[QE-users] HSE Hybrid bands no calculated in an specific k-points path

[ROBERT MIKHAIL GUZMAN ARELLANO]

Dear all.
I am doing a hybrid calculation of bands in the description HSE. My system
is a layer of carbon of 12 atoms at the unit cell. Part of the input is:

!***********************************************
&system
    ibrav           = 8
    celldm(1)       = 9.354145
    celldm(2)       = 1.412121
    celldm(3)       = 2.020202
    nat = 12
    ntyp= 1
    nbnd            = 28
    occupations     = 'smearing'
    smearing        = 'gauss'
    degauss         = 0.01d0
    ecutwfc = 80.d0,
    ecutrho = 320.d0,
    input_dft='HSE'
    nqx1=2, nqx2=2, nqx3=1
 /
 &electrons
    mixing_mode = 'plain'
    mixing_beta = 0.5
    conv_thr =  1.0d-8
 /
ATOMIC_SPECIES
 C  12.0107     C_ONCV_PBE_sr.upf
ATOMIC_POSITIONS {angstrom}
 C   -2.4782021033    0.0532472376    0.0000000000
 C   -0.0032466468   -0.0074780911    0.0000000000
 C   -1.2248455699    0.7566921174    0.0000000000
 C    1.2184243008    0.7566745164    0.0000000000
 C   -0.7508000863    2.0875939495    0.0000000000
 C    0.7444458380    2.0876025878    0.0000000000
 C   -0.7508209917    3.5285256632    0.0000000000
 C    0.7444436433    3.5285398124    0.0000000000
 C   -1.2248187534    4.8594363132    0.0000000000
 C    1.2184363447    4.8594795502    0.0000000000
 C   -0.0032200034    5.6235974470    0.0000000000
 C   -2.4781959721    5.5628888962    0.0000000000
K_POINTS (automatic)
9 6 1 0 0 0
ADDITIONAL_K_POINTS tpiba_b
6
0.0 0.0 0.0 20
0.5 0.0 0.0 10
0.5 0.3540 0.0 20
0.0 0.0 0.0 10
0.0 0.3540 0.0 20
0.5 0.3540 0.0 0
!***********************************************

And when I tried to run, it did not finish the calculation. I was reading
about that in the QE-Forum and some advice suggested that my problem would
be about memory, So I divided my calculation. Now I am doing the same
calculation but I exchange the "ADDITIONAL_K_POINTS" section by:

ADDITIONAL_K_POINTS crystal_b
1
0.4400000000    0.4400000000    0.0000000000 0

So I calculated the bands point by point and then matched the results. This
worked for different kpoints, except for the k-points which are between the
corner of Brillouin Zone and Gamman (along the diagonal path, for example
0.4400000000    0.4400000000    0.0000000000). In these points I received
the following message:

    total energy              =    -136.45081412 Ry
     estimated scf accuracy    <          6.8E-09 Ry
     smearing contrib. (-TS)   =      -0.00080694 Ry
     internal energy E=F+TS    =    -136.45000718 Ry

     convergence has been achieved in  10 iterations

     Using ACE for calculation of exact exchange

     EXX grid:   225657 G-vectors     FFT dimensions: (  54,  75, 108)
     ACE projected onto    28 (nbndproj) and applied to    28 (nbnd) bands
 ZPOTRF exited with INFO=            1

 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     Error in routine ZPOTRF (1):
     Cholesky failed in invchol.
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

     stopping ...

I do not know what the problem is. Maybe some of us can give some advice.
The pseudo potential is PBE Norm Conserving.

Thank you for your attention.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20230703/dfc5f8c6/attachment.html>