problem using ypp -e a

[waynebeibei]

when I use ypp -e a, I got the strange o.exc_weights_at_1 file just like viewtopic.php?f=9&t=159 and viewtopic.php?f=9&t=376. I do have "WRbsWF" in the BSE inuput file.

Below is my output:


# Electron-Hole pairs that contributes to Excitonic State 1 more than 5%
#
# K-point [iku] Weight
# 0.000000 0.000000 0.000000 0.324921
# 0.166667 0.000000 0.000000 0.450354
# 0.333333 0.000000 0.000000 0.333558
# 0.500000 0.000000 0.000000 0.137423
# 0.000000 0.166667 0.000000 0.650715
# 0.166667 0.166667 0.000000 1.000000
# 0.333333 0.166667 0.000000 0.809775
# 0.500000 0.166667 0.000000 0.361165
# 0.000000 0.333333 0.000000 0.568663
# 0.166667 0.333333 0.000000 0.977303
# 0.333333 0.333333 0.000000 0.928625
# 0.500000 0.333333 0.000000 0.475799
# 0.000000 0.500000 0.000000 0.249858
# 0.166667 0.500000 0.000000 0.447708
# 0.333333 0.500000 0.000000 0.458124
# 0.500000 0.500000 0.000000 0.250445
#
# Band_V Band_C K ibz Symm. Weight Energy
#
#
# YPP@node63 x 008 CPUs * 01/07/2014 16:38 [start]
# 01/07/2014 16:38 [end]
#
# Cpu Timing [Min/Max/Average]: 02s/02s/02s

The possible reason maybe relate to 'none of the electron-hole pairs you have satisfy this statement(more than 5%)' according to Dr.Daniele Varsano. But how to lower this threshold in the code?

[waynebeibei]

Hi all

I have fixed this problem after recompiling the code and modifying the threshold in excitons_sort_and_report.F. Now the o.exc_weights_at_1 file is:

# Electron-Hole pairs that contributes to Excitonic State 1 more than 2%
#
# K-point [iku] Weight
# 0.000000 0.000000 0.000000 0.381138
# 0.166667 0.000000 0.000000 0.438987
# 0.333333 0.000000 0.000000 0.257447
# 0.50000 0.00000 0.00000 0.09580
# 0.000000 0.166667 0.000000 0.714074
# 0.166667 0.166667 0.000000 1.000000
# 0.333333 0.166667 0.000000 0.664368
# 0.500000 0.166667 0.000000 0.267265
# 0.000000 0.333333 0.000000 0.507086
# 0.166667 0.333333 0.000000 0.845497
# 0.333333 0.333333 0.000000 0.719606
# 0.500000 0.333333 0.000000 0.342872
# 0.000000 0.500000 0.000000 0.196593
# 0.166667 0.500000 0.000000 0.348025
# 0.333333 0.500000 0.000000 0.333520
# 0.500000 0.500000 0.000000 0.174631
#
# Band_V Band_C K ibz Symm. Weight Energy
#
32.00 33.00 5.000 3.000 0.3182E-1 2.866
32.00 33.00 5.000 1.0000 0.3179E-1 2.866
32.00 34.00 5.000 1.0000 0.2961E-1 2.901
32.00 34.00 5.000 3.000 0.2952E-1 2.901
30.00 33.00 1.0000 1.0000 0.2369E-1 2.901
30.00 34.00 1.0000 1.0000 0.2236E-1 2.922
32.00 33.00 9.000 1.0000 0.2175E-1 3.292
32.00 33.00 9.000 3.000 0.2174E-1 3.292
#
# YPP@node71 x 008 CPUs * 01/07/2014 19:51 [start]
# 01/07/2014 19:51 [end]

Now I have still two questions:
1)what is the meaning of 'Symm.' following 'K ibz' ?
2)according to the output, the excitonic state 1 may arise mainly from the transitions between Band_V(32) to Band_C(33) or from Band_V(32) to Band_C(34). When I calculate the exciton binding energy for this state, which GW gap should be adopted, E(gw gap)[32-33] or E(gw gap)[32-34]?

[Daniele Varsano]

Dear Waynebeibei,

I have fixed this problem after recompiling the code and modifying the threshold in excitons_sort_and_report.F

Nice you solved, we should put the threshold as input variable, we will do it asap.

1)what is the meaning of 'Symm.' following 'K ibz' ?

In the output you have the k-points in the irreducible Brillouin zone, next this k-point are expanded in all the Bz via symmetry operation of your system. You can see in the report the symmetry operation orrespondig to the index reported in the output.

2)according to the output, the excitonic state 1 may arise mainly from the transitions between Band_V(32) to Band_C(33) or from Band_V(32) to Band_C(34). When I calculate the exciton binding energy for this state, which GW gap should be adopted, E(gw gap)[32-33] or E(gw gap)[32-34]?

This a question of definition.
Usually the binding energy is related to the minimum electronic gap of the system, in your case [32-33].

Best,

Daniele

[waynebeibei]

Hi Dr. Daniele

Thanks for your timely and clear reply.
I understand now:)