Re: [QE-users] NEB images with different cell parameters

Hi Apoorv, You can use external code for vcneb like Uspex-9.4.4 in interface with QE. (In my opinion it does not make lots of sense to study photoisomerization with ground state dft, i would rather think of post hf methods to explore pes of the first excited state(s)) Anyhow could you please

Re: [QE-users] NEB images with different cell parameters possible?

Sorry for not being clearer, the 'excited' state in our case is actually a metastable state, i.e. a stable state on its own that has undergone photoisomerisation via broadband light. I figured there wasn't variable cell neb. I thought to simply average the cell parameters and use them to

[QE-users] Virtual atom projected band structure problem

Hello, I have a situation related to the reading of virtual atoms pseudopotentials in QE. I am trying to obtain the orbital projected band structure of InGaAs with a virtual and As atoms. I calculated the virtual atom with 3 set of pseudopotentials. 1. The pseudopotentials: Generated using

Re: [QE-users] NEB images with different cell parameters possible?

Some time ago there was a project of variable-cell NEB, but it went nowhere. Anyway: here the problem is not variable-cell NEB but a NEB between an initial configuration in the ground state and a final one in the excited state, if I understand correctly. This is just plain impossible. Paolo On

Re: [QE-users] igk_k , davcio, evc and wfck2r.f90

On Fri, Sep 6, 2019 at 3:45 PM Aritz Leonardo wrote: > *they do not match and I don't understand it. Is this related to the fact > that evc.dat was saved to disk in a |k+G| moduli order? * > it was saved to disk in a way that does not depend upon the parallelization geometry, not in straight

Re: [QE-users] Runtime GPU memory issue of q-e-gpu- -6.4a1

Dear Xiaoqin Huang, this is likely a "out of memory" issue. It would be of help if you could: 1) run your input with verbosity = 'high', or better 2) try a (much!) smaller input such that the memory footprint reduces from the one reported in your output (28.50 GB) to less than 10 GB. Or even

Re: [QE-users] igk_k , davcio, evc and wfck2r.f90

Dear Paolo Thank you very much for the reply and it is obvious that I was confused. On the other hand, I have just realized that in q-e6.4, the structure of the "outdir" folder has changed. The list of folders with each K-point that used to contain evc.dat, gkvectors.dat and eigenvalues files

Re: [QE-users] limit to take U value for DFT+U calculation

Dear Matteo,Can I use LDA+U method for a study of CeO2 clusters?Or, I need compute the value of U for every size of CeO2 cluster?  06.09.2019, 17:58, "Matteo Cococcioni" :Dear Rohit, you could compute the value of U, for example using the linear-response, DFPT-based hp.x code now included in QE.

Re: [QE-users] NEB images with different cell parameters possible?

Dear Apoorv Jain AFAIK, variable cell NEB is not implemented in QE (yet). HTH Giuseppe "A. Jain" ha scritto: Hi All, I would like to conduct an neb study on my system (ruthenium based crystals) where photoisomerisation takes place. We would like to investigate the mechanism of

[QE-users] optimization lattice constants of charged single layer 2d metallic materials

Dear all, I would like to calculate the effect of charge doping on the lattice constant of a 2d metallic material. According to the manual, we can do this by changing the keywords "tot_charge" to add or remove charges from the neutral system. The excess charge in perfect single layer 2d

Re: [QE-users] limit to take U value for DFT+U calculation

Dear Rohit, you could compute the value of U, for example using the linear-response, DFPT-based hp.x code now included in QE. In the directory HP you will find several examples on how to do that. Details about the implementation can be found in PRB 98, 085127 (2018), arXiv:1805.01805v1 and