[QE-users] actual ecutwfc for Uranium PAW pseudo is way greater than recommended

[Sudip Kumar Mondal]

Dear Malte,
                    My test system is a U bearing compound. Since I
intended to carry on a phonon calculation I used a very small force
convergence threshold (forc_conv_thr) of 1.0D-7 and an even smaller
convergence for total energy (etot_conv_thr) of 1.0D-9 for ionic
minimization. Consequently I had to choose a lower convergence threshold
for self-consistency (conv_thr in &ELECTRONS) of 1.0D-12 . The k-point grid
is 8*8*9 (optimized).
                    The cut-off energy and  k-pt grid are correlated. But
does my choice of thresholds did push the cut-offs higher ?
                    By the way the system I am dealing with is a less
explored one. no electronic properties so no band gap etc are reported. I
chose to go with 1st order methfessel-paxton smearing for not knowing
whether it is an insulator or semiconductor .
Thank you in advance.


-- 
Sudip Kumar Mondal
DST-INSPIRE Junior Research Fellow
High Pressure Temperature Laboratory
Dept. Of Physics/Geological Sciences
Jadavpur University
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20180823/716489e5/attachment.html>