The ABINIT Users Mailing List ( CLOSED )

[Kim Duckyoung <Duck.Young.Kim@fysik.uu.se>]

Hi,

I am using abinit 5.4.4 to calculating electron-phonon coupling.
Tentatively, I am testing a FCC potasium to compare with the one 
published in PRB 2006.

I am a novice on this issue. I tried to buy all data of the tutorial, 
telphon* because it has same structure Al example luckily.
Also, I decide to increase qpoint mesh to 4x4x4 than 2x2x2.

In short, but, the result was terrible. the Tc was estimated to be ~ 
10e15 Kelvin. If I got a reasonable number, then I would have tried to 
test convergence problems or some physical reasons. So, I decide to ask 
it to ABINIT- FORUM.

My question is following,

1. To generate higher mesh qpoint, suppose simple fcc case, how many 
datasets I should try with ? anaddb asked me to generate 64 qpoints in 
general.
But thanks to symmetry, we can recude it ( am I right?). I was not so 
sure in this problem. I tried the smallest set of qpoints.

Grid q points  :        8
 1)   0.00000000E+00  0.00000000E+00  0.00000000E+00
 2)   5.00000000E-01  0.00000000E+00  0.00000000E+00
 3)   0.00000000E+00  5.00000000E-01  0.00000000E+00
 4)   5.00000000E-01  5.00000000E-01  0.00000000E+00
 5)   0.00000000E+00  0.00000000E+00  5.00000000E-01
 6)   5.00000000E-01  0.00000000E+00  5.00000000E-01
 7)   0.00000000E+00  5.00000000E-01  5.00000000E-01
 8)   5.00000000E-01  5.00000000E-01  5.00000000E-01

Anyhow, anaddb accepted it. 8 qpoints was reasonable number 8(3+1) + 1 = 
33 dataset was okay. But was it enough ?
Especially when generating qpoints, could you recommend a good 
literature to follow ? not Monkhost and Pack paper.

2. What could be a critical parameter to check the convergence of the 
calculations ? As far as I know, in addition to some basic convergence 
tests such as ecut , # of kpoints, tolerances, we can check spectral 
functions' behaviors. something more ?


Sincerely,

Ps. 1. Please find my input file attached.
Ps. 2. I used the 19k.9.hgh potential.


ndtset 33

#
# DATASET 1 : make ground state wavefunctions and density
# 
tolwfr1 1.0d-20
prtden1 1
nline1 8    ! This is to expedite the convergence of higher-lying bands
## set default 
rfphon1 0 
tolvrs1 0 
getwfk1 0 


#
#  qpoint Gamma
#
#  DATASET 2: construct the wavefunctions at k+q (same as at k,
#     but kept for similarity to the other qpoints).
#
iscf2    -3  getwfk2 0 rfphon2 0 tolvrs2 0 tolwfr2  1.0d-14 getden2  1  nqpt2 
 1 qpt2   0.0 0.0 0.0
#
#  DATASET 3,4,5: perturbation at q=Gamma along reduced direction 1,2,3
#
rfatpol3  1 1 rfdir3  1 0 0  getwfq3   2  nqpt3  1 qpt3   0.0 0.0 0.0
rfatpol4  1 1 rfdir4  0 1 0  getwfq4   2  nqpt4  1 qpt4   0.0 0.0 0.0
rfatpol5  1 1 rfdir5  0 0 1  getwfq5   2  nqpt5  1 qpt5   0.0 0.0 0.0

#
#  qpoint 1/2 0 0
#
#  DATASET 6: construct the wavefunctions at k+q
#
iscf6    -3  getwfk6 0 rfphon6 0 tolvrs6 0  tolwfr6  1.0d-14 getden6  1  
nqpt6  1 qpt6   0.5 0.0 0.0
#
#  DATASET 7,8,9: perturbation along reduced direction 1,2,3
#
rfatpol7  1 1 rfdir7  1 0 0  getwfq7   6  nqpt7  1 qpt7   0.5 0.0 0.0
rfatpol8  1 1 rfdir8  0 1 0  getwfq8   6  nqpt8  1 qpt8   0.5 0.0 0.0
rfatpol9  1 1 rfdir9  0 0 1  getwfq9   6  nqpt9  1 qpt9   0.5 0.0 0.0

#
#  qpoint X= 1/2 1/2 0
#
#  DATASET 10: construct the wavefunctions at k+q
#
iscf10   -3  getwfk10 0 rfphon10 0 tolvrs10 0 tolwfr10 1.0d-14 getden10 1  
nqpt10 1 qpt10  0.5 0.5 0.0
#
#  DATASET 11,12,13: perturbation along reduced direction 1,2,3
#
rfatpol11 1 1 rfdir11 1 0 0  getwfq11 10  nqpt11 1 qpt11  0.5 0.5 0.0
rfatpol12 1 1 rfdir12 0 1 0  getwfq12 10  nqpt12 1 qpt12  0.5 0.5 0.0
rfatpol13 1 1 rfdir13 0 0 1  getwfq13 10  nqpt13 1 qpt13  0.5 0.5 0.0

#### Additional 
#q = 1/4 0 0 
iscf14   -3  getwfk14 0 rfphon14 0 tolvrs14 0 tolwfr14 1.0d-14 getden14 1  
nqpt14 1 qpt14  0.25 0.0 0.0
rfatpol15 1 1 rfdir15 1 0 0  getwfq15 14  nqpt15 1 qpt15  0.25 0.0 0.0
rfatpol16 1 1 rfdir16 0 1 0  getwfq16 14  nqpt16 1 qpt16  0.25 0.0 0.0
rfatpol17 1 1 rfdir17 0 0 1  getwfq17 14  nqpt17 1 qpt17  0.25 0.0 0.0

#q = 1/4 1/4 0 
iscf18   -3  getwfk18 0 rfphon18 0 tolvrs18 0 tolwfr18 1.0d-14 getden18 1  
nqpt18 1 qpt18  0.25 0.25 0.0
rfatpol19 1 1 rfdir19 1 0 0  getwfq19 18  nqpt19 1 qpt19  0.25 0.25 0.0
rfatpol20 1 1 rfdir20 0 1 0  getwfq20 18  nqpt20 1 qpt20  0.25 0.25 0.0
rfatpol21 1 1 rfdir21 0 0 1  getwfq21 18  nqpt21 1 qpt21  0.25 0.25 0.0

#q = 1/2 1/4 0 
iscf22   -3  getwfk22 0 rfphon22 0 tolvrs22 0 tolwfr22 1.0d-14 getden22 1  
nqpt22 1 qpt22  0.5 0.25 0.0
rfatpol23 1 1 rfdir23 1 0 0  getwfq23 22  nqpt23 1 qpt23  0.5 0.25 0.0
rfatpol24 1 1 rfdir24 0 1 0  getwfq24 22  nqpt24 1 qpt24  0.5 0.25 0.0
rfatpol25 1 1 rfdir25 0 0 1  getwfq25 22  nqpt25 1 qpt25  0.5 0.25 0.0

#q = -1/2 1/2 0 
iscf26   -3  getwfk26 0 rfphon26 0 tolvrs26 0 tolwfr26 1.0d-14 getden26 1  
nqpt26 1 qpt26  -.25 0.25 0.0
rfatpol27 1 1 rfdir27 1 0 0  getwfq27 26  nqpt27 1 qpt27  -.25 0.25 0.0
rfatpol28 1 1 rfdir28 0 1 0  getwfq28 26  nqpt28 1 qpt28  -.25 0.25 0.0
rfatpol29 1 1 rfdir29 0 0 1  getwfq29 26  nqpt29 1 qpt29  -.25 0.25 0.0

#q = -1/4 1/2 1/4 
iscf30   -3  getwfk30 0 rfphon30 0 tolvrs30 0 tolwfr30 1.0d-14 getden30 1  
nqpt30 1 qpt30  -.25 0.5 0.25
rfatpol31 1 1 rfdir31 1 0 0  getwfq31 30  nqpt31 1 qpt31  -.25 0.5 0.25
rfatpol32 1 1 rfdir32 0 1 0  getwfq32 30  nqpt32 1 qpt32  -.25 0.5 0.25
rfatpol33 1 1 rfdir33 0 0 1  getwfq33 30  nqpt33 1 qpt33  -.25 0.5 0.25


#
#  Common data
#
iscf 3 
rfphon 1 
tolvrs 1.0e-8 
getwfk 1
kptopt 3 
#
#  the kpoint grid is minimalistic to keep the calculation
#   manageable.
#
ngkpt 8 8 8

#
#  as is the kinetic energy cutoff
#
ecut 30.0

#
#  use a centered grid for the kpoints
#
nshiftk 1
shiftk 0.0 0.0 0.0
acell 3*5.43 Angstrom 
rprim 
 0.0 0.5 0.5
 0.5 0.0 0.5
 0.5 0.5 0.0
nband 13

#
#  include metallic occupation function with a small smearing
#
occopt 7
tsmear 0.001
natom 1
typat 1
xred 0.00 0.00 0.00
nstep 800
ntypat 1 
znucl 19