The ABINIT Users Mailing List ( CLOSED )

[Timothy Mason <millionsofbillions@gmail.com>]

Sirs and Madams,

I am running anaddb on bulk Aluminum to find thermal properties.    I have a 6 6 6 nqkpt density.   everything seems to run fine except that I get a warning at the end of the log file.   I'm using a ng2qpt interpolated grid of 200 200 200.       

  thm9 : WARNING -
4032   No atomic factor tensor is printed out :
4033   the tolerance level that was asked
4034   has not been match with the grids specified.
4035   Action : in the input file, increase the resolution
4036   of grids ng2qpt, or decrease the accuracy requirement thmtol.

I see where thmtol gets checked and it looks like its no where close to converging.

  thm9 : check if the thermodynamic functions
3254     change within tolerance.
3255  thm9 : free energy relative changes     1.7626E+02
3256         are larger than thmtol     3.2000E-01
3257  thm9 : energy relative changes     1.7629E+02
3258         are larger than thmtol     3.2000E-01
3259  thm9 : iwchan=     4
3260  thm9 : compute g, f, e, s, c

a few questions:

1:  what does this mean?  should I be concerned about this warning?   the actual thermal output does not seem to change much from a 100 100 100 grid to a 200 200 200 grid

2.  why is it that sometimes an out_THERMO file is created and other times not?  

3.   I set my thmtol to 0.320.    how can i interpret that interms of accuracy of the calculation?

4.   when choosing a gpt grid,  is it necessary to pick high resolution qpt grids for metals just because a large kpt grid was required for the electronic structure?


thanks very much!  following is my anaddb input file.

!Input file for the ifc code. Analysis of bulk Al                       
  3
  4 !Flags
  5  ifcflag   1     ! Interatomic force constant flag
  6  thmflag   1     ! Thermodynamical properties flag
  7
  8 !Wavevector grid number 1 (coarse grid, from DDB)
  9   brav    1      ! Bravais Lattice : 1-S.C., 2-F.C., 3-B.C., 4-Hex.)
 10   ngqpt   6 6 6  ! Monkhorst-Pack indices
 11   nqshft  1         ! number of q-points in repeated basic q-cell
 12   q1shft  3*0.0
 13
 14 !Effective charges
 15      asr   1     ! Acoustic Sum Rule. 1 => imposed asymetrically
 16   chneut   1     ! Charge neutrality requirement for effective charges.
 17
 18 !Interatomic force constant info
 19   dipdip  1      ! Dipole-dipole interaction treatment
 20
 21 !Wavevector grid number 2 (series of fine grids, extrapolated from interat f    orces)
 22   ng2qpt   200 200 200  ! sample the BZ up to ngqpt2
 23   ngrids   5         ! number of grids of increasing size
 24   q2shft   3*0.0
 25
 26 !Thermal information
 27   nchan   2500   ! # of channels for the DOS with channel width 1 cm-1
 28   nwchan  5      ! # of different channel widths from this integer down to 1     cm-1
 29   thmtol  0.320  ! Tolerance on thermodynamical function fluctuations
 30   ntemper 60     ! Number of temperatures
 31   temperinc 20.  ! Increment of temperature in K for temperature dependency
 32   tempermin 20.  ! Minimal temperature in Kelvin
 33
 34 # This line added when defaults were changed (v5.3) to keep the previous, ol    d behaviour
 35 symdynmat 0