.E. degauss parameter.

2: I had selected small smearing in DFT just to ease the convergence, even though the system is a bulk semiconductor. But now, I am worried is it make sense to do excitonic calculations using BSE on the system? Because I have enough temp to kill or ionize the excitons?

3: And finally, I saw the symmetry info in the report file! I found this: Spatial inversion 2 is NOT a symmetry. Does this mean that the system doesn't have the center of inversion or inversion symmetry?

Code: Select all

[02.02] Symmetries
  ==================

  DL (S)ymmetries [cc]
  [S1] 1.000000  0.000000  0.000000  0.000000  1.000000  0.000000  0.000000  0.000000  1.000000

  [SYMs] Time-reversal derived K-space symmetries: 2  2
  [SYMs] Spatial inversion 2 is NOT a symmetry
  [SYMs] Group table built correctly

Thanks,

[Daniele Varsano]

Dear Haseeb,
note your degauss is not small, as it is 0.13eV.
1) degauss is not strictly related, Yambo introduce a 300K temperature.
2) Your degauss does not affect your exciton calculation, what matters is the occupation number and until the system is a semiconductor as in your case it does not mae any difference (see the report).
3) Yes, spatial symmetry is not detected, if you think that it should be there you need to check your QE calculation. Still, you have time-reversal symmetry.

Best,
Daniele

[haseebphysics1]

Dear Daniele, Thank you for your answer.

degauss is not strictly related, Yambo introduce a 300K temperature

.

So, no matter which degauss I choose, will Yambo always do calculations on 300K (~ room temp)?

And at this temp (300K), will I be able to see the exciton (if any), because they might have less binding energy and hence not survive at this temp. That's what my concern was in the last post.


Regards,

[Daniele Varsano]

Dear Haseeb,

So, no matter which degauss I choose, will Yambo always do calculations on 300K (~ room temp)?

You can set the temperature by input with the ElecTemp variable.

And at this temp (300K), will I be able to see the exciton (if any), because they might have less binding energy and hence not survive at this temp.

What does change with the temperature is the occupation number in the bands near Fermi. As long as you are in a semiconducting/insulator situation (occupied valence and empty conduction bands) nothing change.

Best,
Daniele

[haseebphysics1]

Dear Daniele,

It has become much clear to me, now. Thank you!



Regards,

[haseebphysics1]

Dear Developers,

I've got the dimension of the BSE matrix which I was not expecting according to the given input! The report file is attached. The dimension is far less than it should have been! I intended to do spin-orbit coupling calculations.

Thanks,

[Daniele Varsano]

Dear Haseeb,
according to the report you have 72 K points in the entire BZ:

Code: Select all

  X BZ K-points :  72

An you are including 31 valence and 120 conduction bands.
This gives a matrix of dimension: 267840

What is not clear?

Daniele