<html><body>
<p>Dear PWSCF users and developers.<br>
<br>
I'm now try to use HSE functional for improvement the band gap calculation<br>
<br>
of some materials. (perovskite structure with Nb, Ba contains).<br>
<br>
<br>
But, the calculation has stopped with the error message,<br>
<br>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%<br>
from cdiaghg : error # 108<br>
problems computing cholesky <br>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%<br>
<br>
<br>
It seems come from the diagonalization part of the calculation, <br>
<br>
In the output file, the iteration part converged, but once converged <br>
<br>
with the message, <br>
<br>
" NOW GO BACK TO REFINE HYBRID CALCULATION" <br>
<br>
then it restarts iteration from iteration 1. After some "NOW GO BACK...." message, <br>
<br>
the diagonalization failed.<br>
<br>
What this loop of " NOW GO BACK TO REFINE HYBRID CALCULATION" do in the calculation <br>
<br>
of HSE functional ? And how should we treat the error message ? <br>
<br>
If I can get reply, I'm very happy.<br>
<br>
My Input file is as follow, <br>
<br>
&control<br>
calculation = 'scf'<br>
restart_mode = 'from_scratch'<br>
pseudo_dir = '/home/okuno/PWSCF/espresso-5.0/pseudo/'<br>
outdir = './'<br>
prefix = 'material'<br>
tprnfor=.TRUE.<br>
/<br>
&system<br>
ibrav=6<br>
celldm(1)=7.80078888085<br>
celldm(3)=1.0<br>
nat=5<br>
ntyp=4<br>
nbnd=28<br>
ecutwfc=50.0<br>
ecutrho=200.0<br>
degauss=0.00<br>
input_dft='hse'<br>
screening_parameter=0.106<br>
x_gamma_extrapolation=.TRUE.<br>
/<br>
&electrons<br>
conv_thr = 1e-12,<br>
mixing_beta=0.3,<br>
/<br>
ATOMIC_SPECIES<br>
.......<br>
ATOMIC_POSITIONS {crystal}<br>
........<br>
K_POINTS {automatic}<br>
4 4 4 0 0 0<br>
<br>
<br>
Sincerely, <br>
<br>
Yukihiro Okuno.<br>
<br>
<br>
</body></html>