From degironc at sissa.it Thu Jun 1 00:37:36 2006 From: degironc at sissa.it (Stefano de Gironcoli) Date: Thu, 1 Jun 2006 00:37:36 +0200 (CEST) Subject: [Pw_forum] question on random_matrix In-Reply-To: <200605311104.37561.giannozz@nest.sns.it> References: <4471E3A5.2060500@polytechnique.edu> <200605311104.37561.giannozz@nest.sns.it> Message-ID: On Wed, 31 May 2006, Paolo Giannozzi wrote: > > ! Create a random hermitian matrix with non zero elements > ! similar to the dynamical matrix of the system > > but this is somewhat misleading. From what I remember the > random matrix is used to find the irreducible representations > of the small group of q (the phonon wavevector). If no symmetry > sends atom na into atom nb, displacements of atoms na and nb > will belong to different irreps. > > Disclaimer: I didn't write, and never really understood, how > symmetry works in the phonon code, so the above might be > all wrong! > I conmfirm what Paolo says: If two atoms are not connected by a symmetry operation then the corresponding "random matrix element" is actually set to zero so that they will eventually belong to different irreducible representations. Irreps are needed as a basis to expand the vibrational displacement patterns. The irreps chosen by the code are just one possible choice among infinite physically equivalent ones; technically it is convenient for efficiency in the calculation of the perturbation to have basis functions where as few as possible atoms are moved, hence the above prescription. stefano From virginie.quequet at polytechnique.edu Fri Jun 2 11:21:50 2006 From: virginie.quequet at polytechnique.edu (=?ISO-8859-1?Q?Virginie_Trinit=E9?=) Date: Fri, 02 Jun 2006 11:21:50 +0200 Subject: [Pw_forum] question on random_matrix In-Reply-To: References: <4471E3A5.2060500@polytechnique.edu> <200605311104.37561.giannozz@nest.sns.it> Message-ID: <4480032E.6080207@polytechnique.edu> Thanks for your explanations Virginie Trinite Stefano de Gironcoli wrote: > > On Wed, 31 May 2006, Paolo Giannozzi wrote: > >> >> ! Create a random hermitian matrix with non zero elements >> ! similar to the dynamical matrix of the system >> >> but this is somewhat misleading. From what I remember the >> random matrix is used to find the irreducible representations >> of the small group of q (the phonon wavevector). If no symmetry >> sends atom na into atom nb, displacements of atoms na and nb >> will belong to different irreps. >> >> Disclaimer: I didn't write, and never really understood, how >> symmetry works in the phonon code, so the above might be >> all wrong! >> > I conmfirm what Paolo says: If two atoms are not connected by a > symmetry operation then the corresponding "random matrix element" > is actually set to zero so that they will eventually belong to > different irreducible representations. > > Irreps are needed as a basis to expand the vibrational displacement > patterns. > The irreps chosen by the code are just one possible choice among > infinite physically equivalent ones; technically it is convenient for > efficiency in the calculation of the perturbation to have basis > functions where as few as possible atoms are moved, hence the above > prescription. > > stefano > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From umari at MIT.EDU Mon Jun 5 16:06:41 2006 From: umari at MIT.EDU (Paolo Umari) Date: Mon, 5 Jun 2006 10:06:41 -0400 (EDT) Subject: [Pw_forum] Re: watch this Micro-Cap Issue Trade In-Reply-To: <88033607.20060605140325@franz-iim.com> References: <88033607.20060605140325@franz-iim.com> Message-ID: From coldwind1978 at yahoo.com.cn Mon Jun 5 16:23:04 2006 From: coldwind1978 at yahoo.com.cn (=?gb2312?q?=D5=EB=D2=B6=20=D6=EC?=) Date: Mon, 5 Jun 2006 22:23:04 +0800 (CST) Subject: [Pw_forum] how to obtained the eigndisplacement by eignvector in PWSCF Message-ID: <20060605142304.96005.qmail@web15809.mail.cnb.yahoo.com> Dear all user: i have calculated the phonon of PbTiO3, and i have obtained eignvector of atom under each mode. for example: omega( 1) = -4.221301 [THz] = -140.808368 [cm-1] ( -0.092855 0.000000 -0.004178 0.000000 0.001020 0.000000 ) ( -0.185613 0.000000 -0.008352 0.000000 0.002040 0.000000 ) ( 0.598322 0.000000 0.026924 0.000000 -0.005371 0.000000 ) ( 0.598322 0.000000 0.021990 0.000000 -0.006576 0.000000 ) ( 0.488678 0.000000 0.026924 0.000000 -0.006576 0.000000 ) now i have kown the eigndisplacement |u> = M^(-1/2) | v>. but i am confused with mass M. for example, for Pb atom in PbTiO3. how much is mass of Pb in equation above? please grive me some construction! Regards zhuzhenye --------------------------------- ????????-3.5G???20M??? -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060605/86868184/attachment.htm From chincheng37 at hotmail.com Mon Jun 5 17:51:36 2006 From: chincheng37 at hotmail.com (Chin-Cheng Su) Date: Mon, 05 Jun 2006 11:51:36 -0400 Subject: [Pw_forum] New user dealing with compilation error Message-ID: Hi all, Lately I downloaded Quantum-Espresso v3.1 to my PC, which has the following specifications, PC Machine: Intel P-4 Operating System: Microsoft Windows XP Unixlike Environment in Windows: cygwin Compiler: GNU g95 & gcc , and after ./configure was successfully done the file make.sys was also generated with following content ------------------------------------------------------------------------------------------------- # make.sys. Generated from make.sys.in by configure. # compilation rules .SUFFIXES : .SUFFIXES : .o .c .f .f90 .f90.o: $(MPIF90) $(F90FLAGS) -c $< .f.o: $(F77) $(FFLAGS) -c $< .c.o: $(CC) $(CFLAGS) -c $< CC = gcc MPICC = gcc CFLAGS = -O3 -fomit-frame-pointer $(DFLAGS) $(IFLAGS) CPP = cpp CPPFLAGS = -P -traditional $(DFLAGS) $(IFLAGS) F90 = g95 MPIF90 = g95 F90FLAGS = $(FFLAGS) $(FDFLAGS) $(IFLAGS) $(MODFLAGS) F90FLAGS_NOOPT = $(FFLAGS_NOOPT) $(FDFLAGS) $(IFLAGS) $(MODFLAGS) F77 = g95 MPIF77 = g95 FFLAGS = -O3 -cpp FFLAGS_NOOPT = -O0 -cpp LD = g95 LDFLAGS = AR = ar ARFLAGS = ruv RANLIB = echo BLAS_LIBS = -lblas LAPACK_LIBS = -llapack FFT_LIBS = MPI_LIBS = MASS_LIBS = # ----------------------------- # application-specific settings # See include/defs.h.README for a list of precompilation options # (possible arguments to -D or -U) and their meaning DFLAGS = -D__LINUX -D__G95 -D__FFTW -D__USE_INTERNAL_FFTW FDFLAGS = $(DFLAGS) IFLAGS = -I../include MODFLAGS = -I. -I../Modules -I../PW -I../PH -I../iotk/src -I../CPV LIBOBJS = ../flib/ptools.a ../flib/flib.a ../clib/clib.a ../iotk/src/libiotk.a # LIBS must contain the location of all needed external libraries LIBS = $(LAPACK_LIBS) $(BLAS_LIBS) $(FFT_LIBS) $(MPI_LIBS) $(MASS_LIBS) # MYLIB can be one of the following (depending on LIBS): # blas : compile the local copy of blas routines # lapack : compile the local copy of lapack routines # blas_and_lapack : all of the above - use this for a quick test # or if you don't have an optimized blas/lapack library # lapack_ibm : compile only lapack routines not present in IBM ESSL # use this together with IBM ESSL # lapack_t3e : compile only lapack routines not present in T3E scilib # use this together with T3E scilib # lapack_mkl : compile only lapack routines not present in Intel MKL # use this together with Intel MKL MYLIB = ------------------------------------------------------------------------------------------------- Then I typed make all, lots of messages were presented and the last message block is as follows, ? ? a - iotk_xtox.o a - iotk_xtox_interf.o echo libiotk.a libiotk.a make[2]: Leaving directory `/cygdrive/c/DFTpackages/PWSCF/Test/espresso-3.1/iotk /src' make[1]: Leaving directory `/cygdrive/c/DFTpackages/PWSCF/Test/espresso-3.1/iotk ' ( cd Modules ; if test "make" = "" ; then make - --unix TLDEPS= all ; \ else make - --unix TLDEPS= all ; fi ) make[1]: Entering directory `/cygdrive/c/DFTpackages/PWSCF/Test/espresso-3.1/Mod ules' g95 -O3 -cpp -D__LINUX -D__G95 -D__FFTW -D__USE_INTERNAL_FFTW -I../include -I. -I../Modules -I../PW -I../PH -I../iotk/src -I../CPV -c atom.f90 In file atom.f90:14 USE kinds, ONLY : DP 1 Fatal Error: Can't open module file 'kinds.mod' at (1) for reading: No such file or directory make[1]: *** [atom.o] Error 1 make[1]: Leaving directory `/cygdrive/c/DFTpackages/PWSCF/Test/espresso-3.1/Modu les' make: *** [mods] Error 2 Does anyone know how to pass through it? Thanks to your kind help. Yours, Chin _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today - it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ From giannozz at nest.sns.it Mon Jun 5 18:32:22 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 5 Jun 2006 18:32:22 +0200 Subject: [Pw_forum] New user dealing with compilation error In-Reply-To: References: Message-ID: <200606051832.22088.giannozz@nest.sns.it> On Monday 05 June 2006 17:51, Chin-Cheng Su wrote: > Fatal Error: Can't open module file 'kinds.mod' at (1) for reading: > No such file or directory there should be a file "Modules/make.depend" that containes the dependencies of files upon other files. The first two lines should be atom.o : kind.o atom.o : parameters.o instructing make to compile kinds.f90 and parameters.f90 first. It is created by script "makedeps.sh" that is executed by "configure" (but you can run it manually). You should find out whether it is there or not there or there but incorrect, and why Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From giannozz at nest.sns.it Mon Jun 5 18:41:09 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 5 Jun 2006 18:41:09 +0200 Subject: [Pw_forum] how to obtained the eigndisplacement by eignvector in PWSCF In-Reply-To: <20060605142304.96005.qmail@web15809.mail.cnb.yahoo.com> References: <20060605142304.96005.qmail@web15809.mail.cnb.yahoo.com> Message-ID: <200606051841.09048.giannozz@nest.sns.it> On Monday 05 June 2006 16:23, ?? ? wrote: > i have calculated the phonon of PbTiO3, and i have obtained eignvector of > atom under each mode. for example: > omega( 1) = -4.221301 [THz] = -140.808368 [cm-1] > ( -0.092855 0.000000 -0.004178 0.000000 0.001020 0.000000 ) > ( -0.185613 0.000000 -0.008352 0.000000 0.002040 0.000000 ) > ( 0.598322 0.000000 0.026924 0.000000 -0.005371 0.000000 ) > ( 0.598322 0.000000 0.021990 0.000000 -0.006576 0.000000 ) > ( 0.488678 0.000000 0.026924 0.000000 -0.006576 0.000000 ) > now i have kown the eigndisplacement |u> = M^(-1/2) | v>. > but i am confused with mass M. for example, for Pb atom in PbTiO3. > how much is mass of Pb in equation above? I think that what is written in the output are the eigendisplacements, not the eigenvectors. The masses are whatever you specified in input (in amu) Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From chincheng37 at hotmail.com Tue Jun 6 03:53:22 2006 From: chincheng37 at hotmail.com (Chin-Cheng Su) Date: Mon, 05 Jun 2006 21:53:22 -0400 Subject: [Pw_forum] New user dealing with compilation error In-Reply-To: <200606051832.22088.giannozz@nest.sns.it> Message-ID: Hi Paolo, I checked the file, Modules/make.depend, which has the content as follows basic_algebra_routines.o : ../include/f_defs.h constraints_module.o : ../include/f_defs.h fft_base.o : ../include/f_defs.h fft_scalar.o : ../include/f_defs.h mp_wave.o : ../include/f_defs.h parser.o : ../include/f_defs.h path_base.o : ../include/f_defs.h path_io_routines.o : ../include/f_defs.h path_reparametrisation.o : ../include/f_defs.h ptoolkit.o : ../include/f_defs.h wannier.o : ../include/f_defs.h wave_base.o : ../include/f_defs.h So the first two lines you mentioned are not there. Then I run "makedeps.sh" manually, and get the same as above. The messages, when "makedeps.sh" was executed, are following, -dThe system cannot find the file specified. directory Modules : ok directory clib : ok -dThe system cannot find the file specified. directory PW : ok -dThe system cannot find the file specified. directory CPV : ok -dThe system cannot find the file specified. directory flib : ok -dThe system cannot find the file specified. directory pwtools : ok -dThe system cannot find the file specified. directory upftools : ok -dThe system cannot find the file specified. directory PP : ok -dThe system cannot find the file specified. directory PWCOND : ok -dThe system cannot find the file specified. directory Gamma : ok -dThe system cannot find the file specified. directory PH : ok -dThe system cannot find the file specified. directory D3 : ok -dThe system cannot find the file specified. directory atomic : ok directory Nmr : ok directory VIB : ok directory VdW : ok all dependencies updated successfully Is it normal or not??!! Chin >From: Paolo Giannozzi >Reply-To: pw_forum at pwscf.org >To: pw_forum at pwscf.org >Subject: Re: [Pw_forum] New user dealing with compilation error >Date: Mon, 5 Jun 2006 18:32:22 +0200 > >On Monday 05 June 2006 17:51, Chin-Cheng Su wrote: > > > Fatal Error: Can't open module file 'kinds.mod' at (1) for reading: > > No such file or directory > >there should be a file "Modules/make.depend" that containes the >dependencies of files upon other files. The first two lines should be > >atom.o : kind.o >atom.o : parameters.o > >instructing make to compile kinds.f90 and parameters.f90 first. >It is created by script "makedeps.sh" that is executed by "configure" >(but you can run it manually). You should find out whether it is >there or not there or there but incorrect, and why > >Paolo >-- >Paolo Giannozzi e-mail: giannozz at nest.sns.it >Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 >Piazza dei Cavalieri 7 I-56126 Pisa, Italy >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today - it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ From akohlmey at cmm.upenn.edu Tue Jun 6 04:06:42 2006 From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer) Date: Mon, 5 Jun 2006 22:06:42 -0400 Subject: [Pw_forum] New user dealing with compilation error In-Reply-To: References: <200606051832.22088.giannozz@nest.sns.it> Message-ID: <7b6913e90606051906s72b6ec9u7b4b9e31185df505@mail.gmail.com> hi, On 6/5/06, Chin-Cheng Su wrote: > Hi Paolo, [...] > directory D3 : ok > -dThe system cannot find the file specified. > directory atomic : ok > all dependencies updated successfully [...] > > Is it normal or not??!! this is not normal. looks like your 'sort' command cannot handle the -d flag. you might get lucky by removing it in the third but last line in moduledep.sh, but your best bet is probably to generate the dependency files on a proper unix/linux machine and copy them over (they should be pretty platform independent). regards, axel. > > Chin > > >From: Paolo Giannozzi > >Reply-To: pw_forum at pwscf.org > >To: pw_forum at pwscf.org > >Subject: Re: [Pw_forum] New user dealing with compilation error > >Date: Mon, 5 Jun 2006 18:32:22 +0200 > > > >On Monday 05 June 2006 17:51, Chin-Cheng Su wrote: > > > > > Fatal Error: Can't open module file 'kinds.mod' at (1) for reading: > > > No such file or directory > > > >there should be a file "Modules/make.depend" that containes the > >dependencies of files upon other files. The first two lines should be > > > >atom.o : kind.o > >atom.o : parameters.o > > > >instructing make to compile kinds.f90 and parameters.f90 first. > >It is created by script "makedeps.sh" that is executed by "configure" > >(but you can run it manually). You should find out whether it is > >there or not there or there but incorrect, and why > > > >Paolo > >-- > >Paolo Giannozzi e-mail: giannozz at nest.sns.it > >Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 > >Piazza dei Cavalieri 7 I-56126 Pisa, Italy > >_______________________________________________ > >Pw_forum mailing list > >Pw_forum at pwscf.org > >http://www.democritos.it/mailman/listinfo/pw_forum > > _________________________________________________________________ > Express yourself instantly with MSN Messenger! Download today - it's FREE! > http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From baroni at sissa.it Tue Jun 6 06:49:48 2006 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 6 Jun 2006 06:49:48 +0200 Subject: [Pw_forum] how to obtained the eigndisplacement by eignvector in PWSCF In-Reply-To: <20060605142304.96005.qmail@web15809.mail.cnb.yahoo.com> References: <20060605142304.96005.qmail@web15809.mail.cnb.yahoo.com> Message-ID: I do not quite get the question. It is the mass of Pb as found in any perdiodic table: 207.2 amu (see www.webelements.com) Atomic masses should also be specified in the "ATOMIC_SPECIES" card whose format, in the present case, should be something like: Pb 207.2 Pb_pseudopotential_file_name Are you confused instead by the negative value of the frequency? (this has nothing to do with masses) Hope this helps - Stefano On Jun 5, 2006, at 4:23 PM, ?? ? wrote: > Dear all user: > i have calculated the phonon of PbTiO3, and i have obtained > eignvector of atom under each mode. > for example: > omega( 1) = -4.221301 [THz] = -140.808368 [cm-1] > ( -0.092855 0.000000 -0.004178 0.000000 0.001020 0.000000 ) > ( -0.185613 0.000000 -0.008352 0.000000 0.002040 0.000000 ) > ( 0.598322 0.000000 0.026924 0.000000 -0.005371 0.000000 ) > ( 0.598322 0.000000 0.021990 0.000000 -0.006576 0.000000 ) > ( 0.488678 0.000000 0.026924 0.000000 -0.006576 0.000000 ) > now i have kown the eigndisplacement |u> = M^(-1/2) | v>. > but i am confused with mass M. for example, for Pb atom in PbTiO3. > how much is mass of Pb in equation above? > please grive me some construction! > > Regards > > zhuzhenye > > ????????-3.5G???20M??? --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060606/6fb8730b/attachment.htm From vsriniva at Princeton.EDU Tue Jun 6 07:04:20 2006 From: vsriniva at Princeton.EDU (Varadharajan Srinivasan) Date: Tue, 6 Jun 2006 01:04:20 -0400 (EDT) Subject: [Pw_forum] how to obtained the eigndisplacement by eignvector in PWSCF In-Reply-To: References: <20060605142304.96005.qmail@web15809.mail.cnb.yahoo.com> Message-ID: Hello, Please correct me if I am wrong but to obtain the eigenvectors of the dynamical matrix the values given by the code which are actually the normalized eigen displacements cannot be directly used. For this purpose I usually modify the code to print out the unnormalized eigen displacements and then multiply each component with the sqaure root of the corresponding mass. The mass is taken in Rydberg units. In these units the electron mass is 1/2. Best, Vardha. On Tue, 6 Jun 2006, Stefano Baroni wrote: > I do not quite get the question. > It is the mass of Pb as found in any perdiodic table: 207.2 amu (see > www.webelements.com) > > Atomic masses should also be specified in the "ATOMIC_SPECIES" card whose > format, in the present case, should be something like: > > Pb 207.2 Pb_pseudopotential_file_name > > Are you confused instead by the negative value of the frequency? (this has > nothing to do with masses) > > Hope this helps - Stefano > > On Jun 5, 2006, at 4:23 PM, ?????? ??? wrote: > >> Dear all user: >> i have calculated the phonon of PbTiO3, and i have obtained eignvector of >> atom under each mode. >> for example: >> omega( 1) = -4.221301 [THz] = -140.808368 [cm-1] >> ( -0.092855 0.000000 -0.004178 0.000000 0.001020 0.000000 ) >> ( -0.185613 0.000000 -0.008352 0.000000 0.002040 0.000000 ) >> ( 0.598322 0.000000 0.026924 0.000000 -0.005371 0.000000 ) >> ( 0.598322 0.000000 0.021990 0.000000 -0.006576 0.000000 ) >> ( 0.488678 0.000000 0.026924 0.000000 -0.006576 0.000000 ) >> now i have kown the eigndisplacement |u> = M^(-1/2) | v>. >> but i am confused with mass M. for example, for Pb atom in PbTiO3. how much >> is mass of Pb in equation above? >> please grive me some construction! >> >> Regards >> >> zhuzhenye >> >> ????????????????????????-3.5G?????????20M????????? > > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > From coldwind1978 at yahoo.com.cn Tue Jun 6 07:59:12 2006 From: coldwind1978 at yahoo.com.cn (=?gb2312?q?=D5=EB=D2=B6=20=D6=EC?=) Date: Tue, 6 Jun 2006 13:59:12 +0800 (CST) Subject: [Pw_forum] the problem about eigndisplacement of phonon? Message-ID: <20060606055912.48054.qmail@web15808.mail.cnb.yahoo.com> Dear Professor Stefano Baroni : Thank you give me answers. i will give you a clear question.i have calculate the phonon of PbTiO3 at q=0. there is the eignvector of atoms under every phonon. for example,omega( 1) = -4.221301 [THz] = -140.808368 [cm-1], eignvector of Pb atom is -0.092855 -0.004178 0.001020.now i want to calculte the eigndisplacement of Pb atom under this phonon. i have konw eigndisplacement |u> =M^(-1/2) | v>. |v> is eignvector. i donnot know how to select M mass of Pb atom. in your letter, M should be 207.2 amu.namely, for Pb atom, eigndisplacement |u> =207.2^(-1/2) | v>. is my opinion true? there is phonons of PbTiO3 as below: omega( 1) = -4.221301 [THz] = -140.808368 [cm-1] ( -0.092855 0.000000 -0.004178 0.000000 0.001020 0.000000 ) ( -0.185613 0.000000 -0.008352 0.000000 0.002040 0.000000 ) ( 0.598322 0.000000 0.026924 0.000000 -0.005371 0.000000 ) ( 0.598322 0.000000 0.021990 0.000000 -0.006576 0.000000 ) ( 0.488678 0.000000 0.026924 0.000000 -0.006576 0.000000 ) omega( 2) = -4.221301 [THz] = -140.808368 [cm-1] ( 0.000638 0.000000 0.008439 0.000000 0.092568 0.000000 ) ( 0.001275 0.000000 0.016869 0.000000 0.185040 0.000000 ) ( -0.004108 0.000000 -0.054377 0.000000 -0.487170 0.000000 ) ( -0.004108 0.000000 -0.044413 0.000000 -0.596476 0.000000 ) ( -0.003356 0.000000 -0.054377 0.000000 -0.596476 0.000000 ) omega( 3) = -4.221301 [THz] = -140.808368 [cm-1] ( -0.004254 0.000000 0.092476 0.000000 -0.008401 0.000000 ) ( -0.008503 0.000000 0.184856 0.000000 -0.016794 0.000000 ) ( 0.027409 0.000000 -0.595882 0.000000 0.044214 0.000000 ) ( 0.027409 0.000000 -0.486685 0.000000 0.054134 0.000000 ) ( 0.022386 0.000000 -0.595882 0.000000 0.054134 0.000000 ) omega( 4) = -0.669952 [THz] = -22.347350 [cm-1] ( -0.041557 0.000000 -0.005329 0.000000 0.392444 0.000000 ) ( -0.052468 0.000000 -0.006728 0.000000 0.495474 0.000000 ) ( -0.047053 0.000000 -0.006034 0.000000 0.440840 0.000000 ) ( -0.047053 0.000000 -0.005986 0.000000 0.444344 0.000000 ) ( -0.046682 0.000000 -0.006034 0.000000 0.444344 0.000000 ) omega( 5) = -0.669952 [THz] = -22.347350 [cm-1] ( -0.169027 0.000000 0.356408 0.000000 -0.013059 0.000000 ) ( -0.213403 0.000000 0.449978 0.000000 -0.016488 0.000000 ) ( -0.191381 0.000000 0.403543 0.000000 -0.014670 0.000000 ) ( -0.191381 0.000000 0.400360 0.000000 -0.014787 0.000000 ) ( -0.189872 0.000000 0.403543 0.000000 -0.014787 0.000000 ) omega( 6) = -0.669952 [THz] = -22.347350 [cm-1] ( 0.354218 0.000000 0.169447 0.000000 0.039810 0.000000 ) ( 0.447213 0.000000 0.213933 0.000000 0.050262 0.000000 ) ( 0.401063 0.000000 0.191857 0.000000 0.044720 0.000000 ) ( 0.401063 0.000000 0.190344 0.000000 0.045075 0.000000 ) ( 0.397900 0.000000 0.191857 0.000000 0.045075 0.000000 ) omega( 7) = 3.378427 [THz] = 112.692954 [cm-1] ( -0.224846 0.000000 -0.197313 0.000000 -0.001099 0.000000 ) ( 0.706263 0.000000 0.619778 0.000000 0.003451 0.000000 ) ( 0.075451 0.000000 0.066211 0.000000 0.000316 0.000000 ) ( 0.075451 0.000000 0.056787 0.000000 0.000369 0.000000 ) ( 0.064711 0.000000 0.066211 0.000000 0.000369 0.000000 ) omega( 8) = 3.378427 [THz] = 112.692954 [cm-1] ( -0.197127 0.000000 0.224701 0.000000 -0.011833 0.000000 ) ( 0.619195 0.000000 -0.705805 0.000000 0.037169 0.000000 ) ( 0.066149 0.000000 -0.075402 0.000000 0.003406 0.000000 ) ( 0.066149 0.000000 -0.064669 0.000000 0.003971 0.000000 ) ( 0.056734 0.000000 -0.075402 0.000000 0.003971 0.000000 ) omega( 9) = 3.378427 [THz] = 112.692954 [cm-1] ( -0.008630 0.000000 0.008170 0.000000 0.298912 0.000000 ) ( 0.027108 0.000000 -0.025663 0.000000 -0.938909 0.000000 ) ( 0.002896 0.000000 -0.002742 0.000000 -0.086028 0.000000 ) ( 0.002896 0.000000 -0.002351 0.000000 -0.100304 0.000000 ) ( 0.002484 0.000000 -0.002742 0.000000 -0.100304 0.000000 ) omega(10) = 6.809520 [THz] = 227.142638 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( -0.706639 0.000000 -0.025709 0.000000 0.000000 0.000000 ) ( 0.706639 0.000000 0.000000 0.000000 0.000444 0.000000 ) ( 0.000000 0.000000 0.025709 0.000000 -0.000444 0.000000 ) omega(11) = 6.809520 [THz] = 227.142638 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( -0.000153 0.000000 -0.008007 0.000000 0.000000 0.000000 ) ( 0.000153 0.000000 0.000000 0.000000 -0.707061 0.000000 ) ( 0.000000 0.000000 0.008007 0.000000 0.707061 0.000000 ) omega(12) = 6.809520 [THz] = 227.142638 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( -0.025712 0.000000 0.706594 0.000000 0.000000 0.000000 ) ( 0.025712 0.000000 0.000000 0.000000 -0.007996 0.000000 ) ( 0.000000 0.000000 -0.706594 0.000000 0.007996 0.000000 ) omega(13) = 15.128567 [THz] = 504.638023 [cm-1] ( 0.007689 0.000000 -0.000587 0.000000 0.000000 0.000000 ) ( 0.010929 0.000000 -0.000834 0.000000 0.000000 0.000000 ) ( 0.362132 0.000000 -0.027649 0.000000 0.000000 0.000000 ) ( 0.362132 0.000000 0.065311 0.000000 0.000000 0.000000 ) ( -0.855422 0.000000 -0.027649 0.000000 0.000000 0.000000 ) omega(14) = 15.128567 [THz] = 504.638023 [cm-1] ( 0.000587 0.000000 0.007689 0.000000 0.000000 0.000000 ) ( 0.000834 0.000000 0.010929 0.000000 0.000000 0.000000 ) ( 0.027649 0.000000 0.362132 0.000000 0.000000 0.000000 ) ( 0.027649 0.000000 -0.855422 0.000000 0.000000 0.000000 ) ( -0.065311 0.000000 0.362132 0.000000 0.000000 0.000000 ) omega(15) = 15.128567 [THz] = 504.638023 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.007711 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.010961 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 -0.857912 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.363186 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.363186 0.000000 ) Regards zhuzhenye --------------------------------- ????????-3.5G???20M??? -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060606/8670996a/attachment.htm From baroni at sissa.it Tue Jun 6 08:36:46 2006 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 6 Jun 2006 08:36:46 +0200 Subject: [Pw_forum] how to obtained the eigndisplacement by eignvector in PWSCF In-Reply-To: References: <20060605142304.96005.qmail@web15809.mail.cnb.yahoo.com> Message-ID: On Jun 6, 2006, at 7:04 AM, Varadharajan Srinivasan wrote: > Hello, > > Please correct me if I am wrong but to obtain the eigenvectors of > the dynamical matrix the values given by the code which are > actually the normalized eigen displacements cannot be directly > used. For this purpose I usually modify the code to print out the > unnormalized eigen displacements and then multiply each component > with the sqaure root of the corresponding mass. I do not remember whether the code outputs eigen-vectors of the dynamical matrix or eigen-displacements. In the latter case (which, I understand, is what you assume), you do not have to do anything in order to obtain displacement patterns. In the second (which, if my memory does not fail is more likely), you have to multiply eigen- vectors components by the square root of the atomic masses in order to obtain atomic displacement pattern. In any event, it seems to me that your distinction between "normalized" and "unnormalized" eigen-anything is somewhat misleading. Only the _direction_ of atomic displacement patterns actually matters, and the direction is, by its very definition, independent on normalization. > The mass is taken in Rydberg units. In these units the electron > mass is 1/2. Again, a change in the mass units would not change the direction of the eigen-displacements nor the relation between eigen-displacements and eigen-vectors of the dynamical matrix. It would affect the calculated frequencies, though. Stefano > > Best, > Vardha. > > > On Tue, 6 Jun 2006, Stefano Baroni wrote: > >> I do not quite get the question. >> It is the mass of Pb as found in any perdiodic table: 207.2 amu >> (see www.webelements.com) >> >> Atomic masses should also be specified in the "ATOMIC_SPECIES" >> card whose format, in the present case, should be something like: >> >> Pb 207.2 Pb_pseudopotential_file_name >> >> Are you confused instead by the negative value of the frequency? >> (this has nothing to do with masses) >> >> Hope this helps - Stefano >> >> On Jun 5, 2006, at 4:23 PM, ?? ? wrote: >> >>> Dear all user: >>> i have calculated the phonon of PbTiO3, and i have obtained >>> eignvector of atom under each mode. >>> for example: >>> omega( 1) = -4.221301 [THz] = -140.808368 [cm-1] >>> ( -0.092855 0.000000 -0.004178 0.000000 0.001020 0.000000 ) >>> ( -0.185613 0.000000 -0.008352 0.000000 0.002040 0.000000 ) >>> ( 0.598322 0.000000 0.026924 0.000000 -0.005371 0.000000 ) >>> ( 0.598322 0.000000 0.021990 0.000000 -0.006576 0.000000 ) >>> ( 0.488678 0.000000 0.026924 0.000000 -0.006576 0.000000 ) >>> now i have kown the eigndisplacement |u> = M^(-1/2) | v>. >>> but i am confused with mass M. for example, for Pb atom in >>> PbTiO3. how much is mass of Pb in equation above? >>> please grive me some construction! >>> Regards >>> zhuzhenye >>> ????????-3.5G???20M??? >> >> --- >> Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - >> Trieste >> [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) >> >> Please, if possible, don't send me MS Word or PowerPoint attachments >> Why? See: http://www.gnu.org/philosophy/no-word-attachments.html >> >> >> --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060606/880cde5f/attachment.htm From degironc at sissa.it Tue Jun 6 09:42:00 2006 From: degironc at sissa.it (degironc) Date: Tue, 06 Jun 2006 09:42:00 +0200 Subject: [Pw_forum] the problem about eigndisplacement of phonon? In-Reply-To: <20060606055912.48054.qmail@web15808.mail.cnb.yahoo.com> References: <20060606055912.48054.qmail@web15808.mail.cnb.yahoo.com> Message-ID: <448531C8.9030602@sissa.it> let's say that (1) \Phi U = M \omega^2 U and (2) M^(-1/2) \Phi M^(-1/2) V = \omega^2 V if V is a solution of (2) U = M^(1/2) V is a solution of (1). What is printed in matdyn is U (NORMALIZED so that = 1). if you want V (normalized such that = 1) you should take U divide by sqrt of M and renormalize. Units of mass are irrelevant (due to the renormalization) mass ratio betwen different species matters. stefano ?? ? wrote: > Dear Professor Stefano Baroni : > Thank you give me answers. i will give you a clear question.i have > calculate the phonon of PbTiO3 at q=0. there is the eignvector of > atoms under every phonon. > for example,omega( 1) = -4.221301 [THz] = -140.808368 [cm-1], > eignvector of Pb atom is -0.092855 -0.004178 0.001020.now i want to > calculte the eigndisplacement of Pb atom under this phonon. i have > konw eigndisplacement |u> =M^(-1/2) | v>. |v> is eignvector. i donnot > know how to select M mass of Pb atom. in your letter, M should be > 207.2 amu.namely, for Pb atom, eigndisplacement |u> =207.2^(-1/2) | > v>. is my opinion true? > there is phonons of PbTiO3 as below: > omega( 1) = -4.22301 [THz] = -140.808368 [cm-1] > ( -0.092855 0.000000 -0.004178 0.000000 0.001020 0.000000 ) > ( -0.185613 0.000000 -0.008352 0.000000 0.002040 0.000000 ) > ( 0.598322 0.000000 0.026924 0.000000 -0.005371 0.000000 ) > ( 0.598322 0.000000 0.021990 0.000000 -0.006576 0.000000 ) > ( 0.488678 0.000000 0.026924 0.000000 -0.006576 0.000000 ) > omega( 2) = -4.221301 [THz] = -140.808368 [cm-1] > ( 0.000638 0.000000 0.008439 0.000000 0.092568 0.000000 ) > ( 0.001275 0.000000 0.016869 0.000000 0.185040 0.000000 ) > ( -0.004108 0.000000 -0.054377 0.000000 -0.487170 0.000000 ) > ( -0.004108 0.000000 -0.044413 0.000000 -0.596476 0.000000 ) > ( -0.003356 0.000000 -0.054377 0.000000 -0.596476 0.000000 ) > omega( 3) = -4.221301 [THz] = -140.808368 [cm-1] > ( -0.004254 0.000000 0.092476 0.000000 -0.008401 0.000000 ) > ( -0.008503 0.000000 0.184856 0.000000 -0.016794 0.000000 ) > ( 0.027409 0.000000 -0.595882 0.000000 0.044214 0.000000 ) > ( 0.027409 0.000000 -0.486685 0.000000 0.054134 0.000000 ) > ( 0.022386 0.000000 -0.595882 0.000000 0.054134 0.000000 ) > omega( 4) = -0.669952 [THz] = -22.347350 [cm-1] > ( -0.041557 0.000000 -0.005329 0.000000 0.392444 0.000000 ) > ( -0.052468 0.000000 -0.006728 0.000000 0.495474 0.000000 ) > ( -0.047053 0.000000 -0.006034 0.000000 0.440840 0.000000 ) > ( -0.047053 0.000000 -0.005986 0.000000 0.444344 0.000000 ) > ( -0.046682 0.000000 -0.006034 0.000000 0.444344 0.000000 ) > omega( 5) = -0.669952 [THz] = -22.347350 [cm-1] > ( -0.169027 0.000000 0.356408 0.000000 -0.013059 0.000000 ) > ( -0.213403 0.000000 0.449978 0.000000 -0.016488 0.000000 ) > ( -0.191381 0.000000 0.403543 0.000000 -0.014670 0.000000 ) > ( -0.191381 0.000000 0.400360 0.000000 -0.014787 0.000000 ) > ( -0.189872 0.000000 0.403543 0.000000 -0.014787 0.000000 ) > omega( 6) = -0.669952 [THz] = -22.347350 [cm-1] > ( 0.354218 0.000000 0.169447 0.000000 0.039810 0.000000 ) > ( 0.447213 0.000000 0.213933 0.000000 0.050262 0.000000 ) > ( 0.401063 0.000000 0.191857 0.000000 0.044720 0.000000 ) > ( 0.401063 0.000000 0.190344 0.000000 0.045075 0.000000 ) > ( 0.397900 0.000000 0.191857 0.000000 0.045075 0.000000 ) > omega( 7) = 3.378427 [THz] = 112.692954 [cm-1] > ( -0.224846 0.000000 -0.197313 0.000000 -0.001099 0.000000 ) > ( 0.706263 0.000000 0.619778 0.000000 0.003451 0.000000 ) > ( 0.075451 0.000000 0.066211 0.000000 0.000316 0.000000 ) > ( 0.075451 0.000000 0.056787 0.000000 0.000369 0.000000 ) > ( 0.064711 0.000000 0.066211 0.000000 0.000369 0.000000 ) > omega( 8) = 3.378427 [THz] = 112.692954 [cm-1] > ( -0.197127 0.000000 0.224701 0.000000 -0.011833 0.000000 ) > ( 0.619195 0.000000 -0.705805 0.000000 0.037169 0.000000 ) > ( 0.066149 0.000000 -0.075402 0.000000 0.003406 0.000000 ) > ( 0.066149 0.000000 -0.064669 0.000000 0.003971 0.000000 ) > ( 0.056734 0.000000 -0.075402 0.000000 0.003971 0.000000 ) > omega( 9) = 3.378427 [THz] = 112.692954 [cm-1] > ( -0.008630 0.000000 0.008170 0.000000 0.298912 0.000000 ) > ( 0.027108 0.000000 -0.025663 0.000000 -0.938909 0.000000 ) > ( 0.002896 0.000000 -0.002742 0.000000 -0.086028 0.000000 ) > ( 0.002896 0.000000 -0.002351 0.000000 -0.100304 0.000000 ) > ( 0.002484 0.000000 -0.002742 0.000000 -0.100304 0.000000 ) > omega(10) = 6.809520 [THz] = 227.142638 [cm-1] > ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) > ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) > ( -0.706639 0.000000 -0.025709 0.000000 0.000000 0.000000 ) > ( 0.706639 0.000000 0.000000 0.000000 0.000444 0.000000 ) > ( 0.000000 0.000000 0.025709 0.000000 -0.000444 0.000000 ) > omega(11) = 6.809520 [THz] = 227.142638 [cm-1] > ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) > ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) > ( -0.000153 0.000000 -0.008007 0.000000 0.000000 0.000000 ) > ( 0.000153 0.000000 0.000000 0.000000 -0.707061 0.000000 ) > ( 0.000000 0.000000 0.008007 0.000000 0.707061 0.000000 ) > omega(12) = 6.809520 [THz] = 227.142638 [cm-1] > ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) > ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) > ( -0.025712 0.000000 0.706594 0.000000 0.000000 0.000000 ) > ( 0.025712 0.000000 0.000000 0.000000 -0.007996 0.000000 ) > ( 0.000000 0.000000 -0.706594 0.000000 0.007996 0.000000 ) > omega(13) = 15.128567 [THz] = 504.638023 [cm-1] > ( 0.007689 0.000000 -0.000587 0.000000 0.000000 0.000000 ) > ( 0.010929 0.000000 -0.000834 0.000000 0.000000 0.000000 ) > ( 0.362132 0.000000 -0.027649 0.000000 0.000000 0.000000 ) > ( 0.362132 0.000000 0.065311 0.000000 0.000000 0.000000 ) > ( -0.855422 0.000000 -0.027649 0.000000 0.000000 0.000000 ) > omega(14) = 15.128567 [THz] = 504.638023 [cm-1] > ( 0.000587 0.000000 0.007689 0.000000 0.000000 0.000000 ) > ( 0.000834 0.000000 0.010929 0.000000 0.000000 0.000000 ) > ( 0.027649 0.000000 0.362132 0.000000 0.000000 0.000000 ) > ( 0.027649 0.000000 -0.855422 0.000000 0.000000 0.000000 ) > ( -0.065311 0.000000 0.362132 0.000000 0.000000 0.000000 ) > omega(15) = 15.128567 [THz] = 504.638023 [cm-1] > ( 0.000000 0.000000 0.000000 0.000000 0.007711 0.000000 ) > ( 0.000000 0.000000 0.000000 0.000000 0.010961 0.000000 ) > ( 0.000000 0.000000 0.000000 0.000000 -0.857912 0.000000 ) > ( 0.000000 0.000000 0.000000 0.000000 0.363186 0.000000 ) > ( 0.000000 0.000000 0.000000 0.000000 0.363186 0.000000 ) > Regards > zhuzhenye > > ------------------------------------------------------------------------ > ????????-3.5G???20M??? From coldwind1978 at yahoo.com.cn Tue Jun 6 09:02:05 2006 From: coldwind1978 at yahoo.com.cn (=?gb2312?q?=D5=EB=D2=B6=20=D6=EC?=) Date: Tue, 6 Jun 2006 15:02:05 +0800 (CST) Subject: [Pw_forum] how to calculte the phonon in fixing atom? Message-ID: <20060606070205.2140.qmail@web15801.mail.cnb.yahoo.com> Dear users: i want to calculte the phonons of SrTiO3 which Sr atom is fixed. i do it by using "nat_todo". there is my inputfile of ph.x: &INPUTPH amass(1) = 87.62000, amass(2) = 47.88000, amass(3) = 15.99960, outdir = 'tmp/' , prefix = 'st' , tr2_ph = 1.0d-14 , fildyn = 'st.dyn5' , nat_todo= 4 epsil = .true., trans = .true., elph = .false., zue = .true., / 0 0 0 2 3 4 5 but in dyn file there is not phonons. by dynamat.x i deal with the dyn file.the results is as follows. omega( 1) = -2.746240 [THz] = -91.605330 [cm-1] ( -0.229443 0.000000 -0.099229 0.000000 0.000000 0.000000 ) ( -0.089673 0.000000 -0.038782 0.000000 0.000000 0.000000 ) ( 0.541628 0.000000 0.234241 0.000000 0.000000 0.000000 ) ( 0.541628 0.000000 0.190990 0.000000 0.000000 0.000000 ) ( 0.441620 0.000000 0.234241 0.000000 0.000000 0.000000 ) omega( 2) = -2.746240 [THz] = -91.605330 [cm-1] ( 0.099229 0.000000 -0.229443 0.000000 0.000000 0.000000 ) ( 0.038782 0.000000 -0.089673 0.000000 0.000000 0.000000 ) ( -0.234241 0.000000 0.541628 0.000000 0.000000 0.000000 ) ( -0.234241 0.000000 0.441620 0.000000 0.000000 0.000000 ) ( -0.190990 0.000000 0.541628 0.000000 0.000000 0.000000 ) omega( 3) = 0.000000 [THz] = -0.000009 [cm-1] ( -0.037890 0.000000 -0.025352 0.000000 0.444884 0.000000 ) ( -0.037890 0.000000 -0.025352 0.000000 0.444884 0.000000 ) ( -0.037890 0.000000 -0.025352 0.000000 0.444884 0.000000 ) ( -0.037890 0.000000 -0.025352 0.000000 0.444884 0.000000 ) ( -0.037890 0.000000 -0.025352 0.000000 0.444884 0.000000 ) omega( 4) = 0.000000 [THz] = 0.000005 [cm-1] ( -0.180783 0.000000 0.408968 0.000000 0.007909 0.000000 ) ( -0.180783 0.000000 0.408968 0.000000 0.007909 0.000000 ) ( -0.180783 0.000000 0.408968 0.000000 0.007909 0.000000 ) ( -0.180783 0.000000 0.408968 0.000000 0.007909 0.000000 ) ( -0.180783 0.000000 0.408968 0.000000 0.007909 0.000000 ) omega( 5) = 0.000000 [THz] = 0.000005 [cm-1] ( 0.407286 0.000000 0.179171 0.000000 0.044898 0.000000 ) ( 0.407286 0.000000 0.179171 0.000000 0.044898 0.000000 ) ( 0.407286 0.000000 0.179171 0.000000 0.044898 0.000000 ) ( 0.407286 0.000000 0.179171 0.000000 0.044898 0.000000 ) ( 0.407286 0.000000 0.179171 0.000000 0.044898 0.000000 ) omega( 6) = 3.509935 [THz] = 117.079598 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 -0.513490 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.574186 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.292212 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.400781 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.400781 0.000000 ) omega( 7) = 4.714264 [THz] = 157.251966 [cm-1] ( 0.384775 0.000000 0.055007 0.000000 0.000000 0.000000 ) ( -0.866552 0.000000 -0.123881 0.000000 0.000000 0.000000 ) ( 0.181436 0.000000 0.025938 0.000000 0.000000 0.000000 ) ( 0.181436 0.000000 0.017608 0.000000 0.000000 0.000000 ) ( 0.123170 0.000000 0.025938 0.000000 0.000000 0.000000 ) omega( 8) = 4.714264 [THz] = 157.251966 [cm-1] ( 0.055007 0.000000 -0.384775 0.000000 0.000000 0.000000 ) ( -0.123881 0.000000 0.866552 0.000000 0.000000 0.000000 ) ( 0.025938 0.000000 -0.181436 0.000000 0.000000 0.000000 ) ( 0.025938 0.000000 -0.123170 0.000000 0.000000 0.000000 ) ( 0.017608 0.000000 -0.181436 0.000000 0.000000 0.000000 ) omega( 9) = 6.891958 [THz] = 229.892506 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000905 0.000000 0.005360 0.000000 0.000000 0.000000 ) ( -0.000905 0.000000 0.000000 0.000000 0.707086 0.000000 ) ( 0.000000 0.000000 -0.005360 0.000000 -0.707086 0.000000 ) omega(10) = 6.891958 [THz] = 229.892506 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( -0.703970 0.000000 0.066527 0.000000 0.000000 0.000000 ) ( 0.703970 0.000000 0.000000 0.000000 0.000397 0.000000 ) ( 0.000000 0.000000 -0.066527 0.000000 -0.000397 0.000000 ) omega(11) = 6.891958 [THz] = 229.892506 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.066522 0.000000 0.703950 0.000000 0.000000 0.000000 ) ( -0.066522 0.000000 0.000000 0.000000 -0.005421 0.000000 ) ( 0.000000 0.000000 -0.703950 0.000000 0.005421 0.000000 ) omega(12) = 13.570374 [THz] = 452.661953 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 -0.020309 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 -0.269344 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 -0.350802 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.634026 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.634026 0.000000 ) omega(13) = 16.727499 [THz] = 557.973034 [cm-1] ( -0.011300 0.000000 0.000932 0.000000 0.000000 0.000000 ) ( -0.019840 0.000000 0.001636 0.000000 0.000000 0.000000 ) ( -0.365336 0.000000 0.030127 0.000000 0.000000 0.000000 ) ( -0.365336 0.000000 -0.070254 0.000000 0.000000 0.000000 ) ( 0.851929 0.000000 0.030127 0.000000 0.000000 0.000000 ) omega(14) = 16.727499 [THz] = 557.973034 [cm-1] ( -0.000932 0.000000 -0.011300 0.000000 0.000000 0.000000 ) ( -0.001636 0.000000 -0.019840 0.000000 0.000000 0.000000 ) ( -0.030127 0.000000 -0.365336 0.000000 0.000000 0.000000 ) ( -0.030127 0.000000 0.851929 0.000000 0.000000 0.000000 ) ( 0.070254 0.000000 -0.365336 0.000000 0.000000 0.000000 ) omega(15) = 23.660653 [THz] = 789.239679 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 -0.049737 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 -0.280773 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.951708 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.080452 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.080452 0.000000 ) in the results, i found eignvector of Sr atom is not zero.but i found the eignvector of Sr atom in SrTiO3 is zero when Sr atom is fixed.please give me some suggestion how to modify the inputfile of ph.x. Regards zhuzhenye --------------------------------- ??????-3.5G???20M?? -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060606/14118ad5/attachment.htm From lanhaiping at gmail.com Tue Jun 6 14:37:02 2006 From: lanhaiping at gmail.com (lan haiping) Date: Tue, 6 Jun 2006 20:37:02 +0800 Subject: [Pw_forum] Questions about ph.x output Message-ID: Dear developers: i try to understand the output of ph.x , but failed. Would you please help me out of these naive problems ? 1 , what are these numbers meaning? "phonons of Nanotube at Gamma 1 32 8 8.0390000 4.0000000 4.0000000 0.0000000 0.0000000 0.0000000 1 'C ' 10946.15867720511 " 2, why the dynamical matrix is 3 x 6 ? Dynamical Matrix in cartesian axes q = ( 0.000000000 0.000000000 0.000000000 ) 1 1 1.23287517 0.00000000 0.00000000 0.00000000 0.02101913 0.00000000 0.00000000 0.00000000 1.21209929 0.00000000 0.00000000 0.00000000 0.02101913 0.00000000 0.00000000 0.00000000 0.33335967 0.00000000 Thank you in advance~ Regards, Hai-Ping -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060606/bd58d92e/attachment.htm From giannozz at nest.sns.it Tue Jun 6 15:13:42 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 6 Jun 2006 15:13:42 +0200 Subject: [Pw_forum] Questions about ph.x output In-Reply-To: References: Message-ID: <200606061513.42409.giannozz@nest.sns.it> On Tuesday 06 June 2006 14:37, lan haiping wrote: > i try to understand the output of ph.x , but failed. > Would you please help me out of these naive problems ? > 1 , what are these numbers meaning? please explain first why you need to know the contents of an internal file. The dynamical matrix file is not meant to be read by humans, but by other programs, such as those in the tools/ directory. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From chincheng37 at hotmail.com Tue Jun 6 16:09:48 2006 From: chincheng37 at hotmail.com (Chin-Cheng Su) Date: Tue, 06 Jun 2006 10:09:48 -0400 Subject: [Pw_forum] New user dealing with compilation error In-Reply-To: <7b6913e90606051906s72b6ec9u7b4b9e31185df505@mail.gmail.com> Message-ID: Hi, By which, did you mean to remove the line "sort -d | uniq" only, right? Chin >From: "Axel Kohlmeyer" >Reply-To: pw_forum at pwscf.org >To: pw_forum at pwscf.org >Subject: Re: [Pw_forum] New user dealing with compilation error >Date: Mon, 5 Jun 2006 22:06:42 -0400 > >hi, > >On 6/5/06, Chin-Cheng Su wrote: >>Hi Paolo, > >[...] > >>directory D3 : ok >>-dThe system cannot find the file specified. >>directory atomic : ok >>all dependencies updated successfully > >[...] > >> >>Is it normal or not??!! > >this is not normal. looks like your 'sort' command >cannot handle the -d flag. you might get lucky >by removing it in the third but last line in moduledep.sh, >but your best bet is probably to generate the dependency >files on a proper unix/linux machine and copy them over >(they should be pretty platform independent). > >regards, > axel. > >> >>Chin >> >> >From: Paolo Giannozzi >> >Reply-To: pw_forum at pwscf.org >> >To: pw_forum at pwscf.org >> >Subject: Re: [Pw_forum] New user dealing with compilation error >> >Date: Mon, 5 Jun 2006 18:32:22 +0200 >> > >> >On Monday 05 June 2006 17:51, Chin-Cheng Su wrote: >> > >> > > Fatal Error: Can't open module file 'kinds.mod' at (1) for reading: >> > > No such file or directory >> > >> >there should be a file "Modules/make.depend" that containes the >> >dependencies of files upon other files. The first two lines should be >> > >> >atom.o : kind.o >> >atom.o : parameters.o >> > >> >instructing make to compile kinds.f90 and parameters.f90 first. >> >It is created by script "makedeps.sh" that is executed by "configure" >> >(but you can run it manually). You should find out whether it is >> >there or not there or there but incorrect, and why >> > >> >Paolo >> >-- >> >Paolo Giannozzi e-mail: giannozz at nest.sns.it >> >Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 >> >Piazza dei Cavalieri 7 I-56126 Pisa, Italy >> >_______________________________________________ >> >Pw_forum mailing list >> >Pw_forum at pwscf.org >> >http://www.democritos.it/mailman/listinfo/pw_forum >> >>_________________________________________________________________ >>Express yourself instantly with MSN Messenger! Download today - it's FREE! >>http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ >> >>_______________________________________________ >>Pw_forum mailing list >>Pw_forum at pwscf.org >>http://www.democritos.it/mailman/listinfo/pw_forum >> >> > > >-- >======================================================================= >Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu > Center for Molecular Modeling -- University of Pennsylvania >Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 >tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 >======================================================================= >If you make something idiot-proof, the universe creates a better idiot. >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum _________________________________________________________________ Don?t just search. Find. Check out the new MSN Search! http://search.msn.click-url.com/go/onm00200636ave/direct/01/ From akohlmey at cmm.upenn.edu Tue Jun 6 16:17:51 2006 From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer) Date: Tue, 6 Jun 2006 10:17:51 -0400 Subject: [Pw_forum] New user dealing with compilation error In-Reply-To: References: <7b6913e90606051906s72b6ec9u7b4b9e31185df505@mail.gmail.com> Message-ID: <7b6913e90606060717g67b77a2cq3514458b92bfcf08@mail.gmail.com> On 6/6/06, Chin-Cheng Su wrote: > Hi, > By which, did you mean to remove the line "sort -d | uniq" only, right? no. not the entire line but only the '-d' flag. this instructs the sort command to use a special ordering which might not be required here. a. > > Chin > > > >From: "Axel Kohlmeyer" > >Reply-To: pw_forum at pwscf.org > >To: pw_forum at pwscf.org > >Subject: Re: [Pw_forum] New user dealing with compilation error > >Date: Mon, 5 Jun 2006 22:06:42 -0400 > > > >hi, > > > >On 6/5/06, Chin-Cheng Su wrote: > >>Hi Paolo, > > > >[...] > > > >>directory D3 : ok > >>-dThe system cannot find the file specified. > >>directory atomic : ok > >>all dependencies updated successfully > > > >[...] > > > >> > >>Is it normal or not??!! > > > >this is not normal. looks like your 'sort' command > >cannot handle the -d flag. you might get lucky > >by removing it in the third but last line in moduledep.sh, > >but your best bet is probably to generate the dependency > >files on a proper unix/linux machine and copy them over > >(they should be pretty platform independent). > > > >regards, > > axel. > > > >> > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From coldwind1978 at yahoo.com.cn Tue Jun 6 16:32:45 2006 From: coldwind1978 at yahoo.com.cn (=?gb2312?q?=D5=EB=D2=B6=20=D6=EC?=) Date: Tue, 6 Jun 2006 22:32:45 +0800 (CST) Subject: [Pw_forum] the problem about eigndisplacement of phonon? Message-ID: <20060606143245.37877.qmail@web15808.mail.cnb.yahoo.com> Dear stefano: Thank you very much. i still am confused with the equations you give. i have three questions: (1) \Phi U = M \omega^2 U in this equation, what is the meaning of \Phi ? (2) in fact, what is printed in the matdyn is eigndisplacement. when i want eignvector, is M selected to be 207.2 amu. how to renormalize the eignvector? please give me suggestions! Regards zhuzhenye --------------------------------- Mp3???-??????? -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060606/cde6e4af/attachment.htm From chincheng37 at hotmail.com Tue Jun 6 16:46:07 2006 From: chincheng37 at hotmail.com (Chin-Cheng Su) Date: Tue, 06 Jun 2006 10:46:07 -0400 Subject: [Pw_forum] New user dealing with compilation error Message-ID: Great, after deleting "-d" in moduledep.sh, and executing makedeps.sh manually, the file Modules/make.depend are added with whole bunch of .o stuff and the first two lines are just as what Paolo had mentioned earlier. Thanks a lot. Chin >From: "Axel Kohlmeyer" >Reply-To: pw_forum at pwscf.org >To: pw_forum at pwscf.org >Subject: Re: [Pw_forum] New user dealing with compilation error >Date: Tue, 6 Jun 2006 10:17:51 -0400 > >On 6/6/06, Chin-Cheng Su wrote: >>Hi, >>By which, did you mean to remove the line "sort -d | uniq" only, right? > >no. not the entire line but only the '-d' flag. this instructs the sort >command to use a special ordering which might not be required >here. > >a. > >> >>Chin >> >> >> >From: "Axel Kohlmeyer" >> >Reply-To: pw_forum at pwscf.org >> >To: pw_forum at pwscf.org >> >Subject: Re: [Pw_forum] New user dealing with compilation error >> >Date: Mon, 5 Jun 2006 22:06:42 -0400 >> > >> >hi, >> > >> >On 6/5/06, Chin-Cheng Su wrote: >> >>Hi Paolo, >> > >> >[...] >> > >> >>directory D3 : ok >> >>-dThe system cannot find the file specified. >> >>directory atomic : ok >> >>all dependencies updated successfully >> > >> >[...] >> > >> >> >> >>Is it normal or not??!! >> > >> >this is not normal. looks like your 'sort' command >> >cannot handle the -d flag. you might get lucky >> >by removing it in the third but last line in moduledep.sh, >> >but your best bet is probably to generate the dependency >> >files on a proper unix/linux machine and copy them over >> >(they should be pretty platform independent). >> > >> >regards, >> > axel. >> > >> >> > >>Pw_forum mailing list >>Pw_forum at pwscf.org >>http://www.democritos.it/mailman/listinfo/pw_forum >> >> > > >-- >======================================================================= >Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu > Center for Molecular Modeling -- University of Pennsylvania >Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 >tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 >======================================================================= >If you make something idiot-proof, the universe creates a better idiot. >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum _________________________________________________________________ Is your PC infected? Get a FREE online computer virus scan from McAfee? Security. http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963 From burde1jb at cmich.edu Tue Jun 6 17:16:14 2006 From: burde1jb at cmich.edu (John B Burdette) Date: Tue, 6 Jun 2006 11:16:14 -0400 (EDT) Subject: [PW_forum] Calculating Raman Intensities Message-ID: <20060606111614.BKZ74769@student-b.csv.cmich.edu> Hello everyone, I am attempting to create the Raman spectrum for the CaF2 crystal using PWSCF. When I run the program ph.x with lraman=.true. I receive the following error: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from raman : error # 1 Vanderbilt not implemented %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Note also when I set lraman=.false. the program ran smoothly, although I could not make a Raman spectrum. When attemping to find where this error was coming from, I found it occurs when the variable 'okvan' is set equal to true which only happens when there is at least one Ultrasoft Pseudopotential being used in the setup.f90 file for the pw.x program. I am using two USPPs and both are in Vanderbilt format. Is there a reason why this error should be coming up even though I believe Vanderbilt is indeed being implemented? Also, is there anything I can do about this? Here is my input file for the ph.x program. CaF2 Phonons at Gamma Testing &inputph amass(1)=40.078, amass(2)=18.9984032, tr2_ph=1.0d-12, outdir='./', prefix='_CaF2K6', fildyn='CaF2.dyn', fildrho='CaF2.drho', epsil=.true., zue=.true., trans=.true., lraman=.true., ldisp=.true., nq1=6, nq2=6, nq3=6, / Thanks ahead of time for anyone willing to assist me. Best, John From giannozz at nest.sns.it Tue Jun 6 17:42:05 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 6 Jun 2006 17:42:05 +0200 Subject: [PW_forum] Calculating Raman Intensities In-Reply-To: <20060606111614.BKZ74769@student-b.csv.cmich.edu> References: <20060606111614.BKZ74769@student-b.csv.cmich.edu> Message-ID: <200606061742.05362.giannozz@nest.sns.it> On Tuesday 06 June 2006 17:16, John B Burdette wrote: > from raman : error # 1 > Vanderbilt not implemented "calculation of Raman coefficients with Ultrasoft Pseudopotentials is not implemented". Is it clear now? Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From burde1jb at cmich.edu Tue Jun 6 18:22:45 2006 From: burde1jb at cmich.edu (John B Burdette) Date: Tue, 6 Jun 2006 12:22:45 -0400 (EDT) Subject: [PW_forum] Calculating Raman Intensities Message-ID: <20060606122245.BKZ84516@student-b.csv.cmich.edu> "calculation of Raman coefficients with Ultrasoft Pseudopotentials is not implemented". Is it clear now? Does this imply that Raman coefficients cannot be calculated using Ultrasoft Pseudopotentials? Or does it mean I haven't turned on a switch somewhere? John From chincheng37 at hotmail.com Tue Jun 6 18:24:35 2006 From: chincheng37 at hotmail.com (Chin-Cheng Su) Date: Tue, 06 Jun 2006 12:24:35 -0400 Subject: [Pw_forum] New user dealing with compilation error In-Reply-To: Message-ID: Hi all, After a few adjustments, the compilation can pass through the error mentioned earlier. Now it hits another kind, which is as follows, ... ... g95 -O3 -cpp -c kvecs_FS.f g95 -o kvecs_FS.x kvecs_FS.o -llapack -lblas ( cd ../bin ; ln -fs ../pwtools/kvecs_FS.x . ) g95 -O3 -cpp -D__LINUX -D__G95 -D__FFTW -D__USE_INTERNAL_FFTW -I../include -I. -I../Modules -I../PW -I../PH -I../iotk/src -I../CPV -c metadyn_pp.f90 g95 -o metadyn_pp.x metadyn_pp.o ../flib/ptools.a ../flib/flib.a ../clib/clib.a ../iotk/src/libiotk.a -llapack -lblas metadyn_pp.o:metadyn_pp.f90:(.data+0x4): undefined reference to `pgopen_' make[1]: *** [metadyn_pp.x] Error 1 make[1]: Leaving directory `/cygdrive/c/My_XCRYSDEN/espresso-3.1/pwtools' make: *** [tools] Error 2 Does anyone has the idea to get pass this? Thanks greatly. Chin >From: "Chin-Cheng Su" >Reply-To: pw_forum at pwscf.org >To: pw_forum at pwscf.org >Subject: Re: [Pw_forum] New user dealing with compilation error >Date: Tue, 06 Jun 2006 10:46:07 -0400 > >Great, after deleting "-d" in moduledep.sh, and executing makedeps.sh >manually, the file Modules/make.depend are added with whole bunch of .o >stuff and the first two lines are just as what Paolo had mentioned earlier. >Thanks a lot. > >Chin > > >_________________________________________________________________ >Is your PC infected? Get a FREE online computer virus scan from McAfee? >Security. http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963 > >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum _________________________________________________________________ On the road to retirement? Check out MSN Life Events for advice on how to get there! http://lifeevents.msn.com/category.aspx?cid=Retirement From lanhaiping at gmail.com Tue Jun 6 18:35:14 2006 From: lanhaiping at gmail.com (lan haiping) Date: Wed, 7 Jun 2006 00:35:14 +0800 Subject: [PW_forum] Calculating Raman Intensities In-Reply-To: <20060606122245.BKZ84516@student-b.csv.cmich.edu> References: <20060606122245.BKZ84516@student-b.csv.cmich.edu> Message-ID: maybe you can perform your calculations with Norm-conserving PP. Regards H.P On 6/7/06, John B Burdette wrote: > > "calculation of Raman coefficients with Ultrasoft > Pseudopotentials is not implemented". Is it clear now? > > Does this imply that Raman coefficients cannot be calculated > using Ultrasoft Pseudopotentials? Or does it mean I haven't > turned on a switch somewhere? > > John > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060607/f3e7284a/attachment.htm From akohlmey at cmm.upenn.edu Tue Jun 6 18:38:07 2006 From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer) Date: Tue, 6 Jun 2006 12:38:07 -0400 Subject: [PW_forum] Calculating Raman Intensities In-Reply-To: <20060606122245.BKZ84516@student-b.csv.cmich.edu> References: <20060606122245.BKZ84516@student-b.csv.cmich.edu> Message-ID: <7b6913e90606060938m4829b3f0g8587230ec28f328f@mail.gmail.com> On 6/6/06, John B Burdette wrote: > "calculation of Raman coefficients with Ultrasoft > Pseudopotentials is not implemented". Is it clear now? > > Does this imply that Raman coefficients cannot be calculated > using Ultrasoft Pseudopotentials? Or does it mean I haven't > turned on a switch somewhere? john, not implemented means 'not programmed'. that can have various reasons: it may not be possible, too complicated, or nobody bothered to do it. it definitely means though cannot do it _right now_, regardless of any input file changes. this is a matter of changing the source code. regards, axel. > > John > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From giannozz at nest.sns.it Tue Jun 6 18:40:39 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 6 Jun 2006 18:40:39 +0200 Subject: [Pw_forum] New user dealing with compilation error In-Reply-To: References: Message-ID: <200606061840.39203.giannozz@nest.sns.it> On Tuesday 06 June 2006 18:24, Chin-Cheng Su wrote: > metadyn_pp.o:metadyn_pp.f90:(.data+0x4): undefined reference to `pgopen_' > [...] > Does anyone has the idea to get pass this? Thanks greatly. replace the line INTEGER, EXTERNAL :: PGOPEN with #if defined __PGPLOT INTEGER, EXTERNAL :: PGOPEN #endif in file pwtools/metadyn_pp.f90 . Some compilers don't like an "external" declaration to a nonexistent subroutine even it is not actually used. It is already fixed in the CVS version. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From akohlmey at cmm.upenn.edu Tue Jun 6 18:48:14 2006 From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer) Date: Tue, 6 Jun 2006 12:48:14 -0400 Subject: [Pw_forum] New user dealing with compilation error In-Reply-To: References: Message-ID: <7b6913e90606060948o5d8f1810g1a34fbec528eaaa9@mail.gmail.com> dear chin, On 6/6/06, Chin-Cheng Su wrote: > Hi all, > > g95 -O3 -cpp -D__LINUX -D__G95 -D__FFTW -D__USE_INTERNAL_FFTW -I../include > -I. -I../Modules -I../PW -I../PH -I../iotk/src -I../CPV -c metadyn_pp.f90 > g95 -o metadyn_pp.x metadyn_pp.o ../flib/ptools.a ../flib/flib.a > ../clib/clib.a ../iotk/src/libiotk.a -llapack -lblas > metadyn_pp.o:metadyn_pp.f90:(.data+0x4): undefined reference to `pgopen_' > make[1]: *** [metadyn_pp.x] Error 1 > make[1]: Leaving directory `/cygdrive/c/My_XCRYSDEN/espresso-3.1/pwtools' > make: *** [tools] Error 2 > > Does anyone has the idea to get pass this? Thanks greatly. no i have not seen this before (i usually compile espresso on linux and have little problems with that), but please have a look at the file pwtools/metadyn_pp.f90. the subroutine PGOPEN is only referenced when the define __PGPLOT is set. looks like your preprocessor did not work correctly, or some system tool is intrinsically defining __PGPLOT. please try to add as the first lines in pwtools/metadyn_pp.f90: #if defined(__PGPLOT) #undef __PGPLOT #endif however this kind of behaviour makes me doubt, whether the preprocessor of this g95 distribution works properly at all. axel. > > Chin > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From akohlmey at cmm.upenn.edu Tue Jun 6 18:52:23 2006 From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer) Date: Tue, 6 Jun 2006 12:52:23 -0400 Subject: [Pw_forum] New user dealing with compilation error In-Reply-To: <200606061840.39203.giannozz@nest.sns.it> References: <200606061840.39203.giannozz@nest.sns.it> Message-ID: <7b6913e90606060952t31d9ed55oca85e9766d1f8787@mail.gmail.com> On 6/6/06, Paolo Giannozzi wrote: > #if defined __PGPLOT > INTEGER, EXTERNAL :: PGOPEN > #endif > > in file pwtools/metadyn_pp.f90 . Some compilers don't like an "external" > declaration to a nonexistent subroutine even it is not actually used. > It is already fixed in the CVS version. ooops. i only looked at the cvs sources. please disregard my previous comment. a. > > Paolo > -- > Paolo Giannozzi e-mail: giannozz at nest.sns.it > Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 > Piazza dei Cavalieri 7 I-56126 Pisa, Italy > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From chincheng37 at hotmail.com Tue Jun 6 19:25:56 2006 From: chincheng37 at hotmail.com (Chin-Cheng Su) Date: Tue, 06 Jun 2006 13:25:56 -0400 Subject: [Pw_forum] New user dealing with compilation error In-Reply-To: <7b6913e90606060952t31d9ed55oca85e9766d1f8787@mail.gmail.com> Message-ID: Hi Paolo & Axel, Thanks greatly to you guys. Now "make all" does not present error message any more. However, it is a little bit concern to me - during the compilation, it occasionally shows the warning sign like follow In file geninv.f90:145 CALL dsvdc(a,ld,n,n,work,work(n+1),u,ld,v,ld,work(2*n+1),11,info) 1 In file geninv.f90:183 SUBROUTINE dsvdc(x,ldx,n,p,s,e,u,ldu,v,ldv,work,job,info) 2 Warning (155): Inconsistent types (COMPLEX(8)/REAL(8)) in actual argument lists at (1) and (2) Is it all right just leave them there and move forward? Chin >From: "Axel Kohlmeyer" >Reply-To: pw_forum at pwscf.org >To: pw_forum at pwscf.org >Subject: Re: [Pw_forum] New user dealing with compilation error >Date: Tue, 6 Jun 2006 12:52:23 -0400 > >On 6/6/06, Paolo Giannozzi wrote: > >>#if defined __PGPLOT >> INTEGER, EXTERNAL :: PGOPEN >>#endif >> >>in file pwtools/metadyn_pp.f90 . Some compilers don't like an "external" >>declaration to a nonexistent subroutine even it is not actually used. >>It is already fixed in the CVS version. > >ooops. i only looked at the cvs sources. >please disregard my previous comment. > >a. > > _________________________________________________________________ FREE pop-up blocking with the new MSN Toolbar ? get it now! http://toolbar.msn.click-url.com/go/onm00200415ave/direct/01/ From reboredofa at ornl.gov Tue Jun 6 20:51:22 2006 From: reboredofa at ornl.gov (Fernando A Reboredo) Date: Tue, 06 Jun 2006 14:51:22 -0400 Subject: [Pw_forum] espresso 3.1 compilation References: <20060606111614.BKZ74769@student-b.csv.cmich.edu> <200606061742.05362.giannozz@nest.sns.it> Message-ID: <004101c6899a$34918a00$250017ac@ornl.gov> I am trying to compile espresso on a 3.1 4.5 TF IBM pSeries System (cheetah NSSC ). I get the following error ./configure checking build system type... /usr/bin/oslevel[7]: /usr/bin/rm_mlcache_file: cannot execute config.sub: too many arguments Try `config.sub --help' for more information. configure: error: /bin/sh ./config.sub powerpc-ibm-aixoslevel: Error processing cache, return code 126. oslevel: Use the -f flag to rebuild cache. failed I suspect that it is because I do not have executable permisions for /usr/bin/rm_mlcache_file Am I wright? what is rm_mlcache_file for? Fernando From giannozz at nest.sns.it Tue Jun 6 21:12:57 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 6 Jun 2006 21:12:57 +0200 Subject: [Pw_forum] New user dealing with compilation error In-Reply-To: References: Message-ID: <200606062112.57646.giannozz@nest.sns.it> On Tuesday 06 June 2006 19:25, Chin-Cheng Su wrote: > during the compilation, it occasionally shows the warning sign like follow > > In file geninv.f90:145 > CALL dsvdc(a,ld,n,n,work,work(n+1),u,ld,v,ld,work(2*n+1),11,info) > 1 > In file geninv.f90:183 > SUBROUTINE dsvdc(x,ldx,n,p,s,e,u,ldu,v,ldv,work,job,info) > 2 > Warning (155): Inconsistent types (COMPLEX(8)/REAL(8)) in actual argument > lists at (1) and (2) > > Is it all right just leave them there and move forward? it is all right (in this case; in general, it isn't!). You can pass a complex array to a subroutine as a real array with double size, if you use "old-style" calling sequences (i.e. you pass a pointer to a subroutine). Of course you can't do this if you pass an array with the new f90 syntax. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From giannozz at nest.sns.it Tue Jun 6 21:17:52 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 6 Jun 2006 21:17:52 +0200 Subject: [Pw_forum] espresso 3.1 compilation In-Reply-To: <004101c6899a$34918a00$250017ac@ornl.gov> References: <20060606111614.BKZ74769@student-b.csv.cmich.edu> <200606061742.05362.giannozz@nest.sns.it> <004101c6899a$34918a00$250017ac@ornl.gov> Message-ID: <200606062117.52011.giannozz@nest.sns.it> On Tuesday 06 June 2006 20:51, Fernando A Reboredo wrote: > I am trying to compile espresso on a 3.1 4.5 TF IBM pSeries System > (cheetah NSSC ). > I get the following error > [...] > /usr/bin/rm_mlcache_file: cannot execute > config.sub: too many arguments > Try `config.sub --help' for more information. > configure: error: /bin/sh ./config.sub powerpc-ibm-aixoslevel: Error > processing cache, return code 126. oslevel: Use the -f flag to rebuild > cache. failed congratulations: you won the nomination in the competition for "most obscure error message ever" ! Seriously: inquire with your system manager. Maybe he disabled interactive execution of scripts. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From aaron at chips.ncsu.edu Tue Jun 6 21:24:47 2006 From: aaron at chips.ncsu.edu (aaron at chips.ncsu.edu) Date: Tue, 6 Jun 2006 15:24:47 -0400 (EDT) Subject: [Pw_forum] espresso 3.1 compilation In-Reply-To: <200606062117.52011.giannozz@nest.sns.it> References: <20060606111614.BKZ74769@student-b.csv.cmich.edu> <200606061742.05362.giannozz@nest.sns.it> <004101c6899a$34918a00$250017ac@ornl.gov> <200606062117.52011.giannozz@nest.sns.it> Message-ID: > Seriously: inquire with your system manager. Maybe he disabled > interactive execution of scripts. I think that's right. I got the same or similar message before on cheetah. Try to use: ./configure.old ibmsp I recall this as an adequate workaround on cheetah. I just lost access to so I can't test this for you... Aaron From degironc at sissa.it Tue Jun 6 23:26:46 2006 From: degironc at sissa.it (degironc) Date: Tue, 06 Jun 2006 23:26:46 +0200 Subject: [Pw_forum] Questions about ph.x output In-Reply-To: References: Message-ID: <4485F316.9010604@sissa.it> lan haiping wrote: > Dear developers: > > i try to understand the output of ph.x , but failed. > Would you please help me out of these naive problems ? > 1 , what are these numbers meaning? > > "phonons of Nanotube at Gamma > 1 32 8 8.0390000 4.0000000 4.0000000 0.0000000 0.0000000 > 0.0000000 ntyp nat ibrav celldm(1) ... celldm(6) > 1 'C ' 10946.15867720511 " ityp label mass_in_atomic_rydberg_units > 2, why the dynamical matrix is 3 x 6 ? it's a 3x3 complex matrix Real_xx Im_xx Real_xy Im_xy Real_xz Im_xz ... ... . stefano From degironc at sissa.it Tue Jun 6 23:44:21 2006 From: degironc at sissa.it (degironc) Date: Tue, 06 Jun 2006 23:44:21 +0200 Subject: [Pw_forum] the problem about eigndisplacement of phonon? In-Reply-To: <20060606143245.37877.qmail@web15808.mail.cnb.yahoo.com> References: <20060606143245.37877.qmail@web15808.mail.cnb.yahoo.com> Message-ID: <4485F735.5090701@sissa.it> ?? ? wrote: > Dear stefano: > Thank you very much. i still am confused with the equations you give. > i have three questions: > (1) \Phi U = M \omega^2 U > in this equation, what is the meaning of \Phi ? before answering this question I would like to ear your guess first. > (2) in fact, what is printed in the matdyn is eigndisplacement. when i > want eignvector, is M selected to be 207.2 amu. how to renormalize the > eignvector? > please give me suggestions! Suppose you write your how code and want normalize a vector ... what would you do ? I think I missed your third question... stefano From reboredofa at ornl.gov Tue Jun 6 23:30:06 2006 From: reboredofa at ornl.gov (Fernando A Reboredo) Date: Tue, 06 Jun 2006 17:30:06 -0400 Subject: [Pw_forum] espresso 3.1 compilation References: <20060606111614.BKZ74769@student-b.csv.cmich.edu> <200606061742.05362.giannozz@nest.sns.it> <004101c6899a$34918a00$250017ac@ornl.gov> <200606062117.52011.giannozz@nest.sns.it> Message-ID: <044401c689b0$61851980$250017ac@ornl.gov> Thanks if anyone has the same problem to compiling in Cheetah at NCCS please change the file 'config.guess' as follows # IBM_REV=`/usr/bin/oslevel` IBM_REV=5.2.0.0 That is comment IBM_REV=`/usr/bin/oslevel` and add the expected output IBM_REV=5.2.0.0 Please also thank Paul Kent. Paolo I hope to brake my record again :) Aaron thanks for the solution Fernando ----- Original Message ----- From: To: Sent: Tuesday, June 06, 2006 3:24 PM Subject: Re: [Pw_forum] espresso 3.1 compilation > > >> Seriously: inquire with your system manager. Maybe he disabled >> interactive execution of scripts. > > I think that's right. I got the same or similar message before on > cheetah. Try to use: > > ./configure.old ibmsp > > I recall this as an adequate workaround on cheetah. I just lost access to > so I can't test this for you... > > Aaron > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From coldwind1978 at yahoo.com.cn Wed Jun 7 04:19:01 2006 From: coldwind1978 at yahoo.com.cn (=?gb2312?q?=D5=EB=D2=B6=20=D6=EC?=) Date: Wed, 7 Jun 2006 10:19:01 +0800 (CST) Subject: [Pw_forum] Re:Re: how to obtained the eigndisplacement by eignvector Message-ID: <20060607021901.49121.qmail@web15804.mail.cnb.yahoo.com> Dear Vardha: i have known the output of dyn file is eigndisplacement. in your letter the normalized eigndisplacement should be converted into unnormalized eigndisplacement by modifying the code. please tell me how to modify the code in order to ouput the unnormalized eigndisplacement. Regards zhuzhenye --------------------------------- ??????-3.5G???20M?? -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060607/bafb080a/attachment.htm From coldwind1978 at yahoo.com.cn Wed Jun 7 14:34:58 2006 From: coldwind1978 at yahoo.com.cn (=?gb2312?q?=D5=EB=D2=B6=20=D6=EC?=) Date: Wed, 7 Jun 2006 20:34:58 +0800 (CST) Subject: [Pw_forum] Re:Re: how to obtained the eigndisplacement by eignvector Message-ID: <20060607123458.90886.qmail@web15804.mail.cnb.yahoo.com> Dear Vardha: i have known the output of dyn file is eigndisplacement. in your letter the normalized eigndisplacement should be converted into unnormalized eigndisplacement by modifying the code. please tell me how to modify the code in order to ouput the unnormalized eigndisplacement. Regards zhuzhenye __________________________________________________ ??????????????? http://cn.mail.yahoo.com -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060607/41679267/attachment.htm From baroni at sissa.it Wed Jun 7 23:10:54 2006 From: baroni at sissa.it (Stefano Baroni) Date: Wed, 7 Jun 2006 23:10:54 +0200 Subject: [PW_forum] Calculating Raman Intensities In-Reply-To: <20060606122245.BKZ84516@student-b.csv.cmich.edu> References: <20060606122245.BKZ84516@student-b.csv.cmich.edu> Message-ID: <0BA223EA-2D1D-43E6-BDF7-D7CC9E474B2E@sissa.it> On Jun 6, 2006, at 6:22 PM, John B Burdette wrote: > "calculation of Raman coefficients with Ultrasoft > Pseudopotentials is not implemented". Is it clear now? > > Does this imply that Raman coefficients cannot be calculated > using Ultrasoft Pseudopotentials? Not at at this time. This will require some more coding. sorry. I think that the dielctric constant can be calculated using US PP's. and so do phonons. So, even without any further coding, (non- resonant) Raman coefficients can be calculated using the present code by finite-difference differentiation of the dielectric constant with respect to nuclear displacements. SB > Or does it mean I haven't > turned on a switch somewhere? > > John > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060607/79805d09/attachment.htm From sbraccia at sissa.it Thu Jun 8 16:46:52 2006 From: sbraccia at sissa.it (carlo sbraccia) Date: Thu, 08 Jun 2006 10:46:52 -0400 Subject: Fwd: [Pw_forum] constraints In-Reply-To: <200606071743.25090.p.giannozzi@sns.it> References: <200606071743.25090.p.giannozzi@sns.it> Message-ID: <4488385C.9020100@sissa.it> Dear Guido, > Subject: [Pw_forum] constraints > Date: Friday 19 May 2006 10:46 > From: Guido Roma > To: pw_forum at pwscf.org > > Dear All, > > I have been using contrained relaxations with older version of PWSCF, > I've seen that in version 3.1 many things are changed in that respect. > The module coping with constraints seems to allow only certain types of > constraints (distances, angles), not center of mass type constraints for > examples, and is not any more so easy to write one's own constrain.f90 > subroutine. It is true that the module implementing constrained MD and relaxations presently allows for few standard constraint-types only. It is also true that one shouldn't restrict the code to only those standard types. For this reason the module implementing SHAKE has been designed in order to be flexible and easily extensible. What you need is a piece of code implementing the constraint and its gradient. The procedure to port the code to QE is described step-by-step in the file Doc/constraints_HOWTO.tex. Of course, if you have a constraint-type that is not problem-specific and if you think it can be useful to other users, I'll be pleased to port it to QE. > But, further more, I don't understand if the allowed > constraints as specified in the CONSTRAINTS input card are working or > not. Looking at the code I have the impression that they should work if > relaxing by damped dynamics (ion_dynamics='damp'), but not with bfgs. > Can someone confirm this or tell me if and where I'm wrong? > As far as I know the constraints as specified in the CONSTRAINTS input card work fine. If you have any indication of the contrary, please be more specific. Constrained relaxation is not allowed for Newton-Raphson relaxation scheme (BFGS) because of technical problems (if you want I can be more specific about the technical problems). Only preconditioned-damped-dynamics is available for constrained relaxations, but this is not a limitation since it doesn't do much worse than BFGS. > Any don't see any hint in the user-guide. > Maybe the documentation is not very detailed, but you can find a hint on how to perform a constrained MD calculation in the file Doc/INPUT_PW: ion_dynamics CHARACTER specify the type of ionic dynamics. For constrained dynamics or constrained optimisations add the CONSTRAINTS card. .... Regards, carlo From mverissi at ictp.it Thu Jun 8 18:53:27 2006 From: mverissi at ictp.it (Marcos Verissimo Alves) Date: Thu, 8 Jun 2006 18:53:27 +0200 (CEST) Subject: [Pw_forum] Pb USPP In-Reply-To: <4488385C.9020100@sissa.it> References: <200606071743.25090.p.giannozzi@sns.it> <4488385C.9020100@sissa.it> Message-ID: <39328.10.50.40.120.1149785607.squirrel@webmail1.ictp.trieste.it> Hi all, Does anyone have a good US PBE pseudo for Pb, with d states as semicore? The PWSCF pseudo database has it only in PZ parametrisation. Thanks, Marcos -- Dr. Marcos Verissimo Alves Post-Doctoral Fellow Condensed Matter and Statistical Physics Sector International Centre for Theoretical Physics Trieste, Italy -------- I have become so addicted to vi that I try to exit OpenOffice by typing :wq! From lukethulin at netscape.net Fri Jun 9 16:55:20 2006 From: lukethulin at netscape.net (Luke Thulin) Date: Fri, 09 Jun 2006 10:55:20 -0400 Subject: [Pw_forum] plotband.x Message-ID: <44898BD8.3000805@netscape.net> Hello, Is there documentation for the input definitions of plotband.x and plotrho.x? Thanks, Luke From giannozz at nest.sns.it Fri Jun 9 17:28:45 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 9 Jun 2006 17:28:45 +0200 Subject: [Pw_forum] Fwd: help Message-ID: <200606091728.45389.giannozz@nest.sns.it> ---------- Forwarded Message ---------- Subject: help Date: Friday 09 June 2006 16:52 From: H?seyin Murat T?T?NC? To: pw_forum-admin at pwscf.org Dear pwscf group coordinator I have been running the code successfuly, but have a small technical problem and would be grateful for your advice. In our computer system, different nodes have their own local disks (with scratch partition), and a rather slow communication link between master and slave nodes. The present version of the pwscf code assigns a single scrtach file space (presumably at the master node). This really slows my calculations, as well as virtually stops other users from using the machine. I would be grateful if you could help me (i) identify scratch and re-start data files, and (ii) advise me how to utilize scratch disks on local (i.e. slave) nodes instead of using one scratch file space on the master node. Thank you very much for your help. _____________________________________________________________________________ ______________ Bu e-posta, SA? CAWIS WebMail arac?l???yla g?nderilmi?tir : http://www.mail.sakarya.edu.tr/ Sakarya ?niversitesi ile ilgili bilgi, haber ve duyurular i?in : http://www.sakarya.edu.tr/ ____________________________________________________________________________ _______________ ------------------------------------------------------- -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From giannozz at nest.sns.it Fri Jun 9 17:43:48 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 9 Jun 2006 17:43:48 +0200 Subject: [Pw_forum] plotband.x In-Reply-To: <44898BD8.3000805@netscape.net> References: <44898BD8.3000805@netscape.net> Message-ID: <200606091743.48245.giannozz@nest.sns.it> On Friday 09 June 2006 16:55, Luke Thulin wrote: > Is there documentation for the input definitions of plotband.x and > plotrho.x? no, input is interactive and it is too simple to deserve a documentation Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From timtro at gmail.com Fri Jun 9 18:49:36 2006 From: timtro at gmail.com (Tim Teatro) Date: Fri, 9 Jun 2006 12:49:36 -0400 Subject: [Pw_forum] Dynamic Temperature Control and Select Stationary Ions. Message-ID: <560bcca0606090949s7d9dd9adva5307bce71ceefcf@mail.gmail.com> Hello. I believe I posted a similar question a year or two ago, but A) I don't think the question was resolved and B) the code has changed, so perhaps what was once a problem is not now. My group has a situation where we need to melt a crystal system, and then cool it back down to make an amorphous structure. Is this possible using any of the codes in the ESPRESSO package? If so, how? Do you happen to have an input example? Also, we have another system in which it would be useful to hold certain ions stationary while others are free to move under the Verlet algorithm. Of course we could cheat and set the mass of these atoms to be extremely high, but I think it would be better to simply stop them completely. Thank you for your time. -- Timothy A.V. Teatro University of Ontario Institute of Technology e. timtro at gmail.com p. 905.579.9248 c. 905.442.9248 -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060609/7f242c05/attachment.htm From lukethulin at netscape.net Fri Jun 9 23:28:29 2006 From: lukethulin at netscape.net (Luke Thulin) Date: Fri, 09 Jun 2006 17:28:29 -0400 Subject: [Pw_forum] plotband.x In-Reply-To: <200606091743.48245.giannozz@nest.sns.it> References: <44898BD8.3000805@netscape.net> <200606091743.48245.giannozz@nest.sns.it> Message-ID: <4489E7FD.3040209@netscape.net> Sorry, I'm rather new to the program, and Linux for that matter, so it is not simple for me yet. My calculation seems to go well until the final plotband.x command (I'm basically following example 5 as a template) when I get this: At line 103 of file error.f90 Traceback: not available, compile with -ftrace = frame or -ftrace=full So most everything seems available except for the postscript file of the band structure, which is what I really want though. I assume that I'm doing something slightly wrong that I can't see, so any help would be much appreciated. Thanks, Luke giannozz at nest.sns.it wrote: >On Friday 09 June 2006 16:55, Luke Thulin wrote: > > > >>Is there documentation for the input definitions of plotband.x and >>plotrho.x? >> >> > >no, input is interactive and it is too simple to deserve a documentation > >Paolo > > -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060609/74df2af4/attachment.htm From lzhuang at whu.edu.cn Sat Jun 10 10:54:39 2006 From: lzhuang at whu.edu.cn (Lin Zhuang) Date: Sat, 10 Jun 2006 16:54:39 +0800 Subject: [Pw_forum] ion calculation Message-ID: <2b80d0af0606100154v16e7896fq3936ee8247854c4@mail.gmail.com> Dear Espresso users, I wonder whether Espresso can deal with ion, if yes, how to define an ion as we usually do for an atom? Any advice will be greatly appreciated. best regards, Lin From marzari at MIT.EDU Sat Jun 10 16:26:59 2006 From: marzari at MIT.EDU (Nicola Marzari) Date: Sat, 10 Jun 2006 10:26:59 -0400 Subject: [Pw_forum] ion calculation In-Reply-To: <2b80d0af0606100154v16e7896fq3936ee8247854c4@mail.gmail.com> References: <2b80d0af0606100154v16e7896fq3936ee8247854c4@mail.gmail.com> Message-ID: <448AD6B3.1080304@mit.edu> Dear Lin, the total energy calculated for an infinite periodic system is well defined only for a neutral unit cell. By controlling and changing the number of electrons in CP or PWSCF, you can describe an ion, but in order to maintain neutrality the code adds a uniform background charge of opposite sign. For an isolated system, you obtain the correct result in the limit of large unit cells; this limit is reached fast if you add your own multipole corrections. For the case of a bulk system, the story is a more complex - start by reading the Makov-Payne PRB of 1995, and then some of the more recent works in PRL by Peter Schultz. nicola Lin Zhuang wrote: > Dear Espresso users, > I wonder whether Espresso can deal with ion, if yes, how to define an > ion as we usually do for an atom? > Any advice will be greatly appreciated. > best regards, > Lin > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From marzari at MIT.EDU Sat Jun 10 16:38:44 2006 From: marzari at MIT.EDU (Nicola Marzari) Date: Sat, 10 Jun 2006 10:38:44 -0400 Subject: [Pw_forum] Dynamic Temperature Control and Select Stationary Ions. In-Reply-To: <560bcca0606090949s7d9dd9adva5307bce71ceefcf@mail.gmail.com> References: <560bcca0606090949s7d9dd9adva5307bce71ceefcf@mail.gmail.com> Message-ID: <448AD974.8010007@mit.edu> Dear Tim, you can do this both with CP or PWSCF. CP would be the ideal code, provided the system remains insulating. On a single state-of-thje-art Pentium I think you can get nowadays 1ps/day, for a system with ~250 electrons. Otherwise, you should use Born-Oppenheimer dynamics (that is the standard mode for PWSCF, and is available as an option in CP). Melting and quenching has been used often - you could look e.g. at the works by Alfredo Pasquarello at EPFL (especially from the ~1998 Nature on SiO2 onwards). Car-Parrinello MD, and to a certain extent Born-Oppenheimer MD, require a good understanding of several concepts - we'll run a school on "Ab-initio molecular dynamics" in Urbana-Champaign, at the beginning of August, and attending that would probably be very useful: http://www.mcc.uiuc.edu/summerschool/2006/ At a minimum, have a close look at the review papers on the web - especially the one by Marx and Hutter (http://www.theochem.ruhr-uni-bochum.de/research/marx/cprev.en.html) or the shorter one by Galli and Pasquarello. The RevModPhys review of Payne from 1992 is another very good reference. As a word of care, look also at the 2002 and 2004 JCP from Tangney/Scandolo and from the Galli group. nicola Tim Teatro wrote: > Hello. > > I believe I posted a similar question a year or two ago, but A) I > don't think the question was resolved and B) the code has changed, so > perhaps what was once a problem is not now. > > My group has a situation where we need to melt a crystal system, and > then cool it back down to make an amorphous structure. Is this > possible using any of the codes in the ESPRESSO package? > > If so, how? Do you happen to have an input example? > > Also, we have another system in which it would be useful to hold > certain ions stationary while others are free to move under the Verlet > algorithm. Of course we could cheat and set the mass of these atoms to > be extremely high, but I think it would be better to simply stop them > completely. > > Thank you for your time. > > -- > Timothy A.V. Teatro > University of Ontario Institute of Technology > e. timtro at gmail.com > p. 905.579.9248 > c. 905.442.9248 From giannozz at nest.sns.it Sat Jun 10 19:28:45 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Sat, 10 Jun 2006 19:28:45 +0200 Subject: [Pw_forum] plotband.x In-Reply-To: <4489E7FD.3040209@netscape.net> References: <44898BD8.3000805@netscape.net> <200606091743.48245.giannozz@nest.sns.it> <4489E7FD.3040209@netscape.net> Message-ID: <200606101928.45299.giannozz@nest.sns.it> On Friday 09 June 2006 23:28, Luke Thulin wrote: > My calculation seems to go well until the final plotband.x command > (I'm basically following example 5 as a template) when I get this: > > At line 103 of file error.f90 > Traceback: not available, compile with -ftrace = frame or -ftrace=full this error is issued by the g95 compiler, I think, and means that yous program ended with an error status. You need to look at the output of plotband.x first. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From timtro at gmail.com Sun Jun 11 08:04:22 2006 From: timtro at gmail.com (Tim Teatro) Date: Sun, 11 Jun 2006 02:04:22 -0400 Subject: [Pw_forum] Dynamic Temperature Control and Select Stationary Ions. In-Reply-To: <448AD974.8010007@mit.edu> References: <560bcca0606090949s7d9dd9adva5307bce71ceefcf@mail.gmail.com> <448AD974.8010007@mit.edu> Message-ID: <560bcca0606102304p43382687l40a905da69e87e3e@mail.gmail.com> Dear Nicola, Thank you for your reply. I am working on large clusters, so calculation time is not a problem. Unfortunately I will not have access to cp.x for a few days because the computer cluster on which I was able to compile it is down. Therefore, I will need to use pw.x for now. Could you give me a list of input parameters I would need for changing the temperature during a MD simulation, and for holding a few specific ions still while others are able to move? It seems I can only find references to temperature rescaling and the Nose thermostat, which require the temperature to be specified ahead of time. And of course I can turn the ion dynamics off easily enough, but not for certain atoms. If you have any examples, this would be great too. Thanks again. Cheers. On 6/10/06, Nicola Marzari wrote: > > > Dear Tim, > > > you can do this both with CP or PWSCF. CP would be the ideal code, > provided the system remains insulating. > On a single state-of-thje-art Pentium I think you can get nowadays > 1ps/day, for a system with ~250 electrons. > > Otherwise, you should use Born-Oppenheimer dynamics (that is the > standard mode for PWSCF, and is > available as an option in CP). > > Melting and quenching has been used often - you could look e.g. at the > works by Alfredo Pasquarello at EPFL (especially > from the ~1998 Nature on SiO2 onwards). > > Car-Parrinello MD, and to a certain extent Born-Oppenheimer MD, require > a good understanding of several concepts - > we'll run a school on "Ab-initio molecular dynamics" in > Urbana-Champaign, at the beginning of > August, and attending that would probably be very useful: > http://www.mcc.uiuc.edu/summerschool/2006/ > > At a minimum, have a close look at the review papers on the web - > especially the one > by Marx and Hutter > (http://www.theochem.ruhr-uni-bochum.de/research/marx/cprev.en.html) or the > shorter one by Galli and Pasquarello. The RevModPhys review of Payne > from 1992 is another very > good reference. As a word of care, look also at the 2002 and 2004 JCP > from Tangney/Scandolo and > from the Galli group. > > > nicola > > > > > > > > Tim Teatro wrote: > > Hello. > > > > I believe I posted a similar question a year or two ago, but A) I > > don't think the question was resolved and B) the code has changed, so > > perhaps what was once a problem is not now. > > > > My group has a situation where we need to melt a crystal system, and > > then cool it back down to make an amorphous structure. Is this > > possible using any of the codes in the ESPRESSO package? > > > > If so, how? Do you happen to have an input example? > > > > Also, we have another system in which it would be useful to hold > > certain ions stationary while others are free to move under the Verlet > > algorithm. Of course we could cheat and set the mass of these atoms to > > be extremely high, but I think it would be better to simply stop them > > completely. > > > > Thank you for your time. > > > > -- > > Timothy A.V. Teatro > > University of Ontario Institute of Technology > > e. timtro at gmail.com > > p. 905.579.9248 > > c. 905.442.9248 > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Timothy A.V. Teatro University of Ontario Institute of Technology e. timtro at gmail.com p. 905.579.9248 c. 905.442.9248 From ezadshojaee at hotmail.com Sun Jun 11 10:27:05 2006 From: ezadshojaee at hotmail.com (Ezad Shojaee) Date: Sun, 11 Jun 2006 08:27:05 +0000 Subject: [Pw_forum] problem with DOS Message-ID: hi i want to plot DOS for TiO2 & in the band structure calculation i have this warning : Band Structure Calculation Davidson diagonalization with overlap WARNING: 2 eigenvalues not converged WARNING: 2 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 2 eigenvalues not converged WARNING: 2 eigenvalues not converged and it never ends ! i have choose ' ecutrho=600' not to have problem in recalculation of rho , so what's the point ? this is my input file for nscf cal. any suggestion will be appreciated ! &control calculation='nscf' prefix='TiO2', pseudo_dir = '/root/pseudo/', outdir='/root/tmp/' / &system ibrav=7, celldm(1) =7.155, celldm(3) =2.5136, nat=6, ntyp=2, ecutwfc = 50.0, ecutrho = 600.0, nbnd=20, / &electrons conv_thr = 1.0e-8 mixing_beta = 0.7 / ATOMIC_SPECIES ..... ATOMIC_POSITIONS ..... K_POINTS {automatic} 12 12 12 0 0 0 _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ From giannozz at nest.sns.it Mon Jun 12 09:43:18 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 12 Jun 2006 09:43:18 +0200 Subject: [Pw_forum] Fwd: help In-Reply-To: <200606091728.45389.giannozz@nest.sns.it> References: <200606091728.45389.giannozz@nest.sns.it> Message-ID: <200606120943.18274.giannozz@nest.sns.it> > From: H?seyin Murat T?T?NC? > > In our computer system, different nodes have their own local disks > (with scratch partition), and a rather slow communication link between > master and slave nodes. The present version of the pwscf code assigns > a single scratch file space (presumably at the master node). This > really slows my calculations, as well as virtually stops other users > from using the machine. > > I would be grateful if you could help me (i) identify scratch and > re-start data files, and (ii) advise me how to utilize scratch disks > on local (i.e. slave) nodes instead of using one scratch file space on > the master node. in parallel execution, each processor writes its own chunk of wavefunctions, unless variable 'wf_collect' is specified in input. Wavefunctions are usually by far the largest file. All the rest (small files, except the charge density that becomes large only for very large systems) is collected on a single processor and written there. If this slows down the calculation, either you are writing through the network (check where the directory pointed by variable "outdir" resides: I/O through NFS must be avoided) or you have really slow communications or disks or both. The above holds for the current version of Pwscf (3.1) -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From giannozz at nest.sns.it Mon Jun 12 09:49:02 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 12 Jun 2006 09:49:02 +0200 Subject: [Pw_forum] problem with DOS In-Reply-To: References: Message-ID: <200606120949.02423.giannozz@nest.sns.it> On Sunday 11 June 2006 10:27, Ezad Shojaee wrote: > i have choose ' ecutrho=600' not to have problem in recalculation > of rho , so what's the point ? and what's your point? a cutoff for the charge density should be set only for ultrasoft PP. In any case, it has to be the same in the scf and non-scf calculations -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From ezadshojaee at hotmail.com Mon Jun 12 17:28:15 2006 From: ezadshojaee at hotmail.com (Ezad Shojaee) Date: Mon, 12 Jun 2006 15:28:15 +0000 Subject: [Pw_forum] problem with DOS Message-ID: hi trying to obtain DOS , i am using ultrasoft pp's and ' ecutrho' is the same in scf & nscf calculation. in the INPUT_PW file ,there is written that "occupations=tetrahedra" is for metals & DOS calclation, so it seems that i have to select this kind of occupation but trying this, some of eigenvalues in nscf calculation do not converge and while not using it, i have different band energies from those i had done for band structure before.so what should i do ? _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ From lukethulin at netscape.net Mon Jun 12 18:39:08 2006 From: lukethulin at netscape.net (Luke Thulin) Date: Mon, 12 Jun 2006 12:39:08 -0400 Subject: [Pw_forum] plotband.x In-Reply-To: <200606101928.45299.giannozz@nest.sns.it> References: <44898BD8.3000805@netscape.net> <200606091743.48245.giannozz@nest.sns.it> <4489E7FD.3040209@netscape.net> <200606101928.45299.giannozz@nest.sns.it> Message-ID: <448D98AC.1010205@netscape.net> I was making a mistake with the plotband.x input file, I had accidentally deleted the line with the ticks and reference energy values. The example 5 seems to have chosen the reference energy to be equal to the Fermi energy. Is the Fermi Energy calculated and stored somewhere so that I can do the same? I currently just set those values to zero. Thanks, Luke giannozz at nest.sns.it wrote: >On Friday 09 June 2006 23:28, Luke Thulin wrote: > > > >>My calculation seems to go well until the final plotband.x command >>(I'm basically following example 5 as a template) when I get this: >> >>At line 103 of file error.f90 >>Traceback: not available, compile with -ftrace = frame or -ftrace=full >> >> > >this error is issued by the g95 compiler, I think, and means that yous >program ended with an error status. You need to look at the output >of plotband.x first. > >Paolo > > -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060612/4671d8d7/attachment.htm From giannozz at nest.sns.it Mon Jun 12 19:15:47 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 12 Jun 2006 19:15:47 +0200 Subject: [Pw_forum] plotband.x In-Reply-To: <448D98AC.1010205@netscape.net> References: <44898BD8.3000805@netscape.net> <200606101928.45299.giannozz@nest.sns.it> <448D98AC.1010205@netscape.net> Message-ID: <200606121915.47385.giannozz@nest.sns.it> On Monday 12 June 2006 18:39, Luke Thulin wrote: > Is the Fermi Energy calculated and stored somewhere > so that I can do the same? in a metal (i.e. a system where occupancies are calculated using either tetrahedra or gaissian broadening) the Fermi energy is printed in the output of the self-consistent run and also saved in the data file, but not in the file containing e(k) that plotbands.x uses Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From giannozz at nest.sns.it Mon Jun 12 19:18:48 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 12 Jun 2006 19:18:48 +0200 Subject: [Pw_forum] problem with DOS In-Reply-To: References: Message-ID: <200606121918.48059.giannozz@nest.sns.it> On Monday 12 June 2006 17:28, Ezad Shojaee wrote: > in the INPUT_PW file, there is written that "occupations=tetrahedra" is > for metals & DOS calculation, so it seems that i have to select this kind > of occupation but trying this, some of eigenvalues in nscf calculation > do not converge and while not using it, i have different band energies > from those i had done for band structure before. small differences may arise if the calculation "before" is not done exactly in the same way as "after". Messages complaining that some eigenvalues have not converged are not necessarily significant. In any event, please provide a test job that can be run Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From lukethulin at netscape.net Mon Jun 12 19:43:03 2006 From: lukethulin at netscape.net (Luke Thulin) Date: Mon, 12 Jun 2006 13:43:03 -0400 Subject: [Pw_forum] plotband.x In-Reply-To: <200606121915.47385.giannozz@nest.sns.it> References: <44898BD8.3000805@netscape.net> <200606101928.45299.giannozz@nest.sns.it> <448D98AC.1010205@netscape.net> <200606121915.47385.giannozz@nest.sns.it> Message-ID: <448DA7A7.8070300@netscape.net> What about non-metals. For example, how was the value of 6.255 chosen for Si in example 5? Thanks, Luke giannozz at nest.sns.it wrote: >On Monday 12 June 2006 18:39, Luke Thulin wrote: > > > >>Is the Fermi Energy calculated and stored somewhere >>so that I can do the same? >> >> > >in a metal (i.e. a system where occupancies are calculated >using either tetrahedra or gaissian broadening) the Fermi >energy is printed in the output of the self-consistent run and >also saved in the data file, but not in the file containing e(k) >that plotbands.x uses > >Paolo > > -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060612/f13c7310/attachment.htm From gtg166n at mail.gatech.edu Mon Jun 12 20:09:36 2006 From: gtg166n at mail.gatech.edu (Sigifredo Sanchez-Carrera) Date: Mon, 12 Jun 2006 14:09:36 -0400 Subject: [Pw_forum] rdiaghg error # 122 in Phonon Calculation Message-ID: <1150135775.448dade00681d@webmail.mail.gatech.edu> Dear PWscf list members: I am trying to complete a SCF calculation with K_POINTS (gamma) to be able to use the phcg.x tool as reported in example09; but I am getting an error message in my output calculation (using C.pz-vbc.UPF and H.vbc.UPF pseudopotentials). The error message looks like this: -------- iteration # 1 ecut= 30.00 ryd beta=0.70 Davidson diagonalization with overlap %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from rdiaghg : error # 122 info =/= 0 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... -------- What looks interesting for me is that I do not get this error message when I use the following pseudopotentials in my input file: C.pw91-van_ak.UPF H.pw91-van_ak.UPF C.pz-rrkjus.UPF H.pz-rrkjus.UPF But these pseudopotentials are not implemented in the phcg.x tool as shown in the following error message: -------- nbndx = 60 nbnd = 60 natomwfc = 96 npwx = 3230 nelec = 96.00 nkb = 160 ngl = 13407 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from cg_readin : error # 1 ultrasoft pseudopotential not implemented %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... -------- Is there any reason for the error # 122? What can I do? I have also added my input file: Thanks, Sigifredo Sanchez-Carrera &CONTROL calculation = 'scf' pseudo_dir = '/theoryfs/ccmst/bredas/home/sigi/QE-31/espresso-3.1/pseudo/' , outdir = '/theoryfs/ccmst/bredas/home/sigi/Naph-PWscf/Naph-LDA/tmp/' , prefix = 'naph-phonon' , verbosity = 'high' , tstress = .true. , tprnfor = .true. , / &SYSTEM ibrav = 12 , celldm(1) = 15.3030010555 , celldm(2) = 1.06841195357 , celldm(3) = 0.735119782662 , celldm(4) = -0.564967003425 , nat = 36 , ntyp = 2 , nbnd = 60 , ecutwfc = 20.0 , nosym = .false. , / &ELECTRONS conv_thr = 1.0e-6 , mixing_beta = 0.6 , mixing_mode = 'plain' / ATOMIC_SPECIES C 12.001 C.pz-vbc.UPF H 1.008 H.vbc.UPF ATOMIC_POSITIONS (crystal) C 0.082335521 0.328301882 0.982725405 C 0.112617924 0.222270006 0.837846777 C 0.048726451 0.037863440 0.894571438 C 0.077065232 0.925858072 0.748744036 C 0.987151132 0.253851526 0.191290154 H 0.132264404 0.472813827 0.933376380 H 0.186798064 0.279232009 0.672069323 H 0.153099286 0.985761499 0.584451900 H 0.962496267 0.339761977 0.307904608 C 0.417664479 0.671698118 0.482725405 C 0.917664479 0.671698118 0.017274595 C 0.582335521 0.328301882 0.517274595 C 0.387382076 0.777729994 0.337846777 C 0.887382076 0.777729994 0.162153222 C 0.612617924 0.222270006 0.662153222 C 0.451273549 0.962136560 0.394571438 C 0.951273549 0.962136560 0.105428562 C 0.548726451 0.037863440 0.605428562 C 0.422934768 0.074141928 0.248744036 C 0.922934768 0.074141928 0.251255964 C 0.577065232 0.925858072 0.751255964 C 0.512848868 0.746148474 0.691290154 C 0.012848868 0.746148474 0.808709846 C 0.487151132 0.253851526 0.308709846 H 0.367735596 0.527186173 0.433376380 H 0.867735596 0.527186173 0.066623620 H 0.632264404 0.472813827 0.566623620 H 0.313201936 0.720767991 0.172069323 H 0.813201936 0.720767991 0.327930677 H 0.686798064 0.279232009 0.827930677 H 0.346900714 0.014238501 0.084451900 H 0.846900714 0.014238501 0.415548100 H 0.653099286 0.985761499 0.915548100 H 0.537503733 0.660238023 0.807904608 H 0.037503733 0.660238023 0.692095392 H 0.462496267 0.339761977 0.192095392 K_POINTS (gamma) Thanks, -- Sigifredo Sanchez-Carrera Georgia Institute of Technology School of Chemistry and Biochemistry Boggs 1-90 770 State St. NW Atlanta, GA 30332-0400 Work (404) 894 6456 Home (404) 206 9531 From eyvaz_isaev at yahoo.com Mon Jun 12 22:15:19 2006 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Mon, 12 Jun 2006 13:15:19 -0700 (PDT) Subject: [Pw_forum] rdiaghg error # 122 in Phonon Calculation In-Reply-To: <1150135775.448dade00681d@webmail.mail.gatech.edu> Message-ID: <20060612201519.81215.qmail@web60321.mail.yahoo.com> Hi, > -------- > iteration # 1 ecut= 30.00 ryd > beta=0.70 > Davidson diagonalization with overlap %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > from rdiaghg : error # 122 > info =/= 0 > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > stopping ... > -------- This is a very famous error and explaind in the forum why it happens. Please have a look at the forum archive. Very briefly, it is due to a fact that an eigenvalue is not converged with a machin precision. A solution is to change the diagonalization scheme to 'cg' (conjugate gradient) method which is believed to be slower than the Davidson method (by default). > > What looks interesting for me is that I do not get > this error message when I use > the following pseudopotentials in my input file: > > C.pw91-van_ak.UPF > H.pw91-van_ak.UPF > > C.pz-rrkjus.UPF > H.pz-rrkjus.UPF > > But these pseudopotentials are not implemented in > the phcg.x tool as shown in > the following error message: > > -------- > nbndx = 60 nbnd = 60 natomwfc = > 96 npwx = 3230 > nelec = 96.00 nkb = 160 ngl = > 13407 > > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > from cg_readin : error # 1 > ultrasoft pseudopotential not implemented > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > stopping ... > -------- > > Is there any reason for the error # 122? What can I > do? I have also added my input file: Plesae see above. Besides, sometime the error served me as a signal of incorrect structure. > > &SYSTEM > ibrav = 12 , > celldm(1) = 15.3030010555 , > celldm(2) = 1.06841195357 , > celldm(3) = 0.735119782662 , > celldm(4) = -0.564967003425 , > nat = 36 , > ntyp = 2 , > nbnd = 60 , > ecutwfc = 20.0 , > nosym = .false. , ecutwfc=20 (Ry) looks small for a norm-conserving pseudopotential, try 30 (but, probably, you did it according to your message). nosym=.false. is taken by default. I have also paid attention that you have 20 C and 16 H atoms which bring total 96 electrons. Thus, 48 bands should be enough for a semiconductor (insulator) system. It seems that empty (conductive) states (bands) converge somewhat worse compared to valence bands. So, either you can remove nbnd=60 (it will be calculated by pw.x) or reduce the number of bands to be calculated. Bests, Eyvaz. > / > &ELECTRONS > conv_thr = 1.0e-6 , > mixing_beta = 0.6 , > mixing_mode = 'plain' > / > ATOMIC_SPECIES > C 12.001 C.pz-vbc.UPF > H 1.008 H.vbc.UPF > ATOMIC_POSITIONS (crystal) > C 0.082335521 0.328301882 0.982725405 > C 0.112617924 0.222270006 0.837846777 > C 0.048726451 0.037863440 0.894571438 > C 0.077065232 0.925858072 0.748744036 > C 0.987151132 0.253851526 0.191290154 > H 0.132264404 0.472813827 0.933376380 > H 0.186798064 0.279232009 0.672069323 > H 0.153099286 0.985761499 0.584451900 > H 0.962496267 0.339761977 0.307904608 > C 0.417664479 0.671698118 0.482725405 > C 0.917664479 0.671698118 0.017274595 > C 0.582335521 0.328301882 0.517274595 > C 0.387382076 0.777729994 0.337846777 > C 0.887382076 0.777729994 0.162153222 > C 0.612617924 0.222270006 0.662153222 > C 0.451273549 0.962136560 0.394571438 > C 0.951273549 0.962136560 0.105428562 > C 0.548726451 0.037863440 0.605428562 > C 0.422934768 0.074141928 0.248744036 > C 0.922934768 0.074141928 0.251255964 > C 0.577065232 0.925858072 0.751255964 > C 0.512848868 0.746148474 0.691290154 > C 0.012848868 0.746148474 0.808709846 > C 0.487151132 0.253851526 0.308709846 > H 0.367735596 0.527186173 0.433376380 > H 0.867735596 0.527186173 0.066623620 > H 0.632264404 0.472813827 0.566623620 > H 0.313201936 0.720767991 0.172069323 > H 0.813201936 0.720767991 0.327930677 > H 0.686798064 0.279232009 0.827930677 > H 0.346900714 0.014238501 0.084451900 > H 0.846900714 0.014238501 0.415548100 > H 0.653099286 0.985761499 0.915548100 > H 0.537503733 0.660238023 0.807904608 > H 0.037503733 0.660238023 0.692095392 > H 0.462496267 0.339761977 0.192095392 > K_POINTS (gamma) > > Thanks, > > -- > Sigifredo Sanchez-Carrera > > Georgia Institute of Technology > School of Chemistry and Biochemistry > Boggs 1-90 > 770 State St. NW > Atlanta, GA 30332-0400 > > Work (404) 894 6456 > Home (404) 206 9531 > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From yonasb at yahoo.com Mon Jun 12 22:38:59 2006 From: yonasb at yahoo.com (Yonas Abraham) Date: Mon, 12 Jun 2006 13:38:59 -0700 (PDT) Subject: [Pw_forum] Dynamic Temperature Control and Select Stationary Ions. In-Reply-To: <560bcca0606102304p43382687l40a905da69e87e3e@mail.gmail.com> Message-ID: <20060612203859.97528.qmail@web30911.mail.mud.yahoo.com> Dear Tim, In CP (QE >= 3), there is a way to change temperature on the fly during the MD simulation. look at Doc/AUTOPILOT file on how to use it. In fact, it will let you change up to 10 CP (like dt, isave, iprint etc) variables on the fly. in addition, you can even PAUSE the simulation while you are deciding what value to use! but the current AUTOPILOT implementation doesn't allow you to fix certain atoms on the fly. But if you are serious, you can extend AUTOPILOT to do so. But I don't know if there is such thing on pw.x. /yonas --- Tim Teatro wrote: > Dear Nicola, > > Thank you for your reply. I am working on large > clusters, so > calculation time is not a problem. Unfortunately I > will not have > access to cp.x for a few days because the computer > cluster on which I > was able to compile it is down. Therefore, I will > need to use pw.x for > now. > > Could you give me a list of input parameters I would > need for changing > the temperature during a MD simulation, and for > holding a few specific > ions still while others are able to move? It seems I > can only find > references to temperature rescaling and the Nose > thermostat, which > require the temperature to be specified ahead of > time. And of course I > can turn the ion dynamics off easily enough, but not > for certain > atoms. > > If you have any examples, this would be great too. > > Thanks again. > > Cheers. > > On 6/10/06, Nicola Marzari wrote: > > > > > > Dear Tim, > > > > > > you can do this both with CP or PWSCF. CP would be > the ideal code, > > provided the system remains insulating. > > On a single state-of-thje-art Pentium I think you > can get nowadays > > 1ps/day, for a system with ~250 electrons. > > > > Otherwise, you should use Born-Oppenheimer > dynamics (that is the > > standard mode for PWSCF, and is > > available as an option in CP). > > > > Melting and quenching has been used often - you > could look e.g. at the > > works by Alfredo Pasquarello at EPFL (especially > > from the ~1998 Nature on SiO2 onwards). > > > > Car-Parrinello MD, and to a certain extent > Born-Oppenheimer MD, require > > a good understanding of several concepts - > > we'll run a school on "Ab-initio molecular > dynamics" in > > Urbana-Champaign, at the beginning of > > August, and attending that would probably be very > useful: > > http://www.mcc.uiuc.edu/summerschool/2006/ > > > > At a minimum, have a close look at the review > papers on the web - > > especially the one > > by Marx and Hutter > > > (http://www.theochem.ruhr-uni-bochum.de/research/marx/cprev.en.html) > or the > > shorter one by Galli and Pasquarello. The > RevModPhys review of Payne > > from 1992 is another very > > good reference. As a word of care, look also at > the 2002 and 2004 JCP > > from Tangney/Scandolo and > > from the Galli group. > > > > > > nicola > > > > > > > > > > > > > > > > Tim Teatro wrote: > > > Hello. > > > > > > I believe I posted a similar question a year or > two ago, but A) I > > > don't think the question was resolved and B) the > code has changed, so > > > perhaps what was once a problem is not now. > > > > > > My group has a situation where we need to melt a > crystal system, and > > > then cool it back down to make an amorphous > structure. Is this > > > possible using any of the codes in the ESPRESSO > package? > > > > > > If so, how? Do you happen to have an input > example? > > > > > > Also, we have another system in which it would > be useful to hold > > > certain ions stationary while others are free to > move under the Verlet > > > algorithm. Of course we could cheat and set the > mass of these atoms to > > > be extremely high, but I think it would be > better to simply stop them > > > completely. > > > > > > Thank you for your time. > > > > > > -- > > > Timothy A.V. Teatro > > > University of Ontario Institute of Technology > > > e. timtro at gmail.com > > > p. 905.579.9248 > > > c. 905.442.9248 > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > -- > Timothy A.V. Teatro > University of Ontario Institute of Technology > e. timtro at gmail.com > p. 905.579.9248 > c. 905.442.9248 > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From giannozz at nest.sns.it Tue Jun 13 00:08:12 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 13 Jun 2006 00:08:12 +0200 Subject: [Pw_forum] rdiaghg error # 122 in Phonon Calculation In-Reply-To: <1150135775.448dade00681d@webmail.mail.gatech.edu> References: <1150135775.448dade00681d@webmail.mail.gatech.edu> Message-ID: <200606130008.12321.giannozz@nest.sns.it> On Monday 12 June 2006 20:09, Sigifredo Sanchez-Carrera wrote: > from rdiaghg : error # 122 > info =/= 0 The paragraph in the users' guide about erros in cdiagh/cdiaghg applies as well to this case. --- Possible reasons: * serious error in data, such as bad atomic positions or bad crystal structure/supercell; * a bad PP (for instance, with a ghost); * a failure of the algorithm performing subspace diagonalization. The LAPACK algorithms used by cdiagh or cdiaghg are very robust and extensively tested. Still, it may seldom happen that such algorithms fail. In at least one case the failures was tracked to the non-positiveness of the S matrix appearing in the US-PP formalism. In other cases, the error is found to be non reproducible on different architectures and disappearing if the calculation is repeated with even minimal changes in its parameters. In both cases, the reasons for such behavior are unclear and the only advice is to use conjugate-gradient diagonalization (diagonalization='cg'), a slower but very robust algorithm, and see what happens. --- You checked that your atomic positions make sense, didn't you? Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From giannozz at nest.sns.it Tue Jun 13 00:14:13 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 13 Jun 2006 00:14:13 +0200 Subject: [Pw_forum] plotband.x In-Reply-To: <448DA7A7.8070300@netscape.net> References: <44898BD8.3000805@netscape.net> <200606121915.47385.giannozz@nest.sns.it> <448DA7A7.8070300@netscape.net> Message-ID: <200606130014.13675.giannozz@nest.sns.it> On Monday 12 June 2006 19:43, Luke Thulin wrote: > What about non-metals. For example, how was the value of 6.255 > chosen for Si in example 5? the position of the highest valence band -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From giannozz at nest.sns.it Tue Jun 13 10:30:47 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 13 Jun 2006 10:30:47 +0200 Subject: [Pw_forum] rdiaghg error # 122 in Phonon Calculation In-Reply-To: <1150135775.448dade00681d@webmail.mail.gatech.edu> References: <1150135775.448dade00681d@webmail.mail.gatech.edu> Message-ID: <200606131030.47738.giannozz@nest.sns.it> On Monday 12 June 2006 20:09, Sigifredo Sanchez-Carrera wrote: > I am trying to complete a SCF calculation with K_POINTS (gamma) > to be able to use the phcg.x tool as reported in example09; but I am > getting an error message in my output calculation (using C.pz-vbc.UPF > and H.vbc.UPF pseudopotentials). it works for me, both at 20 ry and at 30 ry (both are very low cutoffs for norm-conserving C and H, by the way) Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From lukethulin at netscape.net Tue Jun 13 16:48:26 2006 From: lukethulin at netscape.net (Luke Thulin) Date: Tue, 13 Jun 2006 10:48:26 -0400 Subject: [Pw_forum] k point selection Message-ID: <448ED03A.2060403@netscape.net> Hello everyone, I don't have a very good understanding of how to properly choose my k points for sampling. Therefore, I'm currently staying with automatic generation. To do band structure plots, is it necessary to choose the same number of k points for both the scf and nscf calculations, or can less be used for the scf? If I try to enter a list of k points manually so that I can choose my path, how do you determine the weights? Thanks, Luke From baroni at sissa.it Tue Jun 13 18:02:07 2006 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 13 Jun 2006 18:02:07 +0200 Subject: [Pw_forum] k point selection In-Reply-To: <448ED03A.2060403@netscape.net> References: <448ED03A.2060403@netscape.net> Message-ID: On Jun 13, 2006, at 4:48 PM, Luke Thulin wrote: > Hello everyone, > > I don't have a very good understanding of how to properly choose my > k points for sampling. Therefore, I'm currently staying with > automatic generation. I would rather use something to be done automatically, when I understand how it works. You may want to get a general idea from the historic paper by Monkhorst and Pack, Phys. Rev. B 13, 5188 (1976). > To do band structure plots, is it necessary to choose the same > number of k points for both the scf and nscf calculations, NO > or can less be used for the scf? sure: the number ok kpoints in a scf calculation is determined by the accuracy by which you want to calculate integrals over the 1st Brillouin Zone. The number of points in a nscf calculation depends on the points YOU are interested in. (some given symmetry directions, rather than others, etc.) > If I try to enter a list of k points manually so that I can choose > my path, how do you determine the weights? In a nscf calculation, weights are read, but not used, as it has been repeatedly noted in this mailing list. In a SCF, the weights would be all alike if the whole BZ was sampled. In order to save cpu time, only the portion of the BZ which is "irreducible" by symmetry (the "irreduciple wedge") is sampled, and the weight ha to be proportional to the number of points in the BZ which are euivalent by symmetry to the given k-point. Full details can be found in the above paper by Monkhorst and Pack. > Thanks, > Luke You are most welcome. In the future, please do not forget to browse the archives of this mailing list before posting. I am pretty sure that the questions you raised have been answered many times, already. Stefano Baroni > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060613/6e442d70/attachment.htm From eyvaz_isaev at yahoo.com Tue Jun 13 18:26:50 2006 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Tue, 13 Jun 2006 09:26:50 -0700 (PDT) Subject: [Pw_forum] k point selection In-Reply-To: <448ED03A.2060403@netscape.net> Message-ID: <20060613162650.78416.qmail@web60317.mail.yahoo.com> Hi, > Therefore, I'm currently staying with automatic > generation. This one you have to use for SCF-calculations. > To do band structure plots, is it necessary to choose > the same number of k points for both the scf and > nscf calculations, or can less be used for the scf? You can use the same set of k-points (you should extract them from scf_output_file) them manually , but the result would be very ugly. You will not be able to plot because k-points are not aligned along high symmetry directions. So, the answer is "No". You have to generate k-points along a given directions and then calculate band energies for these k-points > If I try to enter a list of k points manually so > that I can choose my > path, how do you determine the weights? Put them as 1.0, they are not used for band structure calculations. Actually, I am going to redraw a small set of programs to get band structure (using Gnuplot) for cubic and hexagonal lattices, at least. Hopefully it would be useful for beginners. Bests, Eyvaz. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From lukethulin at netscape.net Wed Jun 14 02:30:13 2006 From: lukethulin at netscape.net (Luke Thulin) Date: Tue, 13 Jun 2006 20:30:13 -0400 Subject: [Pw_forum] k point selection In-Reply-To: <20060613162650.78416.qmail@web60317.mail.yahoo.com> References: <20060613162650.78416.qmail@web60317.mail.yahoo.com> Message-ID: <448F5895.6050105@netscape.net> Eyvaz and Stefano, Thank you very much for the instruction on k point sampling, I'll look into getting a copy of the Monkhorst Pack paper. One more quick issue: To save on computation time, does the automatic k point sampling for the scf calculation determine by itself which points comprise the irriducible brillouin zone? Or is it at least general practice to reuse the scf calculation results among different nscf calculations along different k paths? Sorry if this question has been asked, but I seem to be incompentent when it comes to searching the archives. Thanks, Luke eyvaz_isaev at yahoo.com wrote: >Hi, > > > >>Therefore, I'm currently staying with automatic >>generation. >> >> >This one you have to use for SCF-calculations. > > > >>To do band structure plots, is it necessary to >> >> >choose > the same number of k points for both the scf >and > > >>nscf calculations, or can less be used for the scf? >> >> >You can use the same set of k-points (you should >extract them from scf_output_file) them manually , but >the result would be very ugly. You will not be able >to plot because k-points are not aligned along high >symmetry directions. > >So, the answer is "No". You have to generate k-points >along a given directions and then calculate band >energies for these k-points > > > >>If I try to enter a list of k points manually so >>that I can choose my >>path, how do you determine the weights? >> >> >Put them as 1.0, they are not used for band structure >calculations. > >Actually, I am going to redraw a small set of programs > to get band structure (using Gnuplot) for cubic and >hexagonal lattices, at least. Hopefully it would be >useful for beginners. > >Bests, >Eyvaz. > > > > >__________________________________________________ >Do You Yahoo!? >Tired of spam? Yahoo! Mail has the best spam protection around >http://mail.yahoo.com >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > > -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060613/8ecfae33/attachment.htm From marzari at MIT.EDU Wed Jun 14 02:32:09 2006 From: marzari at MIT.EDU (Nicola Marzari) Date: Tue, 13 Jun 2006 20:32:09 -0400 Subject: [Pw_forum] ads et al In-Reply-To: <200606120943.18274.giannozz@nest.sns.it> References: <200606091728.45389.giannozz@nest.sns.it> <200606120943.18274.giannozz@nest.sns.it> Message-ID: <448F5909.7090502@mit.edu> Dear All, Is it appropriate to post on this list news that are not related to PWSCF/CP, but are relevant to the community ? My suggestion would be that it could be useful to post notices of electronic-structure schools, especially if Espresso related, and open positions, again if Espresso related. In order to test this theory, I attach a few notes below. nicola ---- Bangalore Summer School on "Electronic Structure Methods and their Applications" in conjunction with Conference on Computational Materials Theory: http://www.jncasr.ac.in/smr1783/ ---- 2006 Summer School on Computational Materials Science Ab Initio Molecular Dynamics Simulation Methods in Chemistry: http://www.mcc.uiuc.edu/summerschool/2006/ (Please note that the Espresso program has not been uploaded yet, but lecturers will include Stefano Baroni and Roberto Car) ---- Postdoc position: Defect evolution and deformation in metastable alloys Applications are invited for one postdoctoral position in the MIT Department of Materials Science and Engineering. This researcher will work with Profs. Sidney Yip, Nicola Marzari, and Krystyn Van Vliet as part of an interdisciplinary, international team that includes strong collaboration with industry partners. The focus of the project is multiscale computational modeling and simulation of metastable metallic alloys under extreme thermal and mechanical environments. The objective is to identify mechanisms of crystalline defect evolution and consequences toward mechanical deformation. Funding of this position is for an initial period of 2 years. Demonstrated experience in scientific programming and the application of ab initio / DFT methods for solid state systems is mandatory; experience with kinetic Monte Carlo and molecular/dislocation dynamics is preferred. The position is open immediately with a mutually agreed starting date. Requests for further information and applications (CV, two recommenders' contact information, and one representative publication) should be sent to Prof. Krystyn J. Van Vliet, Department of Materials Science and Engineering, MIT; E-mail:krystyn at mit.edu --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From liyanpcl at yahoo.com.cn Wed Jun 14 04:27:42 2006 From: liyanpcl at yahoo.com.cn (li yan) Date: Wed, 14 Jun 2006 10:27:42 +0800 (CST) Subject: [Pw_forum] how to find out the stable structure under high pressure Message-ID: <20060614022742.6488.qmail@web15605.mail.cnb.yahoo.com> Dear all, I found soft modes in a system under pressure. Is it true that there must be a global energy miniumu in the subspace spanned by the eignevectors of the soft modes? And by searching this munimum, can i find the stable high-pressure strucure? best regards __________________________________________________ ??????????????? http://cn.mail.yahoo.com -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060614/3f5e28d3/attachment.htm From marzari at MIT.EDU Wed Jun 14 05:01:39 2006 From: marzari at MIT.EDU (Nicola Marzari) Date: Tue, 13 Jun 2006 23:01:39 -0400 Subject: [Pw_forum] how to find out the stable structure under high pressure In-Reply-To: <20060614022742.6488.qmail@web15605.mail.cnb.yahoo.com> References: <20060614022742.6488.qmail@web15605.mail.cnb.yahoo.com> Message-ID: <448F7C13.1030300@mit.edu> Yes, nicola li yan wrote: > Dear all, > I found soft modes in a system under pressure. Is it true that > there must be a global energy miniumu in the subspace spanned by the > eignevectors of the soft modes? And by searching this munimum, can i > find the stable high-pressure strucure? > > > best regards > > __________________________________________________ > ??????????????? > http://cn.mail.yahoo.com > -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From baroni at sissa.it Wed Jun 14 08:32:52 2006 From: baroni at sissa.it (Stefano Baroni) Date: Wed, 14 Jun 2006 08:32:52 +0200 Subject: [Pw_forum] k point selection In-Reply-To: <448F5895.6050105@netscape.net> References: <20060613162650.78416.qmail@web60317.mail.yahoo.com> <448F5895.6050105@netscape.net> Message-ID: <087E9D61-9B58-4985-9019-3C3A9907EE1D@sissa.it> On Jun 14, 2006, at 2:30 AM, Luke Thulin wrote: > Eyvaz and Stefano, > > Thank you very much for the instruction on k point sampling, I'll > look into getting a copy of the Monkhorst Pack paper. One more > quick issue: To save on computation time, does the automatic k > point sampling for the scf calculation determine by itself which > points comprise the irriducible brillouin zone? try by yourself. run a calculation for, say, an fcc structure and see whether or not you find two k points which are equivalent by any bubic symmetry operation. (any operation of the cubic group can be seen as the product of a parmutation of the cartesian coordinates - there are seix such permutations - and the change of sign of any of the three coordinates - there are 2^3=8 such operations - for a total of 48=6x8 symmetry operations) > Or is it at least general practice to reuse the scf calculation > results among different nscf calculations along different k paths? I do not quite get the question > Sorry if this question has been asked, but I seem to be > incompentent when it comes to searching the archives. never mind: the important thing is to keep learning (and learning how to search archives is not that difficult!) S. > > Thanks, > Luke > > eyvaz_isaev at yahoo.com wrote: >> Hi, >> >> >>> Therefore, I'm currently staying with automatic >>> generation. >>> >> This one you have to use for SCF-calculations. >> >> >>> To do band structure plots, is it necessary to >>> >> choose > the same number of k points for both the scf >> and >> >>> nscf calculations, or can less be used for the scf? >>> >> You can use the same set of k-points (you should >> extract them from scf_output_file) them manually , but >> the result would be very ugly. You will not be able >> to plot because k-points are not aligned along high >> symmetry directions. >> >> So, the answer is "No". You have to generate k-points >> along a given directions and then calculate band >> energies for these k-points >> >> >>> If I try to enter a list of k points manually so >>> that I can choose my >>> path, how do you determine the weights? >>> >> Put them as 1.0, they are not used for band structure >> calculations. >> >> Actually, I am going to redraw a small set of programs >> to get band structure (using Gnuplot) for cubic and >> hexagonal lattices, at least. Hopefully it would be >> useful for beginners. >> >> Bests, >> Eyvaz. >> >> >> >> >> __________________________________________________ >> Do You Yahoo!? >> Tired of spam? Yahoo! Mail has the best spam protection around >> http://mail.yahoo.com >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060614/03ad3fec/attachment.htm From baroni at sissa.it Wed Jun 14 08:35:39 2006 From: baroni at sissa.it (Stefano Baroni) Date: Wed, 14 Jun 2006 08:35:39 +0200 Subject: [Pw_forum] ads et al In-Reply-To: <448F5909.7090502@mit.edu> References: <200606091728.45389.giannozz@nest.sns.it> <200606120943.18274.giannozz@nest.sns.it> <448F5909.7090502@mit.edu> Message-ID: I would not object, but I kind of remember that somebody did, when I made a similar proposal in the past. I think that the concern was that somebode could abuse of this possibility. I would not mind giving a try ... S. On Jun 14, 2006, at 2:32 AM, Nicola Marzari wrote: > > > Dear All, > > Is it appropriate to post on this list news that are not related > to PWSCF/CP, but are relevant to the community ? > > My suggestion would be that it could be useful to post notices of > electronic-structure schools, especially if Espresso related, > and open positions, again if Espresso related. > > In order to test this theory, I attach a few notes below. > > nicola > > > ---- > Bangalore Summer School on "Electronic Structure Methods and their > Applications" in conjunction with Conference on Computational > Materials > Theory: http://www.jncasr.ac.in/smr1783/ > ---- > 2006 Summer School on Computational Materials Science > Ab Initio Molecular Dynamics Simulation Methods in Chemistry: > http://www.mcc.uiuc.edu/summerschool/2006/ (Please note that > the Espresso program has not been uploaded yet, but lecturers > will include Stefano Baroni and Roberto Car) > ---- > Postdoc position: > Defect evolution and deformation in metastable alloys > > Applications are invited for one postdoctoral position in the MIT > Department of Materials Science and Engineering. This researcher will > work with Profs. Sidney Yip, Nicola Marzari, and Krystyn Van Vliet as > part of an interdisciplinary, international team that includes strong > collaboration with industry partners. The focus of the project is > multiscale computational modeling and simulation of metastable > metallic > alloys under extreme thermal and mechanical environments. The > objective > is to identify mechanisms of crystalline defect evolution and > consequences toward mechanical deformation. Funding of > this position is for an initial period of 2 years. Demonstrated > experience in scientific programming and the application of ab > initio / > DFT methods for solid state systems is mandatory; experience with > kinetic Monte Carlo and molecular/dislocation dynamics is > preferred. The > position is open immediately with a mutually agreed starting date. > Requests for further information and applications (CV, two > recommenders' > contact information, and one representative publication) should be > sent > to Prof. Krystyn J. Van Vliet, Department of Materials Science and > Engineering, MIT; E-mail:krystyn at mit.edu > > > --------------------------------------------------------------------- > Prof Nicola Marzari Department of Materials Science and Engineering > 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA > tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060614/f27734f9/attachment.htm From cyrille.barreteau at cea.fr Wed Jun 14 08:50:47 2006 From: cyrille.barreteau at cea.fr (Cyrille Barreteau) Date: Wed, 14 Jun 2006 08:50:47 +0200 Subject: [Pw_forum] k point selection In-Reply-To: <087E9D61-9B58-4985-9019-3C3A9907EE1D@sissa.it> References: <20060613162650.78416.qmail@web60317.mail.yahoo.com> <448F5895.6050105@netscape.net> <087E9D61-9B58-4985-9019-3C3A9907EE1D@sissa.it> Message-ID: <448FB1C7.8060005@cea.fr> >> >> >> Thank you very much for the instruction on k point sampling, I'll >> look into getting a copy of the Monkhorst Pack paper. One more quick >> issue: To save on computation time, does the automatic k point >> sampling for the scf calculation determine by itself which points >> comprise the irriducible brillouin zone? > > > try by yourself. run a calculation for, say, an fcc structure and see > whether or not you find two k points which are equivalent by any bubic > symmetry operation. (any operation of the cubic group can be seen as > the product of a parmutation of the cartesian coordinates - there are > seix such permutations - and the change of sign of any of the three > coordinates - there are 2^3=8 such operations - for a total of 48=6x8 > symmetry operations) > The Monkhort Pack procedure generates a set of k points in the First Brillouin zone (in fact they are often outside the FBZ but can be brought back into it by translations of reciprocal vectors). The code then determines by itself (thanks to the good work of the programmers:-) the point that are inside the IBZ. Again they are not truly inside the IBZ (try to plot them in simple cases) but can be brought back inside the IBZ by a succession of symmetries (and translations of reciprocal vectors). The code also determines automatically the weight of each k-points according to their symmetries. cyrille From baroni at sissa.it Wed Jun 14 08:56:55 2006 From: baroni at sissa.it (Stefano Baroni) Date: Wed, 14 Jun 2006 08:56:55 +0200 Subject: [Pw_forum] how to find out the stable structure under high pressure In-Reply-To: <448F7C13.1030300@mit.edu> References: <20060614022742.6488.qmail@web15605.mail.cnb.yahoo.com> <448F7C13.1030300@mit.edu> Message-ID: Nicola: here, I beg to differ. For two reasons. First: even if there was a minimum such as discrebed in the question, this would be a _local_ and not necessarily a global one. Second: there would be such a minimum if there was not any coupling between the soft modes and any other (phonon or elastic) modes. In practice, this is seldom the case, so that the low-simmetry phase seldom results from the distortion of soft modes alone. I would advise Li Yan to browse into a book dealing with the Landau theory of phase transitions, where these questions should be thoroughly discussed. Unfortunately, the only title that occurs to me right now is in French ("Sym?tries bris?es", by Boccara and Boccara - an English translation may exist). In any case, these texts are usually rather awkward and, when I needed to work with these concepts, I preferred to "rediscover" what I needed from scratch. My own efforts in mastering the subject are witnessed in the papers: Phys. Rev. Lett., 69, 1069 (1992) and Phys. Rev. B 51, 8060 (1995). What is true (and what Nicola most probably actually meant) is that if you start an energy minimization from a configuration characterized by a non- vanishing amplitude of the soft mode-distortion, then the system will certainly find a configuration of lower energy. This configuration may or may not be a local minimum (let alone a global one that you will never know for sure) according to whether your supercell and the symmetry of the starting configuration are such as to allow for all the relevant mode-mode couplings eventually responsible for the stability of the low-symmetry phase. Hope to have been not too confusing. Stefano On Jun 14, 2006, at 5:01 AM, Nicola Marzari wrote: > > Yes, > > nicola > > > li yan wrote: >> Dear all, >> I found soft modes in a system under pressure. Is it true that >> there must be a global energy miniumu in the subspace spanned by the >> eignevectors of the soft modes? And by searching this munimum, can i >> find the stable high-pressure strucure? >> >> >> best regards >> >> __________________________________________________ >> ??????????????? >> http://cn.mail.yahoo.com >> > > > > > -- > --------------------------------------------------------------------- > Prof Nicola Marzari Department of Materials Science and Engineering > 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA > tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060614/b95d2da7/attachment.htm From akohlmey at vitae.cmm.upenn.edu Wed Jun 14 12:44:42 2006 From: akohlmey at vitae.cmm.upenn.edu (Axel Kohlmeyer) Date: Wed, 14 Jun 2006 06:44:42 -0400 (EDT) Subject: asking smart(er) questions. was: Re: [Pw_forum] k point selection In-Reply-To: <448F5895.6050105@netscape.net> Message-ID: On Tue, 13 Jun 2006, Luke Thulin wrote: hi luke and everybody else on the list, i hope you're all having a nice day. i'd like to comment a little on the following sentence in a more general way, so please people who are not exactly interested in the detail of this discussion, please read on anyways. LT> different k paths? Sorry if this question has been asked, but I seem to LT> be incompentent when it comes to searching the archives. well, it is not _that_ difficult. if having to find and figure out the integrated search engine is too much of a hassle to you, you might try the following trick by (ab)using a more familiar search enging like google. as the mailing list archives are hosted on the democritos.it web server, you can search google, but restrict the search to only this domain. e.g. in your case i would try: "k points" selection site:democritos.it one more thing. as you may have noticed, the people here in this forum are very open to newcomers and trivial questions (something i personally respect a lot, since it is not always easy to answer in a calm and unoffensive way to some of these questions). however, you will see after a while the responsiveness will slow down if you do not demonstrate some significant effort on your side. which is understandable, since nobody here gets paid to respond to questions so the currency to 'pay' people here is mostly respect for their effort, and one way to pay respect to people is to try asking smarter questions. as stefano already pointed out, you are not expected to become an expert, but you have to demonstrate that you at least made an effort. claiming that you are too stupid, inept or incapable to even try, usually reduces people's attention span even more. for a more detailed discussion i recommend looking at: http://www.catb.org/~esr/faqs/smart-questions.html which demonstrates the IMHO essential point that the 'quality' of the response you get depends on the 'quality' of your question. thanks to everybody that made it this far. best regards, axel. LT> LT> Thanks, LT> Luke LT> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From lukethulin at netscape.net Wed Jun 14 14:19:58 2006 From: lukethulin at netscape.net (Luke Thulin) Date: Wed, 14 Jun 2006 08:19:58 -0400 Subject: asking smart(er) questions. was: Re: [Pw_forum] k point selection In-Reply-To: References: Message-ID: <448FFEEE.3040703@netscape.net> Axel, I appreciate very much the time people take to respond to questions, and I'm very sorry if it appears as if I'm not trying hard enough. The issue is simply that I have asked questions in the past and have been told that they have been asked and answered in the archives. My comment was just to try and let people know that I do take time to search the archives and somehow overlooked those relevant discussions. Sorry for the trouble, Luke akohlmey at vitae.cmm.upenn.edu wrote: >On Tue, 13 Jun 2006, Luke Thulin wrote: > >hi luke and everybody else on the list, > >i hope you're all having a nice day. >i'd like to comment a little on the following sentence >in a more general way, so please people who are not >exactly interested in the detail of this discussion, >please read on anyways. > >LT> different k paths? Sorry if this question has been asked, but I seem to >LT> be incompentent when it comes to searching the archives. > >well, it is not _that_ difficult. if having to find and figure out >the integrated search engine is too much of a hassle to you, you >might try the following trick by (ab)using a more familiar search >enging like google. as the mailing list archives are hosted on >the democritos.it web server, you can search google, but restrict >the search to only this domain. e.g. in your case i would try: > >"k points" selection site:democritos.it > >one more thing. as you may have noticed, the people here in this >forum are very open to newcomers and trivial questions (something >i personally respect a lot, since it is not always easy to answer >in a calm and unoffensive way to some of these questions). >however, you will see after a while the responsiveness will slow >down if you do not demonstrate some significant effort on your >side. which is understandable, since nobody here gets paid to >respond to questions so the currency to 'pay' people here is >mostly respect for their effort, and one way to pay respect to >people is to try asking smarter questions. as stefano already >pointed out, you are not expected to become an expert, but you >have to demonstrate that you at least made an effort. claiming >that you are too stupid, inept or incapable to even try, usually >reduces people's attention span even more. > >for a more detailed discussion i recommend looking at: >http://www.catb.org/~esr/faqs/smart-questions.html >which demonstrates the IMHO essential point that the >'quality' of the response you get depends on the 'quality' >of your question. > >thanks to everybody that made it this far. > >best regards, > axel. > >LT> >LT> Thanks, >LT> Luke >LT> > > > From timtro at gmail.com Wed Jun 14 16:33:39 2006 From: timtro at gmail.com (Tim Teatro) Date: Wed, 14 Jun 2006 10:33:39 -0400 Subject: [Pw_forum] Dynamic Temperature Control and Select Stationary Ions. In-Reply-To: <20060612203859.97528.qmail@web30911.mail.mud.yahoo.com> References: <560bcca0606102304p43382687l40a905da69e87e3e@mail.gmail.com> <20060612203859.97528.qmail@web30911.mail.mud.yahoo.com> Message-ID: <560bcca0606140733va1484d7o1a2d2970cdaa757c@mail.gmail.com> Thank you so very much. I appreciate your help. I can't wait to try this autopilot. One thing though, I don't need to stop atoms dynamically. They can be static from the start of the simulation. What input do I specify to do this? Thanks again. On 6/12/06, Yonas Abraham wrote: > Dear Tim, > > In CP (QE >= 3), there is a way to change temperature > on the fly during the MD simulation. look at > Doc/AUTOPILOT file on how to use it. In fact, it will > let you change up to 10 CP (like dt, isave, iprint > etc) variables on the fly. in addition, you can even > PAUSE the simulation while you are deciding what value > to use! > > but the current AUTOPILOT implementation doesn't allow > you to fix certain atoms on the fly. But if you are > serious, you can extend AUTOPILOT to do so. > > > But I don't know if there is such thing on pw.x. > > /yonas > > --- Tim Teatro wrote: > > > Dear Nicola, > > > > Thank you for your reply. I am working on large > > clusters, so > > calculation time is not a problem. Unfortunately I > > will not have > > access to cp.x for a few days because the computer > > cluster on which I > > was able to compile it is down. Therefore, I will > > need to use pw.x for > > now. > > > > Could you give me a list of input parameters I would > > need for changing > > the temperature during a MD simulation, and for > > holding a few specific > > ions still while others are able to move? It seems I > > can only find > > references to temperature rescaling and the Nose > > thermostat, which > > require the temperature to be specified ahead of > > time. And of course I > > can turn the ion dynamics off easily enough, but not > > for certain > > atoms. > > > > If you have any examples, this would be great too. > > > > Thanks again. > > > > Cheers. > > > > On 6/10/06, Nicola Marzari wrote: > > > > > > > > > Dear Tim, > > > > > > > > > you can do this both with CP or PWSCF. CP would be > > the ideal code, > > > provided the system remains insulating. > > > On a single state-of-thje-art Pentium I think you > > can get nowadays > > > 1ps/day, for a system with ~250 electrons. > > > > > > Otherwise, you should use Born-Oppenheimer > > dynamics (that is the > > > standard mode for PWSCF, and is > > > available as an option in CP). > > > > > > Melting and quenching has been used often - you > > could look e.g. at the > > > works by Alfredo Pasquarello at EPFL (especially > > > from the ~1998 Nature on SiO2 onwards). > > > > > > Car-Parrinello MD, and to a certain extent > > Born-Oppenheimer MD, require > > > a good understanding of several concepts - > > > we'll run a school on "Ab-initio molecular > > dynamics" in > > > Urbana-Champaign, at the beginning of > > > August, and attending that would probably be very > > useful: > > > http://www.mcc.uiuc.edu/summerschool/2006/ > > > > > > At a minimum, have a close look at the review > > papers on the web - > > > especially the one > > > by Marx and Hutter > > > > > > (http://www.theochem.ruhr-uni-bochum.de/research/marx/cprev.en.html) > > or the > > > shorter one by Galli and Pasquarello. The > > RevModPhys review of Payne > > > from 1992 is another very > > > good reference. As a word of care, look also at > > the 2002 and 2004 JCP > > > from Tangney/Scandolo and > > > from the Galli group. > > > > > > > > > nicola > > > > > > > > > > > > > > > > > > > > > > > > Tim Teatro wrote: > > > > Hello. > > > > > > > > I believe I posted a similar question a year or > > two ago, but A) I > > > > don't think the question was resolved and B) the > > code has changed, so > > > > perhaps what was once a problem is not now. > > > > > > > > My group has a situation where we need to melt a > > crystal system, and > > > > then cool it back down to make an amorphous > > structure. Is this > > > > possible using any of the codes in the ESPRESSO > > package? > > > > > > > > If so, how? Do you happen to have an input > > example? > > > > > > > > Also, we have another system in which it would > > be useful to hold > > > > certain ions stationary while others are free to > > move under the Verlet > > > > algorithm. Of course we could cheat and set the > > mass of these atoms to > > > > be extremely high, but I think it would be > > better to simply stop them > > > > completely. > > > > > > > > Thank you for your time. > > > > > > > > -- > > > > Timothy A.V. Teatro > > > > University of Ontario Institute of Technology > > > > e. timtro at gmail.com > > > > p. 905.579.9248 > > > > c. 905.442.9248 > > > > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > > > > -- > > Timothy A.V. Teatro > > University of Ontario Institute of Technology > > e. timtro at gmail.com > > p. 905.579.9248 > > c. 905.442.9248 > > _______________________________________________ > > 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 > -- Timothy A.V. Teatro University of Ontario Institute of Technology e. timtro at gmail.com p. 905.579.9248 c. 905.442.9248 From giannozz at nest.sns.it Wed Jun 14 16:54:20 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Wed, 14 Jun 2006 16:54:20 +0200 Subject: [Pw_forum] Dynamic Temperature Control and Select Stationary Ions. In-Reply-To: <560bcca0606140733va1484d7o1a2d2970cdaa757c@mail.gmail.com> References: <560bcca0606102304p43382687l40a905da69e87e3e@mail.gmail.com> <20060612203859.97528.qmail@web30911.mail.mud.yahoo.com> <560bcca0606140733va1484d7o1a2d2970cdaa757c@mail.gmail.com> Message-ID: <200606141654.20972.giannozz@nest.sns.it> On Wednesday 14 June 2006 16:33, Tim Teatro wrote: > I can't wait to try this autopilot. don't try it with pw.x: it doesn't work at present > One thing though, I don't need to stop atoms dynamically. > They can be static from the start of the simulation. > What input do I specify to do this? see variables "if_pos" in cards ATOMIC_POSITIONS Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From wlyim at puccini.che.pitt.edu Wed Jun 14 17:39:49 2006 From: wlyim at puccini.che.pitt.edu (wlyim at puccini.che.pitt.edu) Date: Wed, 14 Jun 2006 11:39:49 -0400 (EDT) Subject: [Pw_forum] frc_blk question Message-ID: Dear all, I tried to calculate the phonon structures of surfaces in (1x1) and (2x2) supercells. However, I encountered a problem of violating the Hermitian of dynamical matrix on the (2x2) model, while the (1x1) model had no problem. A warning from matdyn.x read like: Max |d(i,j)-d*(j,i)| = 0.008889 Max |d(i,j)-d*(j,i)|/|d(i,j)|: 10.7732% When I comment the lines calling "frc_blk" in "setupmat" subroutine, the warning disappeared. ! call frc_blk (nax_blk,dyn_blk,qp,tau_blk,nat_blk, & ! & nr1,nr2,nr3,nrx,frc,at_blk,bg_blk,rws,nrws) Can anyone help me to clarify what "frc_blk" routine doing? Many thanks in advance. Best regards, William From wlyim at puccini.che.pitt.edu Wed Jun 14 17:47:13 2006 From: wlyim at puccini.che.pitt.edu (wlyim at puccini.che.pitt.edu) Date: Wed, 14 Jun 2006 11:47:13 -0400 (EDT) Subject: [Pw_forum] frc_blk question In-Reply-To: Message-ID: Dear all, I am sorry that I shouldn't comment the frc_blk line since it would do nothing on the real space force constant. I re-define my question: what's the main reason for violating the hermitian conditions? Many thanks in advance. Best regards, William On Wed, 14 Jun 2006 wlyim at puccini.che.pitt.edu wrote: > Dear all, > > I tried to calculate the phonon structures of surfaces in (1x1) and (2x2) > supercells. However, I encountered a problem of violating the Hermitian of > dynamical matrix on the (2x2) model, while the (1x1) model had no problem. > > A warning from matdyn.x read like: > Max |d(i,j)-d*(j,i)| = 0.008889 > Max |d(i,j)-d*(j,i)|/|d(i,j)|: 10.7732% > > When I comment the lines calling "frc_blk" in "setupmat" subroutine, the > warning disappeared. > ! call frc_blk (nax_blk,dyn_blk,qp,tau_blk,nat_blk, & > ! & nr1,nr2,nr3,nrx,frc,at_blk,bg_blk,rws,nrws) > > Can anyone help me to clarify what "frc_blk" routine doing? > > Many thanks in advance. > > Best regards, > William > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Wai-Leung Yim, PhD Universtaet Oldenburg Institut fuer Reine und Angewandte Chemie AG Theoretische Chemie Carl-von-Ossiezky-Str. 9-11 26129 Oldenburg Germany Email: wlyim at puccini.che.pitt.edu Phone: +49-441-798-3950 (office) +49-441-798-5102 (home) Fax: +49-441-798-3964 From giannozz at nest.sns.it Wed Jun 14 17:54:11 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Wed, 14 Jun 2006 17:54:11 +0200 Subject: [Pw_forum] frc_blk question In-Reply-To: References: Message-ID: <200606141754.11785.giannozz@nest.sns.it> On Wednesday 14 June 2006 17:47, wlyim at puccini.che.pitt.edu wrote: > I re-define my question: what's the main reason for violating the > hermitian conditions? insufficient convergence of either the unperturbed (scf) or the linear-response (phonon) calculation, I guess Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From wlyim at puccini.che.pitt.edu Wed Jun 14 18:00:38 2006 From: wlyim at puccini.che.pitt.edu (wlyim at puccini.che.pitt.edu) Date: Wed, 14 Jun 2006 12:00:38 -0400 (EDT) Subject: [Pw_forum] frc_blk question In-Reply-To: <200606141754.11785.giannozz@nest.sns.it> Message-ID: Dear Paolo, How can I know whether the convergence is sufficient or not? I used "conv_thr = 1.0d-11" for both scf and nscf calculations, and "tr2_ph=1.0d-14" for phonon calculation. I thought the thresholds were quite tight. Best regards, William On Wed, 14 Jun 2006, Paolo Giannozzi wrote: > On Wednesday 14 June 2006 17:47, wlyim at puccini.che.pitt.edu wrote: > > > I re-define my question: what's the main reason for violating the > > hermitian conditions? > > insufficient convergence of either the unperturbed (scf) or the > linear-response (phonon) calculation, I guess > > Paolo > From ssaarti at jncasr.ac.in Thu Jun 15 07:59:56 2006 From: ssaarti at jncasr.ac.in (aarti srirangarajan) Date: Thu, 15 Jun 2006 11:29:56 +0530 (IST) Subject: [Pw_forum] about total magnetization convergence Message-ID: <44044.202.41.111.151.1150351196.squirrel@202.41.111.151> Dear all, I am doing a spin polarised slab calculation with vacuum layers for Ni. I have a problem here with the total magnetization values. I hv performed calculations with 7 7 1 , 9 9 1 and 11 11 1 and 13 13 1 k-point meshes for various surfaces and the total magnetisation changes very erratically. The values are not converging. Although the Final energies are converging very well. Also the difference b/w the Total and abs. magnetisation is of the order of 2 bohrs. Could anyone help me with this? thank you Aarti From Steven.Kirk at hv.se Thu Jun 15 16:19:14 2006 From: Steven.Kirk at hv.se (Steven Kirk) Date: Thu, 15 Jun 2006 16:19:14 +0200 Subject: [Pw_forum] Diagnoses for two pw.x error messages Message-ID: <44916C62.3080203@hv.se> Hello, I would be grateful if someone could diagnose the following error messages. Both arise when running the very first calculation (al.cg.band.in) in Example 01 from the test set supplied with Espresso 3.1. In both cases, the system running the program is an Opteron (AMD64) cluster, and the executable (pw.x) has been linked with the ACML and internal FFTW libraries, and using PathScale C,C++,F90 and F77 compilers. Case 1: pw.x compiled as a serial application - output is Program PWSCF v.3.1 starts ... Today is 15Jun2006 at 8:23:20 Ultrasoft (Vanderbilt) Pseudopotentials Current dimensions of program pwscf are: ntypx = 10 npk = 40000 lmax = 3 nchix = 6 ndmx = 2000 nbrx = 14 nqfx = 8 bravais-lattice index = 2 lattice parameter (a_0) = 7.5000 a.u. unit-cell volume = 105.4688 (a.u.)^3 number of atoms/cell = 1 number of atomic types = 1 kinetic-energy cutoff = 15.0000 Ry charge density cutoff = 60.0000 Ry convergence threshold = 1.0E-06 beta = 0.7000 number of iterations used = 8 plain mixing Exchange-correlation = SLA PZ NOGX NOGC (1100) celldm(1)= 7.500000 celldm(2)= 0.000000 celldm(3)= 0.000000 celldm(4)= 0.000000 celldm(5)= 0.000000 celldm(6)= 0.000000 crystal axes: (cart. coord. in units of a_0) a(1) = ( -0.500000 0.000000 0.500000 ) a(2) = ( 0.000000 0.500000 0.500000 ) a(3) = ( -0.500000 0.500000 0.000000 ) reciprocal axes: (cart. coord. in units 2 pi/a_0) b(1) = ( -1.000000 -1.000000 1.000000 ) b(2) = ( 1.000000 1.000000 1.000000 ) b(3) = ( -1.000000 1.000000 -1.000000 ) PSEUDO 1 is Al zval = 3.0 lmax= 1 lloc= 0 (in numerical form: 171 grid points, xmin = 0.00, dx = 0.0000) atomic species valence mass pseudopotential Al 3.00 26.98000 Al( 1.00) 48 Sym.Ops. (with inversion) Cartesian axes site n. atom positions (a_0 units) 1 Al tau( 1) = ( 0.0000000 0.0000000 0.0000000 ) number of k points= 28 cart. coord. in units 2pi/a_0 k( 1) = ( 0.0000000 0.0000000 0.0000000), wk = 0.0714286 k( 2) = ( 0.0000000 0.0000000 0.1000000), wk = 0.0714286 k( 3) = ( 0.0000000 0.0000000 0.2000000), wk = 0.0714286 k( 4) = ( 0.0000000 0.0000000 0.3000000), wk = 0.0714286 k( 5) = ( 0.0000000 0.0000000 0.4000000), wk = 0.0714286 k( 6) = ( 0.0000000 0.0000000 0.5000000), wk = 0.0714286 k( 7) = ( 0.0000000 0.0000000 0.6000000), wk = 0.0714286 k( 8) = ( 0.0000000 0.0000000 0.7000000), wk = 0.0714286 k( 9) = ( 0.0000000 0.0000000 0.8000000), wk = 0.0714286 k( 10) = ( 0.0000000 0.0000000 0.9000000), wk = 0.0714286 k( 11) = ( 0.0000000 0.0000000 1.0000000), wk = 0.0714286 k( 12) = ( 0.0000000 0.0000000 0.0000000), wk = 0.0714286 k( 13) = ( 0.0000000 0.1000000 0.1000000), wk = 0.0714286 k( 14) = ( 0.0000000 0.2000000 0.2000000), wk = 0.0714286 k( 15) = ( 0.0000000 0.3000000 0.3000000), wk = 0.0714286 k( 16) = ( 0.0000000 0.4000000 0.4000000), wk = 0.0714286 k( 17) = ( 0.0000000 0.5000000 0.5000000), wk = 0.0714286 k( 18) = ( 0.0000000 0.6000000 0.6000000), wk = 0.0714286 k( 19) = ( 0.0000000 0.7000000 0.7000000), wk = 0.0714286 k( 20) = ( 0.0000000 0.8000000 0.8000000), wk = 0.0714286 k( 21) = ( 0.0000000 0.9000000 0.9000000), wk = 0.0714286 k( 22) = ( 0.0000000 1.0000000 1.0000000), wk = 0.0714286 k( 23) = ( 0.0000000 0.0000000 0.0000000), wk = 0.0714286 k( 24) = ( 0.1000000 0.1000000 0.1000000), wk = 0.0714286 k( 25) = ( 0.2000000 0.2000000 0.2000000), wk = 0.0714286 k( 26) = ( 0.3000000 0.3000000 0.3000000), wk = 0.0714286 k( 27) = ( 0.4000000 0.4000000 0.4000000), wk = 0.0714286 k( 28) = ( 0.5000000 0.5000000 0.5000000), wk = 0.0714286 G cutoff = 85.4897 ( 869 G-vectors) FFT grid: ( 15, 15, 15) nbndx = 8 nbnd = 8 natomwfc = 9 npwx = 113 nelec = 3.00 nkb = 4 ngl = 31 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from cfft3d : error # 1 no scalar fft driver specified %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... Case 2: pw.x executable compiled and linked with installed HP MPI libraries. The run does not get as far this time, but fails with: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% task # 0 from data_structure : error # 1 number of sticks 0 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% A previous post on the mailing list leads me to think these two crashes might be related, somewhere in the FFTs. Can anyone help diagnose the problem(s) ? Many thanks in advance, Steve Kirk -- Dr. Steven R. Kirk Dept. of Technology, Mathematics & Computer Science (P)+46 520 223215 University West (F)+46 520 223299 P.O. Box 957 Trollhattan 461 29 SWEDEN http://taconet.webhop.org From wlyim at puccini.che.pitt.edu Thu Jun 15 16:20:45 2006 From: wlyim at puccini.che.pitt.edu (wlyim at puccini.che.pitt.edu) Date: Thu, 15 Jun 2006 10:20:45 -0400 (EDT) Subject: [Pw_forum] about total magnetization convergence In-Reply-To: <44044.202.41.111.151.1150351196.squirrel@202.41.111.151> Message-ID: I guess your smearing parameter is too large. Try a smaller degauss value. On Thu, 15 Jun 2006, aarti srirangarajan wrote: > Dear all, > > I am doing a spin polarised slab calculation with vacuum layers for Ni. > I have a problem here with the total magnetization values. I hv performed > calculations with 7 7 1 , 9 9 1 and 11 11 1 and 13 13 1 k-point meshes for > various surfaces and > the total magnetisation changes very erratically. The values are not > converging. Although the Final energies are converging very well. > Also the difference b/w the Total and abs. magnetisation is of the order > of 2 bohrs. Could anyone help me with this? > > thank you > Aarti > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From giannozz at nest.sns.it Thu Jun 15 17:30:05 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 15 Jun 2006 17:30:05 +0200 Subject: [Pw_forum] Diagnoses for two pw.x error messages In-Reply-To: <44916C62.3080203@hv.se> References: <44916C62.3080203@hv.se> Message-ID: <200606151730.05838.giannozz@nest.sns.it> On Thursday 15 June 2006 16:19, Steven Kirk wrote: > from cfft3d : error # 1 > no scalar fft driver specified buggy preprocessor (ifort 9?), or missing preprocessing options: there should be one among those appearing in Modules/fft_scalar.f90 > from data_structure : error # 1 > number of sticks 0 too many processors for the size of the system Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From reboredofa at ornl.gov Thu Jun 15 17:43:17 2006 From: reboredofa at ornl.gov (Fernando A Reboredo) Date: Thu, 15 Jun 2006 11:43:17 -0400 Subject: [Pw_forum] how to find out the stable structure under high pressure References: <20060614022742.6488.qmail@web15605.mail.cnb.yahoo.com> <448F7C13.1030300@mit.edu> Message-ID: <0a5801c69092$6c0e76b0$b7305ba0@ornl.gov> I agree with Stefano: While it is true that 1) in matematical terms ALL the phonon modes of a crystal are a compete basis where any posible atomic displacement can be decomposed, and also 2) any new structure with the same number of atoms can be though as a displacement from the old one (it could be big), the set of soft modes are incomple basis and thus they can only describe a subset of arbitrary structure transitions. Fernando ----- Original Message ----- From: Stefano Baroni To: pw_forum at pwscf.org Sent: Wednesday, June 14, 2006 2:56 AM Subject: Re: [Pw_forum] how to find out the stable structure under high pressure Nicola: here, I beg to differ. For two reasons. First: even if there was a minimum such as discrebed in the question, this would be a _local_ and not necessarily a global one. Second: there would be such a minimum if there was not any coupling between the soft modes and any other (phonon or elastic) modes. In practice, this is seldom the case, so that the low-simmetry phase seldom results from the distortion of soft modes alone. I would advise Li Yan to browse into a book dealing with the Landau theory of phase transitions, where these questions should be thoroughly discussed. Unfortunately, the only title that occurs to me right now is in French ("Sym?tries bris?es", by Boccara and Boccara - an English translation may exist). In any case, these texts are usually rather awkward and, when I needed to work with these concepts, I preferred to "rediscover" what I needed from scratch. My own efforts in mastering the subject are witnessed in the papers: Phys. Rev. Lett., 69, 1069 (1992) and Phys. Rev. B 51, 8060 (1995). What is true (and what Nicola most probably actually meant) is that if you start an energy minimization from a configuration characterized by a non-vanishing amplitude of the soft mode-distortion, then the system will certainly find a configuration of lower energy. This configuration may or may not be a local minimum (let alone a global one that you will never know for sure) according to whether your supercell and the symmetry of the starting configuration are such as to allow for all the relevant mode-mode couplings eventually responsible for the stability of the low-symmetry phase. Hope to have been not too confusing. Stefano On Jun 14, 2006, at 5:01 AM, Nicola Marzari wrote: Yes, nicola li yan wrote: Dear all, I found soft modes in a system under pressure. Is it true that there must be a global energy miniumu in the subspace spanned by the eignevectors of the soft modes? And by searching this munimum, can i find the stable high-pressure strucure? best regards __________________________________________________ ??????????????? http://cn.mail.yahoo.com -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060615/9deca591/attachment.htm From marzari at MIT.EDU Thu Jun 15 20:05:08 2006 From: marzari at MIT.EDU (Nicola Marzari) Date: Thu, 15 Jun 2006 14:05:08 -0400 Subject: [Pw_forum] about total magnetization convergence In-Reply-To: <44044.202.41.111.151.1150351196.squirrel@202.41.111.151> References: <44044.202.41.111.151.1150351196.squirrel@202.41.111.151> Message-ID: <4491A154.3080706@mit.edu> One possibility is that you are converging to different self-consistent states - in a spin-polarized calculation this is fairly common - and the energy differences between different magnetizations could be small. So, what you are seeing might not be an erratic behavior, but the fact that there are two or more potential energy surfaces very close to each other, and corresponding to different magnetizations. As always, check carefully k-point sampling and the amount of degauss used (around 0.01-0.02 Ry, for this case). Note that you can fix the tot amgnetization in the code, if needed. nicola aarti srirangarajan wrote: > Dear all, > > I am doing a spin polarised slab calculation with vacuum layers for Ni. > I have a problem here with the total magnetization values. I hv performed > calculations with 7 7 1 , 9 9 1 and 11 11 1 and 13 13 1 k-point meshes for > various surfaces and > the total magnetisation changes very erratically. The values are not > converging. Although the Final energies are converging very well. > Also the difference b/w the Total and abs. magnetisation is of the order > of 2 bohrs. Could anyone help me with this? > > thank you > Aarti > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From cyrille.barreteau at cea.fr Thu Jun 15 20:20:11 2006 From: cyrille.barreteau at cea.fr (Cyrille Barreteau) Date: Thu, 15 Jun 2006 20:20:11 +0200 (CEST) Subject: [Pw_forum] about total magnetization convergence In-Reply-To: <4491A154.3080706@mit.edu> References: <44044.202.41.111.151.1150351196.squirrel@202.41.111.151> <4491A154.3080706@mit.edu> Message-ID: <2258.81.56.70.96.1150395611.squirrel@dsm-mail.saclay.cea.fr> I fully agree with Nicola. It is very common in magnetic materials to have several metastable solutions. For example in fcc iron you can easily find situations where you calculation converges towards different solutions depending on very small changes of initial conditions (or mixing etc..). I am not so familiar with Ni but maybe there is a similar behavior. As suggested by Nicola it could be a good idea to do some "fixed spin moment" calculation for several values of the moment and find the minima in the E(M) curve. cyrille Le Jeu 15 juin 2006 20:05, Nicola Marzari a ?crit : > > > One possibility is that you are converging to different self-consistent > states - in a spin-polarized calculation this is fairly common - and the > energy differences between different magnetizations could be small. > > So, what you are seeing might not be an erratic behavior, but the fact > that there are two or more potential energy surfaces very close to each > other, and corresponding to different magnetizations. > > As always, check carefully k-point sampling and the amount of degauss > used (around 0.01-0.02 Ry, for this case). Note that you can fix the tot > amgnetization in the code, if needed. > > nicola > > > aarti srirangarajan wrote: >> Dear all, >> >> I am doing a spin polarised slab calculation with vacuum layers for Ni. >> I have a problem here with the total magnetization values. I hv >> performed >> calculations with 7 7 1 , 9 9 1 and 11 11 1 and 13 13 1 k-point meshes >> for >> various surfaces and >> the total magnetisation changes very erratically. The values are not >> converging. Although the Final energies are converging very well. >> Also the difference b/w the Total and abs. magnetisation is of the >> order >> of 2 bohrs. Could anyone help me with this? >> >> thank you >> Aarti >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > > > -- > --------------------------------------------------------------------- > Prof Nicola Marzari Department of Materials Science and Engineering > 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA > tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From stargmoon at yahoo.com Thu Jun 15 20:22:39 2006 From: stargmoon at yahoo.com (stargmoon) Date: Thu, 15 Jun 2006 11:22:39 -0700 (PDT) Subject: [Pw_forum] phonon dispersion In-Reply-To: Message-ID: <20060615182239.43374.qmail@web33201.mail.mud.yahoo.com> Dear pwscf community, This is the first time I try to do phonon dispersion calculation. Could anybody please sent me an example input for phonon dispersion calculation or point me to somewhere I can find it? I just found the input for one k-point phonon mode calculation. You help will be highly appreciated! Thanks in advance! Stargmoon __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060615/d6a450fa/attachment.htm From giannozz at nest.sns.it Thu Jun 15 20:37:17 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 15 Jun 2006 20:37:17 +0200 Subject: [Pw_forum] phonon dispersion In-Reply-To: <20060615182239.43374.qmail@web33201.mail.mud.yahoo.com> References: <20060615182239.43374.qmail@web33201.mail.mud.yahoo.com> Message-ID: <200606152037.17843.giannozz@nest.sns.it> On Thursday 15 June 2006 20:22, stargmoon wrote: > Could anybody please sent me an example input for phonon > dispersion calculation or point me to somewhere I can find it? > I just found the input for one k-point phonon mode calculation. had you looked more carefully, you would have found an example of phonon dispersion calculation in example06 Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From stargmoon at yahoo.com Thu Jun 15 22:01:11 2006 From: stargmoon at yahoo.com (stargmoon) Date: Thu, 15 Jun 2006 13:01:11 -0700 (PDT) Subject: [Pw_forum] phonon dispersion In-Reply-To: <200606152037.17843.giannozz@nest.sns.it> Message-ID: <20060615200111.32595.qmail@web33209.mail.mud.yahoo.com> Dr. Giannozzi, Thanks for your quick reply. I got it! Best regards, Paolo Giannozzi wrote: On Thursday 15 June 2006 20:22, stargmoon wrote: > Could anybody please sent me an example input for phonon > dispersion calculation or point me to somewhere I can find it? > I just found the input for one k-point phonon mode calculation. had you looked more carefully, you would have found an example of phonon dispersion calculation in example06 Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060615/f9f52cfd/attachment.htm From lukethulin at netscape.net Fri Jun 16 00:53:41 2006 From: lukethulin at netscape.net (Luke Thulin) Date: Thu, 15 Jun 2006 18:53:41 -0400 Subject: [Pw_forum] Example 5 High Symmetry Points Message-ID: <4491E4F5.2050202@netscape.net> Hello everyone, I am having some difficulty entering the high symmetry paths for the nscf calculation for my problem. So I'm trying to understand how example 5 was done and cannot. I'm using the source http://www.cryst.ehu.es/cgi-bin/cryst/programs/nph-kv-list?gnum=227 that lists the high symmetry points for the diamond structure. According to it, the CDML k vector labels are: gamma (0,0,0) X (1/2,0,1/2) L (1/2,1/2,1/2) W (1/2,1/4,3/4) Does this mean that the points are located at (n1 a*, n2 b*, n3 c*) where a*, b*, and c* are reciprocal lattice vectors and n1, n2, n3 are the fractional numbers given (1/2, 0, 1/2 for X, etc)? If so, I then multiply those out to represent in 2pi/a coordinates and get: gamma (0,0,0) X (0,1,0) L (1/2,1/2,1/2) W(1/2,1,0) However, the example 5 band structure looks exactly correct, but the k path chosen is (1/2,1/2,1/2) to (0,0,0) to (0,0,1) to (0,1,1) to (0,0,0). So my calculated high symmetry points (at least X and W) don't seem to be correct. Can someone please explain? Thanks, Luke From marzari at MIT.EDU Fri Jun 16 01:14:41 2006 From: marzari at MIT.EDU (Nicola Marzari) Date: Thu, 15 Jun 2006 19:14:41 -0400 Subject: [Pw_forum] Example 5 High Symmetry Points In-Reply-To: <4491E4F5.2050202@netscape.net> References: <4491E4F5.2050202@netscape.net> Message-ID: <4491E9E1.2080701@mit.edu> > (1/2,1/2,1/2) to (0,0,0) to (0,0,1) to (0,1,1) to (0,0,0). S (0,0,1) and (0,1,1) are equivalent, and are both X. So you go from L to Gamma to X (taking a "short path"), then to a "different" X (in doing this you go through W), and finally you go back to Gamma through a "long path", that goes through K. Here, http://cst-www.nrl.navy.mil/bind/kpts/fcc/fcczone.gif the Gamma-K (first part of the "long path") and X-W lines cross in the equivalent X (lets call it XX) mentioned above (Gamma-X path shown is the short path, Gamma-K is the beginning of the long path that would end in XX) nicola --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From ssaarti at jncasr.ac.in Fri Jun 16 06:11:27 2006 From: ssaarti at jncasr.ac.in (aarti srirangarajan) Date: Fri, 16 Jun 2006 09:41:27 +0530 (IST) Subject: [Pw_forum] Re: Pw_forum digest, Vol 1 #1029 - 5 msgs In-Reply-To: <20060615154402.12613.28140.Mailman@democritos.sissa.it> References: <20060615154402.12613.28140.Mailman@democritos.sissa.it> Message-ID: <51834.202.41.111.151.1150431087.squirrel@202.41.111.151> Thanks for the reply. The degauss value I am working with is 0.003. What value should typically be used for Ni? regards aarti Send Pw_forum mailing list submissions to > pw_forum at pwscf.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://www.democritos.it/mailman/listinfo/pw_forum > or, via email, send a message with subject or body 'help' to > pw_forum-request at pwscf.org > > You can reach the person managing the list at > pw_forum-admin at pwscf.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Pw_forum digest..." > > > Today's Topics: > > 1. about total magnetization convergence (aarti srirangarajan) > 2. Diagnoses for two pw.x error messages (Steven Kirk) > 3. Re: about total magnetization convergence > (wlyim at puccini.che.pitt.edu) > 4. Re: Diagnoses for two pw.x error messages (Paolo Giannozzi) > 5. Re: how to find out the stable structure under high pressure > (Fernando A Reboredo) > > --__--__-- > > Message: 1 > From: aarti srirangarajan > Date: Thu, 15 Jun 2006 11:29:56 +0530 (IST) > To: pw_forum at pwscf.org > Subject: [Pw_forum] about total magnetization convergence > Reply-To: pw_forum at pwscf.org > > Dear all, > > I am doing a spin polarised slab calculation with vacuum layers for Ni. > I have a problem here with the total magnetization values. I hv performed > calculations with 7 7 1 , 9 9 1 and 11 11 1 and 13 13 1 k-point meshes for > various surfaces and > the total magnetisation changes very erratically. The values are not > converging. Although the Final energies are converging very well. > Also the difference b/w the Total and abs. magnetisation is of the order > of 2 bohrs. Could anyone help me with this? > > thank you > Aarti > > --__--__-- > > Message: 2 > Date: Thu, 15 Jun 2006 16:19:14 +0200 > From: Steven Kirk > To: pw_forum at pwscf.org > Subject: [Pw_forum] Diagnoses for two pw.x error messages > Reply-To: pw_forum at pwscf.org > > Hello, > > I would be grateful if someone could diagnose the following error > messages. Both arise when running the very first calculation > (al.cg.band.in) in Example 01 from the test set supplied with Espresso > 3.1. > > In both cases, the system running the program is an Opteron (AMD64) > cluster, and the executable (pw.x) has been linked with the ACML and > internal FFTW libraries, and using PathScale C,C++,F90 and F77 compilers. > > Case 1: pw.x compiled as a serial application - output is > Program PWSCF v.3.1 starts ... > Today is 15Jun2006 at 8:23:20 > > Ultrasoft (Vanderbilt) Pseudopotentials > > Current dimensions of program pwscf are: > > ntypx = 10 npk = 40000 lmax = 3 > nchix = 6 ndmx = 2000 nbrx = 14 nqfx = 8 > > > bravais-lattice index = 2 > lattice parameter (a_0) = 7.5000 a.u. > unit-cell volume = 105.4688 (a.u.)^3 > number of atoms/cell = 1 > number of atomic types = 1 > kinetic-energy cutoff = 15.0000 Ry > charge density cutoff = 60.0000 Ry > convergence threshold = 1.0E-06 > beta = 0.7000 > number of iterations used = 8 plain mixing > Exchange-correlation = SLA PZ NOGX NOGC (1100) > celldm(1)= 7.500000 celldm(2)= 0.000000 celldm(3)= 0.000000 > celldm(4)= 0.000000 celldm(5)= 0.000000 celldm(6)= 0.000000 > > crystal axes: (cart. coord. in units of a_0) > a(1) = ( -0.500000 0.000000 0.500000 ) > a(2) = ( 0.000000 0.500000 0.500000 ) > a(3) = ( -0.500000 0.500000 0.000000 ) > > reciprocal axes: (cart. coord. in units 2 pi/a_0) > b(1) = ( -1.000000 -1.000000 1.000000 ) > b(2) = ( 1.000000 1.000000 1.000000 ) > b(3) = ( -1.000000 1.000000 -1.000000 ) > > > PSEUDO 1 is Al zval = 3.0 lmax= 1 lloc= 0 > (in numerical form: 171 grid points, xmin = 0.00, dx = 0.0000) > > atomic species valence mass pseudopotential > Al 3.00 26.98000 Al( 1.00) > > 48 Sym.Ops. (with inversion) > > > Cartesian axes > > site n. atom positions (a_0 units) > 1 Al tau( 1) = ( 0.0000000 0.0000000 > 0.0000000 ) > > number of k points= 28 > cart. coord. in units 2pi/a_0 > k( 1) = ( 0.0000000 0.0000000 0.0000000), wk = > 0.0714286 > k( 2) = ( 0.0000000 0.0000000 0.1000000), wk = > 0.0714286 > k( 3) = ( 0.0000000 0.0000000 0.2000000), wk = > 0.0714286 > k( 4) = ( 0.0000000 0.0000000 0.3000000), wk = > 0.0714286 > k( 5) = ( 0.0000000 0.0000000 0.4000000), wk = > 0.0714286 > k( 6) = ( 0.0000000 0.0000000 0.5000000), wk = > 0.0714286 > k( 7) = ( 0.0000000 0.0000000 0.6000000), wk = > 0.0714286 > k( 8) = ( 0.0000000 0.0000000 0.7000000), wk = > 0.0714286 > k( 9) = ( 0.0000000 0.0000000 0.8000000), wk = > 0.0714286 > k( 10) = ( 0.0000000 0.0000000 0.9000000), wk = > 0.0714286 > k( 11) = ( 0.0000000 0.0000000 1.0000000), wk = > 0.0714286 > k( 12) = ( 0.0000000 0.0000000 0.0000000), wk = > 0.0714286 > k( 13) = ( 0.0000000 0.1000000 0.1000000), wk = > 0.0714286 > k( 14) = ( 0.0000000 0.2000000 0.2000000), wk = > 0.0714286 > k( 15) = ( 0.0000000 0.3000000 0.3000000), wk = > 0.0714286 > k( 16) = ( 0.0000000 0.4000000 0.4000000), wk = > 0.0714286 > k( 17) = ( 0.0000000 0.5000000 0.5000000), wk = > 0.0714286 > k( 18) = ( 0.0000000 0.6000000 0.6000000), wk = > 0.0714286 > k( 19) = ( 0.0000000 0.7000000 0.7000000), wk = > 0.0714286 > k( 20) = ( 0.0000000 0.8000000 0.8000000), wk = > 0.0714286 > k( 21) = ( 0.0000000 0.9000000 0.9000000), wk = > 0.0714286 > k( 22) = ( 0.0000000 1.0000000 1.0000000), wk = > 0.0714286 > k( 23) = ( 0.0000000 0.0000000 0.0000000), wk = > 0.0714286 > k( 24) = ( 0.1000000 0.1000000 0.1000000), wk = > 0.0714286 > k( 25) = ( 0.2000000 0.2000000 0.2000000), wk = > 0.0714286 > k( 26) = ( 0.3000000 0.3000000 0.3000000), wk = > 0.0714286 > k( 27) = ( 0.4000000 0.4000000 0.4000000), wk = > 0.0714286 > k( 28) = ( 0.5000000 0.5000000 0.5000000), wk = > 0.0714286 > > G cutoff = 85.4897 ( 869 G-vectors) FFT grid: ( 15, 15, > 15) > > nbndx = 8 nbnd = 8 natomwfc = 9 npwx = 113 > nelec = 3.00 nkb = 4 ngl = 31 > > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > from cfft3d : error # 1 > no scalar fft driver specified > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > stopping ... > > Case 2: pw.x executable compiled and linked with installed HP MPI > libraries. > > The run does not get as far this time, but fails with: > > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > task # 0 > from data_structure : error # 1 > number of sticks 0 > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > A previous post on the mailing list leads me to think these two crashes > might be related, somewhere in the FFTs. Can anyone help diagnose the > problem(s) ? > > Many thanks in advance, > Steve Kirk > > -- > Dr. Steven R. Kirk > Dept. of Technology, Mathematics & Computer Science (P)+46 520 223215 > University West (F)+46 520 223299 > P.O. Box 957 Trollhattan 461 29 SWEDEN http://taconet.webhop.org > > --__--__-- > > Message: 3 > Date: Thu, 15 Jun 2006 10:20:45 -0400 (EDT) > From: wlyim at puccini.che.pitt.edu > To: aarti srirangarajan > Cc: pw_forum at pwscf.org > Subject: Re: [Pw_forum] about total magnetization convergence > Reply-To: pw_forum at pwscf.org > > I guess your smearing parameter is too large. Try a smaller degauss value. > > On Thu, 15 Jun 2006, aarti srirangarajan wrote: > >> Dear all, >> >> I am doing a spin polarised slab calculation with vacuum layers for Ni. >> I have a problem here with the total magnetization values. I hv >> performed >> calculations with 7 7 1 , 9 9 1 and 11 11 1 and 13 13 1 k-point meshes >> for >> various surfaces and >> the total magnetisation changes very erratically. The values are not >> converging. Although the Final energies are converging very well. >> Also the difference b/w the Total and abs. magnetisation is of the order >> of 2 bohrs. Could anyone help me with this? >> >> thank you >> Aarti >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> > > > > --__--__-- > > Message: 4 > From: Paolo Giannozzi > Organization: Scuola Normale Superiore di Pisa > To: pw_forum at pwscf.org > Subject: Re: [Pw_forum] Diagnoses for two pw.x error messages > Date: Thu, 15 Jun 2006 17:30:05 +0200 > Reply-To: pw_forum at pwscf.org > > On Thursday 15 June 2006 16:19, Steven Kirk wrote: > >> from cfft3d : error # 1 >> no scalar fft driver specified > > buggy preprocessor (ifort 9?), or missing preprocessing options: > there should be one among those appearing in Modules/fft_scalar.f90 > >> from data_structure : error # 1 >> number of sticks 0 > > too many processors for the size of the system > > Paolo > -- > Paolo Giannozzi e-mail: giannozz at nest.sns.it > Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 > Piazza dei Cavalieri 7 I-56126 Pisa, Italy > > --__--__-- > > Message: 5 > Date: Thu, 15 Jun 2006 11:43:17 -0400 > From: Fernando A Reboredo > Subject: Re: [Pw_forum] how to find out the stable structure under high > pressure > To: pw_forum at pwscf.org > Reply-To: pw_forum at pwscf.org > > This is a multi-part message in MIME format. > > ------=_NextPart_000_0A55_01C69070.E4D6B110 > Content-Type: text/plain; > charset="UTF-8" > Content-Transfer-Encoding: quoted-printable > > I agree with Stefano: > While it is true that 1) in matematical terms ALL the phonon modes of a = > crystal are a compete basis where any posible atomic displacement can be = > decomposed, and also 2) any new structure with the same number of atoms = > can be though as a displacement from the old one (it could be big), the = > set of soft modes are incomple basis and thus they can only describe a = > subset of arbitrary structure transitions. > Fernando > > ----- Original Message -----=20 > From: Stefano Baroni=20 > To: pw_forum at pwscf.org=20 > Sent: Wednesday, June 14, 2006 2:56 AM > Subject: Re: [Pw_forum] how to find out the stable structure under = > high pressure > > > Nicola: here, I beg to differ. For two reasons. > > > First: even if there was a minimum such as discrebed in the question, = > this would be a _local_ and not necessarily a global one. > > > Second: there would be such a minimum if there was not any coupling = > between the soft modes and any other (phonon or elastic) modes. In = > practice, this is seldom the case, so that the low-simmetry phase seldom = > results from the distortion of soft modes alone. I would advise Li Yan = > to browse into a book dealing with the Landau theory of phase = > transitions, where these questions should be thoroughly discussed. = > Unfortunately, the only title that occurs to me right now is in French = > ("Sym=C3=A9tries bris=C3=A9es", by Boccara and Boccara - an English = > translation may exist). In any case, these texts are usually rather = > awkward and, when I needed to work with these concepts, I preferred to = > "rediscover" what I needed from scratch. My own efforts in mastering the = > subject are witnessed in the papers: Phys. Rev. Lett., 69, 1069 (1992) = > and Phys. Rev. B 51, 8060 (1995). What is true (and what Nicola most = > probably actually meant) is that if you start an energy minimization = > from a configuration characterized by a non-vanishing amplitude of the = > soft mode-distortion, then the system will certainly find a = > configuration of lower energy. This configuration may or may not be a = > local minimum (let alone a global one that you will never know for sure) = > according to whether your supercell and the symmetry of the starting = > configuration are such as to allow for all the relevant mode-mode = > couplings eventually responsible for the stability of the low-symmetry = > phase. > > > Hope to have been not too confusing. > > > Stefano > > > > > > On Jun 14, 2006, at 5:01 AM, Nicola Marzari wrote: > > > > > Yes, > > > nicola > > > > > li yan wrote: > Dear all, > I found soft modes in a system under pressure. Is it true = > that=20 > there must be a global energy miniumu in the subspace spanned by = > the=20 > eignevectors of the soft modes? And by searching this munimum, = > can i=20 > find the stable high-pressure strucure? =20 > > > > > best regards > > > __________________________________________________ > = > =E8=B5=B6=E5=BF=AB=E6=B3=A8=E5=86=8C=E9=9B=85=E8=99=8E=E8=B6=85=E5=A4=A7=E5= > =AE=B9=E9=87=8F=E5=85=8D=E8=B4=B9=E9=82=AE=E7=AE=B1? > http://cn.mail.yahoo.com > > > > > > > > > > > --=20 > = > --------------------------------------------------------------------- > Prof Nicola Marzari Department of Materials Science and = > Engineering > 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 = > USA > tel 617.4522758 fax 2586534 marzari at mit.edu = > http://quasiamore.mit.edu > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - = > Trieste > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) > > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > > > > ------=_NextPart_000_0A55_01C69070.E4D6B110 > Content-Type: text/html; > charset="UTF-8" > Content-Transfer-Encoding: quoted-printable > > =EF=BB=BF > > > > > > style=3D"WORD-WRAP: break-word; khtml-nbsp-mode: space; = > khtml-line-break: after-white-space"=20 > bgColor=3D#ffffff> >
I agree with Stefano:
>
While it is true that 1) in matematical = > > terms ALL the phonon modes of a crystal are a compete basis where = > any=20 > posible atomic displacement can be decomposed, a face=3DArial=20 > size=3D2>nd also 2) any new structure with the same number of atoms can = > be though=20 > as a displacement from the old one (it could be big), the set of soft=20 > modes are incomple basis and thus they can only describe a = > subset of=20 > arbitrary structure transitions.
>
Fernando
>
 
>
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; = > BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px"> >
----- Original Message -----
> style=3D"BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: = > black">From:=20 > Stefano = > Baroni=20 > > >
Sent: Wednesday, June 14, 2006 = > 2:56=20 > AM
>
Subject: Re: [Pw_forum] how to = > find out=20 > the stable structure under high pressure
>

Nicola: here, I beg to differ. For two reasons. >

>
First: even if there was a minimum such as discrebed in the = > question,=20 > this would be a _local_ and not necessarily a global one.
>

>
Second: there would be such a minimum if there was not any = > coupling=20 > between the soft modes and any other (phonon or elastic) modes. In = > practice,=20 > this is seldom the case, so that the low-simmetry phase seldom results = > from=20 > the distortion of soft modes alone. I would advise Li Yan to browse = > into a=20 > book dealing with the Landau theory of phase transitions, where these=20 > questions should be thoroughly discussed. Unfortunately, the only = > title that=20 > occurs to me right now is in French ("Sym=C3=A9tries bris=C3=A9es", by = > Boccara and=20 > Boccara - an English translation may exist). In any case, these texts = > are=20 > usually rather awkward and, when I needed to work with these concepts, = > I=20 > preferred to "rediscover" what I needed from scratch. My own efforts = > in=20 > mastering the subject are witnessed in the = > papers:   Phys. Rev.=20 > Lett., 69, 1069 (1992) and Phys. Rev. B 51, 8060 (1995). What is = > true=20 > (and what Nicola most probably actually meant) is that if you start an = > energy=20 > minimization from a configuration characterized by a non-vanishing = > amplitude=20 > of the soft mode-distortion, then the system will certainly find a=20 > configuration of lower energy. This configuration may or may not be a = > local=20 > minimum (let alone a global one that you will never know for sure) = > according=20 > to whether your supercell and the symmetry of the starting = > configuration are=20 > such as to allow for all the relevant mode-mode couplings eventually=20 > responsible for the stability of the low-symmetry phase.
>

>
Hope to have been not too confusing.
>

>
Stefano
>

>
 
>

>
>
On Jun 14, 2006, at 5:01 AM, Nicola Marzari wrote:
class=3DApple-interchange-newline> >
>

>
Yes,
>

>
style=3D"WHITE-SPACE: pre"> style=3D"WHITE-SPACE: pre"> style=3D"WHITE-SPACE: pre">nicola
>

>

>
li yan wrote:
>
>
Dear all,
>
class=3DApple-converted-space>  =20 >     I found soft modes in a system under = > pressure. Is it=20 > true that 
>
there must be a global energy miniumu = > in the=20 > subspace spanned by the class=3DApple-converted-space> 
>
eignevectors of the soft modes? class=3DApple-converted-space>  And by searching this = > munimum,=20 > can i 
>
find the stable high-pressure strucure? = > class=3DApple-converted-space> 
>

>

>
best regards
>

> style=3D"MARGIN: = > 0px">__________________________________________________
>
0px">=E8=B5=B6=E5=BF=AB=E6=B3=A8=E5=86=8C=E9=9B=85=E8=99=8E=E8=B6=85=E5=A4= > =A7=E5=AE=B9=E9=87=8F=E5=85=8D=E8=B4=B9=E9=82=AE=E7=AE=B1?
>
= > href=3D"http://cn.mail.yahoo.com">http://cn.mail.yahoo.com
>
0px">
>

>

>

>

>
-- class=3DApple-converted-space> 
> style=3D"MARGIN: = > 0px">--------------------------------------------------------------------= > - >
Prof Nicola Marzari class=3DApple-converted-space>  Department of Materials = > Science=20 > and Engineering
>
13-5066 class=3DApple-converted-space> =20 > MIT   77 = > Massachusetts=20 > Avenue   Cambridge = > MA=20 > 02139-4307 USA
>
tel 617.4522758 fax 2586534 = > marzari at mit.edu = > href=3D"http://quasiamore.mit.edu">http://quasiamore.mit.edu
> style=3D"MARGIN: = > 0px">_______________________________________________ >
Pw_forum mailing list
>
href=3D"mailto:Pw_forum at pwscf.org">Pw_forum at pwscf.org
>
= > href=3D"http://www.democritos.it/mailman/listinfo/pw_forum">http://www.de= > mocritos.it/mailman/listinfo/pw_forum

>
style=3D"WORD-SPACING: 0px; FONT: 14px Helvetica; TEXT-TRANSFORM: = > none; COLOR: rgb(0,0,0); TEXT-INDENT: 0px; WHITE-SPACE: normal; = > LETTER-SPACING: normal; BORDER-COLLAPSE: separate; border-spacing: 0px = > 0px; khtml-text-decorations-in-effect: none; apple-text-size-adjust: = > auto; orphans: 2; widows: 2"> class=3DApple-style-span=20 > style=3D"WORD-SPACING: 0px; FONT: 14px Helvetica; TEXT-TRANSFORM: = > none; COLOR: rgb(0,0,0); TEXT-INDENT: 0px; WHITE-SPACE: normal; = > LETTER-SPACING: normal; BORDER-COLLAPSE: separate; border-spacing: 0px = > 0px; khtml-text-decorations-in-effect: none; apple-text-size-adjust: = > auto; orphans: 2; widows: 2"> class=3DApple-style-span=20 > style=3D"WORD-SPACING: 0px; FONT: 14px Helvetica; TEXT-TRANSFORM: = > none; COLOR: rgb(0,0,0); TEXT-INDENT: 0px; WHITE-SPACE: normal; = > LETTER-SPACING: normal; BORDER-COLLAPSE: separate; border-spacing: 0px = > 0px; khtml-text-decorations-in-effect: none; apple-text-size-adjust: = > auto; orphans: 2; widows: 2"> >
size=3D3> class=3DApple-style-span style=3D"FONT-SIZE: 12px"> class=3DApple-style-span=20 > style=3D"FONT-SIZE: 12px"> style=3D"FONT-SIZE: 12px">---
>
size=3D3> class=3DApple-style-span style=3D"FONT-SIZE: 12px"> class=3DApple-style-span=20 > style=3D"FONT-SIZE: 12px"> style=3D"FONT-SIZE: 12px">Stefano Baroni -=20 > SISSA size=3D3> class=3DApple-style-span style=3D"FONT-SIZE: 12px"> class=3DApple-style-span=20 > style=3D"FONT-SIZE: 12px"> style=3D"FONT-SIZE: 12px">  class=3DApple-style-span size=3D3> style=3D"FONT-SIZE: 12px"> style=3D"FONT-SIZE: 12px"> style=3D"FONT-SIZE: 12px">& class=3DApple-style-span size=3D3> style=3D"FONT-SIZE: 12px"> style=3D"FONT-SIZE: 12px"> style=3D"FONT-SIZE: 12px">  class=3DApple-style-span size=3D3> style=3D"FONT-SIZE: 12px"> style=3D"FONT-SIZE: 12px"> style=3D"FONT-SIZE: 12px">DEMOCRITOS National Simulation Center -=20 > Trieste
>
size=3D3> class=3DApple-style-span style=3D"FONT-SIZE: 12px"> class=3DApple-style-span=20 > style=3D"FONT-SIZE: 12px"> style=3D"FONT-SIZE: 12px">[+39] 040 3787 406 (tel) -528 (fax) / = > stefanobaroni=20 > (skype)
>

> style=3D"FONT-SIZE: 12px">
>
size=3D3> class=3DApple-style-span style=3D"FONT-SIZE: 12px"> class=3DApple-style-span=20 > style=3D"FONT-SIZE: 12px"> style=3D"FONT-SIZE: 12px">Please, if possible,=20 > don't size=3D3> class=3DApple-style-span style=3D"FONT-SIZE: 12px"> class=3DApple-style-span=20 > style=3D"FONT-SIZE: 12px"> style=3D"FONT-SIZE: 12px">  class=3DApple-style-span size=3D3> style=3D"FONT-SIZE: 12px"> style=3D"FONT-SIZE: 12px"> style=3D"FONT-SIZE: 12px">send me MS Word or PowerPoint=20 > attachments
>
size=3D3> class=3DApple-style-span style=3D"FONT-SIZE: 12px"> class=3DApple-style-span=20 > style=3D"FONT-SIZE: 12px"> style=3D"FONT-SIZE: 12px">Why? See: class=3DApple-style-span size=3D3> style=3D"FONT-SIZE: 12px"> style=3D"FONT-SIZE: 12px"> style=3D"FONT-SIZE: 12px">  href=3D"http://www.gnu.org/philosophy/no-word-attachments.html"> class=3DApple-style-span color=3D#0023e9 size=3D3> class=3DApple-style-span=20 > style=3D"FONT-SIZE: 12px; COLOR: rgb(0,35,233); = > khtml-text-decorations-in-effect: underline"> class=3DApple-style-span=20 > style=3D"FONT-SIZE: 12px; COLOR: rgb(0,35,233); = > khtml-text-decorations-in-effect: underline"> class=3DApple-style-span=20 > style=3D"FONT-SIZE: 12px; COLOR: rgb(0,35,233); = > khtml-text-decorations-in-effect: = > underline">http://www.gnu.org/philosophy/no-word-attachments.html<= > /SPAN>
>

= > class=3DApple-interchange-newline>

BLOCKQUOTE> > > ------=_NextPart_000_0A55_01C69070.E4D6B110-- > > > > --__--__-- > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > End of Pw_forum Digest > From marzari at MIT.EDU Fri Jun 16 06:23:00 2006 From: marzari at MIT.EDU (Nicola Marzari) Date: Fri, 16 Jun 2006 00:23:00 -0400 Subject: [Pw_forum] Re: Pw_forum digest, Vol 1 #1029 - 5 msgs In-Reply-To: <51834.202.41.111.151.1150431087.squirrel@202.41.111.151> References: <20060615154402.12613.28140.Mailman@democritos.sissa.it> <51834.202.41.111.151.1150431087.squirrel@202.41.111.151> Message-ID: <44923224.6050901@mit.edu> aarti srirangarajan wrote: > > Thanks for the reply. > The degauss value I am working with is 0.003. What value should typically > be used for Ni? > regards > aarti Such small value will require a large amount of kpoints to converge forces and energies. My rule of thumb is between 0.1 eV and 0.5 eV (REMEMBER TO divide by 13.5068 to obtain the correct degauss for pwscf). 0.1eV is really small, to be used for difficult/magnetic cases. 0.5eV is large, for simple metals like Al. The larger the smearing, the fewer kpoints needed for an accurate integration (functions are smoother). But, the larger the smearing, the further away you move from the zero-temperature target. So it's a tradeoff, and you need to find the sweet spot. You should use either methfessel-paxton or marzari-vanderbilt, as a smearing scheme. I prefer the latter. nicola --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From ssaarti at jncasr.ac.in Fri Jun 16 08:22:28 2006 From: ssaarti at jncasr.ac.in (aarti srirangarajan) Date: Fri, 16 Jun 2006 11:52:28 +0530 (IST) Subject: [Pw_forum] Re: Pw_forum digest, Vol 1 #1031 - 1 msg In-Reply-To: <20060616054610.7722.12262.Mailman@democritos.sissa.it> References: <20060616054610.7722.12262.Mailman@democritos.sissa.it> Message-ID: <44304.202.41.111.151.1150438948.squirrel@202.41.111.151> Thanks a lot for the info. I am using methfessel-paxton. Also I could not find any info on how to fix the spin moments in INPUT_PW file. thanks again regards aarti end Pw_forum mailing list submissions to > pw_forum at pwscf.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://www.democritos.it/mailman/listinfo/pw_forum > or, via email, send a message with subject or body 'help' to > pw_forum-request at pwscf.org > > You can reach the person managing the list at > pw_forum-admin at pwscf.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Pw_forum digest..." > > > Today's Topics: > > 1. Re: Re: Pw_forum digest, Vol 1 #1029 - 5 msgs (Nicola Marzari) > > --__--__-- > > Message: 1 > Date: Fri, 16 Jun 2006 00:23:00 -0400 > From: Nicola Marzari > Organization: Massachusetts Institute of Technology > To: pw_forum at pwscf.org > Subject: Re: [Pw_forum] Re: Pw_forum digest, Vol 1 #1029 - 5 msgs > Reply-To: pw_forum at pwscf.org > > aarti srirangarajan wrote: >> >> Thanks for the reply. >> The degauss value I am working with is 0.003. What value should >> typically >> be used for Ni? >> regards >> aarti > > > Such small value will require a large amount of kpoints to > converge forces and energies. My rule of thumb is between 0.1 eV > and 0.5 eV (REMEMBER TO divide by 13.5068 to obtain the correct degauss > for pwscf). > > 0.1eV is really small, to be used for difficult/magnetic cases. 0.5eV > is large, for simple metals like Al. > > The larger the smearing, the fewer kpoints needed for an accurate > integration (functions are smoother). But, the larger the smearing, > the further away you move from the zero-temperature target. > > So it's a tradeoff, and you need to find the sweet spot. > > You should use either methfessel-paxton or marzari-vanderbilt, as a > smearing scheme. I prefer the latter. > > > nicola > > > > --------------------------------------------------------------------- > Prof Nicola Marzari Department of Materials Science and Engineering > 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA > tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu > > > --__--__-- > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > End of Pw_forum Digest > From giannozz at nest.sns.it Fri Jun 16 09:50:38 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 16 Jun 2006 09:50:38 +0200 Subject: [Pw_forum] Re: Pw_forum digest, Vol 1 #1031 - 1 msg In-Reply-To: <44304.202.41.111.151.1150438948.squirrel@202.41.111.151> References: <20060616054610.7722.12262.Mailman@democritos.sissa.it> <44304.202.41.111.151.1150438948.squirrel@202.41.111.151> Message-ID: <200606160950.38324.giannozz@nest.sns.it> On Friday 16 June 2006 08:22, aarti srirangarajan wrote: > Also I could not find any info on how to fix the spin moments > in INPUT_PW file. this is the documentation of the relevant variables (for LSDA), slightly modified wrt the official version (admittedly not a marvel of clarity). ------------------------------------------------------------------------------- starting_magnetization(i) REAL starting spin polarization (values between -1 and 1) on atomic type 'i' in a spin-polarized calculation. Breaks the symmetry and provides a starting point for self-consistency. The default value is zero, BUT a value MUST be specified for AT LEAST one atomic type in spin polarized calculations. Note that if start from zero initial magnetization, you will get zero final magnetization in any case. If you desire to start from an antiferromagnetic state, you may need to define two different atomic species corresponding to sublattices of the same atomic type. If you fix the magnetization with "nelup/neldw" or with "multiplicity" or with "tot_magnetization", you should not specify starting_magnetization nelup, neldw REAL number of spin-up and spin-down electrons, respectively Note that this fixes the final value of the magnetization. The sum must yield nelec that must also be specified explicitly in this case. Not valid for spin-unpolarized or noncollinear calculations, only for LSDA. Obsolescent: use multiplicity or tot_magnetization instead. multiplicity INTEGER ( default = 0 [unspecified] ) spin multiplicity (2s+1). 1 is singlet, 2 for doublet etc. Note that this fixes the final value of the magnetization. if unspecified or a non-zero value is specified in nelup/neldw then multiplicity variable is ignored. Do not specify both multiplicity and tot_magnetization. tot_magnetization INTEGER ( default = -1 [unspecified] ) majority spin - minority spin (nelup - neldw). if unspecified or a non-zero value is specified in nelup/neldw then tot_magnetization variable is ignored. Do not specify both multiplicity and tot_magnetization. YES, there is redundancy! nelup/neldw are enough to specify the spin state. However these variables are not very convenient and will be eliminated from the input in future versions. It is recommended to use either 'multiplicity' or equivalently 'tot_magnetization' to specify the spin state. ------------------------------------------------------------------------------- -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From reinout.declerck at UGent.be Fri Jun 16 09:59:47 2006 From: reinout.declerck at UGent.be (Reinout Declerck) Date: Fri, 16 Jun 2006 09:59:47 +0200 Subject: [Pw_forum] Re: Pw_forum digest, Vol 1 #1031 - 1 msg References: <20060616054610.7722.12262.Mailman@democritos.sissa.it> <44304.202.41.111.151.1150438948.squirrel@202.41.111.151> <200606160950.38324.giannozz@nest.sns.it> Message-ID: <003301c6911a$d68670f0$9162c19d@UGent.be> Hi, > If you fix the magnetization with "nelup/neldw" or with > "multiplicity" or with "tot_magnetization", you should > not specify starting_magnetization if you did specify the starting_magnetization, will it affect the results? Regards, Reinout Declerck ----- Original Message ----- From: "Paolo Giannozzi" To: Sent: Friday, June 16, 2006 9:50 AM Subject: Re: [Pw_forum] Re: Pw_forum digest, Vol 1 #1031 - 1 msg > On Friday 16 June 2006 08:22, aarti srirangarajan wrote: > >> Also I could not find any info on how to fix the spin moments >> in INPUT_PW file. > > this is the documentation of the relevant variables (for LSDA), > slightly modified wrt the official version (admittedly not a marvel > of clarity). > ------------------------------------------------------------------------------- > starting_magnetization(i) > REAL > starting spin polarization (values between -1 and 1) > on atomic type 'i' in a spin-polarized calculation. > Breaks the symmetry and provides a starting point for > self-consistency. The default value is zero, BUT a value > MUST be specified for AT LEAST one atomic type in spin > polarized calculations. Note that if start from zero > initial magnetization, you will get zero final magnetization > in any case. If you desire to start from an > antiferromagnetic > state, you may need to define two different atomic species > corresponding to sublattices of the same atomic type. > If you fix the magnetization with "nelup/neldw" or with > "multiplicity" or with "tot_magnetization", you should > not specify starting_magnetization > nelup, neldw REAL > number of spin-up and spin-down electrons, respectively > Note that this fixes the final value of the magnetization. > The sum must yield nelec that must also be specified > explicitly in this case. Not valid for spin-unpolarized > or noncollinear calculations, only for LSDA. Obsolescent: > use multiplicity or tot_magnetization instead. > > multiplicity INTEGER ( default = 0 [unspecified] ) > spin multiplicity (2s+1). 1 is singlet, 2 for doublet etc. > Note that this fixes the final value of the magnetization. > if unspecified or a non-zero value is specified in > nelup/neldw > then multiplicity variable is ignored. > Do not specify both multiplicity and tot_magnetization. > > tot_magnetization INTEGER ( default = -1 [unspecified] ) > majority spin - minority spin (nelup - neldw). > if unspecified or a non-zero value is specified in > nelup/neldw > then tot_magnetization variable is ignored. > Do not specify both multiplicity and tot_magnetization. > YES, there is redundancy! nelup/neldw are enough to specify > the spin state. However these variables are not very > convenient > and will be eliminated from the input in future versions. > It is recommended to use either 'multiplicity' or > equivalently > 'tot_magnetization' to specify the spin state. > ------------------------------------------------------------------------------- > -- > Paolo Giannozzi e-mail: giannozz at nest.sns.it > Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 > Piazza dei Cavalieri 7 I-56126 Pisa, Italy > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From giannozz at nest.sns.it Fri Jun 16 11:01:29 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 16 Jun 2006 11:01:29 +0200 Subject: [Pw_forum] Re: Pw_forum digest, Vol 1 #1031 - 1 msg In-Reply-To: <003301c6911a$d68670f0$9162c19d@UGent.be> References: <20060616054610.7722.12262.Mailman@democritos.sissa.it> <200606160950.38324.giannozz@nest.sns.it> <003301c6911a$d68670f0$9162c19d@UGent.be> Message-ID: <200606161101.29048.giannozz@nest.sns.it> On Friday 16 June 2006 09:59, Reinout Declerck wrote: > > If you fix the magnetization with "nelup/neldw" or with > > "multiplicity" or with "tot_magnetization", you should > > not specify starting_magnetization > > if you did specify the starting_magnetization, will it affect the results? no, it shouldn't. -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From serenechan at charteredsemi.com Fri Jun 16 14:00:37 2006 From: serenechan at charteredsemi.com (serenechan at charteredsemi.com) Date: Fri, 16 Jun 2006 20:00:37 +0800 Subject: [Pw_forum] CHAN Serene is out of the office. Message-ID: I will be out of the office starting 06/16/2006 and will not return until 06/23/2006. Please approach my colleague Wang Xinzi for all tapeout related matters. From Steven.Kirk at hv.se Fri Jun 16 16:42:33 2006 From: Steven.Kirk at hv.se (Steven Kirk) Date: Fri, 16 Jun 2006 16:42:33 +0200 Subject: [Pw_forum] Re: Diagnoses for two pw.x error messages Message-ID: <4492C359.5080009@hv.se> >From: Paolo Giannozzi >Organization: Scuola Normale Superiore di Pisa >To: pw_forum at pwscf.org >Subject: Re: [Pw_forum] Diagnoses for two pw.x error messages >Date: Thu, 15 Jun 2006 17:30:05 +0200 >Reply-To: pw_forum at pwscf.org > >On Thursday 15 June 2006 16:19, Steven Kirk wrote: > >> > from cfft3d : error # 1 >> > no scalar fft driver specified > >buggy preprocessor (ifort 9?), or missing preprocessing options: >there should be one among those appearing in Modules/fft_scalar.f90 OK, tracked down and fixed, thanks! > >> > from data_structure : error # 1 >> > number of sticks 0 > >too many processors for the size of the system I get this error message even when running on 1,2 and 4 processors using MPI, running a calculation on a 120-atom unit cell. Can you suggest at a deeper level what might be causing the code to make this judgement, and could I force it to go ahead anyway ? Many thanks in advance, Steve Kirk -- Dr. Steven R. Kirk Dept. of Technology, Mathematics & Computer Science (P)+46 520 223215 University West (F)+46 520 223299 P.O. Box 957 Trollhattan 461 29 SWEDEN http://taconet.webhop.org From lukethulin at netscape.net Fri Jun 16 17:59:56 2006 From: lukethulin at netscape.net (Luke Thulin) Date: Fri, 16 Jun 2006 11:59:56 -0400 Subject: [Pw_forum] Band Gap Significance Message-ID: <4492D57C.1070204@netscape.net> Hello everyone, As I understand it, DFT has a known inability to predict the band gap properly compared to measured values. I have observed for the material I am currently studying. Is the calculated value of no physical significance? Or would relative values be valid, if for example, I wanted to calculate the percent change in band gap due to stress? Thank you, Luke From eylvisaker at physics.ucdavis.edu Fri Jun 16 20:19:38 2006 From: eylvisaker at physics.ucdavis.edu (Erik Ylvisaker) Date: Fri, 16 Jun 2006 11:19:38 -0700 Subject: [Pw_forum] Use of FHI generated pseudopotential Message-ID: <00b201c69171$6f532610$11f4a8c0@ErikLaptop> I have some pseudopotentials which were generated with the FHI code that I'm trying to use with pwscf. I used the fhi2upf utility, and I enter in the information (which I believe is correct) for converting the pseudopotential (output attached as fhi2upf.output.txt). One issue I notice is when I run pw.x, it outputs this: PSEUDO 1 is Li zval = 1.0 lmax= 3 lloc= 0 (in numerical form: 433 grid points, xmin = 0.00, dx = 0.0000) However, the local potential for this pseudopotential should be 2, as was entered into fhi2upf. I can fix this manually in the UPF file, but I don't know what part of the file needs to be modified to do that. Can someone tell me what needs to be changed? It gives reasonable results for relaxation of atoms, but I think dynamical properties (phonons, el-ph coupling) may be more sensitive to this type of issue? -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060616/8d7a34b0/attachment.htm -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: fhi2upf.output.txt Url: /pipermail/attachments/20060616/8d7a34b0/attachment.txt From akohlmey at cmm.upenn.edu Fri Jun 16 23:19:17 2006 From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer) Date: Fri, 16 Jun 2006 17:19:17 -0400 Subject: [Pw_forum] Use of FHI generated pseudopotential In-Reply-To: <00b201c69171$6f532610$11f4a8c0@ErikLaptop> References: <00b201c69171$6f532610$11f4a8c0@ErikLaptop> Message-ID: <7b6913e90606161419y4cd8828bj2f01510a62b0d0f2@mail.gmail.com> On 6/16/06, Erik Ylvisaker wrote: erik, > However, the local potential for this pseudopotential should be 2, as was > entered into fhi2upf. I can fix this manually in the UPF file, but I don't > know what part of the file needs to be modified to do that. Can someone in UPF format, the potential is aready stored in separable form, i.e. you have the local potential stored explicitely in the section and the non-local part with the required indicators for the angular momentum in . so the potential will work fine without specifying the angular momentum of the local potential explicitely. actually not even LMAX is needed, so the output should be removed (the corresponding numbers are only set when using ultrasoft pseudopotentials). > tell me what needs to be changed? It gives reasonable results for > relaxation of atoms, but I think dynamical properties (phonons, el-ph > coupling) may be more sensitive to this type of issue? nothing. this is a cosmetic issue. regards, axel. -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From ezadshojaee at hotmail.com Mon Jun 19 12:37:57 2006 From: ezadshojaee at hotmail.com (Ezad Shojaee) Date: Mon, 19 Jun 2006 10:37:57 +0000 Subject: [Pw_forum] problem wuth DOS Message-ID: hi i want to plot DOS for TiO2 & still(!) in the 'nscf' calculation i have this error(that never ends!): WARNING: 8 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 8 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 7 eigenvalues not converged can anyone tell me what's wrong in my input file: &control calculation='nscf' prefix='TiO2', pseudo_dir = '/root/pseudo/', outdir='/root/tmp/' / &system ibrav=7, celldm(1) =7.155, celldm(3) =2.5136, nat=6, ntyp=2, ecutwfc = 50.0, ecutrho = 300.0, nbnd=60, occupations='tetrahedra' / &electrons conv_thr = 1.0e-6 mixing_beta = 0.7 / ATOMIC_SPECIES .... ATOMIC_POSITIONS .... K_POINTS { automatic } 10 10 10 0 0 0 _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ From giannozz at nest.sns.it Mon Jun 19 14:13:03 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 19 Jun 2006 14:13:03 +0200 Subject: [Pw_forum] Re: Diagnoses for two pw.x error messages In-Reply-To: <4492C359.5080009@hv.se> References: <4492C359.5080009@hv.se> Message-ID: <200606191413.03037.giannozz@nest.sns.it> On Friday 16 June 2006 16:42, Steven Kirk wrote: >> from data_structure : error # 1 > > number of sticks 0 > I get this error message even when running on 1,2 and 4 processors > using MPI, running a calculation on a 120-atom unit cell. it looks like your compiler is buggy: try with a different one. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From robert at oeffner.net Mon Jun 19 15:18:33 2006 From: robert at oeffner.net (robert at oeffner.net) Date: Mon, 19 Jun 2006 09:18:33 -0400 Subject: [Pw_forum] (no subject) Message-ID: <380-220066119131833281@M2W022.mail2web.com> Hi, Sorry if this topic has been covered before but so far I haven't found it. I'm a new user of the espresso-3.1 package which I recently downloaded. The computer system I'm working on is Windows XP Mingw32 + MSYS1.0 + G95 From eyvaz_isaev at yahoo.com Mon Jun 19 15:31:05 2006 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Mon, 19 Jun 2006 06:31:05 -0700 (PDT) Subject: [Pw_forum] problem wuth DOS In-Reply-To: Message-ID: <20060619133105.75181.qmail@web60315.mail.yahoo.com> Hi, > > WARNING: 8 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 8 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 7 eigenvalues not converged This is not an error, but just a warning. Please have a look at the forum archive. So, go ahead and plot your DOS. Bests, Eyvaz. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From ezadshojaee at hotmail.com Mon Jun 19 16:00:01 2006 From: ezadshojaee at hotmail.com (Ezad Shojaee) Date: Mon, 19 Jun 2006 14:00:01 +0000 Subject: [Pw_forum] about DOS Message-ID: hi dear Eyvaz i have two questions: is "occupations=tetrahedra" right for obtaining DOS for a semiconductor like TiO2?because when i do an nscf calculation without it ,my DOS is resonable according the band structure ,but when i use this occup-mode,DOS is completely different & i have not the gap too. second, when i do an nscf calc. then use dos.x , the integrated DOS is not right in the gap,but when i use dos.x right after scf calculation ,i have the right integrated DOS but it's obvious that the i have no DOS after the gap.so what should i do to have the right DOS and integrated DOS ? thanks _________________________________________________________________ FREE pop-up blocking with the new MSN Toolbar - get it now! http://toolbar.msn.click-url.com/go/onm00200415ave/direct/01/ From giannozz at nest.sns.it Mon Jun 19 16:46:37 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 19 Jun 2006 16:46:37 +0200 Subject: [Pw_forum] about DOS In-Reply-To: References: Message-ID: <200606191646.37444.giannozz@nest.sns.it> On Monday 19 June 2006 16:00, Ezad Shojaee wrote: > is "occupations=tetrahedra" right for obtaining DOS for a semiconductor > like TiO2? yes it is > because when i do an nscf calculation without it ,my DOS is resonable > according the band structure ,but when i use this occup-mode, DOS > is completely different & i have not the gap too. you are doing something wrong. Please note that under some conditions, running a non-selfconsistent calculation may destroy the self-consistent charge density. Save the directory containing output self-consistent data and work on a copy of it Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From giannozz at nest.sns.it Mon Jun 19 17:12:50 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 19 Jun 2006 17:12:50 +0200 Subject: [Pw_forum] (no subject) In-Reply-To: <380-220066119131833281@M2W022.mail2web.com> References: <380-220066119131833281@M2W022.mail2web.com> Message-ID: <200606191712.50566.giannozz@nest.sns.it> On Monday 19 June 2006 15:18, robert at oeffner.net wrote: > (although a few warnings about REAL(8)/COMPLEX(8) inconsistencies > are issued by the compiler). they are irrelevant. See: http://www.democritos.it/pipermail/pw_forum/2006-June/004300.html > When I try running a few examples they fail as follows all of them, or do some examples work? > from readpp : error # 3 > file /c/Users/Oeffner/Tmp/espresso-3.1/pseudo/Si.vbc.UPF not found > But on my system the Si.vbc.UPF is present as evident from the > following command: > $ ls $PSEUDO_DIR/Si.vbc.UPF > /c/Users/Oeffner/Tmp/espresso-3.1/pseudo/Si.vbc.UPF the file opening that fails is performed in PW/readin.f90. Try to yield the same command with "shell" or whatever executes an OS command from fortran to verify that the code "sees" the file. Experiment a bit with the part of the code that opens a file. Opening a file will request a translation of file names from unix style to Windows style; maybe there are restrictions on file names (length, format, whatever) Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From eyvaz_isaev at yahoo.com Mon Jun 19 17:16:46 2006 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Mon, 19 Jun 2006 08:16:46 -0700 (PDT) Subject: [Pw_forum] about DOS In-Reply-To: Message-ID: <20060619151646.9885.qmail@web60323.mail.yahoo.com> Hi, > i have two questions: > is "occupations=tetrahedra" right for obtaining DOS > for a semiconductor like TiO2? Yes, it is. Nevertheless, I remember it was mentioned in the forum (by Paolo?) that it may not work for noncubic structures. You can try the smearing technique also, and you can compare is your DOS OK or not. > because when i do an nscf calculation without it ,my > DOS is resonable according the band structure ,but > when i use this occup-mode, DOS is completely > different & i have not the gap too. If I understood correctly, you did also nscf calculations, but did not specify occupations. Did you specify in this case a smearing (by default smearing="gaussian", and degauss=0.0 (Ry))? I do not think that with degauss=0 you will obtain a correct result. > second, when i do an nscf calc. then use dos.x , the > integrated DOS is not right in the gap,but when i > use dos.x right after scf calculation ,i have > the right integrated DOS but it's obvious that the i > have no DOS after the gap.so what should i do to > have the right DOS and integrated DOS ? I am not sure that I understood correctly the first part of this question, but for the last question an answer is that try a smearing with a nonzero degauss. If you are not sure on your results, send us your DOS output files. Bests, Eyvaz. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From ezadshojaee at hotmail.com Mon Jun 19 17:33:19 2006 From: ezadshojaee at hotmail.com (Ezad Shojaee) Date: Mon, 19 Jun 2006 15:33:19 +0000 Subject: [Pw_forum] about DOS Message-ID: hi dear Eyvaz by the INPUT_PW, i think that 'smearing' occupation is for metals( with a non-zero degauss ) .should i use this for a semi-conductor? thanks _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ From eyvaz_isaev at yahoo.com Mon Jun 19 17:46:14 2006 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Mon, 19 Jun 2006 08:46:14 -0700 (PDT) Subject: [Pw_forum] about DOS In-Reply-To: Message-ID: <20060619154614.5774.qmail@web60322.mail.yahoo.com> Hi again, Using the smearing technique you allow your system to be "metallic". In this case for a semiconductor (insulator) you will find a tail of the DOS in the band gap (due to a smearing). So, yes, you can use nonzero degauss even for semiconductors. But be careful for a small band gap semicondutor using degauss. Bests, Eyvaz. --- Ezad Shojaee wrote: > hi dear Eyvaz > by the INPUT_PW, i think that 'smearing' occupation > is for metals( with a > non-zero degauss ) .should i use this for a > semi-conductor? > thanks > > _________________________________________________________________ > Express yourself instantly with MSN Messenger! > Download today it's FREE! > http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From frederic.bouyer at cea.fr Tue Jun 20 10:18:29 2006 From: frederic.bouyer at cea.fr (=?iso-8859-1?Q?BOUYER_Fr=E9d=E9ric_153746?=) Date: Tue, 20 Jun 2006 10:18:29 +0200 Subject: [Pw_forum] postdoc position Message-ID: <3789CD0DE868404E9070AA3F4109F95E576164@YACA.intra.cea.fr> Dear Everyone, A postdoc position is opened in our lab in theoretical calculations. Here, a brief summary of the subject. For more informations, send me an email. Best regards, Fr?d?ric ================================================================================= Ab initio studies of the alteration processes in glasses for nuclear confinement ================================================================================= A postdoctoral position for 1 year (with the possibility of renewal for one more year)) is opened in the Laboratoire du Comportement a Long Terme (LCLT) of CEA (Commissari at a l'Energie Atomique -France) in collaboration with the Modelisation Group of the Laboratoire des Colloides, Verres et Nanomateriaux (LCVN-Univ. Montpellier 2, France). The research will cover the study of hydrolysis mechanisms within in the bulk as well as the gel formation at the surfaces of nuclear waste confinement glasses. The focus of the study will be the microscopic description of the solvation proc ess by a combination of Ab initio and classical Molecular Dynamics simulations. The theoretical results hence obtained will be compared to experimental data in this field, already available (as WASX, EXAFS, chemical-physical data) or as IR, Raman and Hyper-Raman scattering data. Candidates must have a PhD in computational condensed matter physics or theoretical chemistry, and experience in Molecular Dynamics calculations. They should have a previous postdoctoral experience (no more than 2 years) OR they should defend their PhD before December 31 st, 2006. This study will by carried out at LCTL (CEA Marcoule), and regular meetings and stays will be organized at LCVN (Montpellier) Applications, including - a detailed curriculum vitae (indicating date of birth, nationality, family status, dependents)- - list of publications - the address of at least two academic references that can be contacted confidentially - statement of research, should be sent to: F. Bouyer Email : frederic.bouyer at cea.fr S. Ispas Email : simona.ispas at lcvn.univ-montp2.fr S. Dennler Email : samuel.dennler at lcvn.univ-montp2.fr Deadline for receipt of applications: October 15th, 2006 ______________________________________ CEA VRH MARCOULE F. Bouyer DTCD / SECM / LCLT - B?t 208 BP 17171 30207 Bagnols-sur-C?ze Cedex - France Tel : (+33) 4 66 79 69 04 Fax : (+33) 4 66 79 66 20 Email : frederic.bouyer at cea.fr -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060620/38eb2cc4/attachment.htm From fengjie.ma at gmail.com Tue Jun 20 14:58:15 2006 From: fengjie.ma at gmail.com (Fengjie Ma) Date: Tue, 20 Jun 2006 20:58:15 +0800 Subject: [Pw_forum] About full Fully-Relativistic Calculation of GaAs Message-ID: Dear all: When I use espresso-3.1 to generate full relativistic pseudopotentials of Ga and As, I find some problems: 1. I choose nlcc=.False., and get the pseudopotentials of Ga and As. However, when I use them to calculate the band structure of GaAs, I find that I must write both lspinorb=.true. and noncolin=.true. in the input file. If not, the band split at Gamma point will not appear. Why? 2. Use the pseudopotntials, whether choose nlcc=.true. or not, the gap between valence band and conduct band becomes only 0.6 eV. I know that should be 1.4eV in GaAs, Why? ( I also re-calculate with changing the matching radius, but get the same problems) Could anyone tell me, is there any critical mistakes and how should I solve this problem? Thank you in advance, Ma Fengjie ------------------------------------------------------------------------ &input title='As', zed=33., rel=2, rlderiv=2.50, eminld=-4.0, emaxld=4.0, deld=0.02, nld=5, iswitch=3, config='[Ar] 3d10 4s2 4p3 4d-1' dft='LDA', & &inputp pseudotype=2, lloc=2, nlcc =.true., file_pseudopw='Asrel.RRKJ3', & 5 4S 1 0 2.00 0.00 2.20 2.20 0.50 4P 2 1 2.00 0.00 2.20 2.20 0.50 4P 2 1 1.00 0.00 2.20 2.20 1.50 4D 3 2 0.00 0.05 2.30 2.30 1.50 4D 3 2 0.00 0.05 2.30 2.30 2.50 -------------------------------------------------------------------------------- &input title='Ga', zed=31., rel=2, rlderiv=2.50, eminld=-4.0, emaxld=4.0, deld=0.02, nld=5, iswitch=3, config='[Ar] 3d10 4s2 4p0.33 4p0.67 4d-1' dft='LDA', & &inputp pseudotype=2, lloc=2, nlcc =.true., file_pseudopw='Ga.RRKJ3', & 5 4S 1 0 2.00 0.00 2.40 2.40 0.50 4P 2 1 0.33 0.00 1.60 1.60 0.50 4P 2 1 0.67 0.00 1.60 1.60 1.50 4D 3 2 0.00 0.05 2.70 2.70 1.50 4D 3 2 0.00 0.05 2.70 2.70 2.50 -------------------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060620/d1987db4/attachment.htm From gtg166n at mail.gatech.edu Tue Jun 20 16:11:57 2006 From: gtg166n at mail.gatech.edu (Sigifredo Sanchez-Carrera) Date: Tue, 20 Jun 2006 10:11:57 -0400 Subject: [Pw_forum] Rigid-Body approximation with PWscf Message-ID: <1150812717.4498022db7b42@webmail.mail.gatech.edu> Dear PWscf users, I was wondering if the "rigid-body approximation" can be explicitly treated with PWscf. I am trying to set a lattice dynamics calculation of molecular organic crystals, within the rigid-body approximation in which the molecules are constrained to maintain their equilibrium intramolecular bond lengths and bond angles. I have seen examples in which you can actually set either an atomic position (example 1) or the distance & angles between atoms (example 2). My molecular systems include 40 ? 50 atoms per unit cell, thus several constrains in terms of bonds and angles are required; I would be interested in knowing if this approach can be fulfilled with PWscf according to your experience, and what are the general guidelines to implement it? Thanks in advance, Sigifredo S?nchez-Carrera =================== Example 1: ATOMIC_POSITIONS { bohr } C 0.000000000 0.000000000 0.000000000 0 0 0 C 2.100000000 0.000000000 0.000000000 H -1.600000000 0.400000000 0.000000000 H 3.600000000 -0.400000000 0.000000000 =================== =================== Example 2: ATOMIC_POSITIONS { bohr } C 0.000000000 0.000000000 0.000000000 0 0 0 C 2.100000000 0.000000000 0.000000000 H -1.600000000 0.400000000 0.000000000 H 3.600000000 -0.400000000 0.000000000 CONSTRAINTS 2 1.D-8 planar_angle 1 2 4 distance 1 2 =================== -- Sigifredo Sanchez-Carrera Georgia Institute of Technology School of Chemistry and Biochemistry Boggs 1-90 770 State St. NW Atlanta, GA 30332-0400 Work (404) 894 6456 Home (404) 206 9531 From giannozz at nest.sns.it Tue Jun 20 16:38:29 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 20 Jun 2006 16:38:29 +0200 Subject: [Pw_forum] Band Gap Significance In-Reply-To: <4492D57C.1070204@netscape.net> References: <4492D57C.1070204@netscape.net> Message-ID: <200606201638.29418.giannozz@nest.sns.it> On Friday 16 June 2006 17:59, Luke Thulin wrote: > As I understand it, DFT has a known inability to predict the band > gap properly compared to measured values. [....] would relative > values be valid, if for example, I wanted to calculate the percent > change in band gap due to stress? the conventional wisdom is that changes in the band gaps are better reproduced by DFT than band gaps themselves. It is debatable whether the conventional wisdom is based on the comparison of actual calculations with experiments, or on clear theoretical arguments, or on the lack of better approximations for realistic materials! Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From giannozz at nest.sns.it Tue Jun 20 16:39:49 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 20 Jun 2006 16:39:49 +0200 Subject: [Pw_forum] About full Fully-Relativistic Calculation of GaAs In-Reply-To: References: Message-ID: <200606201639.49194.giannozz@nest.sns.it> On Tuesday 20 June 2006 14:58, Fengjie Ma wrote: > 2. Use the pseudopotntials, whether choose nlcc=.true. or not, the > gap between valence band and conduct band becomes only 0.6 eV. > I know that should be 1.4eV in GaAs, Why? never heard about the "band gap problem" of DFT? Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From giannozz at nest.sns.it Tue Jun 20 16:55:42 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 20 Jun 2006 16:55:42 +0200 Subject: [Pw_forum] Rigid-Body approximation with PWscf In-Reply-To: <1150812717.4498022db7b42@webmail.mail.gatech.edu> References: <1150812717.4498022db7b42@webmail.mail.gatech.edu> Message-ID: <200606201655.42998.giannozz@nest.sns.it> On Tuesday 20 June 2006 16:11, Sigifredo Sanchez-Carrera wrote: > I am trying to set a lattice dynamics calculation of molecular > organic crystals, within the rigid-body approximation in which > the molecules are constrained to maintain their equilibrium > intramolecular bond lengths and bond angles. if you plan to calculate intermolecular modes in molecular crystals using DFT, my advice is to change plan: you won't get anything sensible Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From fengjie.ma at gmail.com Tue Jun 20 17:31:33 2006 From: fengjie.ma at gmail.com (Fengjie Ma) Date: Tue, 20 Jun 2006 23:31:33 +0800 Subject: [Pw_forum] About full Fully-Relativistic Calculation of GaAs Message-ID: Hi Paolo, In the output file : ---------------------------------------------------------------------------------------------------------- End of self-consistent calculation k = 0.0000 0.0000 0.0000 ( 869 PWs) bands (ev): -7.8656 -7.8656 4.6130 4.6130 4.9727 4.9727 4.9727 4.9727 5.5949 5.5949 8.5592 8.5592 8.7645 8.7645 8.7645 8.7645 ---------------------------------------------------------------------------------------------------------- Is there any mistakes in the inputfile of pseudopotential generation? Is the reason why appear such problems using wrong generated pseudopotentials? Thank you very much. Ma Fengjie >>* 2. Use the pseudopotntials, whether choose nlcc=.true. or not, the *>>* gap between valence band and conduct band becomes only 0.6 eV. *>*> I know that should be 1.4eV in GaAs, Why? * >never heard about the "band gap problem" of DFT? >Paolo -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20060620/571f82ee/attachment.htm From giannozz at nest.sns.it Tue Jun 20 18:03:43 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 20 Jun 2006 18:03:43 +0200 Subject: [Pw_forum] About full Fully-Relativistic Calculation of GaAs In-Reply-To: References: Message-ID: <200606201803.43650.giannozz@nest.sns.it> On Tuesday 20 June 2006 17:31, Fengjie Ma wrote: > k = 0.0000 0.0000 0.0000 ( 869 PWs) bands (ev): > > -7.8656 -7.8656 4.6130 4.6130 4.9727 4.9727 4.9727 4.9727 > 5.5949 5.5949 8.5592 8.5592 8.7645 8.7645 8.7645 8.7645 what should we get from the above numbers? that your band gap is 0.6 eV ? it is well know that band gaps are strongly underestimated in DFT. Before attempting difficult calculations with spin-orbit, you should first understand the basics -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From robert at oeffner.net Tue Jun 20 18:54:42 2006 From: robert at oeffner.net (Robert Oeffner) Date: Tue, 20 Jun 2006 17:54:42 +0100 Subject: [Pw_forum] (no subject) References: <380-220066119131833281@M2W022.mail2web.com> <200606191712.50566.giannozz@nest.sns.it> Message-ID: <000501c6948a$3b05ee80$0202a8c0@Calculus.oeffner.net> Well, it seems that the readin.f90 file makes heavy use of forward-slashes, "/", when denoting files in folders. This is unlike Windows that uses back-slashes, "\". I'll need to experiment a bit more to verify this. But I've successfully run other Fortran code compiled with G95 + Mingw32 that opened files with no problems although these files wre located in the same folder as the executable program. Will let you know soon later. thanks, rob ----- Original Message ----- From: "Paolo Giannozzi" To: Sent: Monday, June 19, 2006 4:12 PM Subject: Re: [Pw_forum] (no subject) > On Monday 19 June 2006 15:18, robert at oeffner.net wrote: > >> (although a few warnings about REAL(8)/COMPLEX(8) inconsistencies >> are issued by the compiler). > > they are irrelevant. See: > http://www.democritos.it/pipermail/pw_forum/2006-June/004300.html > >> When I try running a few examples they fail as follows > > all of them, or do some examples work? > >> from readpp : error # 3 >> file /c/Users/Oeffner/Tmp/espresso-3.1/pseudo/Si.vbc.UPF not found > >> But on my system the Si.vbc.UPF is present as evident from the >> following command: > >> $ ls $PSEUDO_DIR/Si.vbc.UPF >> /c/Users/Oeffner/Tmp/espresso-3.1/pseudo/Si.vbc.UPF > > the file opening that fails is performed in PW/readin.f90. > Try to yield the same command with "shell" or whatever executes > an OS command from fortran to verify that the code "sees" the file. > Experiment a bit with the part of the code that opens a file. > Opening a file will request a translation of file names from unix > style to Windows style; maybe there are restrictions on file names > (length, format, whatever) > > Paolo > -- > Paolo Giannozzi e-mail: giannozz at nest.sns.it > Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 > Piazza dei Cavalieri 7 I-56126 Pisa, Italy > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From adeagbo at thp.Uni-Duisburg.de Wed Jun 21 18:15:13 2006 From: adeagbo at thp.Uni-Duisburg.de (Adeagbo Waheed Adeniyi) Date: Wed, 21 Jun 2006 18:15:13 +0200 (CEST) Subject: [Pw_forum] primitive orthorohmbic face centred phonons In-Reply-To: <20060621054618.22902.36689.Mailman@democritos.sissa.it> References: <20060621054618.22902.36689.Mailman@democritos.sissa.it> Message-ID: Dear all, I am trying to do phonon calculation of a primitive orthorohmbic face centred unit cell with 2*2*2 uniform q grid, the nscf and the phonon calculation are right but when I performed q2r.x after the runs the program stopped with error messages %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from init : error # 1 missing q-point(s)! %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... 2 Here is my input file &input nr1=2,nr2=2, nr3=2, zasr=.true., fild='system222.fc' / 5 system.dynG Q1 Q2 Q3 Q4 ================= where q-values obtained from c/a and b/a ratios using kpoints.x are 0.0000000 0.0000000 0.0000000 for system.dynG 0.5000000 -0.5216606 0.4487578 for Q1 1.0000000 0.0000000 0.0000000 for Q2 0.0000000 0.0000000 0.8975155 for Q3 0.0000000 1.0433213 0.0000000 for Q4 I used these values in phonon calculations. I never encounter any problem for other lattice such as fcc of the same system calculated so far. Any advice will be highly appreciated. Best regards Adeagbo From giannozz at nest.sns.it Thu Jun 22 12:14:35 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 22 Jun 2006 12:14:35 +0200 Subject: [Pw_forum] primitive orthorohmbic face centred phonons In-Reply-To: References: <20060621054618.22902.36689.Mailman@democritos.sissa.it> Message-ID: <200606221214.35361.giannozz@nest.sns.it> On Wednesday 21 June 2006 18:15, Adeagbo Waheed Adeniyi wrote: > I am trying to do phonon calculation of a primitive orthorohmbic face > centred unit cell with 2*2*2 uniform q grid, the nscf and the phonon > calculation are right but when I performed q2r.x after the runs the program > stopped with error messages > from init : error # 1 > missing q-point(s)! > [...] q-values obtained from c/a and b/a ratios using kpoints.x are > > 0.0000000 0.0000000 0.0000000 for system.dynG > 0.5000000 -0.5216606 0.4487578 for Q1 > 1.0000000 0.0000000 0.0000000 for Q2 > 0.0000000 0.0000000 0.8975155 for Q3 > 0.0000000 1.0433213 0.0000000 for Q4 > > I used these values in phonon calculations. there might be a mismatch between these points and what the code expects. Even a small difference might do the job. Please try to calculate the phonon dispersion with ldisp.true. and nq1=nq2=nq3=2 Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From lukethulin at netscape.net Thu Jun 22 16:18:07 2006 From: lukethulin at netscape.net (Luke Thulin) Date: Thu, 22 Jun 2006 10:18:07 -0400 Subject: [Pw_forum] xcrysden chdens plots Message-ID: <449AA69F.8010909@netscape.net> Hello everyone, I'm trying to use xcrysden to view 3D charge density plots. I set iflag = 3 and output_format = 5 for pp.x input. The pp.x output file doesn't report any problems and says it wrote the data file in xcrysden format for a 3D plot. When I try to open the data file in xcrysden it gives a parse error and won't open it. There seemed to be a similar discussion about this in the archives but I didn't see if the problem was resolved. According to the input_pp documentation, it looks to me that if I choose this output format, I don't need to include any of the e1, e2, e3, xo, nx, ny, and nz variables, is this correct? Thanks, Luke From lukethulin at netscape.net Thu Jun 22 19:04:44 2006 From: lukethulin at netscape.net (Luke Thulin) Date: Thu, 22 Jun 2006 13:04:44 -0400 Subject: [Pw_forum] xcrysden chdens plots In-Reply-To: <449AA69F.8010909@netscape.net> References: <449AA69F.8010909@netscape.net> Message-ID: <449ACDAC.6080808@netscape.net> Nevermind, it's working fine. Luke lukethulin at netscape.net wrote: > Hello everyone, > > I'm trying to use xcrysden to view 3D charge density plots. I set > iflag = 3 and output_format = 5 for pp.x input. The pp.x output file > doesn't report any problems and says it wrote the data file in > xcrysden format for a 3D plot. When I try to open the data file in > xcrysden it gives a parse error and won't open it. There seemed to be > a similar discussion about this in the archives but I didn't see if > the problem was resolved. According to the input_pp documentation, it > looks to me that if I choose this output format, I don't need to > include any of the e1, e2, e3, xo, nx, ny, and nz variables, is this > correct? > > Thanks, > Luke > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From proffess at yandex.ru Fri Jun 23 00:12:18 2006 From: proffess at yandex.ru (Sergey Lisenkov) Date: Fri, 23 Jun 2006 02:12:18 +0400 (MSD) Subject: [Pw_forum] force calculation in 3.1 on IBM machine Message-ID: <449B15C2.000001.03555@webmail12.yandex.ru> Dear developers, I found tha there is a problem with computing the forces during the relaxation (I got "segmentation fault" error). It happened on our IBM PowerPC cluster under Linux; compiler is xlf 9.1, xlc 7.0, ESSL, pwscf 3.1 version (serial and parallel). For tests, I used the example from example03 directory. Both input and output are attached. BTW, if I recompile the code without optimization, this problem dissapears. I do not believe that this is a compiler bug, because I have never had such problems (also with this test job) with pwscf versions <=3.0 Thanks, Sergey -------------- next part -------------- A non-text attachment was scrubbed... Name: file.tar.gz Type: application/octet-stream Size: 17024 bytes Desc: not available Url : /pipermail/attachments/20060623/9876bbed/attachment.obj From prasenjit at jncasr.ac.in Fri Jun 23 06:53:17 2006 From: prasenjit at jncasr.ac.in (PRASENJIT GHOSH) Date: Fri, 23 Jun 2006 10:23:17 +0530 (IST) Subject: [Pw_forum] use of starting_magnetization value when restarting an lsda cal. Message-ID: <35237.202.41.111.151.1151038397.squirrel@202.41.111.151> Dear all, I'm using PWscf to do lsda calculations on clusters. What I'm worried about is when restarting the cal. is it correct to provide a value of the initial_magnetization because on doing so one may loose the information regarding the magnetization from the previous run? I also tried to restart the calculation without the "initial_magnetization" flag but the code is asking for an initial value. Can anyone please throw some light on this issue? With regards, Prasenjit. PRASENJIT GHOSH, Ph. D STUDENT, THEORETICAL SCIENCES UNIT, JAWAHARLAL NEHRU CENTRE, JAKKUR P. O., BANGALORE - 560064, INDIA. PHONE:+91-80-22082835 (OFFICE) +91 9880519401 (MOBILE) From degironc at sissa.it Fri Jun 23 11:37:12 2006 From: degironc at sissa.it (Stefano de Gironcoli) Date: Fri, 23 Jun 2006 11:37:12 +0200 (CEST) Subject: [Pw_forum] use of starting_magnetization value when restarting an lsda cal. In-Reply-To: <35237.202.41.111.151.1151038397.squirrel@202.41.111.151> References: <35237.202.41.111.151.1151038397.squirrel@202.41.111.151> Message-ID: It is ok to restart a calculation w/o removing starting_magnetization definitions. The starting magnetization info is used when the starting potential is computed from atomic charges. This is usual case when starting from scratch. When you restart a calculation the first potential is read from file and the starting_magnetization keywords are not used and make no harm. stefano On Fri, 23 Jun 2006, PRASENJIT GHOSH wrote: > > Dear all, > > I'm using PWscf to do lsda calculations on clusters. > What I'm worried about is when restarting the cal. is it correct to > provide a value of the initial_magnetization because on doing so one may > loose the information regarding the magnetization from the previous run? > > I also tried to restart the calculation without the > "initial_magnetization" flag but the code is asking for an initial value. > > Can anyone please throw some light on this issue? > > With regards, > Prasenjit. > > PRASENJIT GHOSH, > Ph. D STUDENT, > THEORETICAL SCIENCES UNIT, > JAWAHARLAL NEHRU CENTRE, > JAKKUR P. O., > BANGALORE - 560064, > INDIA. > > PHONE:+91-80-22082835 (OFFICE) > +91 9880519401 (MOBILE) > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From prasenjit at jncasr.ac.in Fri Jun 23 12:14:26 2006 From: prasenjit at jncasr.ac.in (PRASENJIT GHOSH) Date: Fri, 23 Jun 2006 15:44:26 +0530 (IST) Subject: [Pw_forum] use of starting_magnetization value when restarting an lsda cal. In-Reply-To: References: <35237.202.41.111.151.1151038397.squirrel@202.41.111.151> Message-ID: <56817.202.41.111.151.1151057666.squirrel@202.41.111.151> Dear Stefano, Thanks a lot for ur reply. With regards, Prasenjit. On Fri, June 23, 2006 15:07, Stefano de Gironcoli said: > It is ok to restart a calculation w/o removing starting_magnetization > definitions. > The starting magnetization info is used when the starting potential is > computed from atomic charges. This is usual case when starting from > scratch. > When you restart a calculation the first potential is read from file and > the starting_magnetization keywords are not used and make no harm. > > stefano > > On Fri, 23 Jun 2006, PRASENJIT GHOSH wrote: > >> >> Dear all, >> >> I'm using PWscf to do lsda calculations on clusters. >> What I'm worried about is when restarting the cal. is it correct to >> provide a value of the initial_magnetization because on doing so one may >> loose the information regarding the magnetization from the previous run? >> >> I also tried to restart the calculation without the >> "initial_magnetization" flag but the code is asking for an initial >> value. >> >> Can anyone please throw some light on this issue? >> >> With regards, >> Prasenjit. >> >> PRASENJIT GHOSH, >> Ph. D STUDENT, >> THEORETICAL SCIENCES UNIT, >> JAWAHARLAL NEHRU CENTRE, >> JAKKUR P. O., >> BANGALORE - 560064, >> INDIA. >> >> PHONE:+91-80-22082835 (OFFICE) >> +91 9880519401 (MOBILE) >> _______________________________________________ >> 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 > PRASENJIT GHOSH, Ph. D STUDENT, THEORETICAL SCIENCES UNIT, JAWAHARLAL NEHRU CENTRE, JAKKUR P. O., BANGALORE - 560064, INDIA. PHONE:+91-80-22082835 (OFFICE) +91 9880519401 (MOBILE) From giannozz at nest.sns.it Fri Jun 23 12:33:46 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 23 Jun 2006 12:33:46 +0200 Subject: [Pw_forum] force calculation in 3.1 on IBM machine In-Reply-To: <449B15C2.000001.03555@webmail12.yandex.ru> References: <449B15C2.000001.03555@webmail12.yandex.ru> Message-ID: <200606231233.46592.giannozz@nest.sns.it> On Friday 23 June 2006 00:12, Sergey Lisenkov wrote: > I found that there is a problem with computing the forces during the > relaxation (I got "segmentation fault" error). It happened on our IBM > PowerPC cluster under Linux; compiler is xlf 9.1, xlc 7.0, ESSL, > pwscf 3.1 version (serial and parallel). For tests, I used the example > from example03 directory. Both input and output are attached. > > BTW, if I recompile the code without optimization, this problem > disappears. I do not believe that this is a compiler bug well, I do (unless you have evidence of the opposite) > because I have never had such problems (also with this test job) > with pwscf versions <=3.0 this doesn't qualify as evidence of the opposite Paolo -- Paolo Giannozzi e-mail: p.giannozzi at sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From etabel at hotmail.com Mon Jun 26 12:40:44 2006 From: etabel at hotmail.com (Eric Abel) Date: Mon, 26 Jun 2006 06:40:44 -0400 Subject: [Pw_forum] Phonon atomic displacement units Message-ID: Hello, I have a question about the units of the atomic displacments in the output of matdyn.x. This issue has been addressed before, and I have read over those threads, but I have a add-on question. First I assume that matdyn.x has the same output units as ph.x, which is 1/m^(1/2), where m is the mass in atomic rydberg units. This is fine and dandy, however, I was wondering how to get from this to a unit of length, as in the end, that is the meaning of the atomic displacement eigenvector. Perhaps it would help to indicate that what I would like to do in the end is calculate a inelastic x-ray scattering cross section which to first order goes like modsquared[Sum on i {(Zi*exp(q * ui)}]. Clearly ui needs to have the units of length in order to get the proper phase in the exponential. Thank you in advance for your advice. Eric From sir_puding at tut.by Mon Jun 26 13:18:08 2006 From: sir_puding at tut.by (sir_puding at tut.by) Date: Mon, 26 Jun 2006 15:18:08 +0400 Subject: [Pw_forum] Charged systems calculation In-Reply-To: References: <35237.202.41.111.151.1151038397.squirrel@202.41.111.151> Message-ID: <1988487089.20060626151808@tut.by> Hi, All. Maybe this is an rtfm-question, then sorry. How can i calculate electronic structure of such charged system: 1. in general system in neutral, for example this it nanotube (closed) 2. we charge it by adding additional electrons from substrate/contacts or other (for example polarize it by an external field. Field emission)... Then i want to calculate the conductance of this system. Maybe someone know the way how to do this with espresso? Sergey. From marzari at MIT.EDU Mon Jun 26 14:25:38 2006 From: marzari at MIT.EDU (Nicola Marzari) Date: Mon, 26 Jun 2006 20:25:38 +0800 Subject: [Pw_forum] Charged systems calculation In-Reply-To: <1988487089.20060626151808@tut.by> References: <35237.202.41.111.151.1151038397.squirrel@202.41.111.151> <1988487089.20060626151808@tut.by> Message-ID: <449FD242.6060608@mit.edu> 1) have a look at papers on charged systems, from the Makov-Payne PRB 1995 to the PRLs et al from Peter Schulz (including one in the last month) 2) charging is a different problem than polarizing. Understanding where an extra electron would go in a contacted nanotube is complicate, and it is not evident to me that GGAs would give a correct answer. Polarization by an external field is much simpler (see e.g. PRL 2006 from our group. 3) for transport and quantum-espresso, have a look at the various papers by Calzolari, Nardelli, ourselves, and the www.wannier-transport.org (and www.wannier.org) websites. My advice would be to master one of these topics (1, 2 or 3) before trying to combine all of them together. nicola sir_puding at tut.by wrote: > Hi, All. > > Maybe this is an rtfm-question, then sorry. > > How can i calculate electronic structure of such charged system: > 1. in general system in neutral, for example this it nanotube (closed) > 2. we charge it by adding additional electrons from substrate/contacts > or other (for example polarize it by an external field. Field > emission)... > > Then i want to calculate the conductance of this system. > > Maybe someone know the way how to do this with espresso? > > > Sergey. > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From Steven.Kirk at hv.se Tue Jun 27 12:34:43 2006 From: Steven.Kirk at hv.se (Steven Kirk) Date: Tue, 27 Jun 2006 12:34:43 +0200 Subject: [Pw_forum] Strange pp.x output after vc-relax (v3.1) Message-ID: <44A109C3.4080602@hv.se> Hello, I have been trying to run a vc-relax calculation using pw.x (Q-E v3.1) to get a relaxed configuration of a unit cell at a specified external pressure, then dumping the relaxed cell configuration and charge density distribution using pp.x. My input data is below pw.x input ========== &CONTROL title='iceXI calculation', calculation='vc-relax', restart_mode='from_scratch', prefix='iceXI', dt=30.0, nstep = 100, tstress = .true., tprnfor = .true. / &SYSTEM ibrav = 8, A = 4.383, B = 7.623, C = 7.163, cosAB = 0, cosAC = 0, cosBC = 0, nat = 24, ntyp = 2, ecutwfc = 100.0 / &ELECTRONS / &IONS ion_dynamics='damp' / &CELL cell_dynamics='damp-w', press = 100.0 wmass = 0.0004 / ATOMIC_SPECIES O 15.9994 O.pbe-rrkjus.UPF H 1.00794 H.pbe-rrkjus.UPF ATOMIC_POSITIONS {crystal} O 0.00000 0.66600 0.05900 O 0.50000 0.16600 0.05900 O 0.00000 0.33400 0.55900 O 0.50000 0.83400 0.55900 O 0.50000 0.83300 0.93400 O 0.00000 0.33300 0.93400 O 0.50000 0.16700 0.43400 O 0.00000 0.66700 0.43400 H 0.68300 0.76800 0.98100 H 0.18300 0.26800 0.98100 H 0.31700 0.23200 0.48100 H -0.18300 0.73200 0.48100 H 0.68300 0.23200 0.48100 H 0.18300 0.73200 0.48100 H 0.31700 0.76800 0.98100 H -0.18300 0.26800 0.98100 H 0.00000 0.66400 0.19800 H 0.00000 0.54100 0.01700 H 0.50000 0.16400 0.19800 H 0.50000 0.04100 0.01700 H 0.00000 0.33600 0.69800 H 0.00000 0.45900 0.51700 H 0.50000 0.83600 0.69800 H 0.50000 0.95900 0.51700 K_POINTS automatic 4 4 4 0 0 0 pp.x input ========== &INPUTPP prefix = 'iceXI' , filplot = 'iceXI.chdens' , plot_num = 0, / &PLOT nfile = 1 , filepp(1) = 'iceXI.chdens', weight(1) = 1.0, fileout = 'iceXI.xsf' , iflag = 3 , output_format = 3 , / =========== The problem is that the output cell parameters (at least) in the output .XSF file are the ORIGINAL cell parameters, not the RELAXED cell parameters. What has gone wrong here ? Why doesn't my output XSF file contain the details for the relaxed cell ? Are the XSF lattice parameters , the XSF atomic coordinates or the 3D charge density grid wrong, or possibly all three? Many thanks! Steve Kirk -- Dr. Steven R. Kirk Dept. of Technology, Mathematics & Computer Science (P)+46 520 223215 University West (F)+46 520 223299 P.O. Box 957 Trollhattan 461 29 SWEDEN http://taconet.webhop.org From giannozz at nest.sns.it Tue Jun 27 17:13:57 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 27 Jun 2006 17:13:57 +0200 Subject: [Pw_forum] Strange pp.x output after vc-relax (v3.1) In-Reply-To: <44A109C3.4080602@hv.se> References: <44A109C3.4080602@hv.se> Message-ID: <200606271713.57111.giannozz@nest.sns.it> On Tuesday 27 June 2006 12:34, Steven Kirk wrote: > The problem is that the output cell parameters (at least) in the output > .XSF file are the ORIGINAL cell parameters, not the RELAXED cell > parameters. are you really sure you are reading the final data files and not something else? Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From jcconesa at icp.csic.es Tue Jun 27 18:19:49 2006 From: jcconesa at icp.csic.es (=?ISO-8859-1?Q?Jos=E9_C._Conesa?=) Date: Tue, 27 Jun 2006 18:19:49 +0200 Subject: [Pw_forum] use of CASTEP USPP parameters Message-ID: <44A176C5.5799.41104394@localhost> Hi all, I am a licensed user of Accelrys' CASTEP package, which includes a fairly extense library of ultrasoft pseudopotentials. The files in it have (apart from a propietary format) a heading with pseudopotential- determining parameters, but the format and contents of this heading is to my knowledge not documented. I should like to have a general procedure allowing to obtain, from the data in that heading, a parameter file suitable as input to Vanderbilt's USPP generating code, so that in the end a pseudopotential utilizable in Espresso, and equivalent to one in the CASTEP library, can be obtained. I know from Accelrys that a licensed CASTEP user is allowed to do this. Has any of you experience in such a conversion? Can you facilitate to me such a procedure? Thanks in advance, Jose C. Conesa Instituto de Catalisis y Petroleoquimica, CSIC Campus de Cantoblanco 28049 Madrid - Spain Phone Nr. 34-91-5854766 Fax Nr. 34-91-5854760 From vranjan at ncsu.edu Wed Jun 28 05:20:55 2006 From: vranjan at ncsu.edu (Vivek Ranjan) Date: Tue, 27 Jun 2006 23:20:55 -0400 (EDT) Subject: [Pw_forum] warning with configure on origin Message-ID: <45343.70.150.217.227.1151464855.squirrel@webmail.ncsu.edu> I am trying to compile pwscf on multiprocessor sgi origin machine. I run configure, and I get the following warning messages : Cannot open ../iotk/src/iotk_attr+COMP: No such file or directory pw_export.o : @iotk_module@ WARNING: dependencies not found in directory PP Cannot open ../iotk/src/iotk_at: No such file or directory directory PWCOND : ok Cannot open ../iotk/src/iotk: No such file or directory directory Gamma : ok directory PH : ok allocate_d3.o : @phcom@ bcast_d3_input.o : @phcom@ ch_psi_all2.o : @phcom@ close_open.o : @phcom@ d0rhod2v.o : @phcom@ d3dyn_cc.o : @phcom@ d3_exc.o : @phcom@ d3_init.o : @phcom@ d3matrix.o : @phcom@ d3_readin.o : @phcom@ d3_recover.o : @phcom@ d3_setup.o : @phcom@ d3_summary.o : @phcom@ d3toten.o : @phcom@ d3_valence.o : @phcom@ d3vrho.o : @phcom@ davcio_drho2.o : @phcom@ dpsi_corr.o : @phcom@ dpsidpsidv.o : @phcom@ dpsidvdpsi.o : @phcom@ dqrhod2v.o : @phcom@ drho_cc.o : @phcom@ drhod2v.o : @phcom@ drho_drc.o : @phcom@ dvdpsi.o : @phcom@ dvscf.o : @phcom@ gen_dpdvp.o : @phcom@ gen_dwfc.o : @phcom@ incdrhoscf2.o : @phcom@ openfild3.o : @phcom@ psymd0rho.o : @phcom@ set_d3irr.o : @phcom@ set_efsh.o : @phcom@ solve_linter_d3.o : @phcom@ stop_d3.o : @phcom@ sym_def1.o : @phcom@ write_aux.o : @phcom@ writed3dyn_5.o : @phcom@ write_igk.o : @phcom@ WARNING: dependencies not found in directory D3 directory atomic : ok directory Nmr : ok directory VIB : ok directory VdW : ok configure: creating ./config.status config.status: creating make.sys config.status: creating configure.msg -------------------------------------------------------------------------- Parallel environment detected successfully. Configured for compilation of parallel executables. For more info, read the ESPRESSO User's Guide (Doc/users-guide.tex). -------------------------------------------------------------------- configure: success Can anyone help me with their advice and how to deal with it ? Thanks in advance, Vivek From akohlmey at vitae.cmm.upenn.edu Wed Jun 28 06:04:00 2006 From: akohlmey at vitae.cmm.upenn.edu (Axel Kohlmeyer) Date: Wed, 28 Jun 2006 00:04:00 -0400 (EDT) Subject: [Pw_forum] warning with configure on origin In-Reply-To: <45343.70.150.217.227.1151464855.squirrel@webmail.ncsu.edu> Message-ID: On Tue, 27 Jun 2006, Vivek Ranjan wrote: VR> I am trying to compile pwscf on multiprocessor sgi origin machine. I run VR> configure, and I get the following warning messages : VR> hmm, this looks very similar to the report we had recently from a windows machine. please try changing at the very end of modulep.sh 'sort -d' into 'sort' and then re-run ./makedeps.sh and then 'make clean' and finally 'make all'. let is know whether this helps. cheers, axel. VR> VR> VR> Cannot open ../iotk/src/iotk_attr+COMP: No such file or directory VR> pw_export.o : @iotk_module@ VR> WARNING: dependencies not found in directory PP VR> Cannot open ../iotk/src/iotk_at: No such file or directory VR> directory PWCOND : ok VR> Cannot open ../iotk/src/iotk: No such file or directory VR> directory Gamma : ok VR> directory PH : ok VR> allocate_d3.o : @phcom@ VR> bcast_d3_input.o : @phcom@ VR> ch_psi_all2.o : @phcom@ VR> close_open.o : @phcom@ VR> d0rhod2v.o : @phcom@ VR> d3dyn_cc.o : @phcom@ VR> d3_exc.o : @phcom@ VR> d3_init.o : @phcom@ VR> d3matrix.o : @phcom@ VR> d3_readin.o : @phcom@ VR> d3_recover.o : @phcom@ VR> d3_setup.o : @phcom@ VR> d3_summary.o : @phcom@ VR> d3toten.o : @phcom@ VR> d3_valence.o : @phcom@ VR> d3vrho.o : @phcom@ VR> davcio_drho2.o : @phcom@ VR> dpsi_corr.o : @phcom@ VR> dpsidpsidv.o : @phcom@ VR> dpsidvdpsi.o : @phcom@ VR> dqrhod2v.o : @phcom@ VR> drho_cc.o : @phcom@ VR> drhod2v.o : @phcom@ VR> drho_drc.o : @phcom@ VR> dvdpsi.o : @phcom@ VR> dvscf.o : @phcom@ VR> gen_dpdvp.o : @phcom@ VR> gen_dwfc.o : @phcom@ VR> incdrhoscf2.o : @phcom@ VR> openfild3.o : @phcom@ VR> psymd0rho.o : @phcom@ VR> set_d3irr.o : @phcom@ VR> set_efsh.o : @phcom@ VR> solve_linter_d3.o : @phcom@ VR> stop_d3.o : @phcom@ VR> sym_def1.o : @phcom@ VR> write_aux.o : @phcom@ VR> writed3dyn_5.o : @phcom@ VR> write_igk.o : @phcom@ VR> WARNING: dependencies not found in directory D3 VR> directory atomic : ok VR> directory Nmr : ok VR> directory VIB : ok VR> directory VdW : ok VR> configure: creating ./config.status VR> config.status: creating make.sys VR> config.status: creating configure.msg VR> -------------------------------------------------------------------------- VR> Parallel environment detected successfully. VR> Configured for compilation of parallel executables. VR> VR> For more info, read the ESPRESSO User's Guide (Doc/users-guide.tex). VR> -------------------------------------------------------------------- VR> configure: success VR> VR> VR> Can anyone help me with their advice and how to deal with it ? VR> VR> Thanks in advance, VR> VR> Vivek VR> VR> VR> VR> VR> _______________________________________________ VR> Pw_forum mailing list VR> Pw_forum at pwscf.org VR> http://www.democritos.it/mailman/listinfo/pw_forum VR> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From etabel at hotmail.com Wed Jun 28 09:42:41 2006 From: etabel at hotmail.com (Eric Abel) Date: Wed, 28 Jun 2006 03:42:41 -0400 Subject: [Pw_forum] Neutrons, Xrays, and Phonons, Oh, My! Message-ID: Hello pwscf users, I recently made a rather lengthy cry for help on the phonon eigendisplacements, which fell, it seems, on deaf ears. So I want to place the question again plain and simple: Has anyone successfully calculated the Neutron/Xray scattering cross-section based on the results of ph.x, matdyn.x, or dynmat.x? If so I would like to pick your brain a little. Eric P.S. Stefano, Paolo, and friends...PWscf is a great piece of software. I applaud you for bringing ab-initio to the masses! From marzari at MIT.EDU Wed Jun 28 10:39:35 2006 From: marzari at MIT.EDU (Nicola Marzari) Date: Wed, 28 Jun 2006 16:39:35 +0800 Subject: [Pw_forum] Neutrons, Xrays, and Phonons, Oh, My! In-Reply-To: References: Message-ID: <44A24047.9080409@mit.edu> Hi Eric, you could look at "Neutron static structure factor" in Paolo Umari's thesis: http://library.epfl.ch/en/theses/?nr=2774 It's CP, not PWSCF, but the ingredients are all there. nicola Eric Abel wrote: > Hello pwscf users, > > I recently made a rather lengthy cry for help on the phonon > eigendisplacements, which fell, it seems, on deaf ears. So I want to > place the question again plain and simple: > > Has anyone successfully calculated the Neutron/Xray scattering > cross-section based on the results of ph.x, matdyn.x, or dynmat.x? If > so I would like to pick your brain a little. > > Eric > > P.S. Stefano, Paolo, and friends...PWscf is a great piece of software. > I applaud you for bringing ab-initio to the masses! > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From giannozz at nest.sns.it Wed Jun 28 11:34:15 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Wed, 28 Jun 2006 11:34:15 +0200 Subject: [Pw_forum] Neutrons, Xrays, and Phonons, Oh, My! In-Reply-To: References: Message-ID: <200606281134.15728.giannozz@nest.sns.it> On Wednesday 28 June 2006 09:42, Eric Abel wrote: > I recently made a rather lengthy cry for help on the phonon > eigendisplacements, which fell, it seems, on deaf ears. my ears are deaf to such questions because the ratio between "effort required to answer" versus "information provided" is unfavourable. All you need is to know whether what is printed or calculated is the eigenvector of the dynamical matrix or the eigendisplacements. This piece of information is usually written inside the various codes that write/calculate/process the dynamical matrix. > Has anyone successfully calculated the Neutron/Xray scattering > cross-section based on the results of ph.x, matdyn.x, or dynmat.x? IR and Raman cross sections are calculated by dynmat.x Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From giannozz at nest.sns.it Wed Jun 28 11:49:42 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Wed, 28 Jun 2006 11:49:42 +0200 Subject: [Pw_forum] warning with configure on origin In-Reply-To: <45343.70.150.217.227.1151464855.squirrel@webmail.ncsu.edu> References: <45343.70.150.217.227.1151464855.squirrel@webmail.ncsu.edu> Message-ID: <200606281149.42575.giannozz@nest.sns.it> On Wednesday 28 June 2006 05:20, Vivek Ranjan wrote: > I am trying to compile pwscf on multiprocessor sgi origin machine. > I run configure, and I get the following warning messages : > Cannot open ../iotk/src/iotk_attr+COMP: No such file or directory somebody else had the same problem on a sgi origin. It is a bug in some shell command used by the script that finds dependencies. Get the */make.depend files from another machine and overwrite those in the sgi origin. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From kajewski at us.edu.pl Wed Jun 28 16:54:42 2006 From: kajewski at us.edu.pl (Dariusz Kajewski) Date: Wed, 28 Jun 2006 17:54:42 +0300 Subject: [Pw_forum] external static electric field Message-ID: <1151506482.13476.1.camel@localhost.localdomain> Hello! I'm a fresh user of Q-E so please forgive me for any silly mistakes. I was browsing the forum archive but I couldn't find any solution for my problem - i try to make some calculations aboout ferroelectric materials - polarization via Berry Phase method, which seems to be verry good at the moment for me. But I need to aplay an electric field and relax the structure under it's influence to calculte polarization after that. I try to do that for BaTiO3 and without field it works great but i don't know what's wrong with my input file, which is: &CONTROL calculation = "relax", restart_mode='from_scratch', prefix = "bto", pseudo_dir = "../pseudo", outdir = "./", lelfield = .true., gdir = 3, nppstr = 7, nberrycyc=1 / &SYSTEM ibrav=6, celldm(1)=7.541801934, celldm(3)=1.011087586, nat=5, ntyp=3, degauss=0.00, nbnd=20, occupations='fixed', ecutwfc = 20.0, ecutrho =120.0 / &ELECTRONS conv_thr = 1.0d-12, mixing_beta = 0.7, efield=0.0001 / &IONS ion_dynamics = 'bfgs' pot_extrapolation = 'second_order' / ATOMIC_SPECIES O 15.9994 O.pw91-van_ak.UPF Ti 47.867 Ti.pw91-nsp-van.UPF Ba 137.327 Ba.pw91-nsp-van.UPF ATOMIC_POSITIONS crystal Ba 0.0 0.0 0.0 Ti 0.5 0.5 0.5224 O 0.5 0.5 -0.0244 O 0.5 0.0 0.4895 O 0.0 0.5 0.4895 K_POINTS automatic 4 4 4 1 1 1 When i tray to make calculations with that program gaves me such an answer after " Self-consistent Calculation iteration # 1 ecut= 20.00 ryd beta=0.70 Davidson diagonalization with overlap error hepsiher: translated G= -0.145833333333333 -0.145833333333333 -0.144234125067513 with crystal coordinates 0 0 0 corresponds to ng= 1 but G(ng)= 0.000000000000000E+000 0.000000000000000E+000 0.000000000000000E+000 probably because G_par is NOT a reciprocal lattice vector Possible choices as smallest G_par: " what I'm doing wrong? Please help me :) thank you in advance Dariusz Kajewski From jasonsun98 at hotmail.com Wed Jun 28 17:34:07 2006 From: jasonsun98 at hotmail.com (sun jason) Date: Wed, 28 Jun 2006 15:34:07 +0000 Subject: [Pw_forum] about nonlinear core-correction for pseudopotentials In-Reply-To: <20060628054617.32245.51340.Mailman@democritos.sissa.it> Message-ID: Dear all, 1) on the pwscf website, why all the ultrasoft pseudopotentials with GGA for boron are all include nonlinear core-correction, while this correction is not considered for carbon and nitrogen ? Is that means this correction is very important for boron but not for carbon and nitrogen ? 2) Is there any relation between PBE exchange-correlation ? I find the LDA norm-conserving PPs are not include NLCC. 3) I will very appreciate anyone who would send me a GGA pseudopotential without nonlinear core-correction for boron. I want to exemplify the changes induced by this approximation. Thank you very much in advance. Best regards, =============================================== Jian SUN Physics Dept. of Nanjing University National Lab. of Solid State Microstructures 22 Hankou Road, Gulou District Nanjing, Jiangsu Province 210093 China =============================================== From etabel at hotmail.com Thu Jun 29 00:14:38 2006 From: etabel at hotmail.com (Eric Abel) Date: Wed, 28 Jun 2006 18:14:38 -0400 Subject: [Pw_forum] Neutrons, Xrays, and Phonons, Oh, My! In-Reply-To: <200606281134.15728.giannozz@nest.sns.it> Message-ID: >From: Paolo Giannozzi >Reply-To: pw_forum at pwscf.org >To: pw_forum at pwscf.org >Subject: Re: [Pw_forum] Neutrons, Xrays, and Phonons, Oh, My! >Date: Wed, 28 Jun 2006 11:34:15 +0200 > >On Wednesday 28 June 2006 09:42, Eric Abel wrote: > > > I recently made a rather lengthy cry for help on the phonon > > eigendisplacements, which fell, it seems, on deaf ears. > >my ears are deaf to such questions because the ratio between >"effort required to answer" versus "information provided" is >unfavourable. "unfavourable" us subjective...to me the information is valuable, therefore the ratio is "favourable". >All you need is to know whether what is printed >or calculated is the eigenvector of the dynamical matrix or the >eigendisplacements. This piece of information is usually written >inside the various codes that write/calculate/process the >dynamical matrix. I couldn't agree with you more...and if this were my only problem, I never would have written you. Clearly, from the snippet of code below, what is written by the subroutine dyndia is |v>*M^-1/2, thus making the output of the program the eigendisplacement right? Code snippet: -- write displacements onto matrix dyn ! unorm = 0.d0 do mu = 1, 3 * nat na = (mu - 1) / 3 + 1 dyn (mu, nu_i) = z (mu, nu_i) / sqrt (amass (ityp (na) ) ) unorm = unorm + dyn (mu, nu_i) * conjg (dyn (mu, nu_i) ) enddo However, this doesn't really address my question. What's bothering me is the following. I always thought of eigenvectors as being normalized unitvectors which provide only the "direction" rather than the "magnitude", such as the pauli spinors. The units are attached to the prefactor on the vector. In this case the prefactor is 1/sqrt(M). Unless I'm mistaken, |u> should have the unit length, and none of my physics books are telling me how to convert from [1/sqrt(mass)] to [length]. The other possibilty is that |v> has the unit of length, then |u> has the units [lenght/sqrt(mass)], meaning to get the displacement length, all I have to do is multiply the output of dyndia by sqrt(M), but I didn't find this scenerio so likely. Perhaps the simplest, most transparent way to ask my question is this: If matdyn/dynmat/ph (I believe use the dyndia subroutine) spits out the value 0.4 for one of the elements in the eigendisplacement vector, what, then does this correspond to in Angstroms, for example? > > Has anyone successfully calculated the Neutron/Xray scattering > > cross-section based on the results of ph.x, matdyn.x, or dynmat.x? > >IR and Raman cross sections are calculated by dynmat.x I noticed this...unfortunately for me, however, the interaction interaction in neutron/x-ray scattering is entirely different from that of Raman/IR. Thsnk you, Paolo, for taking the time to address my "unfavorable" question. Eric >Paolo >-- >Paolo Giannozzi e-mail: giannozz at nest.sns.it >Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 >Piazza dei Cavalieri 7 I-56126 Pisa, Italy >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum From degironc at sissa.it Thu Jun 29 11:30:08 2006 From: degironc at sissa.it (degironc) Date: Thu, 29 Jun 2006 11:30:08 +0200 Subject: [Pw_forum] Neutrons, Xrays, and Phonons, Oh, My! In-Reply-To: References: Message-ID: <44A39DA0.6040609@sissa.it> dear Eric, My understanfing of the matter is the following: Let's have a system that is described by an harmonic hamiltonian H = \sum_i 1/(2M_i) P_i^2 + 1/2 \sum_ij U_i Phi_ij U_j and let's consider the eigenvalue problem \sum_j (Phi_ij/sqrt(M_i M_j) v_j^nu = (omega^nu) ^2 v_i^nu The canonical transformation from ordinary cartesian coordinates (U_i,P_i) to normal mode coordinates (csi_nu, pi_nu) U_i = \sum_nu v_i^nu / sqrt(M_i) csi_nu P_i = \sum_nu v_i^nu * sqrt(M_i) pi_nu with its inverse csi_nu = sum_i v_i^nu * sqrt(M_i) U_i pi_nu = sum_i v_i^nu / sqrt(M_i) P_i the unitary matrix v_i^nu has no "units" .... it is a rotation in coordinate space. For instance, using this change of variables the hamiltonian is H = sum_nu 1/2 pi_nu ^2 + 1/2 (omega^nu)^2 csi_nu^2 In order to calculate whatever cross-section you are interested in, you must similarly use the same tranformation to write things in terms of normal-mode amplitudes and then plug in the value corresponding to the temperature you are interested (for instance at zero temperature: 1/2 pi_nu^2 = 1/2 omega_nu^2 csi_nu^2 = hbar omega^nu / 4) . At least that would be what I would try to do. stefano From giannozz at nest.sns.it Thu Jun 29 11:48:13 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 29 Jun 2006 11:48:13 +0200 Subject: [Pw_forum] about nonlinear core-correction for pseudopotentials In-Reply-To: References: Message-ID: <200606291148.13903.giannozz@nest.sns.it> On Wednesday 28 June 2006 17:34, sun jason wrote: > 1) on the pwscf website, why all the ultrasoft pseudopotentials with GGA > for boron are all include nonlinear core-correction, while this correction > is not considered for carbon and nitrogen ? Is that means this correction > is very important for boron but not for carbon and nitrogen ? typically the core correction is more important for atoms with few valence electrons > 2) Is there any relation between PBE exchange-correlation ? I find the LDA > norm-conserving PPs are not include NLCC. no direct relationship, but gradient-corrected functionals tend to be ill-behaved when rho(r) => 0, i.e. at large r but also at small r for pseudopotentials. This may lead to instabilities in the pseudopotential generation procedure, or to pseudopotentials with an unpleasant "spike" at r=>0. With the core correction there is no such problem. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From jcconesa at icp.csic.es Thu Jun 29 12:27:38 2006 From: jcconesa at icp.csic.es (=?ISO-8859-1?Q?Jos=E9_C._Conesa?=) Date: Thu, 29 Jun 2006 12:27:38 +0200 Subject: [Pw_forum] warning with configure on origin In-Reply-To: <200606281149.42575.giannozz@nest.sns.it> References: <45343.70.150.217.227.1151464855.squirrel@webmail.ncsu.edu> Message-ID: <44A3C73A.7832.43606AE@localhost> On 28 Jun 2006 at 11:49, Paolo Giannozzi wrote: > On Wednesday 28 June 2006 05:20, Vivek Ranjan wrote: > > > I am trying to compile pwscf on multiprocessor sgi origin machine. > > I run configure, and I get the following warning messages : > > > Cannot open ../iotk/src/iotk_attr+COMP: No such file or directory > > somebody else had the same problem on a sgi origin. It is a bug in > some shell command used by the script that finds dependencies. > Get the */make.depend files from another machine and overwrite > those in the sgi origin. > > Paolo I had that problem (message sent on May 23th to Pw_forum). A reply by Andrea Ferretti on May 24th gave the clue for a way to solve it without having to generate the make.depend files elsewhere: Right after having executed configure, execute from the main directory: grep @ */make.depend Any lines revealed there which have a name bracketed between two @ characters indicate a dependency that should have been solved but was not. You should change these offending lines in their respective make.depend files in order to have, instead of @name@, a reference to an object file. Normally the name of this file is name.o preceded by the appropriate path. The path where such file is to be located must thus be identified. At the time of making such change in the make.depend files, i.e. before running make, the name.o file does not exist yet, but a corresponding name.f90 file should be already in the corresponding directory, that is how you can locate it. Once these @name@ occurrencies are substituted, running make should find no more this problem. In my case, I had offending files in the D3 and PP directories. I cured the problem with the following: cd D3 cp make.depend make.depend.tmp sed "s/@phcom@/..\/PH\/phcom.o/" make.depend.tmp >make.depend rm make.depend.tmp cd ../PP cp make.depend make.depend.tmp sed "s/@iotk_module@/..\/iotk\/src\/iotk_module.o/" make.depend.tmp >make.depend rm make.depend.tmp cd .. The first sed execution substituted many lines; the second one only one (thus in that case it was easier to do it with the vi text editor) Hope this helps Jose C. Conesa Instituto de Catalisis y Petroleoquimica, CSIC Campus de Cantoblanco 28049 Madrid - Spain Phone Nr. 34-91-5854766 Fax Nr. 34-91-5854760 From marzari at MIT.EDU Thu Jun 29 13:11:56 2006 From: marzari at MIT.EDU (Nicola Marzari) Date: Thu, 29 Jun 2006 19:11:56 +0800 Subject: [Pw_forum] wiki for Neutrons, Xrays, and Phonons, Oh, My! In-Reply-To: <44A39DA0.6040609@sissa.it> References: <44A39DA0.6040609@sissa.it> Message-ID: <44A3B57C.3060508@mit.edu> Stefano, mi controlli questo: mancano delle parentesi, ma sono di frettissima. nicola http://www.quantum-espresso.org/wiki/index.php/Dynamical_matrix --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From marzari at MIT.EDU Thu Jun 29 13:16:28 2006 From: marzari at MIT.EDU (Nicola Marzari) Date: Thu, 29 Jun 2006 19:16:28 +0800 Subject: [Pw_forum] Pretty please In-Reply-To: <44A3B57C.3060508@mit.edu> References: <44A39DA0.6040609@sissa.it> <44A3B57C.3060508@mit.edu> Message-ID: <44A3B68C.5030900@mit.edu> Dear All, can we change the default Reply-to field of pw_forum, so that the default reply goes only to the sender, and one needs a conscious or semiconscious effort to send it also to pw_forum ? Of course, as it is now it's more fun... nicola --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From akohlmey at cmm.upenn.edu Thu Jun 29 17:57:04 2006 From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer) Date: Thu, 29 Jun 2006 11:57:04 -0400 Subject: [Pw_forum] Pretty please In-Reply-To: <44A3B68C.5030900@mit.edu> References: <44A39DA0.6040609@sissa.it> <44A3B57C.3060508@mit.edu> <44A3B68C.5030900@mit.edu> Message-ID: <7b6913e90606290857p1b9bfbf1j1453713ebfd40f99@mail.gmail.com> On 6/29/06, Nicola Marzari wrote: > > > Dear All, dear nicola, > can we change the default Reply-to field of pw_forum, so that > the default reply goes only to the sender, and one needs a > conscious or semiconscious effort to send it also to pw_forum ? in my opinion that would be worse. because that would have many reponses go only to the people asking and thus defeat the purpose of a mailing list, that everybody can read the solution of a problem and learn from it. > Of course, as it is now it's more fun... ...and you get lessons in italian for free. ;-) ciao, axel. > > nicola > > > --------------------------------------------------------------------- > Prof Nicola Marzari Department of Materials Science and Engineering > 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA > tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From lukethulin at netscape.net Thu Jun 29 19:15:04 2006 From: lukethulin at netscape.net (Luke Thulin) Date: Thu, 29 Jun 2006 13:15:04 -0400 Subject: [Pw_forum] Meaning of Press variable Message-ID: <44A40A98.6010703@netscape.net> Hello everyone, Can someone please explain more fully the meaning of the variable press under &Cell? I don't understand what is meant by 'target pressure.' Thanks, Luke From etabel at hotmail.com Fri Jun 30 04:08:08 2006 From: etabel at hotmail.com (Eric Abel) Date: Thu, 29 Jun 2006 22:08:08 -0400 Subject: [Pw_forum] Neutrons, Xrays, and Phonons, Oh, My! In-Reply-To: <44A39DA0.6040609@sissa.it> Message-ID: Dear all, Thank you all for your input. I found the solution to my problem. First of all, it seems that "cross-section" may have been a poor choice of words. What I really want to calculate is the dynamical structure factor, S(Q,\omega). This is simply (for mode s): S(Q,\omega)=\sum_j b_j/sqrt(m_j) * (Q \dot v_{js})exp(i Q \dot R_j)exp(W_j) where b_j is the neutron scattering cross-section of atom j, m_j is the mass of atom j, Q is the reciprocal space vector, v_{js} is the eigenvector and R_j is the position in the unit cell of atom j. The exp(W_j) is the debye waller factor. For x-rays I simply need to replace the b_j with the x-ray form factor f_j(Q). In the end, the unitless v_{js} of pwscf is what I need. Thank you for your valuable input. I learned a lot from this discussion. In the end, this is what being a grad student is all about, right? Eric >From: degironc >Reply-To: pw_forum at pwscf.org >To: pw_forum at pwscf.org >Subject: Re: [Pw_forum] Neutrons, Xrays, and Phonons, Oh, My! >Date: Thu, 29 Jun 2006 11:30:08 +0200 > >dear Eric, > >My understanfing of the matter is the following: > >Let's have a system that is described by an harmonic hamiltonian > >H = \sum_i 1/(2M_i) P_i^2 + 1/2 \sum_ij U_i Phi_ij U_j > >and let's consider the eigenvalue problem >\sum_j (Phi_ij/sqrt(M_i M_j) v_j^nu = (omega^nu) ^2 v_i^nu > >The canonical transformation from ordinary cartesian coordinates >(U_i,P_i) to normal mode coordinates (csi_nu, pi_nu) > >U_i = \sum_nu v_i^nu / sqrt(M_i) csi_nu >P_i = \sum_nu v_i^nu * sqrt(M_i) pi_nu > >with its inverse > >csi_nu = sum_i v_i^nu * sqrt(M_i) U_i >pi_nu = sum_i v_i^nu / sqrt(M_i) P_i > >the unitary matrix v_i^nu has no "units" .... it is a rotation in >coordinate space. > >For instance, using this change of variables the hamiltonian is > >H = sum_nu 1/2 pi_nu ^2 + 1/2 (omega^nu)^2 csi_nu^2 > >In order to calculate whatever cross-section you are interested in, you >must similarly use the same tranformation to write things in terms of >normal-mode amplitudes and then plug in the value corresponding to the >temperature you are interested (for instance at zero temperature: >1/2 pi_nu^2 = 1/2 omega_nu^2 csi_nu^2 = hbar omega^nu / 4) . > >At least that would be what I would try to do. > >stefano > >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum From giannozz at nest.sns.it Fri Jun 30 11:41:15 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 30 Jun 2006 11:41:15 +0200 Subject: [Pw_forum] Meaning of Press variable In-Reply-To: <44A40A98.6010703@netscape.net> References: <44A40A98.6010703@netscape.net> Message-ID: <200606301141.15399.giannozz@nest.sns.it> On Thursday 29 June 2006 19:15, Luke Thulin wrote: > Can someone please explain more fully the meaning of the variable press > under &Cell? I don't understand what is meant by 'target pressure.' external pressure applied to the system. "Target" because during a simulation the actual pressure will oscillate around that value Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From cesards at msi.umn.edu Fri Jun 30 23:42:19 2006 From: cesards at msi.umn.edu (cesards at msi.umn.edu) Date: Fri, 30 Jun 2006 16:42:19 -0500 (CDT) Subject: [Pw_forum] Meaning of press parameter Message-ID: <51088.128.101.191.226.1151703739.squirrel@www.msi.umn.edu> Press is the external pressure. In variable cell shape MD, the external pressure is a parameter in the Lagrangian of the system. If you are optimizing cell shape with VCS-MD in damped mode, pwscf will find the cell shape that makes the diagonal components of the stress tensor equal the external pressure and the non-diagonal ones equal zero. This is why it is called ?target? pressure. Cesar R. S. da Silva