Page 1 of 1

Number of states in QE vs GW for PDOS calculation

Posted: Fri Jan 09, 2026 2:36 pm
by Dhanjit
Dear all,
I saw a strange difference in the DOS plots of DFT vs GW, particularly in the number of states/eV (y axis value). The values differ significantly. The files are attached for reference. Am I doing something wrong in the GW_pdos input file? DFT I did with nspin =2, and as the states are symmetric (non-magnetic), I did the GW calculation with no spin.
DOS input

Code: Select all

# .-Input file  ypp.in
# | infver                           # [R] Input file variables verbosity
# | electrons                        # [R] Electronic properties
# | dos                              # [R] DOS
# | ElecTemp= 0.025878         eV    # Electronic Temperature
# | BoseTemp= 0.025861         eV    # Bosonic Temperature
# | PROJECT_mode= "ATOM"             # Instruct ypp how to project the DOS. ATOM, LINE, PLANE.
# | PROJECT_path= "none"             # Project path. PLANE mode: 3 points in the plane. LINE mode: any series of points.
# | PROJECT_steps=-1                 # Project LINE/PLANE-axis steps.
# | PROJECT_depth= 1                 # Project perpendicular direction divisions.
# | INTERP_mode= "NN"                # Interpolation mode (NN=nearest point, BOLTZ=boltztrap aproach)
# | INTERP_Shell_Fac= 20.00000       # The bigger it is a higher number of shells is used
# | INTERP_NofNN= 1                  # Number of Nearest sites in the NN method
# | % INTERP_Grid
# | -1 |-1 |-1 |                             # Interpolation BZ Grid
# | %
# | GfnQPdb= "E < ./all_Bz/ndb.QP"   # [EXTQP G] Database action
# | % DOSERange
# | -9.000000 | 9.000000 |         eV    # Energy range
# | %
# | DOSESteps= 1000                  # Energy steps
# | DOS_broad= 0.100000        eV    # Broadening of the DOS
# | % DOS_bands
# |    1 | 200 |                         # DOS bands
# | %
# | #DOSUsesOcc                    # Weight the DOS with the occupation factors
PDOS input for a 2s orbital

Code: Select all

# .-Input file  ypp.in
# | infver                           # [R] Input file variables verbosity
# | electrons                        # [R] Electronic properties
# | dos                              # [R] DOS
# | ElecTemp= 0.025878         eV    # Electronic Temperature
# | BoseTemp= 0.025861         eV    # Bosonic Temperature
# | PROJECT_mode= "ATOM"             # Instruct ypp how to project the DOS. ATOM, LINE, PLANE.
# | PROJECT_path= "none"             # Project path. PLANE mode: 3 points in the plane. LINE mode: any series of points.
# | PROJECT_steps=-1                 # Project LINE/PLANE-axis steps.
# | PROJECT_depth= 1                 # Project perpendicular direction divisions.
# | INTERP_mode= "NN"                # Interpolation mode (NN=nearest point, BOLTZ=boltztrap aproach)
# | INTERP_Shell_Fac= 20.00000       # The bigger it is a higher number of shells is used
# | INTERP_NofNN= 1                  # Number of Nearest sites in the NN method
# | % INTERP_Grid
# | -1 |-1 |-1 |                             # Interpolation BZ Grid
# | %
# | GfnQPdb= "E < ./all_Bz/ndb.QP"   # [EXTQP G] Database action
# | % DOSERange
# | -9.000000 | 9.000000 |         eV    # Energy range
# | %
# | DOSESteps= 1000                  # Energy steps
# | DOS_broad= 0.100000        eV    # Broadening of the DOS
# | % DOS_bands
# |    1 | 200 |                         # DOS bands
# | %
# | #DOSUsesOcc                    # Weight the DOS with the occupation factors
# | % PDOS_kinds
# |  1 | 1 |                             # Select atom kind   range
# | %
# | % PDOS_l
# | 0 |0 |                               # Select angular momentum l range
# | %

Re: Number of states in QE vs GW for PDOS calculation

Posted: Mon Jan 12, 2026 9:17 am
by Daniele Varsano
Dear Dhanjit,

unfortunately, at moment I do not have access to my workstation and I cannot do a check on your inputs and I do not recall by heart the details of the DOS/PDOS calculation.

Anyway, it is not clear from your post what it is exactly plotted in the graphs, what input has been used for the two and which code. I suggest you to check if there is a normalization issue comparing QE and Yambo at GGA level (DOS and PDOS).

Best,

Daniele