HXC collisions not found

[sunxl]

Dear all,

Currently, I am calculating the BSE-SHG for the bulk structure. First of all, I successfully completed the calculation of the BSE absorption spectrum. Then, I removed the symmetry and performed the collision calculation. The calculation was shown to have been completed successfully with command "yambo_rt -F collision.in -J coll -C coll_files", and the output files from the collision calculation process were stored in the "coll" folder. When I used the command "yambo_nl -F shg-bse.in -J shg-bse,coll -C shg-bse_files" to calculate SHG, the following error message was displayed. The relevant input and output files have been placed in the attachment.

P1-cn82100: [WARNING] HXC collisions not found. Potential is computed on the fly.
P1-cn82100: [ERROR] STOP signal received while in[05.01] Messaging
P1-cn82100: [ERROR] HF / SEX need Collisions. Cannot be computed on the fly.

Best，
sunxl

[Daniele Varsano]

Dear Sunil,
this happens as it seems there are some inconsistencies between the collision parameters stored in the database ad the actual calculation:

Code: Select all

[RD./coll//ndb.COLLISIONS_HXC_header]-------------------------------------------
   Brillouin Zone Q/K grids (IBZ/BZ)                :   576   576   576   576
   *WRN*RL vectors                                  :   8067 [WF]
   COLLISIONS name                                  : HARTREE+SEX
   Total number of COLLISIONS                       :   98496
   COLLISIONS kind                                  :  1327
   Bands                                            :   17   34
   Bands mixing                                     :  100.0000
   CV only scattering                               : no
   Collisions Cutoff                                :  0.000000
   *ERR*Hartree  RL vectors                         :  227
   *ERR*Exchange RL vectors                         :  227

You can see that the number of G vector in the first calculation has been slightly changed by Yambo to close g vectors shell:

Code: Select all

| HXC_Potential= "SEX+HARTREE"     # [SC] SC HXC Potential
 | HARRLvcs=  6075            mRy   # [HA] Hartree     RL components
 | EXXRLvcs=  6075            mRy   # [XX] Exchange    RL components
 | CORRLvcs=  6075            mRy   # [GW] Correlation RL components

You can try to set the same number in the shg input file.

Best,

Daniele