On Mon, Aug 6, 2012 at 4:33 PM, Yue-Wen Fang <yuewen.fang at gmail.com> wrote: > > Dear Dr. Axel, > > You said ld1.x cannot generate van-style ultrasoft pseudopotential, but > can ld1.x generate van-style norm-conserving pseudopotential? I think ld1.x > can't make it.
i don't understand what you are asking here. there is a difference between multi-projector norm-conserving pseudopotentials and vanderbilt style ultra-soft potentials. that being said, there is nothing inherently bad about rrkj ultrasofts. axel. > > Regards! > Yuewen > > > 2012/8/6 Yue-Wen Fang <yuewen.fang at gmail.com> >> >> >>> three comments: >>> 1: it makes *very* little sense to create a virtual Ti-O pseudopotential >> >> Dr. Axel, I want to create a virtual F-O pp in fact, Ti-O was proposed as >> an example. >>> >>> 2: last time i checked, ld1.x could not create vanderbilt style ultasoft >>> pseudopotentials. both potentials have to be of the same type *and* >>> then use the same valence configurations and pseudization >> >> In my latest work, I need the F-O virtual vanderbilt style pp, but >> present PPs of F and O downloaded from the quantum espresso website cannot >> be used by virtual.x to create a new virtual pp because of different nql and >> nqlc. So I have to use ld1.x to create new pp of F and O. >> I checked the Input file description of ld1.x in quantum espresso, it >> tells us some details about "tm" index that >> >> * .true. for Troullier-Martins pseudization [PRB 43, 1993 (1991)] >> >> * .false. for Rabe-Rappe-Kaxiras-Joannopoulos pseudization >> [PRB 41, 1227 (1990), erratum PRB 44, 13175 (1991)] >> >> Is it means that we only can get tm style and rrkj style pp using ld1.x? >> If it is, is any other way to get vanderbilt pp of F and O? >>> >>> 3: pseudopotential generation is a bit of a black art. it is highly >>> recommended that you spend more time learning and understanding >>> the process before messing with this. there are many ways to >>> mess up a pseudopotential, even if you follow the general >>> instructions. >> >> Yeah, the pp generation is actually a challenge thing for me and need >> much experience. The process is involved too many parameters. >> Thank you very much >> Yuewen >> >>> >>> axel. >>> >>> >>> how about just reading through the sources? >>> a quick survey with grep reveals that nqlc is >>> 2*lmax+ with lmax being the maximum l of >>> the beta functions in the pseudopotential. >> >> Sorry, Dr Axel, I know it is far more than enough by just reading the >> sources because I am a newcomer. I haven't been skilled in using grep index, >> thank you for your instruction and i will take more efforts to make it. >> >> >>> >>> axel. >>> >>> >>> >>> >>> > _______________________________________________ >>> > Pw_forum mailing list >>> > Pw_forum at pwscf.org >>> > http://www.democritos.it/mailman/listinfo/pw_forum >>> > >>> >>> >>> >>> -- >>> Dr. Axel Kohlmeyer akohlmey at gmail.com http://goo.gl/1wk0 >>> International Centre for Theoretical Physics, Trieste. Italy. >>> _______________________________________________ >>> Pw_forum mailing list >>> Pw_forum at pwscf.org >>> http://www.democritos.it/mailman/listinfo/pw_forum >> >> > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Dr. Axel Kohlmeyer akohlmey at gmail.com http://goo.gl/1wk0 International Centre for Theoretical Physics, Trieste. Italy.
