From yuwen_66 at yahoo.com Fri Jul 1 04:48:33 2005 From: yuwen_66 at yahoo.com (W. YU) Date: Thu, 30 Jun 2005 19:48:33 -0700 (PDT) Subject: [Pw_forum] Noncollinear magnetism Message-ID: <20050701024833.58158.qmail@web51010.mail.yahoo.com> Dear Pw users, I have several questions about non collinear magnetism calculation as shown in example13. The following is quoted from the calculation for Fe with penalty functional: ----------------------------------------------------- # self-consistent calculation with penalty functional &system ibrav = 3, celldm(1) =5.217, nat= 1, ntyp= 1, ecutwfc = 25.0,ecutrho = 200.0, report=1, occupations='smearing', smearing='gaussian', degauss=0.05 noncolin = .true. starting_magnetization(1) = 0.5 angle1(1) = 90.0 angle2(1) = 0.0 mcons(1,1) = 0.0 mcons(2,1) = 0.5 mcons(3,1) = 0.0 i_cons = 1 lambda = 1 / ================================================= I have these questions: 1.From the input menu, I know the mcons(i,j) is the i'th component of j'th atom type where i runs from 1 to 3. But if angle1(1)=90.0 and angle2(1)=0.0, the magnetic moment should be along X axis, but the constraint seems to leave the Y axis unchanged is 2 represents Y axis. It seems that 1,2 and 3 represent Z, X and Y axses, am I right? --------------------------------------------------- 2. I hope to know how to choose lambda. why it is 1 here in the example. =================================================== The following is quoted from the calcualtion fo Ni: # self-consistent calculation cat > ni.scf.in << EOF &control calculation='scf' restart_mode='from_scratch', pseudo_dir = '$PSEUDO_DIR/', outdir='$TMP_DIR/' prefix='ni' / &system ibrav=2, celldm(1) =6.48, nat=1, ntyp=1, starting_magnetization(1)=0.7, ecutwfc = 24.0, ecutrho = 288.0, occupations='smearing', smearing='methfessel-paxton', degauss=0.02 noncolin = .true. starting_magnetization(1) = 0.1 angle1(1) = 90.0 angle2(1) = 0.0 / ==================================================== 3. Now the question is: Why there are two starting manetisation when there is only one type of Ni atom? 4. On what basis do we choose smearing scheme such as "methfessel-paxton" or "Gaussian"? 5. Last but most important question is: can we do calculations on spiral? If the answer is positive, do we have to use a very large supper cell? ====================================================== Your answers and comments would be much a preciated. W. YU __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From jasonsun98 at hotmail.com Fri Jul 1 05:41:37 2005 From: jasonsun98 at hotmail.com (sun jason) Date: Fri, 01 Jul 2005 03:41:37 +0000 Subject: [Pw_forum] about structural optimization In-Reply-To: <20050630181201.14994.52275.Mailman@democritos.sissa.it> Message-ID: dear all, I have some questions about structural optimization 1) can I just optimize the lattice constants but not change the atomic fractional coordinates? 2) does ion_dynamics = 'bfgs' can be use to vc-relax? I can run a vc-relax calculation when I put ion_dynamics = 'damp', and it says does not support 'bfgs' when I put ion_dynamics='bfgs' 3) I run an example to calculate the effective charges of cubic BN, to my knowledge, the absolute value of the effective charge for cation and anion should be equal in this case. but I get 1.759 and -1.736 for B and N atom, respectively. why? or just because I have not relax the cell? any help would be appreciate. ============================================== Jian Sun Physics Dept. of Nanjing University National Lab. of Solid State Microstructures 22 Hankou Road, Gulou District Nanjing, Jiangsu Province 210093 China ============================================== >From: pw_forum-request at pwscf.org >Reply-To: pw_forum at pwscf.org >To: pw_forum at pwscf.org >Subject: Pw_forum digest, Vol 1 #678 - 12 msgs >Date: Thu, 30 Jun 2005 20:12:01 +0200 > >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. i/o error in davcio in the phonon calculation (Jian ZHOU) > 2. Re: Fwd: Re: Segmentation fault with Setting 2 --- > follow-up of Re. ATOMIC_POSITIONS (Gerardo Ballabio) > 3. Re: RE: Pw_forum digest, Vol 1 #668 - 2 msgs (Gerardo Ballabio) > 4. about pwgui (sun jason) > 5. Re: Fwd: Re: Segmentation fault with Setting 2 --- > follow-up of Re. ATOMIC_POSITIONS (Guido Roma) > 6. Re: Fwd: Re: Segmentation fault with Setting 2 --- follow-up of Re. ATOMIC_POSITIONS (Eyvaz Isaev) > 7. Re: i/o error in davcio in the phonon calculation (Eyvaz Isaev) > 8. Re: how to find the space group and coordinates (Eduardo Ariel Menendez P) > 9. Understanding "eband" & "deband" in "total energy" (Mousumi Upadhyay Kahaly) > 10. Re: Understanding "eband" & "deband" in "total energy" (Konstantin Kudin) > >--__--__-- > >Message: 1 >Date: Thu, 30 Jun 2005 15:37:20 +0800 >From: Jian ZHOU >To: pw_forum at pwscf.org >Subject: [Pw_forum] i/o error in davcio in the phonon calculation >Reply-To: pw_forum at pwscf.org > >Dear all, > >I have compiled the pwscf v2.1.4 successfully and also have run the >example02 without any error. However, when I tried to do another >phonon calculation of Gamma point, it stops during running the ph.x. > >I firstly run a scf calculation from scratch and then do a gamma point >phonon calculation, reading the previous output files(The two outdirs >are the same in the two calculation). The following is the screen >output message of ph.x. > >---------------------------------------------------------------------------= >- >... > Atomic displacements: > There are 17 irreducible representations > > Representation 1 3 modes - To be done > > Representation 2 3 modes - To be done > > Representation 3 2 modes - To be done > > Representation 4 3 modes - To be done > > Representation 5 3 modes - To be done > > Representation 6 3 modes - To be done > > Representation 7 3 modes - To be done > > Representation 8 1 modes - To be done > > Representation 9 3 modes - To be done > > Representation 10 2 modes - To be done > > Representation 11 3 modes - To be done > > Representation 12 2 modes - To be done > > Representation 13 1 modes - To be done > > Representation 14 3 modes - To be done > > Representation 15 3 modes - To be done > > Representation 16 1 modes - To be done > > Representation 17 3 modes - To be done > IOS =3D 253 > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%= >%%%% > from davcio : error # 20 > i/o error in davcio > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%= >%%%% > > stopping ... >[0] MPI Abort by user Aborting program ! >[0] Aborting program! >p0_5498: p4_error: : 0 > >-----------------------------------------------stop here > >There is also a CRASH file with the content: > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%= >%%%% > task # 2 > from set_irr : error # 0 > npert > max_irr_dim > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%= >%%%% > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%= >%%%% > task # 3 > from set_irr : error # 0 > npert > max_irr_dim > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%= >%%%% > > >------------------------------------------------------------- > >I have searched the pw forum and some one suggests that one should >increase the max_irr_dim in the phcom.f90. However, it seems not work >after I increased the value. > >http://www.democritos.it/pipermail/pw_forum/2005-March/002167.html > > >Thank you in advance. > >Best wishes, > >Jian > >--__--__-- > >Message: 2 >From: Gerardo Ballabio >Subject: Re: [Pw_forum] Fwd: Re: Segmentation fault with Setting 2 --- > follow-up of Re. ATOMIC_POSITIONS >To: pw_forum at pwscf.org >Date: Thu, 30 Jun 2005 10:11:38 +0200 (MEST) >Reply-To: pw_forum at pwscf.org > >On 06/29/05 22:55:07, Eyvaz Isaev wrote: > > I can suggest that errors are due to your system > > configuration. > > May be your local network is slow or it configured > > improperly. > > > --- Yong Jiang wrote: > > > BTW, I use the serial code of PWscf. > >He's running on single processor. I can't see how the network can >have any effect in this case. Unless he's using a remotely mounted >disk or something like. > >Gerardo > > >--__--__-- > >Message: 3 >From: Gerardo Ballabio >Subject: Re: [Pw_forum] RE: Pw_forum digest, Vol 1 #668 - 2 msgs >To: pw_forum at pwscf.org >Date: Thu, 30 Jun 2005 10:13:17 +0200 (MEST) >Reply-To: pw_forum at pwscf.org > >On 06/26/05 10:27:47, sun jason wrote: > > Thank you very much! I've solve the problems of display. > >So is PWgui actually faster now? > >Gerardo > > >--__--__-- > >Message: 4 >From: "sun jason" >To: pw_forum at pwscf.org >Date: Thu, 30 Jun 2005 09:11:45 +0000 >Subject: [Pw_forum] about pwgui >Reply-To: pw_forum at pwscf.org > >dear all, > >1)I've solve the problem about 3D accelerate of my display card, now it >runs xcrysden well. but I got some problem of pwgui.I can launch it and >open some file, such as input file of pw.x but I cann't specify the value >in the file. when I want to type something, it always jump out and display >some error imformation like this: >$ pwgui > > ================================================== > This is PWgui version: 2.1.3 > -------------------------------------------------- > > > PWgui: using the system default "wish" interpreter > > PWGUI : /home/sunjian/espresso-2.1.3/PWgui-2.1.3/ > GUIB engine : /home/sunjian/espresso-2.1.3/PWgui-2.1.3/lib/Guib-0.3.2 > >/home/sunjian/espresso-2.1.3/PWgui-2.1.3/pwgui: line 56: 3762 ????????? >wish $PWGUI/pwgui.tcl > >what's wrong? > > >2) I don't know how to write a input file for pw.x to execute vc-relax >when I follow example03 and change the system, it always gets error >can anyone give me a templet of the input file for crystal variable cell >relaxation? > >Thank you very much! > > > > >============================================== >Jian Sun >Physics Dept. of Nanjing University >National Lab. of Solid State Microstructures >22 Hankou Road, Gulou District >Nanjing, Jiangsu Province >210093 >China >============================================== > > > > > >From: pw_forum-request at pwscf.org > >Reply-To: pw_forum at pwscf.org > >To: pw_forum at pwscf.org > >Subject: Pw_forum digest, Vol 1 #677 - 1 msg > >Date: Thu, 30 Jun 2005 07:36:31 +0200 > > > >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. Understanding "eband" & "deband" in "total energy" (Mousumi >Upadhyay Kahaly) > > > >-- __--__-- > > > >Message: 1 > >From: Mousumi Upadhyay Kahaly > >Date: Thu, 30 Jun 2005 10:14:57 +0530 (IST) > >To: pw_forum at pwscf.org > >Cc: eyvaz_isaev at yahoo.com, p.giannozzi at sns.it, > > axel.kohlmeyer at theochem.ruhr-uni-bochum.de > >Subject: [Pw_forum] Understanding "eband" & "deband" in "total energy" > >Reply-To: pw_forum at pwscf.org > > > >Dear Sir, > > > > I wanted to know how the final "total energy" is calculated for >the > > PWSCF output file and what each of those terms line "one-electron > >contribution" etc meant. > > > > As the source code "electrons.f90" in PW directory shows, > > > > etot = eband + ( etxc - etxcc ) + ewld + ehart + deband + demet > > > > After writing "etot", also other terms like "one-electron > >contribution", "hartree contribution" etc are written by, > >" > > WRITE( stdout, 9060 ) & > > eband, ( eband + deband ), ehart, ( etxc - etxcc ), ewld " > > > > I am having problem to understand the terms "eband" and "deband". > >How the "deband" value is assigned? When we solve the Kohn-Sham > >equation, the > >energy eigen values obtained are band energies, right? From K-S > >equation how do I obtain all above terms? > > > > May I please know (1) from where can I find the details of all > >these terms and their explanations and (2) how each of them will > >contribute to the cohesive energy of a system? > > > > Best regards, mousumi. > > > > > > > > > On Tuesday 28 June 2005 10:14, Mousumi Upadhyay Kahaly wrote: > > >> [...] total energy is being obtained if we add other terms except >"band > >energy sum" i.e. total energy = "one-electron contribution" + "hartree > >contribution" + "xc contribution" + "ewald contribution" + "correction for > >metals ". > > >> Any help to understand this is appreciated. regards. mousumi. > > > For the definiiton of the various terms, see > > > http://www.democritos.it/pipermail/pw_forum/2003-May/000291.html > > > and > > > http://www.democritos.it/pipermail/pw_forum/2003-May/000293.html > > > 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 > > > > > > > > > > > > > > > > > > > > > > > >-- __--__-- > > > >_______________________________________________ > >Pw_forum mailing list > >Pw_forum at pwscf.org > >http://www.democritos.it/mailman/listinfo/pw_forum > > > > > >End of Pw_forum Digest > > > >--__--__-- > >Message: 5 >Subject: Re: [Pw_forum] Fwd: Re: Segmentation fault with Setting 2 --- > follow-up of Re. ATOMIC_POSITIONS >From: Guido Roma >To: pw_forum at pwscf.org >Date: Thu, 30 Jun 2005 11:17:01 +0200 >Reply-To: pw_forum at pwscf.org > >On Wed, 2005-06-29 at 22:06, Yong Jiang wrote: > > (This is a resubmission of my last email) > > > > Dear All, > > > > During phonon calc at G and K vector (by following example02), I used Setting 1 > > of ATOMIC_POSITIONS and got reasonable vibrational frequencies. Setting 2 is > > equivalent as Setting 1. However, a similar phonon calc using Setting 2 failed. > > The code always stops with error "Segmentation fault". > > > >Memory limits ...? >You can check it with limit or ulimit according to your shell. > >Guido > > >--__--__-- > >Message: 6 >Date: Thu, 30 Jun 2005 02:29:32 -0700 (PDT) >From: Eyvaz Isaev >Subject: Re: [Pw_forum] Fwd: Re: Segmentation fault with Setting 2 --- follow-up of Re. ATOMIC_POSITIONS >To: pw_forum at pwscf.org >Reply-To: pw_forum at pwscf.org > >But I used just output file provided, and I saw >something like: > >pw.x XXXXX >pw.x XXXXX >pw.x XXXXX > >Bests, >Eyvaz. > >--- Gerardo Ballabio wrote: > > > On 06/29/05 22:55:07, Eyvaz Isaev wrote: > > > I can suggest that errors are due to your system > > > configuration. > > > May be your local network is slow or it configured > > > improperly. > > > > > --- Yong Jiang wrote: > > > > BTW, I use the serial code of PWscf. > > > > He's running on single processor. I can't see how > > the network can > > have any effect in this case. Unless he's using a > > remotely mounted > > disk or something like. > > > > Gerardo > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > >____________________________________________________ >Yahoo! Sports >Rekindle the Rivalries. Sign up for Fantasy Football >http://football.fantasysports.yahoo.com > >--__--__-- > >Message: 7 >Date: Thu, 30 Jun 2005 03:22:03 -0700 (PDT) >From: Eyvaz Isaev >Subject: Re: [Pw_forum] i/o error in davcio in the phonon calculation >To: pw_forum at pwscf.org >Reply-To: pw_forum at pwscf.org > >Hi, > >I am a "someone who advised increase max_irr_dim". But >in your particular case it may not work because you >have davcio-error meaning that something was wrong in >your previous calculations (davcio - davidson >input-output). Probably, it is due to a network error >because there is a strange message "task # 2" and >"task # 3". Besides, there is also "MPI abort" >message. > >Bests, >Eyvaz > >--- Jian ZHOU wrote: > > > Dear all, > > > > I have compiled the pwscf v2.1.4 successfully and > > also have run the > > example02 without any error. However, when I tried > > to do another > > phonon calculation of Gamma point, it stops during > > running the ph.x. > > > > I firstly run a scf calculation from scratch and > > then do a gamma point > > phonon calculation, reading the previous output > > files(The two outdirs > > are the same in the two calculation). The following > > is the screen > > output message of ph.x. > > > > >---------------------------------------------------------------------------- > > ... > > Atomic displacements: > > There are 17 irreducible representations > > > > Representation 1 3 modes - To be done > > > > Representation 2 3 modes - To be done > > > > Representation 3 2 modes - To be done > > > > Representation 4 3 modes - To be done > > > > Representation 5 3 modes - To be done > > > > Representation 6 3 modes - To be done > > > > Representation 7 3 modes - To be done > > > > Representation 8 1 modes - To be done > > > > Representation 9 3 modes - To be done > > > > Representation 10 2 modes - To be done > > > > Representation 11 3 modes - To be done > > > > Representation 12 2 modes - To be done > > > > Representation 13 1 modes - To be done > > > > Representation 14 3 modes - To be done > > > > Representation 15 3 modes - To be done > > > > Representation 16 1 modes - To be done > > > > Representation 17 3 modes - To be done > > IOS = 253 > > > > > > >%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > from davcio : error # 20 > > i/o error in davcio > > > > >%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > > > stopping ... > > [0] MPI Abort by user Aborting program ! > > [0] Aborting program! > > p0_5498: p4_error: : 0 > > > > -----------------------------------------------stop > > here > > > > There is also a CRASH file with the content: > > > > > > >%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > task # 2 > > from set_irr : error # 0 > > npert > max_irr_dim > > > > >%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > > > >%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > task # 3 > > from set_irr : error # 0 > > npert > max_irr_dim > > > > >%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > > > > > >------------------------------------------------------------- > > > > I have searched the pw forum and some one suggests > > that one should > > increase the max_irr_dim in the phcom.f90. However, > > it seems not work > > after I increased the value. > > > > >http://www.democritos.it/pipermail/pw_forum/2005-March/002167.html > > > > > > Thank you in advance. > > > > Best wishes, > > > > Jian > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > >__________________________________ >Yahoo! Mail >Stay connected, organized, and protected. Take the tour: >http://tour.mail.yahoo.com/mailtour.html > > >--__--__-- > >Message: 8 >Date: Thu, 30 Jun 2005 10:35:38 -0400 (CLT) >From: Eduardo Ariel Menendez P >To: PWSCF Forum >Subject: [Pw_forum] Re: how to find the space group and coordinates >Reply-To: pw_forum at pwscf.org > >Hello, >Concerning my question on Fri, 28 Jan 2005 09:43:11 -0300 (CLST) with the >the subject 'how to find the space group and coordinates' nobody answered. >An answer can be found on the article J. Appl. Cryst. (2005). vol 38, >p237-238, FINDSYM: program for identifying the space-group symmetry of a >crystal, Harold T. Stokes and Dorian M. Hatch. There are references to >other programs also. The code is available at >http://stokes.byu.edu/findsym.html >Enjoy it. > >Eduardo A. Menendez Proupin >Department of Physics >Faculty of Science >University of Chile >Las Palmeras 3425 >=D1u=F1oa, Santiago >Chile >Phone: 56+2+678 74 11 >http://fisica.ciencias.uchile.cl/~emenendez/ > >--__--__-- > >Message: 9 >From: Mousumi Upadhyay Kahaly >Date: Thu, 30 Jun 2005 21:19:57 +0530 (IST) >To: pw_forum at pwscf.org >Subject: [Pw_forum] Understanding "eband" & "deband" in "total energy" >Reply-To: pw_forum at pwscf.org > >This is a resubmission of my last mail, since I'm in real need of help. >========================================================================= > >Dear All, > > I wanted to know how the final "total energy" is calculated for the > PWSCF output file and what each of those terms line "one-electron >contribution" etc meant. > > As the source code "electrons.f90" in PW directory shows, > > etot = eband + ( etxc - etxcc ) + ewld + ehart + deband + demet > > After writing "etot", also other terms like "one-electron >contribution", "hartree contribution" etc are written by, >" > WRITE( stdout, 9060 ) & > eband, ( eband + deband ), ehart, ( etxc - etxcc ), ewld " > > I am having problem to understand the terms "eband" and "deband". >How the "deband" value is assigned? When we solve the Kohn-Sham >equation, the >energy eigen values obtained are band energies, right? From K-S >equation how do I obtain all above terms? > > May I please know (1) from where can I find the details of all >these terms and their explanations and (2) how each of them will >contribute to the cohesive energy of a system? > > Best regards, mousumi. > > > > > On Tuesday 28 June 2005 10:14, Mousumi Upadhyay Kahaly wrote: > >> [...] total energy is being obtained if we add other terms except "band >energy sum" i.e. total energy = "one-electron contribution" + "hartree >contribution" + "xc contribution" + "ewald contribution" + "correction for >metals ". > >> Any help to understand this is appreciated. regards. mousumi. > > For the definiiton of the various terms, see > > http://www.democritos.it/pipermail/pw_forum/2003-May/000291.html > > and > > http://www.democritos.it/pipermail/pw_forum/2003-May/000293.html > > 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 > > > > > > > > > > > > >--__--__-- > >Message: 10 >Date: Thu, 30 Jun 2005 11:11:26 -0700 (PDT) >From: Konstantin Kudin >Subject: Re: [Pw_forum] Understanding "eband" & "deband" in "total energy" >To: pw_forum at pwscf.org >Reply-To: pw_forum at pwscf.org > > Things look to be rather tricky in electrons.f90 ... > > There is some density that is generated from new bands, and then there >is also this density mixed with the older ones via the Broyden scheme >for which the potential is actually calculated. > > In principle, one should be able to get the total energy and all its >components for a given density. However, it is not clear if the >subroutine "electrons" actually does this. > > So I would also like to know what is the total energy in "electrons", >and which density it corresponds to. > > Kostya > > > >--- Mousumi Upadhyay Kahaly wrote: > > > Dear Sir, > > > > I wanted to know how the final "total energy" is calculated > > for the > > PWSCF output file and what each of those terms line "one-electron > > contribution" etc meant. > > > > As the source code "electrons.f90" in PW directory shows, > > > > etot = eband + ( etxc - etxcc ) + ewld + ehart + deband + > > demet > > > > After writing "etot", also other terms like "one-electron > > contribution", "hartree contribution" etc are written by, > > " > > WRITE( stdout, 9060 ) & > > eband, ( eband + deband ), ehart, ( etxc - etxcc ), ewld > > " > > > > I am having problem to understand the terms "eband" and > > "deband". > > How the "deband" value is assigned? When we solve the Kohn-Sham > > equation, the > > energy eigen values obtained are band energies, right? From K-S > > equation how do I obtain all above terms? > > > > May I please know (1) from where can I find the details of all > > these terms and their explanations and (2) how each of them will > > contribute to the cohesive energy of a system? > > > > Best regards, mousumi. > > > > > > > > > On Tuesday 28 June 2005 10:14, Mousumi Upadhyay Kahaly wrote: > > >> [...] total energy is being obtained if we add other terms except > > "band > > energy sum" i.e. total energy = "one-electron contribution" + > > "hartree > > contribution" + "xc contribution" + "ewald contribution" + > > "correction for > > metals ". > > >> Any help to understand this is appreciated. regards. > > mousumi. > > > For the definiiton of the various terms, see > > > http://www.democritos.it/pipermail/pw_forum/2003-May/000291.html > > > and > > > http://www.democritos.it/pipermail/pw_forum/2003-May/000293.html > > > 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 > > > > > > > > > > > > > > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > >____________________________________________________ >Yahoo! Sports >Rekindle the Rivalries. Sign up for Fantasy Football >http://football.fantasysports.yahoo.com > > >--__--__-- > >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > > >End of Pw_forum Digest From degironc at sissa.it Fri Jul 1 09:24:25 2005 From: degironc at sissa.it (Stefano de Gironcoli) Date: Fri, 01 Jul 2005 09:24:25 +0200 Subject: [Pw_forum] about structural optimization References: Message-ID: <42C4EFA9.6000400@sissa.it> sun jason wrote > > I have some questions about structural optimization > 1) can I just optimize the lattice constants but not change the atomic > fractional coordinates? Why do you want to do that ? If positions are fixed by symmetry (like in diamond structure for instance) they will not move even if they are allowed to. If they are not fixed by symmetry then it is incorrect to fix them. Anyhow, in the input file there is the possibility to lock certain atomic positions by specifying three additional numbers per atoms as in the following example (check the Doc/INPUT_PW file for an explanation) Si 0.00 0.00 0.00 0 0 0 Si 0.25 0.25 0.25 0 0 0 It should work also in combination with cell-dinamics (the force on the locked atoms are zeroed before entering MD routines) Another option is obviously to manually change the lattice parameter w/o changing the atomic positions (in crystal coordinates) > > 2) does ion_dynamics = 'bfgs' can be use to vc-relax? > I can run a vc-relax calculation when I put ion_dynamics = 'damp', and > it says does not support 'bfgs' when I put ion_dynamics='bfgs' No. vc-relax and vc-md employ routines contributed by Renata Wenzcovitch and coworkers and do not support bfgs minimization. > > 3) I run an example to calculate the effective charges of cubic BN, to > my knowledge, the absolute value of the effective charge for cation > and anion should be equal in this case. but I get 1.759 and -1.736 for > B and N atom, respectively. why? or just because I have not relax the > cell? > any help would be appreciate. effective charges should satisfy the so called acustic sum rule ( \sum_i Z_i = 0) . However in practice this is never realized due to incomplete BZ sampling, other technical limitation like not-optimized structure, finite cutoff or FFT grids are not relevant for ASR violation. One can easily IMPOSE the ASR obtetaining accurate results even with modest BZ sampling. This isssue is discussed in P.Giannozzi et al, PRB 43, 7231 (1991) . Stefano de Gironcoli From dalcorso at sissa.it Fri Jul 1 09:48:07 2005 From: dalcorso at sissa.it (Andrea Dal Corso) Date: Fri, 01 Jul 2005 09:48:07 +0200 Subject: [Pw_forum] Noncollinear magnetism In-Reply-To: <20050701024833.58158.qmail@web51010.mail.yahoo.com> References: <20050701024833.58158.qmail@web51010.mail.yahoo.com> Message-ID: <1120204087.3737.52.camel@dhpc-5-33.sissa.it> On Thu, 2005-06-30 at 19:48 -0700, W. YU wrote: > Dear Pw users, > > I have several questions about non collinear magnetism > calculation as shown in example13. > > The following is quoted from the calculation for Fe > with penalty functional: > ----------------------------------------------------- > # self-consistent calculation with penalty functional > &system > ibrav = 3, celldm(1) =5.217, nat= 1, ntyp= 1, > ecutwfc = 25.0,ecutrho = 200.0, > report=1, > occupations='smearing', smearing='gaussian', > degauss=0.05 > noncolin = .true. > starting_magnetization(1) = 0.5 > angle1(1) = 90.0 > angle2(1) = 0.0 > mcons(1,1) = 0.0 > mcons(2,1) = 0.5 > mcons(3,1) = 0.0 > i_cons = 1 > lambda = 1 > / > ================================================= > I have these questions: > > 1.From the input menu, I know the mcons(i,j) is the > i'th component of j'th atom type where i runs from 1 > to 3. But if angle1(1)=90.0 and angle2(1)=0.0, the > magnetic moment should be along X axis, but the > constraint seems to leave the Y axis unchanged is 2 > represents Y axis. It seems that 1,2 and 3 represent > Z, X and Y axses, am I right? You are right: these examples should be updated. Some problems have been already noticed and updated in the cvs version (where however the input variables for the constraint have been changed). For the present 2.1.3, or 2.1.4 version of PWscf: 1,2,3 are X,Y,Z. angle1(1) and angle2(1) are the theta and phi angles of the starting magnetization, which can be different from those of the magnetization specified by the constraint. In this particular example, starting with a magnetization along x, the magnetization remains along x for symmetry reasons The example can be corrected starting with a magnetization which has a component along y to break the initial the symmetry, for instance angle2(1)=10... > --------------------------------------------------- > 2. I hope to know how to choose lambda. why it is 1 > here in the example. ...and it is better to increase lambda. > =================================================== > The following is quoted from the calcualtion fo Ni: > > # self-consistent calculation > cat > ni.scf.in << EOF > &control > calculation='scf' > restart_mode='from_scratch', > pseudo_dir = '$PSEUDO_DIR/', > outdir='$TMP_DIR/' > prefix='ni' > / > &system > ibrav=2, celldm(1) =6.48, nat=1, ntyp=1, > starting_magnetization(1)=0.7, > ecutwfc = 24.0, ecutrho = 288.0, > occupations='smearing', > smearing='methfessel-paxton', degauss=0.02 > noncolin = .true. > starting_magnetization(1) = 0.1 > angle1(1) = 90.0 > angle2(1) = 0.0 > / > ==================================================== > 3. Now the question is: Why there are two starting > manetisation when there is only one type of Ni atom? This will be updated, thank you for noticing it. > 4. On what basis do we choose smearing scheme such as > "methfessel-paxton" or "Gaussian"? Both should work with the noncollinear version. There is no particular reason to use Gaussian as in the example. > 5. Last but most important question is: can we do > calculations on spiral? If the answer is positive, do > we have to use a very large supper cell? Presently you need a supercell. > ====================================================== > > Your answers and comments would be much a preciated. > > W. YU > > __________________________________________________ > 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 -- Andrea Dal Corso Tel. 0039-040-3787428 SISSA, Via Beirut 2/4 Fax. 0039-040-3787528 34014 Trieste (Italy) e-mail: dalcorso at sissa.it From vdiep at ictp.it Fri Jul 1 13:50:58 2005 From: vdiep at ictp.it (Diep Vinh Quang) Date: Fri, 1 Jul 2005 13:50:58 +0200 (CEST) Subject: [Pw_forum] VC-Relaxation Message-ID: <32990.10.41.38.18.1120218658.squirrel@webmail2.ictp.trieste.it> Hi all, How can we using the Variable Cell Relaxation programe. I have the following input file for pw.x but after some time, the propram stop. And on the other hand, how can we fix the pressure for the cell. did i do the right thing in the following input? (Does it means the pressure acting on the cell after minimization =12.8 kba?) &control calculation = 'vc-relax' restart_mode='from_scratch', prefix='Polyethylene', tstress = .true. tprnfor = .true. pseudo_dir = '/afs/ictp.trieste.it/home/v/vdiep/espresso-2.1.3/Polyethylene/', outdir='/afs/ictp.trieste.it/home/v/vdiep/espresso-2.1.3/Polyethylene/' / &system ibrav=8, celldm(1) = 13.48016706 celldm(2) = 0.7173398007 celldm(3) = 0.3609829952 nat=12 , ntyp= 2, ecutwfc =40.0, / &electrons diagonalization='cg' mixing_mode = 'plain' mixing_beta = 0.7 conv_thr = 1.0d-8 / &ions ion_dynamics='damp' / &cell cell_dynamics='damp-w' press=12.85 / ATOMIC_SPECIES C 12.0107 C.pbe-van_bm.UPF H 1.00794 H.pbe-van_bm.UPF ATOMIC_POSITIONS {angstrom} C -0.3176744885 -0.2775159117 0.6437500001 H -0.2369597346 -1.384577396 0.6437500001 H -1.403859394 -0.0488188374 0.6437500001 C 0.3176744885 0.2775159117 -0.6437500001 H 0.2369597346 1.384577396 -0.6437500001 H 1.403859394 0.0488188374 -0.6437500001 C 3.248975510 2.836015910 1.931250000 H 2.162790604 2.607318836 1.931250000 H 3.329690265 3.943077394 1.931250000 C 3.884324490 2.280984090 0.643750000 H 4.970509396 2.509681164 0.643750000 H 3.803609735 1.173922606 0.643750000 K_POINTS {automatic} 2,2,2,0,0,0 Thank you very much for your help. Best regards, Vinh ------------------------------------------------- Diep Vinh Quang C/o Diploma Program, The Abdus Salam International Center for Theoretical Physics(ICTP), 11 Strada Costiera, 34014, Trieste,Italy. http://www.ictp.it From patricia at fcq.unc.edu.ar Fri Jul 1 13:36:19 2005 From: patricia at fcq.unc.edu.ar (Patricia Paredes) Date: Fri, 01 Jul 2005 13:36:19 Subject: [Pw_forum] espresso 2.1.4 and ifc8.1.024 Message-ID: <3.0.1.32.20050701133619.00eb92d8@mail.fcq.unc.edu.ar> A non-text attachment was scrubbed... Name: not available Type: text/enriched Size: 1905 bytes Desc: not available Url : /pipermail/attachments/20050701/e32889f4/attachment.bin From axel.kohlmeyer at theochem.ruhr-uni-bochum.de Fri Jul 1 19:05:42 2005 From: axel.kohlmeyer at theochem.ruhr-uni-bochum.de (Axel Kohlmeyer) Date: Fri, 1 Jul 2005 19:05:42 +0200 (CEST) Subject: [Pw_forum] espresso 2.1.4 and ifc8.1.024 In-Reply-To: <3.0.1.32.20050701133619.00eb92d8@mail.fcq.unc.edu.ar> Message-ID: On Fri, 1 Jul 2005, Patricia Paredes wrote: PP> Hi, I was using until now pwscf-v2.0. This week I decided to upgrade to PP> espresso. Also, I installed the latest version of intel compiler PP> (ifc8.1.024) .For the compilation of pwscf-v2.0 we used ifc7, but PP> espresso with ifc7 gives us error 0 0. In the system we had installed PP> cluster mkl 7.0. I changed the make.sys to include more optimization PP> (-O3, -tpp7). patricia, you have to be careful here. higher optimization levels can create problems sometimes, since the compiler may rearrange the code a little bit, and that may not always be optimal for numerical codes. it may slow down and sometimes even produce incorrect results. however this very much depends on the compiler/hardware combination, the specific code and the problem. on pc-style hardware i found that using, ' -O2 -unroll -pc64' is a quite fast and secure choice. PP> PP> The compilation went OK, so I began testing with and old input file. But PP> the new pw.x runs slower than the old version (pwscf.v2.0). PP> PP> PP> For example: I obtained for following results for a SCH3 radical in a PP> box: PP> can you please post this example input, so we can try to reproduce your problem. some flag's names or default settings may have changed since v2.0. PP> PP> (single node, P4 HT, 3GHz, running Fedora and mpi6.0) ok, now i have a bunch of questions, since based the provided information, there can be quite a few number of reasons for the slowdown you are seeing: do you have hyper-threading enabled or disabled? are you running in parallel or serial? if the former, does the slowdown only happen in parallel? do you see the same problem with some of the integrated (and smaller) test examples? did you run the two jobs on the exact same machine with the same communication library and hardware? PP> PP> first scf iteration PP> PP> pwscf v2.0 47.55 secs PP> PP> espresso v2.1.4 418.15 secs PP> PP> PP> time for the fist ionic step PP> PP> pwscf v2.0 358.88 secs PP> PP> espresso v2.1.4 2103.36 secs PP> PP> PP> Also, I tried with a periodic system with metallic atoms withe SCH3l PP> adsorbed on the surface, running on 6 nodes: PP> PP> the new executable was slower as in the case of the radical, and the PP> results were completely different. Since the default criterion for ionic PP> convergence is PP> PP> setting by default in 1e-4 and 1e-3 and there are some changes in the PP> value of the energy, the results are: PP> PP> Pwscf-v2.0 converges in 12 iterations with a total force of 2 e-3. PP> PP> espresso 2.1.4 has sligtly differents results (the total energy differs PP> in the last 4 digits) and after iteration 30 something happens and got an PP> error and the run stopped. what error? PP> I'm aware of the problems with ifc as described in the archives and in PP> Axel Kohlmeyer web page, but I?m still in trouble. PP> PP> PP> Did someone face the same problem?? Suppose the problem has to do with PP> the compiler, but I don?t know at this stage how to solve it. it need not be the compiler. regards, axel. PP> PP> Many thanks, PP> PP> Patricia. PP> PP> PP> PP> Prof. Dr. Patricia Paredes Olivera PP> PP> Facultad de Ciencias Qu?micas PP> PP> Universidad Nacional de C?rdoba PP> PP> ph/fax: 054-0351-434-4972 PP> _______________________________________________ PP> Pw_forum mailing list PP> Pw_forum at pwscf.org PP> http://www.democritos.it/mailman/listinfo/pw_forum PP> PP> -- ======================================================================= Dr. Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/ ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From patricia at fcq.unc.edu.ar Fri Jul 1 19:00:23 2005 From: patricia at fcq.unc.edu.ar (Patricia Paredes) Date: Fri, 01 Jul 2005 19:00:23 Subject: [Pw_forum] espresso 2.1.4 and ifc8.1.024 Message-ID: <3.0.1.32.20050701190023.012f7aa0@mail.fcq.unc.edu.ar> A non-text attachment was scrubbed... Name: not available Type: text/enriched Size: 4174 bytes Desc: not available Url : /pipermail/attachments/20050701/32867027/attachment.bin From patricia at fcq.unc.edu.ar Fri Jul 1 19:01:07 2005 From: patricia at fcq.unc.edu.ar (Patricia Paredes) Date: Fri, 01 Jul 2005 19:01:07 Subject: [Pw_forum] espresso 2.1.4 and ifc8.1.024 Message-ID: <3.0.1.32.20050701190107.012f7aa0@mail.fcq.unc.edu.ar> A non-text attachment was scrubbed... Name: not available Type: text/enriched Size: 6069 bytes Desc: not available Url : /pipermail/attachments/20050701/02563016/attachment.bin From axel.kohlmeyer at theochem.ruhr-uni-bochum.de Sat Jul 2 16:38:49 2005 From: axel.kohlmeyer at theochem.ruhr-uni-bochum.de (Axel Kohlmeyer) Date: Sat, 2 Jul 2005 16:38:49 +0200 (CEST) Subject: [Pw_forum] espresso 2.1.4 and ifc8.1.024 In-Reply-To: <3.0.1.32.20050701190023.012f7aa0@mail.fcq.unc.edu.ar> Message-ID: On Fri, 1 Jul 2005, Patricia Paredes wrote: patricia, PP> I use the same input for the old and new pwscf. i have run your input with a serial 32-bit binary on an opteron 246 machine and i need 42.85 seconds for the first scf iteration and 403 seconds for the first ionic step. so i guess it cannot be the intel compiler. PP> There are some differences in the energies through the PP> scf iterations, in the last 4 digits, PP> as a result of these small diferences, it took more ionic iterations PP> to converge for the expresso version of the code. i did not follow all the details, but there have been changes and corrections to several parts of the code that affect the total energies and also the way the potential and wavefunctions are extrapolated, between the ionic steps since version 2.0. in some cases, such 'improvements' can result in requiring a few more steps in the optimization and differences in the final total energies. it is not always easy to tell whether this is significant or not. PP> It was running on a single node, using the parallel code. Note that it PP> was a Gamma point calculation. [...] PP> >do you have hyper-threading enabled or disabled? PP> PP> It is enabled this is a point you should watch. for numerical codes this is actually harming performance more often than helping. that would, however, be totally independent from the compiler and the pwscf version. PP> >are you running in parallel or serial? PP> parallel version, but only in one node (since it was PP> a Gamma point input) you don't need k-points for running in parallel. pw.x also supports distributed memory parallelization of the G and R space. this is less efficient than parallelizing across k-points but with a fast network like myrinet or infiniband it should even scale reasonably well up to quite a few nodes. PP> >the same communication library and hardware? PP> PP> PP> Yes!! ok, you may have a problem, if the MPI was compiled with ifc7 and you compile espresso with ifort8.1 for that you may want to test compiling pwscf2.0 with ifort8.1 to see. PP> >what error? PP> PP> p4_10080: (24418.614286) net_recv failed for fd = 8 PP> PP> p4_10080: p4_error: net_recv read, errno = : 104 ouch. that does not look too good. there is something fishy going on. perhaps you have a bad network connection (and thus a lot of re-transmits) somewhere. i hope this will help you a little more to figure out what is going wrong. locating the problem when you change several parameters simultaneously is never simple. best regards, axel. -- ======================================================================= Dr. Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/ ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From zjspam at gmail.com Sun Jul 3 11:03:06 2005 From: zjspam at gmail.com (Jian ZHOU) Date: Sun, 3 Jul 2005 17:03:06 +0800 Subject: [Pw_forum] i/o error in davcio in the phonon calculation In-Reply-To: <20050630102204.70499.qmail@web60316.mail.yahoo.com> References: <8d38992a05063000377cae5fc@mail.gmail.com> <20050630102204.70499.qmail@web60316.mail.yahoo.com> Message-ID: <8d38992a05070302035e856615@mail.gmail.com> Dear Eyvaz, Thank you for your reply. I found it seems a parallel problems. When I do parallel calculation for the scf and phonon calculation, it stops with errors. However, if I use only 1 cpu to do the scf and phonon calculation, it seems works well. The serial ph.x has been running for 2 days, I am not sure that if it is abnormal, since the scf calculation is quite fast with 1 hour by 1 cpu. In the output file, it said he found 17 "Representations", It has done more that 11 Representations now. On 6/30/05, Eyvaz Isaev wrote: > Hi, > > I am a "someone who advised increase max_irr_dim". But > in your particular case it may not work because you > have davcio-error meaning that something was wrong in > your previous calculations (davcio - davidson > input-output). Probably, it is due to a network error > because there is a strange message "task # 2" and > "task # 3". Besides, there is also "MPI abort" > message. > > Bests, > Eyvaz > > --- Jian ZHOU wrote: > > > Dear all, > > > > I have compiled the pwscf v2.1.4 successfully and > > also have run the > > example02 without any error. However, when I tried > > to do another > > phonon calculation of Gamma point, it stops during > > running the ph.x. > > > > I firstly run a scf calculation from scratch and > > then do a gamma point > > phonon calculation, reading the previous output > > files(The two outdirs > > are the same in the two calculation). The following > > is the screen > > output message of ph.x. > > > > > ---------------------------------------------------------------------------- > > ... > > Atomic displacements: > > There are 17 irreducible representations > > > > Representation 1 3 modes - To be done > > > > Representation 2 3 modes - To be done > > > > Representation 3 2 modes - To be done > > > > Representation 4 3 modes - To be done > > > > Representation 5 3 modes - To be done > > > > Representation 6 3 modes - To be done > > > > Representation 7 3 modes - To be done > > > > Representation 8 1 modes - To be done > > > > Representation 9 3 modes - To be done > > > > Representation 10 2 modes - To be done > > > > Representation 11 3 modes - To be done > > > > Representation 12 2 modes - To be done > > > > Representation 13 1 modes - To be done > > > > Representation 14 3 modes - To be done > > > > Representation 15 3 modes - To be done > > > > Representation 16 1 modes - To be done > > > > Representation 17 3 modes - To be done > > IOS = 253 > > > > > > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > from davcio : error # 20 > > i/o error in davcio > > > > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > > > stopping ... > > [0] MPI Abort by user Aborting program ! > > [0] Aborting program! > > p0_5498: p4_error: : 0 > > > > -----------------------------------------------stop > > here > > > > There is also a CRASH file with the content: > > > > > > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > task # 2 > > from set_irr : error # 0 > > npert > max_irr_dim > > > > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > > > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > task # 3 > > from set_irr : error # 0 > > npert > max_irr_dim > > > > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > > > > > > ------------------------------------------------------------- > > > > I have searched the pw forum and some one suggests > > that one should > > increase the max_irr_dim in the phcom.f90. However, > > it seems not work > > after I increased the value. > > > > > http://www.democritos.it/pipermail/pw_forum/2005-March/002167.html > > > > > > Thank you in advance. > > > > Best wishes, > > > > Jian > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > __________________________________ > Yahoo! Mail > Stay connected, organized, and protected. Take the tour: > http://tour.mail.yahoo.com/mailtour.html > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From vdiep at ictp.it Sun Jul 3 15:38:04 2005 From: vdiep at ictp.it (Diep Vinh Quang) Date: Sun, 3 Jul 2005 15:38:04 +0200 (CEST) Subject: [Pw_forum] VC-Relaxation In-Reply-To: References: <32990.10.41.38.18.1120218658.squirrel@webmail2.ictp.trieste.it> Message-ID: <1931.10.41.38.15.1120397884.squirrel@webmail3.ictp.trieste.it> Dear Axel, Thank you very much for your reply, I change the input file a little bit (by inscrease nstep=100 (instead of 50) and the outdir='/tmp') but I still get the error. The error message is "FORTRAN STOP 2" (and for sure it run longer than the previous one). But the result is reasonable when i checked separately by the SCF method. I mean after getting the results from the VC-relaxation (with Stop after sometime), using that result as an input for "scf" and the total energy is very good in the sensen that is agree very well with other calculation. Thank you very much. Vinh > > On Fri, 1 Jul 2005, Diep Vinh Quang wrote: > > VQ> Hi all, > > hi vinh, > > VQ> How can we using the Variable Cell Relaxation programe. I have the > > please have a look at the Doc/INPUT_PW file of the espresso distribution, > it should have a lot of useful information about the various flags > and parameters. the only are a little scattered throughout the file. > > VQ> following input file for pw.x but after some time, the propram stop. > And > > was there an error message, or did the program just stop? > > VQ> on the other hand, how can we fix the pressure for the cell. did i do > the > VQ> right thing in the following input? (Does it means the pressure acting > on > VQ> the cell after minimization =12.8 kba?) > > that should be written in the output. > > VQ> > VQ> &control > VQ> calculation = 'vc-relax' > VQ> restart_mode='from_scratch', > VQ> prefix='Polyethylene', > VQ> > VQ> tstress = .true. > VQ> tprnfor = .true. > VQ> pseudo_dir = > VQ> '/afs/ictp.trieste.it/home/v/vdiep/espresso-2.1.3/Polyethylene/', > VQ> > outdir='/afs/ictp.trieste.it/home/v/vdiep/espresso-2.1.3/Polyethylene/' > > hmmm... it probably is not a good idea to use AFS for > storing your scratch files. running a job with pw.x may create a > substantial amount of i/o and you ideally want to keep that > on a local scratch disk. actually, you should also check whether > the job stops, because you're running out of disk quota. > > VQ> / > > also you don't specify a value for nstep. perhaps > your job just stopped becaus it reached the default > limit of 50(?) ionic steps. > > VQ> &system > VQ> ibrav=8, > VQ> celldm(1) = 13.48016706 > VQ> celldm(2) = 0.7173398007 > VQ> celldm(3) = 0.3609829952 > VQ> nat=12 , ntyp= 2, > VQ> ecutwfc =40.0, > > hmmm. for a variable cell calculation with ultra-soft pseudopotentials, > you may have to also increase the density cutoff here from the default > 4*ecutwfc. i remember there were some discussions about that, please > check the mailing list archive. > > hope this helps, > axel. > > > -- > > ======================================================================= > Dr. Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de > Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 > Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 > D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/ > ======================================================================= > If you make something idiot-proof, the universe creates a better idiot. > ------------------------------------------------- Diep Vinh Quang C/o Diploma Program, The Abdus Salam International Center for Theoretical Physics(ICTP), 11 Strada Costiera, 34014, Trieste,Italy. http://www.ictp.it From cbarreteau at cea.fr Mon Jul 4 09:09:10 2005 From: cbarreteau at cea.fr (Cyrille Barreteau) Date: Mon, 4 Jul 2005 09:09:10 +0200 (CEST) Subject: [Pw_forum] FLUSH--FAILED:: Bad address Message-ID: <40673.132.166.21.1.1120460950.squirrel@dsm-mail> Dear pwscf users, I wanted to report a strange behaviour observed on the cvs version of pwscf. When using a script to perform a series of calculation (for example doing a calculation at various lattice parameters..) like the following: ================================================== #!/bin/sh #################################################################### # include the needed environment variables #.. ./environment_variables PW_ROOT=$HOME/SOFTWARE/espresso.cvs PSEUDO_DIR=$HOME/SOFTWARE/espresso.cvs/pseudo TMP_DIR=/tmp/orbit-barreto/espresso #rm -fr $TMP_DIR/* #mkdir results cd results rm -f fefcc.scf.out for lattice_const in 6.6 6.7 6.8 6.9 7.0 7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.7 ; do # self-consistent calculation cat > fefcc.scf.in << EOF &control pseudo_dir = '$PSEUDO_DIR/', outdir='$TMP_DIR/' prefix='fe' / &system ibrav = 2, celldm(1) =$lattice_const, nat= 1, ntyp= 1, nbnd=10, nspin = 2, starting_magnetization(1)=0.30, ecutwfc = 35, ecutrho = 250.0, occupations='smearing', smearing='methfessel-paxton', degauss=0.005 / &electrons conv_thr = 1.0e-8 mixing_beta = 0.3 / ATOMIC_SPECIES Fe 55.847 Fe_us_gga_d2.1.8.pseudo.UPF ATOMIC_POSITIONS Fe 0.0 0.0 0.0 K_POINTS automatic 20 20 20 0 0 0 EOF $PW_ROOT/bin/pw.x < fefcc.scf.in >> fefcc.scf.out done =================================================== A strange warning appears at the second call of pw.x saying: "FLUSH--FAILED:: Bad address" (three times per itetation) . Moreover in the output file many exotic characters appear depending on the editor used (^@ with xemacs, or with nedit etc..) By looking on the web I discovered that this "bug" is a combination of several factors: the use of FLUSH routine (machine dependant) the compilation with ifc7 the double redirection >> as described on the following web site: http://softwareforums.intel.com/ids/board/message?board.id=11&message.id=762 Any comment is very welcome. cyrille -- ================================================ Cyrille Barreteau | phone:+33 (0)1 69 08 29 51 CEA Saclay | fax : +33 (0)1 69 08 84 46 DSM/DRECAM/SPCSI | email: cbarreteau at cea.fr Batiment 462 | 91191 Gif sur Yvette Cedex FRANCE ~~~~~~~~~~~~~~~~~~~~~~~~ http://www-drecam.cea.fr/spcsi/index.php http://www-drecam.cea.fr/Images/Pisp/cbarreteau/cbarreteau_fr.html ================================================ From mousumi at jncasr.ac.in Tue Jul 5 05:43:40 2005 From: mousumi at jncasr.ac.in (Mousumi Upadhyay Kahaly) Date: Tue, 5 Jul 2005 09:13:40 +0530 (IST) Subject: [Pw_forum] Understanding "eband" & "deband" in "total energy" Message-ID: <56621.202.41.111.151.1120535020.squirrel@202.41.111.151> This is a resubmission of my last mail, since I'm in real need of help. ========================================================================= Dear All, I wanted to know how the final "total energy" is calculated for the PWSCF output file and what each of those terms line "one-electron contribution" etc meant. As the source code "electrons.f90" in PW directory shows, etot = eband + ( etxc - etxcc ) + ewld + ehart + deband + demet After writing "etot", also other terms like "one-electron contribution", "hartree contribution" etc are written by, " WRITE( stdout, 9060 ) & eband, ( eband + deband ), ehart, ( etxc - etxcc ), ewld " I am having problem to understand the terms "eband" and "deband". How the "deband" value is assigned? When we solve the Kohn-Sham equation, the energy eigen values obtained are band energies, right? From K-S equation how do I obtain all above terms? May I please know (1) from where can I find the details of all these terms and their explanations and (2) how each of them will contribute to the cohesive energy of a system? Best regards, mousumi. > On Tuesday 28 June 2005 10:14, Mousumi Upadhyay Kahaly wrote: >> [...] total energy is being obtained if we add other terms except "band energy sum" i.e. total energy = "one-electron contribution" + "hartree contribution" + "xc contribution" + "ewald contribution" + "correction for metals ". >> Any help to understand this is appreciated. regards. mousumi. > For the definiiton of the various terms, see > http://www.democritos.it/pipermail/pw_forum/2003-May/000291.html > and > http://www.democritos.it/pipermail/pw_forum/2003-May/000293.html > 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 zhupwscf at yahoo.com.cn Tue Jul 5 06:44:30 2005 From: zhupwscf at yahoo.com.cn (=?gb2312?q?=C1=D9=20=D7=A1?=) Date: Tue, 5 Jul 2005 12:44:30 +0800 (CST) Subject: [Pw_forum] dynamics (vc-md) error! Message-ID: <20050705044430.42718.qmail@web15502.mail.cnb.yahoo.com> Dear all: I have performed the Molecular dynamics with Varible-Cell , becuase I want relax the surface,but there is an error in the output as follwing: ******************************************************************************** G cutoff = 1028.0180 ( 150039 G-vectors) FFT grid: ( 72, 72, 81) G cutoff = 356.9507 ( 30737 G-vectors) smooth grid: ( 40, 40, 48) nbndx = 1152 nbnd = 288 natomwfc = 336 npwx = 3876 nelec = 480.00 nkb = 816 ngl = 150039 warning: negative or imaginary core charge -0.000035 0.000000 Initial potential from superposition of free atoms starting charge 479.98858, renormalised to 480.00000 Starting wfc are atomic IOS = 161 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from davcio : error # 10 i/o error in davcio %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... *********************************************************************************** Can someone tell me how to relax the surface with pwscf? Best wishes! --------------------------------- DO YOU YAHOO!? ????G??????????????????? -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20050705/e5c847d1/attachment.htm From swblelia at sw.ehu.es Tue Jul 5 09:30:11 2005 From: swblelia at sw.ehu.es (Aritz Leonardo) Date: Tue, 05 Jul 2005 09:30:11 +0200 Subject: [Pw_forum] negative lambda Message-ID: <42CA3703.40007@sw.ehu.es> Hello all: I have been searching the archive to know why am I obtaining negative values of lambda and gamma and I don't really understand what is going on. As far as I see from the formulas (for gamma and lambda) Shouldn't they be positively defined??? for example for gamma, we have: the square of the matrix element, 2 deltas, ph-frequencies, and BZ volume. Even though I may be far from convergence in my opinion all the values should be positive. Thanks a lot Aritz From yjiang at asu.edu Tue Jul 5 19:22:40 2005 From: yjiang at asu.edu (Yong Jiang) Date: Tue, 05 Jul 2005 10:22:40 -0700 (MST) Subject: [Pw_forum] Re. from total energy to cohesive energy Message-ID: <1120584160.42cac1e064fb7@webmail.asu.edu> Dear All, My question is about the cohesive energy. In example01 ni.scf.cg.out, total energy is calculated as -85.72249139 ryd (or ~1165.8eV) for one atom in fcc structure (nat=1 and ibrav=2). How to use this info to get cohesive energy of Ni, which is about 5eV? Thanks very much in advance for your help. Best regards, Yong From fratesi at sissa.it Wed Jul 6 00:49:01 2005 From: fratesi at sissa.it (Guido Fratesi) Date: Wed, 6 Jul 2005 00:49:01 +0200 (CEST) Subject: [Pw_forum] Re: Understanding "eband" & "deband" in "total energy" Message-ID: The one-electron contribution is the kinetic + ionic term: \sum + Integral[rho(r)*V_ion(r),dr] This contribution is computed as (eband+deband), since: 1) the sum of the eigenvalues (eband) gives (from KS equations): \sum eig = \sum + Integral[rho(r)*V_ion(r),dr] + Integral[rho(r)*(V_H(r)+V_XC(r)),dr] 2) deband is defined as: deband = -Integral[rho(r)*(V_H(r)+V_XC(r)),dr] Guido > From: Mousumi Upadhyay Kahaly > Date: July 5, 2005 5:43:40 AM GMT+02:00 > To: pw_forum at pwscf.org > Subject: [Pw_forum] Understanding "eband" & "deband" in "total energy" > Reply-To: pw_forum at pwscf.org > > > > This is a resubmission of my last mail, since I'm in real need of help. > ========================================================================= > > Dear All, > > I wanted to know how the final "total energy" is calculated for the > PWSCF output file and what each of those terms line "one-electron > contribution" etc meant. > > As the source code "electrons.f90" in PW directory shows, > > etot = eband + ( etxc - etxcc ) + ewld + ehart + deband + demet > > After writing "etot", also other terms like "one-electron > contribution", "hartree contribution" etc are written by, > " > WRITE( stdout, 9060 ) & > eband, ( eband + deband ), ehart, ( etxc - etxcc ), ewld " > > I am having problem to understand the terms "eband" and "deband". > How the "deband" value is assigned? When we solve the Kohn-Sham > equation, the > energy eigen values obtained are band energies, right? From K-S > equation how do I obtain all above terms? > > May I please know (1) from where can I find the details of all > these terms and their explanations and (2) how each of them will > contribute to the cohesive energy of a system? > > Best regards, mousumi. > > > > >> On Tuesday 28 June 2005 10:14, Mousumi Upadhyay Kahaly wrote: >> >>> [...] total energy is being obtained if we add other terms except "band >>> > energy sum" i.e. total energy = "one-electron contribution" + "hartree > contribution" + "xc contribution" + "ewald contribution" + "correction for > metals ". > >>> Any help to understand this is appreciated. regards. mousumi. >>> >> For the definiiton of the various terms, see >> http://www.democritos.it/pipermail/pw_forum/2003-May/000291.html >> and >> http://www.democritos.it/pipermail/pw_forum/2003-May/000293.html >> 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 >> > > > > > > > > > > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From physicalreviewb at yahoo.com Wed Jul 6 06:02:43 2005 From: physicalreviewb at yahoo.com (Jigang Zhang) Date: Tue, 5 Jul 2005 21:02:43 -0700 (PDT) Subject: [Pw_forum] configure problems Message-ID: <20050706040243.10916.qmail@web33606.mail.mud.yahoo.com> Hello, Dear PWscf users, I got some problems in the very beginning. I tried to do ./configure on a cluster with F90 and MPIF90 but failed. The log file says f77 can not make executables. Sounds like PWscf doesn't like F90? Then I tried to do the same thing under ifc 8.0 and mpiifort, this time it worked , but could not compile some subroutines (e.g path_tangent ). My knowledge in compilers is limited, but I need to use PWscf urgently . Does anybody can help? Thank you very much! __________________________________________________ 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/20050705/acb8e0ba/attachment.htm From swblelia at sw.ehu.es Wed Jul 6 09:36:57 2005 From: swblelia at sw.ehu.es (Aritz Leonardo) Date: Wed, 06 Jul 2005 09:36:57 +0200 Subject: [Pw_forum] lambda Message-ID: <42CB8A19.9070507@sw.ehu.es> Hello all: I have been searching the archive to know why am I obtaining negative values of lambda and gamma and I don't really understand what is going on. As far as I see from the formulas (for gamma and lambda) Shouldn't they be positively defined??? for example for gamma, we have: the square of the matrix element, 2 deltas, ph-frequencies, and BZ volume. Even though I may be far from convergence in my opinion all the values should be positive. Thanks a lot Aritz From zjspam at gmail.com Wed Jul 6 09:39:44 2005 From: zjspam at gmail.com (Jian ZHOU) Date: Wed, 6 Jul 2005 15:39:44 +0800 Subject: [Pw_forum] pseudopotential of Ba and Ti Message-ID: <8d38992a05070600395a94d698@mail.gmail.com> Dear all, I am trying to do some calculations of BaTiO3, However, I can not download the suitable pseudopotentials(PP) of Ba and Ti. From jasonsun98 at hotmail.com Wed Jul 6 11:38:55 2005 From: jasonsun98 at hotmail.com (sun jason) Date: Wed, 06 Jul 2005 09:38:55 +0000 Subject: [Pw_forum] about energy gap In-Reply-To: <20050706053627.30456.4804.Mailman@democritos.sissa.it> Message-ID: Dear all, I'm a freshman in computational materials, I'm puzzled by a simple question about energy gap, when I calculated the C2-BN system which was construtured by mixing cubic BN and diamond. Well, the computed LDA band gap of c-BN and diamond is around 4.5 and 4.1 eV, respectively. Therefore, by mixing two compounds with such a wide band gap (roughly of the order of 4 eV) one might expect to get a BC2N compound with an Eg between 4.5 and 4.1eV. Hence, why is our calculated Eg (about 1.0 to 2.0 eV) is much lower than expected for a C2-BN system? the unit cell I calculated is crystallined in diamond-like structure and the atoms are sp3 hybridization. Thank you very much! From eyvaz_isaev at yahoo.com Wed Jul 6 14:12:54 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Wed, 6 Jul 2005 05:12:54 -0700 (PDT) Subject: [Pw_forum] configure problems In-Reply-To: <20050706040243.10916.qmail@web33606.mail.mud.yahoo.com> Message-ID: <20050706121254.87623.qmail@web60313.mail.yahoo.com> Hi, --- Jigang Zhang wrote: > I tried to do ./configure on a cluster with F90 and > MPIF90 but failed. The log file says f77 can not > make executables. Sounds like PWscf doesn't like > F90? This usually means that f90 compiler is installed improperly. And f77 is really not suitable to compile the code. > Then I tried to do the same thing under ifc > 8.0 and mpiifort, this time it worked , but could > not compile some subroutines (e.g path_tangent ). This is very famous compiling error using ifc-8.0. So, upgrade your compiler to 8.1. Bests, Eyvaz. __________________________________ Yahoo! Mail Stay connected, organized, and protected. Take the tour: http://tour.mail.yahoo.com/mailtour.html From jasonsun98 at hotmail.com Wed Jul 6 17:14:34 2005 From: jasonsun98 at hotmail.com (sun jason) Date: Wed, 06 Jul 2005 15:14:34 +0000 Subject: [Pw_forum] about phonon calculation In-Reply-To: <20050706053627.30456.4804.Mailman@democritos.sissa.it> Message-ID: Dear all, I executed a test of phonon calculation of cubic BN, the input for scf is like this =========================================================================== &SYSTEM ibrav = 2, celldm(1)= 6.792, nat = 2, ntyp = 2, ecutwfc = 50.0 , / &ELECTRONS conv_thr = 1.0d-9 , mixing_beta = 0.7 , / ATOMIC_SPECIES B 10.81100 B.pz-vbc.UPF N 14.00674 N.pz-vbc.UPF ATOMIC_POSITIONS B 0.000000000 0.000000000 0.000000000 N 0.250000000 0.250000000 0.250000000 K_POINTS automatic 6 6 6 1 1 1 =========================================================================== and the result of dielectric tensor and effective charge is like this: =========================================================================== Dielectric constant in cartesian axis ( 4.637335813 0.000000000 0.000000000 ) ( 0.000000000 4.637335813 0.000000000 ) ( 0.000000000 0.000000000 4.637335813 ) Effective charges E-U in cartesian axis atom 1 ( 1.85986 0.00000 0.00000 ) ( 0.00000 1.85986 0.00000 ) ( 0.00000 0.00000 1.85986 ) atom 2 ( -1.82935 0.00000 0.00000 ) ( 0.00000 -1.82935 0.00000 ) ( 0.00000 0.00000 -1.82935 ) =========================================================================== and the frequency at gamma point is like this: =========================================================================== 0.0000 0.0001 0.0001 1055.5528 1055.5528 1274.3693 =========================================================================== so my question is: 1)why the LO mode is so small(the exact value is about 1305 cm-1) when the TO mode is correct? and how to improve it? 2) why the absolute value of effective charge for B atom and N atom is not equal? any help would be appreciate much! jason From eyvaz_isaev at yahoo.com Wed Jul 6 18:51:30 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Wed, 6 Jul 2005 09:51:30 -0700 (PDT) Subject: [Pw_forum] about phonon calculation In-Reply-To: Message-ID: <20050706165130.7988.qmail@web60315.mail.yahoo.com> Hi, In order to get correct LO mode you should calculate phonons using dynmat.x from /pwtools directory. Effective charges for B and N are not equal due to numerical errors in our computer calculations. Bests, Eyvaz. --- sun jason wrote: > Dear all, > > I executed a test of phonon calculation of cubic BN, > the input for scf is > like this > =========================================================================== > &SYSTEM > ibrav = 2, > celldm(1)= 6.792, > nat = 2, > ntyp = 2, > ecutwfc = 50.0 , > / > &ELECTRONS > conv_thr = 1.0d-9 , > mixing_beta = 0.7 , > / > ATOMIC_SPECIES > B 10.81100 B.pz-vbc.UPF > N 14.00674 N.pz-vbc.UPF > ATOMIC_POSITIONS > B 0.000000000 0.000000000 0.000000000 > > N 0.250000000 0.250000000 0.250000000 > > K_POINTS automatic > 6 6 6 1 1 1 > =========================================================================== > > and the result of dielectric tensor and effective > charge is like this: > > =========================================================================== > Dielectric constant in cartesian axis > > ( 4.637335813 0.000000000 > 0.000000000 ) > ( 0.000000000 4.637335813 > 0.000000000 ) > ( 0.000000000 0.000000000 > 4.637335813 ) > > Effective charges E-U in cartesian axis > > atom 1 > ( 1.85986 0.00000 > 0.00000 ) > ( 0.00000 1.85986 > 0.00000 ) > ( 0.00000 0.00000 > 1.85986 ) > atom 2 > ( -1.82935 0.00000 > 0.00000 ) > ( 0.00000 -1.82935 > 0.00000 ) > ( 0.00000 0.00000 > -1.82935 ) > > =========================================================================== > and the frequency at gamma point is like this: > =========================================================================== > 0.0000 0.0001 0.0001 1055.5528 1055.5528 > 1274.3693 > =========================================================================== > > so my question is: > > 1)why the LO mode is so small(the exact value is > about 1305 cm-1) when the > TO mode is correct? and how to improve it? > > 2) why the absolute value of effective charge for B > atom and N atom is not > equal? > > any help would be appreciate much! > > > jason > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ____________________________________________________ Sell on Yahoo! Auctions ? no fees. Bid on great items. http://auctions.yahoo.com/ From jasonsun98 at hotmail.com Thu Jul 7 09:42:15 2005 From: jasonsun98 at hotmail.com (sun jason) Date: Thu, 07 Jul 2005 07:42:15 +0000 Subject: [Pw_forum] RE: Pw_forum digest, Vol 1 #685 - 6 msgs In-Reply-To: <20050707053622.3872.41908.Mailman@democritos.sissa.it> Message-ID: about phonon calculation >Message: 6 >Date: Wed, 6 Jul 2005 09:51:30 -0700 (PDT) >From: Eyvaz Isaev >Subject: Re: [Pw_forum] about phonon calculation >To: pw_forum at pwscf.org >Reply-To: pw_forum at pwscf.org > >Hi, > >In order to get correct LO mode you should calculate >phonons using dynmat.x from /pwtools directory. ========================================================================================== the result I presented yesterday(0.0000 0.0001 0.0001 1055.5528 1055.5528 1274.3693) was exactly the one, which have imported the nonlinear term using dynmat.x but the LO mode is also too small after recalculatetion using q2r.x + dynmat.x did you get the LO mode near 1305 cm-1 ? my input file for dynmat.x is like this: ============================================ &input asr=.true., amass(1)=10.81100, amass(2)=14.00674, flfrc='bn111.fc', flfrq='bn.freq' / 9 0.000 0.0 0.0 0.0 0.125 0.0 0.0 0.0 0.250 0.0 0.0 0.0 0.375 0.0 0.0 0.0 0.500 0.0 0.0 0.0 0.625 0.0 0.0 0.0 0.750 0.0 0.0 0.0 0.875 0.0 0.0 0.0 1.000 0.0 0.0 0.0 ============================================= best, jason >Effective charges for B and N are not equal due to >numerical errors in our computer calculations. > >Bests, >Eyvaz. > >--- sun jason wrote: > > > Dear all, > > > > I executed a test of phonon calculation of cubic BN, > > the input for scf is > > like this > > >=========================================================================== > > &SYSTEM > > ibrav = 2, > > celldm(1)= 6.792, > > nat = 2, > > ntyp = 2, > > ecutwfc = 50.0 , > > / > > &ELECTRONS > > conv_thr = 1.0d-9 , > > mixing_beta = 0.7 , > > / > > ATOMIC_SPECIES > > B 10.81100 B.pz-vbc.UPF > > N 14.00674 N.pz-vbc.UPF > > ATOMIC_POSITIONS > > B 0.000000000 0.000000000 0.000000000 > > > > N 0.250000000 0.250000000 0.250000000 > > > > K_POINTS automatic > > 6 6 6 1 1 1 > > >=========================================================================== > > > > and the result of dielectric tensor and effective > > charge is like this: > > > > >=========================================================================== > > Dielectric constant in cartesian axis > > > > ( 4.637335813 0.000000000 > > 0.000000000 ) > > ( 0.000000000 4.637335813 > > 0.000000000 ) > > ( 0.000000000 0.000000000 > > 4.637335813 ) > > > > Effective charges E-U in cartesian axis > > > > atom 1 > > ( 1.85986 0.00000 > > 0.00000 ) > > ( 0.00000 1.85986 > > 0.00000 ) > > ( 0.00000 0.00000 > > 1.85986 ) > > atom 2 > > ( -1.82935 0.00000 > > 0.00000 ) > > ( 0.00000 -1.82935 > > 0.00000 ) > > ( 0.00000 0.00000 > > -1.82935 ) > > > > >=========================================================================== > > and the frequency at gamma point is like this: > > >=========================================================================== > > 0.0000 0.0001 0.0001 1055.5528 1055.5528 > > 1274.3693 > > >=========================================================================== > > > > so my question is: > > > > 1)why the LO mode is so small(the exact value is > > about 1305 cm-1) when the > > TO mode is correct? and how to improve it? > > > > 2) why the absolute value of effective charge for B > > atom and N atom is not > > equal? > > > > any help would be appreciate much! > > > > > > jason > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > >____________________________________________________ >Sell on Yahoo! Auctions ?no fees. Bid on great items. >http://auctions.yahoo.com/ > > >--__--__-- > >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > > >End of Pw_forum Digest From degironc at sissa.it Thu Jul 7 11:27:01 2005 From: degironc at sissa.it (Stefano de Gironcoli) Date: Thu, 07 Jul 2005 11:27:01 +0200 Subject: [Pw_forum] about energy gap References: Message-ID: <42CCF565.4090309@sissa.it> 1) LDA strongly underestimate band gaps how large are your gaps for the parents materials compared to literature values (Exp and Theory) 2) N doping is known to produce large reduction (giant negative bowing) in GaAs. A few % of N actually reduces the gap. 3) disorder tends to close the gaps 4) are your valence-band top and conduction -band bottom states localized on different atoms species. is your cell a kind of nano-heterojunction ? in that case you might have some band-lineup effects that could reduce the gap Dig better in the literature and in your calculations. Compare with known results in similar systems. best regards, Stefano de Gironcoli sun jason wrote: > Dear all, > > I'm a freshman in computational materials, I'm puzzled by a simple > question about energy gap, when I calculated the C2-BN system which > was construtured by mixing cubic BN and diamond. > Well, the computed LDA band gap of c-BN and diamond is around 4.5 and > 4.1 eV, respectively. Therefore, by mixing two compounds with such a > wide band gap (roughly of the order of 4 > eV) one might expect to get a BC2N compound with an Eg between 4.5 and > 4.1eV. Hence, why is our calculated Eg (about 1.0 to 2.0 eV) is much > lower than expected for a C2-BN system? > the unit cell I calculated is crystallined in diamond-like structure > and the atoms are sp3 hybridization. > > Thank you very much! > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From degironc at sissa.it Thu Jul 7 12:00:10 2005 From: degironc at sissa.it (Stefano de Gironcoli) Date: Thu, 07 Jul 2005 12:00:10 +0200 Subject: [Pw_forum] Re. from total energy to cohesive energy References: <1120584160.42cac1e064fb7@webmail.asu.edu> Message-ID: <42CCFD2A.2060808@sissa.it> Cohesive energy is the difference between the energy/atom in the solid and the energy of the isolated atom that you canc compute considering a single atom in a large enough supercell (keep the same cutoff, Gamma-only is ok). stefano Yong Jiang wrote: >Dear All, > >My question is about the cohesive energy. > >In example01 ni.scf.cg.out, total energy is calculated as -85.72249139 ryd (or >~1165.8eV) for one atom in fcc structure (nat=1 and ibrav=2). How to use this >info to get cohesive energy of Ni, which is about 5eV? > >Thanks very much in advance for your help. > >Best regards, >Yong >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > > From degironc at sissa.it Thu Jul 7 12:33:56 2005 From: degironc at sissa.it (Stefano de Gironcoli) Date: Thu, 07 Jul 2005 12:33:56 +0200 Subject: [Pw_forum] about phonon calculation References: Message-ID: <42CD0514.9050805@sissa.it> sun jason wrot > > =========================================================================== > > and the frequency at gamma point is like this: > =========================================================================== > > 0.0000 0.0001 0.0001 1055.5528 1055.5528 1274.3693 > =========================================================================== > > > so my question is: > > 1)why the LO mode is so small(the exact value is about 1305 cm-1) when > the TO mode is correct? and how to improve it? Has the lattice parameter been optimized ? is 50 ryd cutoff sufficient ? is LDA adequate in this case ? is 30 cm-1 out of 1300 really bad compared with present experimental/theoretical accuracy ? > > 2) why the absolute value of effective charge for B atom and N atom is > not equal? > because the ASR is only exactly satisfied in calculations with very large k-point samplig. The violation you report is not large and in any case you can impose ASR in the input of dynmat.x obtaining results equivalent to calculations with better sampling. stefano de Gironcoli From swblelia at sw.ehu.es Thu Jul 7 12:42:43 2005 From: swblelia at sw.ehu.es (Aritz Leonardo) Date: Thu, 07 Jul 2005 12:42:43 +0200 Subject: [Pw_forum] lambda Message-ID: <42CD0723.3030300@sw.ehu.es> Hello all: I have been searching the archive to know why am I obtaining negative values of lambda and gamma and I don't really understand what is going on. As far as I see from the formulas (for gamma and lambda) Shouldn't they be positively defined??? for example for gamma, we have: the square of the matrix element, 2 deltas, ph-frequencies, and BZ volume. Even though I may be far from convergence in my opinion all the values should be positive. Thanks a lot Aritz From giannozz at nest.sns.it Thu Jul 7 14:38:40 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 7 Jul 2005 14:38:40 +0200 Subject: [Pw_forum] lambda In-Reply-To: <42CD0723.3030300@sw.ehu.es> References: <42CD0723.3030300@sw.ehu.es> Message-ID: <200507071438.40514.giannozz@nest.sns.it> On Thursday 07 July 2005 12:42, Aritz Leonardo wrote: > I have been searching the archive to know why am I obtaining negative > values of lambda and gamma you didn't search very well: http://www.democritos.it/pipermail/pw_forum/2004-September/001316.html The calculation of electron-phonon coefficients is quite tricky and involves complex symmetrization operations. I don't know exactly where negative values can come from, but I have never noticed anything wrong in the calculation. In any event, if you get negative values of lambda and gamma, you are seriously far from convergence. 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 patricia at fcq.unc.edu.ar Thu Jul 7 17:59:57 2005 From: patricia at fcq.unc.edu.ar (Patricia Paredes) Date: Thu, 07 Jul 2005 17:59:57 Subject: [Pw_forum] espresso 2.1.4 and ifc8.1.024 In-Reply-To: References: <3.0.1.32.20050701190023.012f7aa0@mail.fcq.unc.edu.ar> Message-ID: <3.0.1.32.20050707175957.011fa2c0@mail.fcq.unc.edu.ar> A non-text attachment was scrubbed... Name: not available Type: text/enriched Size: 2651 bytes Desc: not available Url : /pipermail/attachments/20050707/332559d6/attachment.bin From rjxiao at blem.ac.cn Fri Jul 8 11:35:00 2005 From: rjxiao at blem.ac.cn (Ruijuan Xiao) Date: Fri, 8 Jul 2005 17:35:00 +0800 Subject: [Pw_forum] pseudopotential in phonon calculation Message-ID: <20050708093426.252D611272D@democritos.sissa.it> Dear users, I have a question about how to choose the pseudopotential correctly in the phonon calculation.Because I found that different pseudopotentials influence the results greatly. There are three kinds of elements in the system I studied. Fisrt, I use the ultrasoft pseudopotetials:Na.pbe-n-van.UPF,Co.pbe-nd-rrkjus.UPF and O.pbe-rrkjus.UPF. Almost all the freqencies I got are less than 600cm-1(the maximum is 600.31cm-1). Then I changed to use Norm-Conserving pseudopotential:Na.pz-n-vbc.UPF and O.pz-mt.UPF. Co.pz-nd-rrkjus.UPF is used for the potential of Co,because there is no Norm-Conserving pseudopotential of Co in the pseudopotential download page. And all the other parameters are same with the first calculation. But I got a very different results in the second calculation: the maximum freqence is 703.89cm-1, and it seems that all the frenqencies are enlarged especially for the high freqencies. I am puzzled why the difference is so large? Then how to choose a suitable potential in the calculation? Would you like to give me any information about that? Thank you very much! Best regards, Sincerely, Ruijuan Xiao From roma at srmp19.saclay.cea.fr Fri Jul 8 14:11:57 2005 From: roma at srmp19.saclay.cea.fr (Guido Roma) Date: Fri, 08 Jul 2005 14:11:57 +0200 Subject: [Pw_forum] pseudopotential in phonon calculation In-Reply-To: <20050708093426.252D611272D@democritos.sissa.it> References: <20050708093426.252D611272D@democritos.sissa.it> Message-ID: <1120824717.21431.335.camel@srmp19.saclay.cea.fr> Hello, I would say that, more than from the pseudopotential itself, the difference comes from the exchange correlation functional (gradient corrected PBE versus LDA) for which the pseudopotential has been generated; by the way, you have probably different equilibrium lattice parameters: which one is closer to the experimental value? And bulk modulus? And keep in mind that conergence parameters (wfc and rho cutoff) are not necessarily the same. Guido On Fri, 2005-07-08 at 11:35, Ruijuan Xiao wrote: > Dear users, > > I have a question about how to choose the pseudopotential correctly in the phonon calculation.Because I found that different pseudopotentials influence the results greatly. There are three kinds of elements in the system I studied. Fisrt, I use the ultrasoft pseudopotetials:Na.pbe-n-van.UPF,Co.pbe-nd-rrkjus.UPF and O.pbe-rrkjus.UPF. Almost all the freqencies I got are less than 600cm-1(the maximum is 600.31cm-1). Then I changed to use Norm-Conserving pseudopotential:Na.pz-n-vbc.UPF and O.pz-mt.UPF. Co.pz-nd-rrkjus.UPF is used for the potential of Co,because there is no Norm-Conserving pseudopotential of Co in the pseudopotential download page. And all the other parameters are same with the first calculation. But I got a very different results in the second calculation: the maximum freqence is 703.89cm-1, and it seems that all the frenqencies are enlarged especially for the high freqencies. I am puzzled why the difference is so large? Then how to choose a suitable potential! in > the calculation? Would you like to give me any information about that? > > Thank you very much! > > > > Best regards, > > Sincerely, > Ruijuan Xiao > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- Guido Roma -- CEA-Saclay - DEN/DMN/SRMP Bat.520/130 Phone: [+33]-1-69085738 -- Fax: ...6867 -- Mobile: [+33]-6-20069085 From zjspam at gmail.com Fri Jul 8 16:32:32 2005 From: zjspam at gmail.com (Jian ZHOU) Date: Fri, 8 Jul 2005 22:32:32 +0800 Subject: [Pw_forum] how to calculate the polarization of a crystal by berry phase Message-ID: <8d38992a05070807323e3b4ba0@mail.gmail.com> Dear all, In the example10 of pwscf, it run a berry phase calculation and get the Born effective charge of Pb. However, I am now trying to calculate the polarization of a Crystal, for example for a tetragonal BaTiO3. I firstly do a scf calculation with the following input file: &control calculation = 'scf' restart_mode = 'from_scratch' pseudo_dir = '/home/zj/espresso-2.1.4/pseudo/' outdir = '/root/zj/compute14/pwscf/tet_batio3' tstress = .true., / &system ibrav=6 celldm(1)= 7.5435 celldm(3)=1.0110 nat=5 ntyp=3 ecutwfc=30.0 occupations = 'fixed' degauss=0.00 / &electrons conv_thr = 1e-9, mixing_beta = 0.3d0 / 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 O 0 0.5 0.48500000 O 0.5 0 0.48500000 O 0.5 0.5 -0.02500000 Ti 0.5 0.5 0.50135000 Ba 0 0 0 K_POINTS {automatic} 4 4 4 1 1 1 And then do a berry phase calculation with the input file: &control calculation = 'nscf' pseudo_dir = '/home/zj/espresso-2.1.4/pseudo/' outdir = '/root/zj/compute14/pwscf/tet_batio3' lberry = .true. gdir = 3 nppstr = 7 / &system ibrav=6 celldm(1)= 7.5435 celldm(3)=1.0110 nat=5 ntyp=3 ecutwfc=30.0 occupations = 'fixed' degauss=0.00 / &electrons conv_thr = 1e-9 mixing_beta = 0.3d0 / 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 O 0 0.5 0.48500000 O 0.5 0 0.48500000 O 0.5 0.5 -0.02500000 Ti 0.5 0.5 0.50135000 Ba 0 0 0 K_POINTS {automatic} 4 4 7 1 1 1 The final output is : SUMMARY OF PHASES ~~~~~~~~~~~~~~~~~ Ionic Phase: -0.44095 (mod 2) Electronic Phase: 0.63353 (mod 2) TOTAL PHASE: 0.19258 (mod 2) VALUES OF POLARIZATION ~~~~~~~~~~~~~~~~~~~~~~ The calculation of phases done along the direction of vector 3 of the reciprocal lattice gives the following contribution to the polarization vector (in different units, and being Omega the volume of the unit cell): P = 1.4687319 (mod 15.2529570) (e/Omega).bohr P = 0.0033843 (mod 0.0351467) e/bohr^2 P = 0.1934880 (mod 2.0093958) C/m^2 The polarization direction is: ( 0.00000 , 0.00000 , 1.00000 ) It is said that the final polarization is 0.1934880 C/m^2. And the experiment value is about 0.26 C/m^2. A more high cutoff energy and more k points is tested, and found the polarization does not change much. So, I am not sure If the procedure is correct. If it is correct, why the value is much smaller than the experiment one. Thank you. Best wishes, Jian From vdiep at ictp.it Fri Jul 8 17:31:06 2005 From: vdiep at ictp.it (Diep Vinh Quang) Date: Fri, 8 Jul 2005 17:31:06 +0200 (CEST) Subject: [Pw_forum] What is the different between Born effective charge and effective In-Reply-To: <8d38992a05070807323e3b4ba0@mail.gmail.com> References: <8d38992a05070807323e3b4ba0@mail.gmail.com> Message-ID: <33745.10.41.38.21.1120836666.squirrel@webmail1.ictp.trieste.it> Dear all, Right now I am calculating the IR spectrum of Polyethylene (12 atom in a unit cell) by using the example09. I also deal with the effective charge and the dynamical matrix to calculate IR. But I get the result for the effective charge like the following (it is a tensor): atom 1 = C 2 = H 3 = H 4 = C 5 = H 6 = H ........ Effective Charges E-U: Z_{alpha}{s,beta} atom # 1 -0.099427475790 -0.248752090473 0.001043795428 -0.069615072695 -0.159907527964 0.007634655521 -0.001753156183 0.006377735929 -1.312054019801 atom # 2 0.077956360437 -0.030538688100 0.000934947973 -0.199637665554 -0.046995649249 0.001576613923 -0.000850707804 0.003202247825 0.059757957907 atom # 3 0.032830541054 0.147773617226 0.000600323072 0.175304985018 -0.199147711217 -0.004431652344 0.003510661051 -0.002839150183 0.046754448589 atom # 4 -0.022464377689 -0.205009989734 0.000887740549 -0.077166162241 -0.161294181774 0.011823645473 -0.002437562298 0.010399149740 -1.323252940006 atom # 5 -0.058408457609 -0.067254988611 -0.000085228083 -0.254215018726 -0.028258016054 0.001283680975 -0.001137843246 0.001977691596 0.058849948380 atom # 6 0.008147259866 0.125032673064 0.001112013451 0.140717895758 -0.170289293702 -0.004346617517 0.003720246553 -0.002531134573 0.059304718408 atom # 7 -0.102543543931 -0.249140592599 0.001444711453 -0.071204041408 -0.156518468321 0.005522539460 0.005068839433 -0.000136373867 -1.310914795339 atom # 8 0.080427330490 -0.029766382005 0.000949440595 -0.197677260138 -0.048161049366 0.001682025760 -0.000794687824 0.003304342703 0.060723469122 atom # 9 0.032606029639 0.148487590659 0.000617626001 0.175284391800 -0.197008577608 -0.004454326407 0.003562264841 -0.002896819928 0.047484400215 atom # 10 -0.023549648226 -0.204184590745 0.000862655075 -0.076420841674 -0.158655937842 0.011795766380 0.000669693064 0.011438938194 -1.322555094138 atom # 11 -0.056614516798 -0.066225081854 -0.000066149431 -0.252832726795 -0.029751471027 0.001395607163 -0.001131325363 0.002098938278 0.059573058213 atom # 12 0.007792801562 0.126138888731 0.001135706575 0.141185264261 -0.168958178617 -0.004372196571 0.003739948221 -0.002524338993 0.059935664421 Therefore my question is: What is the different between Born effective charge and the above effective charge? Or are they the same? And on the other hand, how can we impose sysmetry condition for the caculation of Z* (sum of Z* of each atom in one direction must = zero) Thank you in advance. Vinh From yjiang at asu.edu Sat Jul 9 00:27:47 2005 From: yjiang at asu.edu (Yong Jiang) Date: Fri, 08 Jul 2005 15:27:47 -0700 (MST) Subject: [Pw_forum] SCF not stop when it reaches the default conv_thr In-Reply-To: <33745.10.41.38.21.1120836666.squirrel@webmail1.ictp.trieste.it> References: <8d38992a05070807323e3b4ba0@mail.gmail.com> <33745.10.41.38.21.1120836666.squirrel@webmail1.ictp.trieste.it> Message-ID: <1120861667.42cefde31ed1e@webmail.asu.edu> Dear All, This is a question about conv_thr: I use the default value for conv_thr (default = 1.D-6)in my supercell (including 32 Ni atoms). Why the scf calculation is still running when it reaches ethr = 6.39E-07 since interation #22 (pls see below)? ---------------------------------- iteration # 21 ecut= 27.00 ryd beta=0.30 Conjugate-gradient style diagonalization ethr = 1.52E-06, avg # of iterations = 3.0 total cpu time spent up to now is 31997.23 secs total energy = -2748.70062200 ryd estimated scf accuracy < 0.00204526 ryd total magnetization = 22.86 Bohr mag/cell absolute magnetization = 26.19 Bohr mag/cell iteration # 22 ecut= 27.00 ryd beta=0.30 Conjugate-gradient style diagonalization ethr = 6.39E-07, avg # of iterations = 3.1 total cpu time spent up to now is 33483.75 secs total energy = -2748.68225286 ryd estimated scf accuracy < 0.02254775 ryd total magnetization = 22.86 Bohr mag/cell absolute magnetization = 26.22 Bohr mag/cell iteration # 23 ecut= 27.00 ryd beta=0.30 Conjugate-gradient style diagonalization ethr = 6.39E-07, avg # of iterations = 3.1 total cpu time spent up to now is 35054.16 secs total energy = -2748.69429201 ryd --More--(98%) -------------------------------------------- Thanks very much in advance for your kind help. Good weekend, Yong From rjxiao at blem.ac.cn Sat Jul 9 08:26:56 2005 From: rjxiao at blem.ac.cn (Ruijuan Xiao) Date: Sat, 9 Jul 2005 14:26:56 +0800 Subject: [Pw_forum] pseudopotential in phonon calculation Message-ID: <20050709062541.B6A87112732@democritos.sissa.it> Dear Guido, Thank you very much. I forgot to mention that I use the wfc and rho cutoff as 30Ry and 300Ry in the fisrt calculation since "for a reasonable ultrasoft psp should be between 25 and 30 Ry, and unlikely to be more than 40 Ry".In the second calculation, I use the wfc and rho cutoff as 50Ry and 500Ry. The freqencies are calculated at Gamma point.I use the same lattice parameters in the two calculations, and fix the lattice parameter, then optimize the positions of atoms. I do the calculation to compare with the results of Li (Physical Review B,70,144518). I found that none results of mine is similar with his. I agree with you that the difference has some relationship to the GGA or LDA. Do you mean that we should judge which potential is more suitable by comparing the results with experiments? Best regards, Sincerely, Ruijuan Xiao From Steven.Kirk at htu.se Sat Jul 9 09:56:23 2005 From: Steven.Kirk at htu.se (Steven Kirk) Date: Sat, 09 Jul 2005 09:56:23 +0200 Subject: [Pw_forum] Functionals, pseudopotentials, etc. for H2O ice Message-ID: <42CF8327.3010804@htu.se> Hello, Does anyone on this list have some recommendations for good functionals, pseudopotentials and other simulation configuration data (e.g. cutoffs) for PWSCF simulations of low-pressure tetrahedrally-coordinated H2O ice? I already have some crystal structure data to work with, but I need recommendations for the other simulation setup information. One important requirement is that the topology of the charge density be accurate, so that charge density distribution on both covalent (sigma) and 'H-bond' bonding interactions is as accurately represented as possible. This suggests the use of ultrasoft pseudopotentials, but I'd be grateful for some advice on this. Many thanks in advance, Steve Kirk -- Dr. Steven R. Kirk Dept. of Computer Science, TMD http://taconet.webhop.org Univ. of Trollhattan/Uddevalla Ph: +46 520 475318 Fax:+46 520 475399 P.O. Box 957 Trollhattan 461 29 SWEDEN From vdiep at ictp.it Sat Jul 9 14:36:19 2005 From: vdiep at ictp.it (Diep Vinh Quang) Date: Sat, 9 Jul 2005 14:36:19 +0200 (CEST) Subject: [Pw_forum] Acousic Sum rule In-Reply-To: <42CF8327.3010804@htu.se> References: <42CF8327.3010804@htu.se> Message-ID: <33988.10.41.38.21.1120912579.squirrel@webmail3.ictp.trieste.it> Dear all, I am calculating the effect charge. And I using the example09 for calculating the phonon at Gamma (q=0). The result in example09 give like the following (SiH4): z*( 1) -0.4485 0.0000 0.0000 0.0000 -0.4485 0.0000 0.0000 0.0000 -0.4485 z*( 2) -1.1347 -0.5787 -0.5787 -0.5787 -1.1347 -0.5787 -0.5787 -0.5787 -1.1347 z*( 3) -1.1347 -0.5787 0.5787 -0.5787 -1.1347 0.5787 0.5787 0.5787 -1.1347 z*( 4) -1.1347 0.5787 -0.5787 0.5787 -1.1347 0.5787 -0.5787 0.5787 -1.1347 z*( 5) -1.1347 0.5787 0.5787 0.5787 -1.1347 -0.5787 0.5787 -0.5787 -1.1347 I just want to ask whether this result is reasonable or not? because it doesn't satify the Acoustic Sum Rule (Sham 1969 or Vogl 1978) in that: the sum of the effective charge Z_{ij} over all atom = 0. In this case we can see that it not true for the case Z_{33}: Z_{33}(1)+Z_{33}(2)+Z_{33}(3)+Z_{33}(4)+Z_{33}(5)= -0.4485+ 4*( -1.1347) differ from zero? How can we impose this condition in the above calculation? Thank you very much. Vinh ------------------------------------------------- Diep Vinh Quang C/o Diploma Program, The Abdus Salam International Center for Theoretical Physics(ICTP), 11 Strada Costiera, 34014, Trieste,Italy. http://www.ictp.it From jasonsun98 at hotmail.com Sat Jul 9 17:07:22 2005 From: jasonsun98 at hotmail.com (sun jason) Date: Sat, 09 Jul 2005 15:07:22 +0000 Subject: [Pw_forum] about band gap In-Reply-To: <20050708053634.10888.36235.Mailman@democritos.sissa.it> Message-ID: Much appreciation for your kind answers! 1) we mentioned that LDA underestimate band gaps, the experimental gaps for c-BN and diamond are 6.4 eV and 5.6 eV, and the LDA results are about 4.5 and 4.1, respectively. why the LDA gap of B-C-N system is not in the range of 4.1~4.5 eV? 2)N doping may couse gap reduction, but should I looks my B-C-N system as N doping in boron carbide? doping may couse impurity (donor or acceptor), but my systems are perfect crystals. they are in the case of doping? 3)as above one, are there disorders in perfect crystals( in computational model)? 4)yes, the top of the valence band or the bottom of conduction band is a little localized, but very small, within 0.5 eV. the gap is still small compared with the parents materials even we add the lineup value. Thank you very much! ==================================================== Message: 2 Date: Thu, 07 Jul 2005 11:27:01 +0200 From: Stefano de Gironcoli To: pw_forum at pwscf.org Subject: Re: [Pw_forum] about energy gap Reply-To: pw_forum at pwscf.org 1) LDA strongly underestimate band gaps how large are your gaps for the parents materials compared to literature values (Exp and Theory) 2) N doping is known to produce large reduction (giant negative bowing) in GaAs. A few % of N actually reduces the gap. 3) disorder tends to close the gaps 4) are your valence-band top and conduction -band bottom states localized on different atoms species. is your cell a kind of nano-heterojunction ? in that case you might have some band-lineup effects that could reduce the gap Dig better in the literature and in your calculations. Compare with known results in similar systems. best regards, Stefano de Gironcoli sun jason wrote: > Dear all, > > I'm a freshman in computational materials, I'm puzzled by a simple > question about energy gap, when I calculated the C2-BN system which > was construtured by mixing cubic BN and diamond. > Well, the computed LDA band gap of c-BN and diamond is around 4.5 and > 4.1 eV, respectively. Therefore, by mixing two compounds with such a > wide band gap (roughly of the order of 4 > eV) one might expect to get a BC2N compound with an Eg between 4.5 and > 4.1eV. Hence, why is our calculated Eg (about 1.0 to 2.0 eV) is much > lower than expected for a C2-BN system? > the unit cell I calculated is crystallined in diamond-like structure > and the atoms are sp3 hybridization. > > Thank you very much! From jasonsun98 at hotmail.com Sat Jul 9 17:22:35 2005 From: jasonsun98 at hotmail.com (sun jason) Date: Sat, 09 Jul 2005 15:22:35 +0000 Subject: [Pw_forum] RE: Pw_forum digest, Vol 1 #686 - 7 msgs-about phonon calculation In-Reply-To: <20050708053634.10888.36235.Mailman@democritos.sissa.it> Message-ID: Much thank to stefano de Gironcoli, > 1)why the LO mode is so small(the exact value is about 1305 cm-1) when > the TO mode is correct? and how to improve it? Has the lattice parameter been optimized ? is 50 ryd cutoff sufficient ? is LDA adequate in this case ? is 30 cm-1 out of 1300 really bad compared with present experimental/theoretical accuracy ? ------------------------------ the optimized lattice is not reach the correct phonon modes, yet. much high cutoff and k-points till to 80 ryd and 16*16*16 are tested. the LO mode is about 30 cm-1 smaller than experimental value. I think 30 cm-1 is a big gap,hehe:) the result from direct method provide by professor K.Parlinski is about 1054 and 1304 cm-1 > > 2) why the absolute value of effective charge for B atom and N atom is > not equal? > because the ASR is only exactly satisfied in calculations with very large k-point samplig. The violation you report is not large and in any case you can impose ASR in the input of dynmat.x obtaining results equivalent to calculations with better sampling. Yes, when I use 16*16*16, they are equal in principle. stefano de Gironcoli From yay451 at mail.usask.ca Sat Jul 9 21:04:12 2005 From: yay451 at mail.usask.ca (yay451 at mail.usask.ca) Date: Sat, 09 Jul 2005 13:04:12 -0600 Subject: [Pw_forum] problems while running examples in LAM-MPI Message-ID: <1120935852.42d01fac604e6@webmail.usask.ca> Dear PWscf users, I got a problem in running examples. My machine is a 128-cpu Beowulf cluster. I installed ifc 8.1 and LAM-MPI 7.1.1 yesterday. Examples can be executed in single processor. But when I try to run them on multiple CPUs, they always finishes in error. For the use of LAM-MPI, I modified hostfile myself to assign 6 nodes and used lamboot command.The output files look like this: /home/yao/newproject/espresso-2.1.4/examples/example02 : starting This example shows how to use pw.x and ph.x to calculate phonon frequencies at Gamma and X for Si, C in diamond structure and fcc-Ni. executables directory: /home/yao/newproject/espresso-2.1.4/bin pseudo directory: /home/yao/newproject/espresso-2.1.4/pseudo temporary directory: /home/yao/tmp checking that needed directories and files exist... done running pw.x as: mpirun C /home/yao/newproject/espresso-2.1.4/bin/pw.x running ph.x as: mpirun C /home/yao/newproject/espresso-2.1.4/bin/ph.x cleaning /home/yao/tmp... done running the scf calculation...----------------------------------------------------------------------------- One of the processes started by mpirun has exited with a nonzero exit code. This typically indicates that the process finished in error. If your process did not finish in error, be sure to include a "return 0" or "exit(0)" in your C code before exiting the application. PID 8005 failed on node n1 (192.168.20.1) with exit status 1. Could anyone be so kind to tell me what is wrong here? I guess there is sth. wrong in my configuration of LAM-MPI. I appreciate if you could tell me a little bit about how to configure LAM-MPI. By the way, mpich is also installed by the root user; I am not sure whether this is also a problem. Thank you very much! Yansun Yao Department of Physics University of Saskatchewan From axel.kohlmeyer at theochem.ruhr-uni-bochum.de Sun Jul 10 15:54:16 2005 From: axel.kohlmeyer at theochem.ruhr-uni-bochum.de (Axel Kohlmeyer) Date: Sun, 10 Jul 2005 15:54:16 +0200 (CEST) Subject: [Pw_forum] Functionals, pseudopotentials, etc. for H2O ice In-Reply-To: <42CF8327.3010804@htu.se> Message-ID: On Sat, 9 Jul 2005, Steven Kirk wrote: SK> Hello, steve, SK> Does anyone on this list have some recommendations for good functionals, SK> pseudopotentials and other simulation configuration data (e.g. cutoffs) SK> for PWSCF simulations of low-pressure tetrahedrally-coordinated H2O ice? SK> I already have some crystal structure data to work with, but I need SK> recommendations for the other simulation setup information. i assume, that you are aware of the fact, that the investigation of the structure ice and even more so of liquid water from various models (classical or ab initio or anthing in between) is and has been a controversially discussed topic for decades. so _the_ best set of parameters does not exist. every choice so far had some advantages and some shortcomings, so people tend to use what works well enough for what they want to know. SK> One important requirement is that the topology of the charge density be SK> accurate, so that charge density distribution on both covalent (sigma) SK> and 'H-bond' bonding interactions is as accurately represented as SK> possible. This suggests the use of ultrasoft pseudopotentials, but I'd SK> be grateful for some advice on this. your main problem here should be the representation of the dispersion interaction contribution at all, which is not very good in plain DFT. there have been serveral suggestions on how to overcome this limitation: using an additional and empirically fitted attractive lennard-jones potential, modifying the nonlocal component of pseudopotentials using a pertubation theory based optimization scheme, and most rigorously using hybrid functionals, that include hartree-fock or exact-exchange. to the best of my knowledge this is currently being implemented in pwscf and in at least one other similar package i know about, but not yet available in a released version. unfortunately, using hybrid functionals, however successful and efficient they are with local (gaussian) basis sets, pose a severe performance impact on calculations with pseudopotentials and a plane wave basis set. in the end, you are probably best off starting with some simple benchmark calculations to see how well the available choices (pseudopotentials, functionals, parameters) fit your requirements and how much performance you are willing to sacrifice for how much additional accuracy. with best regards, axel. SK> SK> Many thanks in advance, SK> Steve Kirk SK> -- ======================================================================= Dr. Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/ ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From jasonsun98 at hotmail.com Sun Jul 10 16:06:59 2005 From: jasonsun98 at hotmail.com (sun jason) Date: Sun, 10 Jul 2005 14:06:59 +0000 Subject: [Pw_forum] about phonon and structural optimization In-Reply-To: <20050710053629.21987.69477.Mailman@democritos.sissa.it> Message-ID: Hi, eyvaz, Thank you to answer my questions:) did you calculate the phonon mode of cubic BN before? today, I recalculate the phonon using PBE potentials, the result is better now, but the LO mode is 1294 cm-1, about 10 cm-1 less than the experimental value. I calculated the LO/TO spliting using q2r.x + matdyn.x. ~~~~~~~~~~~~ but the user manual says, the dynmat.x should be use to calculate LO/TO spliting and IR. ~~~~~~~~~~~ so which program do you use to get LO/TO spliting? if you use pwtools dynmat.x, how to use it? Thank you in advance. another question is about cell optimization. 2) does pw.x can calculate from the optimized cell automatically? or I should read the output file of vc-relax and write the coordinate into the input file of pw.x manually? 3) I perform a vc-relaxtion using PBE USPP, the optimized lattice constants is larger than the experimental value. why? it's reasonable? jason Message: 6 Date: Wed, 6 Jul 2005 09:51:30 -0700 (PDT) From: Eyvaz Isaev Subject: Re: [Pw_forum] about phonon calculation To: pw_forum at pwscf.org Reply-To: pw_forum at pwscf.org Hi, In order to get correct LO mode you should calculate phonons using dynmat.x from /pwtools directory. Effective charges for B and N are not equal due to numerical errors in our computer calculations. Bests, Eyvaz. From marzari at MIT.EDU Sun Jul 10 16:14:33 2005 From: marzari at MIT.EDU (Nicola Marzari) Date: Sun, 10 Jul 2005 10:14:33 -0400 Subject: [Pw_forum] Functionals, pseudopotentials, etc. for H2O ice In-Reply-To: References: Message-ID: <42D12D49.5050304@mit.edu> > > in the end, you are probably best off starting with some > simple benchmark calculations to see how well the available > choices (pseudopotentials, functionals, parameters) fit your > requirements and how much performance you are willing to sacrifice > for how much additional accuracy. Thanks Axel for all the comments. Steve - the H and O PBE RRKJ3 pseudopotentials from the pwscf distribution are as accurate as they come. DFT using local/semi-local GGAs is only partially satisfactory in the description of water or ice; even more than exchange-correlatione effects, though, zero-point motion can be very important, and sometimes proton tunnelling. As Axel mentioned, there is a wide body of literature on water and ice - you could check the papers by the Parrinello group on both systems, and *especially* two recent (2004) JCPs of Galli and coworkers. If you'd like to compare some of your own results, we have a recent JCP (Sit and Marzari 2005) with Espresso (pwscf and cp) calculations, using the above-mentioned pseudopotentials. Best, 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 617.2586534 marzari at mit.edu http://nnn.mit.edu From marzari at MIT.EDU Sun Jul 10 16:30:52 2005 From: marzari at MIT.EDU (Nicola Marzari) Date: Sun, 10 Jul 2005 10:30:52 -0400 Subject: [Pw_forum] espresso 2.1.4 and ifc8.1.024 In-Reply-To: <3.0.1.32.20050707175957.011fa2c0@mail.fcq.unc.edu.ar> References: <3.0.1.32.20050701190023.012f7aa0@mail.fcq.unc.edu.ar> <3.0.1.32.20050707175957.011fa2c0@mail.fcq.unc.edu.ar> Message-ID: <42D1311C.2020507@mit.edu> Thanks a lot Patricia - these are *exceedingly* useful tests. We have even a standard test for CP: http://www.democritos.it/pipermail/pw_forum/2004-October/001483.html Axel had kindly provided some results, still in the repository, and that I should add to the report. If you'd like, we can add yours as a standard test for PWSCF. Of course, you should switch to a "marzari-vanderbilt" smearing, then :-) . nicola Patricia Paredes wrote: > Hi Axel, > I made some experiments compiling all the versions of pwscf since 2.0.3 > using ifc8.1.024, cluster mkl 7.1 and compiling mpi with the same compiler. > first iteration > 2.0.3 117.11 secs > 2.0.4 109.89 secs > 2.1.1 110.03 secs > 2.1.2 164.93 secs > 2.1.3 152.52 secs > 2.1.4 154.43 secs > > So, the breaking point was version 2.1.1 (btw, v2.1.0 didn?t work). In > the web page it is stated that it was a modification in soubroutine > electrons.f90 to make a double check... (By carlo sbraccia), perhaps > there is the point: the double check. I was taking a look to the final > statistics and the major differences were at electrons.f90 (Sorry, I > don?t have the numbers now since I overwrite the files :). > > The input file was: > > cuatricapa de Au > Au > &control > calculation='relax' > restart_mode='from_scratch', > prefix='prueba', > pseudo_dir = '/home/pwscf/pseudo/', > outdir='/home/patricia/plata/temp/' > / > &system > ibrav=0, > nat=17, > ntyp=4, > ecutwfc = 22.0, > ecutrho = 180.0 > celldm(1)=7.89304630 > degauss=0.03 > occupations='smearing' > smearing='methfessel-paxton' > / > &electrons > mixing_mode = 'local-TF' > conv_thr = 1.0d-6 > mixing_beta = 0.5 > diagonalization = 'cg' > / > &ions > ion_dynamics = 'bfgs' > / > ATOMIC_SPECIES > S 32.06 S.gpbe.vdb > C 12.01 C.gpbe.vdb > H 1 H.gpbe.vdb > Au 196.96 Au.pbe-nd-van.UPF > ATOMIC_POSITIONS {alat} > H -0.187034332 0.221830025 4.921552258 > H -0.013559323 0.486812870 4.628026988 > H 0.242342701 0.256878798 4.888036613 > C 0.014538967 0.257636565 4.754725119 > S 0.016846606 -0.082426513 4.477818864 > Au 0.383806249 0.011800463 4.017799873 > Au 1.076105787 -0.011642255 3.950466565 > Au 1.768205765 0.011736229 4.018611550 > Au 0.706660429 -0.205310574 3.426597457 > Au 1.418423038 -0.205689922 3.424408560 > Au 1.062125184 0.398360581 3.405678032 > Au 0.701778921 0.206802972 2.843998759 > Au 1.435470249 0.207226266 2.845135398 > Au 1.067569501 -0.420576720 2.851998036 > Au 0.353553300 0.000000000 2.267949200 0 0 0 > Au 1.060660200 0.000000000 2.267949200 0 0 0 > Au 1.767767000 0.000000000 2.267949200 0 0 0 > CELL_PARAMETERS {cubic} > 1.0606602 0.6123724 0.0 > 1.0606602 -0.6123724 0.0 > 0.0000000 0.0 7.0 > K_POINTS {automatic} > 4 4 1 0 0 0 > > > *_Prof. Dr. Patricia Paredes Olivera > _*Facultad de Ciencias Qu?micas > Universidad Nacional de C?rdoba > ph/fax: 054-0351-434-4972 > _______________________________________________ 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 617.2586534 marzari at mit.edu http://nnn.mit.edu From konstantin_kudin at yahoo.com Sun Jul 10 16:59:38 2005 From: konstantin_kudin at yahoo.com (Konstantin Kudin) Date: Sun, 10 Jul 2005 07:59:38 -0700 (PDT) Subject: [Pw_forum] Functionals, pseudopotentials, etc. for H2O ice In-Reply-To: Message-ID: <20050710145938.37073.qmail@web52014.mail.yahoo.com> --- Axel Kohlmeyer wrote: > your main problem here should be the representation of the dispersion > interaction contribution at all, which is not very good in plain DFT. > there have been serveral suggestions on how to overcome this > limitation: > using an additional and empirically fitted attractive lennard-jones > potential, modifying the nonlocal component of pseudopotentials using > > a pertubation theory based optimization scheme, and most rigorously > using hybrid functionals, that include hartree-fock or > exact-exchange. Well, it is not that useful to lump all types of weak interactions into dispersion that DFT supposedly does not describe. As a matter of fact, hydrogen bonds do have a rather strong Coulomb and density overlap component, so plain DFT description of that can be quite reasonable. Such bonds are quite different from those in something like He dimer, where the lowest level of physically proper description is some kind of perturbation theory like MP2. Hybrid functionals do work somewhat better, however, the expense is much greater as well. I think there was a recent paper assessing a new meta-GGA functional TPSS for hydrogen bonds, and it probably also contains data for other functionals as well (Staroverov, Scuseria, JCP, 2004 [or 2003]). If I recall correctly they concluded that TPSS could be as good as hybrids for hydrogen bonds while being cheaper computationally. Kostya ____________________________________________________ Sell on Yahoo! Auctions ? no fees. Bid on great items. http://auctions.yahoo.com/ From axel.kohlmeyer at theochem.ruhr-uni-bochum.de Sun Jul 10 19:40:23 2005 From: axel.kohlmeyer at theochem.ruhr-uni-bochum.de (Axel Kohlmeyer) Date: Sun, 10 Jul 2005 19:40:23 +0200 (CEST) Subject: [Pw_forum] Functionals, pseudopotentials, etc. for H2O ice Message-ID: On Sun, 10 Jul 2005, Konstantin Kudin wrote: kostya, thank you for proving the point that this _is_ a controversial topic. ;-) KK> Well, it is not that useful to lump all types of weak interactions KK> into dispersion that DFT supposedly does not describe. As a matter of KK> fact, hydrogen bonds do have a rather strong Coulomb and density KK> overlap component, so plain DFT description of that can be quite KK> reasonable. Such bonds are quite different from those in something like KK> He dimer, where the lowest level of physically proper description is KK> some kind of perturbation theory like MP2. Hybrid functionals do work KK> somewhat better, however, the expense is much greater as well. it was not was not my intention to make plain DFT look bad. it really depends on what you want to investigate, so you have to test what works well. the representation of hydrogen bonds can vary a lot between functionals. one example of comparing the performance of functionals and methods, which also includes a water dimer, is in X. Wu et al., JCP 115(2001), p8748 KK> I think there was a recent paper assessing a new meta-GGA functional KK> TPSS for hydrogen bonds, and it probably also contains data for other KK> functionals as well (Staroverov, Scuseria, JCP, 2004 [or 2003]). If I KK> recall correctly they concluded that TPSS could be as good as hybrids KK> for hydrogen bonds while being cheaper computationally. many seem to be satisfied with BLYP, which gets a positive rating for the water dimer in the paper above, another example would be the HCTH functional which - to the best of my knowledge - is still continously refined, another alternative, as one example for a method where the potential is modified, would be the method presented in PRL 93 (2004), 153004, and this list is definitely not complete. my main point remains, that there still is not _the_ method/functional/potential that works well for water under _all_ circumstances. so that one has to be careful in validating whatever approach one uses. not to mention the issues addressed by nicola (it is particularly embarrassing that i forgot to think about proton tunneling, considering who has signed my current contract. ;-)) ) best regards, axel. KK> KK> Kostya KK> KK> KK> KK> ____________________________________________________ KK> Sell on Yahoo! Auctions ? no fees. Bid on great items. KK> http://auctions.yahoo.com/ KK> _______________________________________________ KK> Pw_forum mailing list KK> Pw_forum at pwscf.org KK> http://www.democritos.it/mailman/listinfo/pw_forum KK> KK> -- ======================================================================= Dr. Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/ ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From eyvaz_isaev at yahoo.com Mon Jul 11 00:20:20 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Sun, 10 Jul 2005 15:20:20 -0700 (PDT) Subject: [Pw_forum] about phonon and structural optimization In-Reply-To: Message-ID: <20050710222020.7684.qmail@web60312.mail.yahoo.com> Hi, First, I would like to answer questions about phonons. 1. No, I did nothing with cubic BN, but I remember a paper of Parlinski. 2. You can calculate LO/TO splitting for semiconductors and insulators using both dynmat.x and matdyn.x. The result will be the same. Input file for dynmat.x looks like phonons of Ti2SiC at 0.0000000,0.0000000,0.0000000 &input amass(1)=47.87, amass(2)=28.09, amass(3)=12.01, asr=.true., fildyn='Ti2SiC.dynG', filout='Ti2SiC.out', filmol='Ti2SiC.molden' &end 0.0000000,0.0000000,0.0000000 Command line as usual $BIN_DIR/dynmat.x < input_file > output_file The manual says correct, of course, but in this particular case you have a choice. 3. For GGA in most cases larger lattice parameter will be obtained, so, phonon frequences reduced. 4. 10cm^-1 of 1304cm^-1 is about 10*100/1300=0.8%. I guess you have got good agreement with experimental data. Bests, Eyvaz. --- sun jason wrote: > Hi, eyvaz, > > Thank you to answer my questions:) > > did you calculate the phonon mode of cubic BN > before? > > today, I recalculate the phonon using PBE > potentials, the result is better > now, but the LO mode is 1294 cm-1, about 10 cm-1 > less than the experimental > value. > > I calculated the LO/TO spliting using q2r.x + > matdyn.x. > > ~~~~~~~~~~~~ > > but the user manual says, the dynmat.x should be use > to calculate LO/TO > spliting and IR. > ~~~~~~~~~~~ > > so which program do you use to get LO/TO spliting? > > > if you use pwtools dynmat.x, how to use it? > > Thank you in advance. > > > > another question is about cell optimization. > > 2) does pw.x can calculate from the optimized cell > automatically? or I > should read the output file of vc-relax and write > the coordinate into the > input file of pw.x manually? > > 3) I perform a vc-relaxtion using PBE USPP, the > optimized lattice constants > is larger than the experimental value. why? it's > reasonable? > > > jason > > Message: 6 > Date: Wed, 6 Jul 2005 09:51:30 -0700 (PDT) > From: Eyvaz Isaev > Subject: Re: [Pw_forum] about phonon calculation > To: pw_forum at pwscf.org > Reply-To: pw_forum at pwscf.org > > Hi, > > In order to get correct LO mode you should calculate > phonons using dynmat.x from /pwtools directory. > > Effective charges for B and N are not equal due to > numerical errors in our computer calculations. > > Bests, > Eyvaz. > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > __________________________________ Yahoo! Mail Stay connected, organized, and protected. Take the tour: http://tour.mail.yahoo.com/mailtour.html From leoant21 at hotmail.com Mon Jul 11 03:12:35 2005 From: leoant21 at hotmail.com (=?gb2312?B?1cUguumx8g==?=) Date: Mon, 11 Jul 2005 01:12:35 +0000 Subject: [Pw_forum] Questions on phonon calculation of MgB2 Message-ID: Dear all: Now I am trying to do some calculation on MgB2, but after I have do phonon calculation on an automatic generated 6*6*6 MP k-points and try to use q2r.x to get the dynamic matrix, the program give me the error info like below: nqs=0.00000 0.0000 0.28867(which was generated automatically by previous scf calculation) ****************************************************************************************8 q point not allowed ***************************************************************************************** stopping...... so may someone tell what's wrong? And I want to know is there any rule when one choose a "uniform grid" which can be used for later q2r calculation when doing phonon calculation? Bests _________________________________________________________________ ???? MSN Explorer: http://explorer.msn.com/lccn/ From giannozz at nest.sns.it Mon Jul 11 08:49:19 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 11 Jul 2005 08:49:19 +0200 Subject: [Pw_forum] Questions on phonon calculation of MgB2 In-Reply-To: References: Message-ID: <200507110849.19157.giannozz@nest.sns.it> On Monday 11 July 2005 03:12, ? ?? wrote: > [...] after I have do phonon calculation on an automatic generated > 6*6*6 MP k-points and try to use q2r.x to get the dynamic matrix, > the program give me the error info like below: [...] > q point not allowed beware truncation errors: when you write k-points in input data, use at least 6 significant figures. The code may not be able to figure out that 0.28867 is actually sqrt(3)/6. 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 giannozz at nest.sns.it Mon Jul 11 08:55:08 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 11 Jul 2005 08:55:08 +0200 Subject: [Pw_forum] about phonon and structural optimization In-Reply-To: <20050710222020.7684.qmail@web60312.mail.yahoo.com> References: <20050710222020.7684.qmail@web60312.mail.yahoo.com> Message-ID: <200507110855.08193.giannozz@nest.sns.it> On Monday 11 July 2005 00:20, Eyvaz Isaev wrote: > 2. You can calculate LO/TO splitting for semiconductors and > insulators using both dynmat.x and matdyn.x. The result will > be the same. the logic is slightly different in the two cases: in dynmat.x one diagonalises the dynamical matrix at q=0 and supplies the direction of q; in matdyn.x the direction of q is calculated from the list of q vectors at which the dynamical matrix is diagonalised. This is explained in the headers of dynmat.f90 and matdyn.f90, respectively 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 Jul 11 08:57:10 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 11 Jul 2005 08:57:10 +0200 Subject: [Pw_forum] What is the different between Born effective charge and effective In-Reply-To: <33745.10.41.38.21.1120836666.squirrel@webmail1.ictp.trieste.it> References: <8d38992a05070807323e3b4ba0@mail.gmail.com> <33745.10.41.38.21.1120836666.squirrel@webmail1.ictp.trieste.it> Message-ID: <200507110857.10198.giannozz@nest.sns.it> On Friday 08 July 2005 17:31, Diep Vinh Quang wrote: > What is the different between Born effective charge and the > above effective charge? Or are they the same? they are meant to be the same > And on the other hand, how can we impose sysmetry condition > for the calculation of Z* (sum of Z* of each atom in one direction > must = zero) use code "dynmat.x" if you are interested in the q=0 case only 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 Jul 11 09:50:33 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 11 Jul 2005 09:50:33 +0200 Subject: [Pw_forum] espresso 2.1.4 and ifc8.1.024 In-Reply-To: <3.0.1.32.20050707175957.011fa2c0@mail.fcq.unc.edu.ar> References: <3.0.1.32.20050701190023.012f7aa0@mail.fcq.unc.edu.ar> <3.0.1.32.20050707175957.011fa2c0@mail.fcq.unc.edu.ar> Message-ID: <200507110950.33753.giannozz@nest.sns.it> On Thursday 07 July 2005 19:59, Patricia Paredes wrote: > I made some experiments compiling all the versions of pwscf since > 2.0.3 using ifc8.1.024, cluster mkl 7.1 and compiling mpi with the > same compiler. > > first iteration > > 2.0.3 117.11 secs > 2.0.4 109.89 secs > 2.1.1 110.03 secs > 2.1.2 164.93 secs > 2.1.3 152.52 secs > 2.1.4 154.43 secs this is what I got with your input - not sure about pseudopotentials - for serial execution (2.4GHz dual Xeon, ifc7.1+mkl 7.01) of a single SCF step: Version CPU Time Output File 2.0.4 2h58m http://web1.sns.it/~giannozz/pw/au.204 2.1.4 2h54m http://web1.sns.it/~giannozz/pw/au.214 CVS 2h37m http://web1.sns.it/~giannozz/pw/au.cvs I don't see any overall performance degradation, even if the first scf iteration is actually slower than it used to be, due to different requirements for the threshold of diagonalization 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 degironc at sissa.it Mon Jul 11 09:54:39 2005 From: degironc at sissa.it (Stefano de Gironcoli) Date: Mon, 11 Jul 2005 09:54:39 +0200 Subject: [Pw_forum] SCF not stop when it reaches the default conv_thr References: <8d38992a05070807323e3b4ba0@mail.gmail.com> <33745.10.41.38.21.1120836666.squirrel@webmail1.ictp.trieste.it> <1120861667.42cefde31ed1e@webmail.asu.edu> Message-ID: <42D225BF.3060203@sissa.it> Yong Jiang wrote: >Dear All, > >This is a question about conv_thr: > >I use the default value for conv_thr (default = 1.D-6)in my supercell >(including 32 Ni atoms). Why the scf calculation is still running when it >reaches ethr = 6.39E-07 since interation #22 (pls see below)? > > because conv_thr is the threshold for total energy self consistent accuracy and your estimated scf accuracy if far above 1.D-6 Stefano de Gironcoli >---------------------------------- > iteration # 21 ecut= 27.00 ryd beta=0.30 > Conjugate-gradient style diagonalization > ethr = 1.52E-06, avg # of iterations = 3.0 > > total cpu time spent up to now is 31997.23 secs > > total energy = -2748.70062200 ryd > estimated scf accuracy < 0.00204526 ryd > > total magnetization = 22.86 Bohr mag/cell > absolute magnetization = 26.19 Bohr mag/cell > > iteration # 22 ecut= 27.00 ryd beta=0.30 > Conjugate-gradient style diagonalization > ethr = 6.39E-07, avg # of iterations = 3.1 > > total cpu time spent up to now is 33483.75 secs > > total energy = -2748.68225286 ryd > estimated scf accuracy < 0.02254775 ryd > > total magnetization = 22.86 Bohr mag/cell > absolute magnetization = 26.22 Bohr mag/cell > > iteration # 23 ecut= 27.00 ryd beta=0.30 > Conjugate-gradient style diagonalization > ethr = 6.39E-07, avg # of iterations = 3.1 > > total cpu time spent up to now is 35054.16 secs > > total energy = -2748.69429201 ryd > --More--(98%) >-------------------------------------------- > >Thanks very much in advance for your kind help. > >Good weekend, > >Yong >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > > From vdiep at ictp.it Mon Jul 11 10:09:58 2005 From: vdiep at ictp.it (Diep Vinh Quang) Date: Mon, 11 Jul 2005 10:09:58 +0200 (CEST) Subject: [Pw_forum] About phonon calculation and IR In-Reply-To: <200507110855.08193.giannozz@nest.sns.it> References: <20050710222020.7684.qmail@web60312.mail.yahoo.com> <200507110855.08193.giannozz@nest.sns.it> Message-ID: <32810.10.41.38.18.1121069398.squirrel@webmail3.ictp.trieste.it> Dear all, Thank you very much for your reply and your help. I would like to ask some question about phonon calculation at gamma: 1. Why the propram phcg.x cannot use with ultra-solf pseudopotentials ? what is the different between phcp.x and ph.x 2. When i calculated the phonon at gamma by using ph.x, the effective charge in the dynamical matrix does not satisfy the ASR (the geometry of Polyethylene is this case for sure is correct). I don't know the reason. I thought that dues to the number of K sampling is a few in my input file: ______________________ &control calculation='scf', restart_mode='from_scratch', prefix='pe', pseudo_dir = '/afs/ictp.trieste.it/home/v/vdiep/espresso-2.1.3/pseudo/' outdir='/tmp/' / &system ibrav=8, celldm(1) = 13.479600 celldm(2) = 0.717398 celldm(3) = 0.360998 nat=12 , ntyp= 2, ecutwfc =40.0, / &electrons mixing_beta = 0.7 conv_thr = 1.0d-8 / ATOMIC_SPECIES C 12.0107 C.pbe-van_bm.UPF H 1.00794 H.pbe-van_bm.UPF ATOMIC_POSITIONS {angstrom} C -0.245695461 -0.316717589 0.644009287 H 0.039558649 -1.384963029 0.644168352 H -1.351501461 -0.307736442 0.644163171 C 0.245572450 0.316642176 -0.644005418 H -0.039700505 1.384882483 -0.644164826 H 1.351378971 0.307677348 -0.644159390 C 3.320929864 2.875316272 1.931505401 H 2.215124940 2.866341694 1.931659329 H 3.606194345 3.943558228 1.931664829 C 3.812198945 2.241955578 0.643490878 H 4.918003969 2.250929351 0.643336802 H 3.526935295 1.173713930 0.643331585 K_POINTS {automatic} 2 2 2 0 0 0 _____________________ and phonon input file &inputph tr2_ph=1.0d-14, prefix='pe', epsil=.true. amass(1)=12.0107, amass(2)=1.00794, outdir='/tmp/' fildyn='pe.dynG', / 0.0 0.0 0.0 ------------------------------- But when i try to increase the number of k points (4 4 4 0 0 0) in scf inplut, the ph.x doesnot recognize this change, It's still using 8 k's points as in the case (2 2 2 0 0 0). How can we solve this problem. Thank you very much. Vinh ------------------------------------------------- Diep Vinh Quang C/o Diploma Program, The Abdus Salam International Center for Theoretical Physics(ICTP), 11 Strada Costiera, 34014, Trieste,Italy. http://www.ictp.it From eyvaz_isaev at yahoo.com Mon Jul 11 12:03:39 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Mon, 11 Jul 2005 03:03:39 -0700 (PDT) Subject: [Pw_forum] FS construction In-Reply-To: <200507110855.08193.giannozz@nest.sns.it> Message-ID: <20050711100339.45109.qmail@web60322.mail.yahoo.com> Dear Paolo, In my home PC I have integrated some small programs for the Fermi Surface construction into /pwtools. I.e. now "make tools" will produce two auxiliary executable files used in Fermi_Surface.sh. I have also changed a path for these files in the script. So, I suggest, we can add this functionality to PWSCF. I have checked the FS for NiAl, and again it is very similar to the one published in PHYSICAL REVIEW B 70, 064301 (2004). Later I will add the spin-polarized case. Bests, Eyvaz. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From eyvaz_isaev at yahoo.com Mon Jul 11 15:48:06 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Mon, 11 Jul 2005 06:48:06 -0700 (PDT) Subject: [Pw_forum] Fermi Surface In-Reply-To: <42D225BF.3060203@sissa.it> Message-ID: <20050711134806.45470.qmail@web60320.mail.yahoo.com> Dear Paolo, I am not sure that you received my previous mail because it did not appear on the forum, too. I would like to note that in my home PC I have integrated some programs for FS construction into /pwtools. Now "make tools" compile also two small auxiliary files for FS construction, and a path for these executable files are also corrected in Fermi_Surface.sh script. So, I suggest, now we can submit these files into PWSCF distribution in order to give Fermi surface construction functionality to PWSCF. The last FS I checked was the one for B2 NiAl. It is in very good agreement with PHYSICAL REVIEW B 70, 064301 (2004). Best regards, Eyvaz. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From lee.atkinson at targacept.com Mon Jul 11 16:09:41 2005 From: lee.atkinson at targacept.com (Atkinson, Lee) Date: Mon, 11 Jul 2005 10:09:41 -0400 Subject: [Pw_forum] Autopilot Feature Suite Message-ID: Hello, I would like to introduce into the Espresso package an update that may useful to the ab-initio community. The Autopilot Feature Suite is used for automating long segmented runs at Targacept, Inc in Winston-Salem, NC, USA, for drug discovery. Many of the features may well suit other research interest areas. The code is fully developed and tested under Intel 8.1 for single Linux-32, or PGI(with MPI) for Multiple Linux-64 processors. The Autopilot Feature Suite is a user level enhancement focused on directing Car-Parrinello simulations based on CP.X packaged in ESPRESSO. The following features are incorporated: I. Auto Restart Mode II. Auto Pilot Course Configuration (Dynamic Rules) III. Auto Pilot Course Correction (Steering) Auto Pilot Course Correction (or Steering), provides a runtime method of changing Autopilot variables on-the-fly, after the simulation is underway. Course Correction contains the following sub features: New Course (power steering), Manual Steering, and Pause. Read more about this feature suite at this temporary link: http://66.56.97.9/macchiato/Doc/autopilot_doc.html Or contact me for more information. NOTE: This is a first time to add code to the project and I would like advisement on the proper procedure and current usefulness of the proposed interjection. Please contact me for discussion. Lee Atkinson, Sr. Sw. Eng., CTR Targacept, Inc latkinson at targacept.com USA-PH (336-480-2232) -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20050711/42da867f/attachment.htm From giannozz at nest.sns.it Mon Jul 11 17:37:20 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 11 Jul 2005 17:37:20 +0200 Subject: [Pw_forum] Fermi Surface In-Reply-To: <20050711134806.45470.qmail@web60320.mail.yahoo.com> References: <20050711134806.45470.qmail@web60320.mail.yahoo.com> Message-ID: <200507111737.20648.giannozz@nest.sns.it> On Monday 11 July 2005 15:48, Eyvaz Isaev wrote: > I am not sure that you received my previous mail > because it did not appear on the forum, too. this can be simply verified by checking the mailing list archive: http://www.democritos.it/pipermail/pw_forum/2005-July/002686.html > I would like to note that in my home PC I have > integrated some programs for FS construction into > /pwtools. Now "make tools" compile also two small > auxiliary files for FS construction, and a path for > these executable files are also corrected in > Fermi_Surface.sh script. > So, I suggest, now we can submit these files into > PWSCF distribution in order to give Fermi surface > construction functionality to PWSCF. very good. Please submit the additional and modified files to pwscf at pwscf.org and sooner or later somebody will insert them into the distribution. Thank you for your contribution. 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 mousumi at jncasr.ac.in Mon Jul 11 18:42:01 2005 From: mousumi at jncasr.ac.in (Mousumi Upadhyay Kahaly) Date: Mon, 11 Jul 2005 22:12:01 +0530 (IST) Subject: [Pw_forum] ??? marks in ELF analysis Message-ID: <48467.202.41.111.151.1121100121.squirrel@202.41.111.151> Dear All, I want to do elf analysis for Cd-di-acetate system. For that, I am using ULTRASOFT PSEUDOPOTENTIALS for all the atoms. After relaxation, I did the scf pw.x run(it is done correctly, I am sure) and then with files in the tmp/ directory, I have done pp.x run. The output elf file looks to be incomplete with lots of question marks like "4.615327952E-12????????????????? 1.189272540E-09 2.367439860E-10????????????????? ????????????????? 4.568158721E-08????????????????? 1.745761642E-12 4.550483202E-08 ????????????????? " What is the mistake I hav made? What should I do now to get correct ELF analysis of my system? Best regards, mousumi. Input file for my pp.x run, ================================================================ &inputpp prefix = 'Cd_di_acetate' outdir = '/export/mousumi/PWSCF_2.0.1/Cd_2CH3COO/ELF/tmp//' filplot = 'Cd_2CH3COOelf' plot_num= 8 / ================================================================= Then the output file after pp.x execution writes ========================================================================== Program POST-PROC v.2.0.1 starts ... Today is 11Jul2005 at 22: 2:36 nbndx = 44 nbnd = 37 natomwfc = 44 npwx = 23683 nelec = 58.00 nkb = 88 ngl = 11161 Calling punch_plot, plot_num = 8 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from do_elf : error # -1 elf + US not fully implemented %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Writing data on file Cd_2CH3COOelf ========================================================================== And the file Cd_2CH3COOelf writes- ========================================================================== 200 60 60 200 60 60 15 4 8 50.00000000 0.30000000 0.30000000 0.00000000 0.00000000 0.00000000 9498.8609664692 6.0000000000 25.0000000000 8 1 C 4.00 2 H 1.00 3 O 6.00 4 Cd 12.00 1 -0.082463418 -0.010889725 -0.018561191 3 2 -0.074898581 -0.017885652 0.065519357 3 3 -0.102378413 -0.016333137 0.025443499 1 4 -0.159067520 -0.020123028 0.030276227 1 5 -0.169237202 -0.042777034 0.063521107 2 6 -0.174509697 0.018232762 0.035611759 2 7 -0.175604491 -0.036018176 -0.004303162 2 8 -0.005204872 -0.008307511 0.017425116 4 9 0.092022107 -0.001191275 0.009297931 1 10 0.065358644 0.039253646 0.013394188 3 11 0.071264196 -0.045139461 0.009064170 3 12 0.148772494 0.002887309 0.005618279 1 13 0.159816129 0.039913367 -0.009472608 2 14 0.164797233 -0.000978230 0.043907778 2 15 0.163980748 -0.027908380 -0.017686596 2 3.323774911E-01 1.452789067E-01 9.538560487E-02 9.032122583E-02 9.775089880E-02 1.142089558E-01 1.595044445E-01 2.505376772E-01 3.865973450E-01 4.855420241E-01 5.345963002E-01 4.520594869E-01 2.646483550E-01 1.364132552E-01 1.780586327E-01 4.554727587E-01 7.760721604E-01 9.086906599E-01 9.442795876E-01 9.476321327E-01 9.096693830E-01 8.105565828E-01 5.942526205E-01 3.099656660E-01 1.505911825E-01 1.352057934E-01 2.384930089E-01 4.194133884E-01 5.641333943E-01 5.882099675E-01 5.669442522E-01 5.364892012E-01 4.110423394E-01 3.425590596E-01 2.371541964E-01 1.756398958E-01 9.689860769E-02 9.900749195E-02 2.517728549E-02 2.770832418E-02 2.929513975E-02 8.470112825E-03 7.982682914E-04 3.597589793E-04 4.392355790E-05 7.191333644E-05 5.934434930E-06 6.190322287E-04 2.592244117E-07 1.704007690E-08 4.615327952E-12?????????????????????????????????? 4.568158721E-08????????????????? 1.745761642E-12 4.550483202E-08 ????????????????? 3.804524614E-10 1.418421202E-08????????????????? 3.239593949E-09 6.195890609E-08?????????????????????????????????? 1.129576610E-07????????????????? 5.830903906E-10 3.768694819E-11??????????????????????????????????????????????????? 3.665181926E-10 2.975430105E-12????????????????? 2.465696404E-12 1.090467859E-09 1.189272540E-09 2.367439860E-10????????????????? and so on.... =========================================================================== From cbarreteau at cea.fr Tue Jul 12 10:13:57 2005 From: cbarreteau at cea.fr (Cyrille Barreteau) Date: Tue, 12 Jul 2005 10:13:57 +0200 (CEST) Subject: [Pw_forum] FLUSH--FAILED:: Bad address Message-ID: <56864.132.166.21.1.1121156037.squirrel@dsm-mail> Dear Pwscf users, I recently posted an email on PWscf forum reporting a very strange behaviour of the cvs version with ifc7 ** I have recompiled with ifort (version 8.1) and the problem disappears.. ** see email on the 4th of july cyrille -- ================================================ Cyrille Barreteau | phone:+33 (0)1 69 08 29 51 CEA Saclay | fax : +33 (0)1 69 08 84 46 DSM/DRECAM/SPCSI | email: cbarreteau at cea.fr Batiment 462 | 91191 Gif sur Yvette Cedex FRANCE ~~~~~~~~~~~~~~~~~~~~~~~~ http://www-drecam.cea.fr/spcsi/index.php http://www-drecam.cea.fr/Images/Pisp/cbarreteau/cbarreteau_fr.html ================================================ From leoant21 at hotmail.com Tue Jul 12 11:05:45 2005 From: leoant21 at hotmail.com (=?gb2312?B?1cUguumx8g==?=) Date: Tue, 12 Jul 2005 09:05:45 +0000 Subject: [Pw_forum] RE: Questions on phonon calculation of MgB2 Message-ID: Dear Paolo: Thank you for your reply. But, I am sorry for that, I have missed the 'correct' error in my last post, here is what it is:(when I doing q2r calculation) q= 0.00000000 0.00000000 0.28610020(this is the second kpoints, while using the first giving the same error message) %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from init : error # 1 q not allowed %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... And the same to the third one. So the delete the first three 'wrong' kpoints in my input file, and run the q2r.x again, it goes wrong again: reading dyn.mat. from file mgb.dynG Dielectric Tensor not found nqs= 1 q= 0.00000000 0.00000000 0.00000000 reading dyn.mat. from file dyn.0.0000000,0.1924501,0.0000000 nqs= 6 q= 0.00000000 0.19245010 0.00000000 q= 0.00000000 -0.19245010 0.00000000 q= -0.16666668 0.09622505 0.00000000 q= 0.16666668 0.09622505 0.00000000 q= -0.16666668 -0.09622505 0.00000000 q= 0.16666668 -0.09622505 0.00000000 reading dyn.mat. from file dyn.0.0000000,0.1924501,0.1430501 nqs= 12 q= 0.00000000 0.19245010 0.14305010 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from init : error # 1 q not allowed %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... Plz give me some suggestion on how to correct this error, and my input file is attatched. Best Regards. _________________________________________________________________ ?????????????? MSN Messenger: http://messenger.msn.com/cn -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: run_example Url: /pipermail/attachments/20050712/dd7b2a13/attachment.txt From degironc at sissa.it Tue Jul 12 12:13:31 2005 From: degironc at sissa.it (Stefano de Gironcoli) Date: Tue, 12 Jul 2005 12:13:31 +0200 Subject: [Pw_forum] RE: Questions on phonon calculation of MgB2 References: Message-ID: <42D397CB.2020204@sissa.it> in your input to q2r.x you are declaring a 6x6x1 grid but, as far as I understand, you actually computed a 6x6x6 one Stefano de Girocoli From yjiang at asu.edu Tue Jul 12 15:39:14 2005 From: yjiang at asu.edu (Yong Jiang) Date: Tue, 12 Jul 2005 06:39:14 -0700 (MST) Subject: [Pw_forum] why phonon calc stops at the first mode Message-ID: <1121175554.42d3c802bc4de@webmail.asu.edu> Dear All, I am doing the scf and phonon (at G) calculation on a 2x2x2 fcc Ni supercell (a serial calc using parallel code on single node). The scf finished properly, but the phonon calc stops at the first mode without any error/reason. Message gives "running the phonon calculation at Gamma... done". If interested, pls check my input and .out info as attached below. Could you give me some hints on possible causes of such problem and how to fix it? Thanks very much in advance for your kind help... Best regards, Yong -----------------input ------------------ # self-consistent calculation for Ni with US-PP cat > Ni.scf.in << EOF &control calculation='scf', restart_mode='from_scratch', tprnfor = .true., prefix='Ni', pseudo_dir = '$PSEUDO_DIR/', outdir='$TMP_DIR/' / &system ibrav=1, celldm(1) =13.31444, nat= 32, ntyp= 1, nspin=2, starting_magnetization(1)=0.5, degauss=0.003, smearing='m-p', occupations='smearing', ecutwfc =27.0, ecutrho =400.0, / &electrons diagonalization='cg', mixing_beta = 0.3, / ATOMIC_SPECIES Ni 58.6934 Ni.pbe-nd-rrkjus.UPF ATOMIC_POSITIONS {angstrom} Ni 0.000 0.000 0.000 Ni 0.000 1.762 1.762 Ni 1.762 0.000 1.762 Ni 1.762 1.762 0.000 Ni 0.000 0.000 3.523 Ni 0.000 1.762 5.285 Ni 1.762 0.000 5.285 Ni 1.762 1.762 3.523 Ni 0.000 3.523 0.000 Ni 0.000 5.285 1.762 Ni 1.762 3.523 1.762 Ni 1.762 5.285 0.000 Ni 0.000 3.523 3.523 Ni 0.000 5.285 5.285 Ni 1.762 3.523 5.285 Ni 1.762 5.285 3.523 Ni 3.523 0.000 0.000 Ni 3.523 1.762 1.762 Ni 5.285 0.000 1.762 Ni 5.285 1.762 0.000 Ni 3.523 0.000 3.523 Ni 3.523 1.762 5.285 Ni 5.285 0.000 5.285 Ni 5.285 1.762 3.523 Ni 3.523 3.523 0.000 Ni 3.523 5.285 1.762 Ni 5.285 3.523 1.762 Ni 5.285 5.285 0.000 Ni 3.523 3.523 3.523 Ni 3.523 5.285 5.285 Ni 5.285 3.523 5.285 Ni 5.285 5.285 3.523 K_POINTS {automatic} 2 2 2 1 1 1 EOF $ECHO " running the scf calculation for Ni31Al1...\c" $PW_COMMAND < Ni.scf.in > Ni.scf.out $ECHO " done" # phonon calculation at Gamma cat > Ni.phG.in << EOF phonons of Ni at Gamma &inputph tr2_ph=1.0d-14, prefix='Ni', amass(1)=58.6934, outdir='$TMP_DIR/', fildyn='Ni.dynG', / 0.0 0.0 0.0 EOF $ECHO " running the phonon calculation at Gamma...\c" $PH_COMMAND < Ni.phG.in > Ni.phG.out $ECHO " done" --------------------scf.out------------------- .... Writing file Ni.save for program phonon PWSCF : 18h26m CPU time init_run : 406.61s CPU electrons : 65265.58s CPU forces : 714.60s CPU electrons : 65265.58s CPU c_bands : 50356.50s CPU ( 44 calls,1144.466 s avg) sum_band : 9366.84s CPU ( 44 calls, 212.883 s avg) v_of_rho : 440.35s CPU ( 89 calls, 4.948 s avg) newd : 5037.68s CPU ( 45 calls, 111.948 s avg) mix_rho : 113.48s CPU ( 44 calls, 2.579 s avg) c_bands : 50356.50s CPU ( 44 calls,1144.466 s avg) init_us_2 : 75.39s CPU ( 720 calls, 0.105 s avg) ccgdiagg : 37242.80s CPU ( 352 calls, 105.803 s avg) sum_band : 9366.84s CPU ( 44 calls, 212.883 s avg) becsum : 3.73s CPU ( 352 calls, 0.011 s avg) addusdens : 4798.03s CPU ( 44 calls, 109.046 s avg) wfcrot1 : 13339.67s CPU ( 352 calls, 37.897 s avg) h_1psi : 33256.44s CPU ( 205814 calls, 0.162 s avg) s_1psi : 12248.24s CPU ( 138230 calls, 0.089 s avg) cdiaghg : 117.43s CPU ( 352 calls, 0.334 s avg) h_1psi : 33256.44s CPU ( 205814 calls, 0.162 s avg) init : 22.85s CPU ( 205814 calls, 0.000 s avg) firstfft : 2715.70s CPU ( 205814 calls, 0.013 s avg) secondfft : 2408.59s CPU ( 205814 calls, 0.012 s avg) add_vuspsi : 9194.51s CPU ( 205814 calls, 0.045 s avg) General routines ccalbec : 18167.98s CPU ( 344404 calls, 0.053 s avg) cft3 : 350.84s CPU ( 2331 calls, 0.151 s avg) cft3s : 5431.43s CPU ( 479390 calls, 0.011 s avg) interpolate : 36.53s CPU ( 178 calls, 0.205 s avg) davcio : 74.37s CPU ( 1072 calls, 0.069 s avg) Parallel routines -----------------------phG.out------------------------------- ... Representation 62 1 modes - To be done Representation 63 2 modes - To be done Representation 64 1 modes - To be done PHONON : 9h19m CPU time Alpha used in Ewald sum = 2.8000 Representation # 1 mode # 1 Self-consistent Calculation Pert. # 1: Fermi energy shift (Ryd) = 0.0000 0.0000 iter # 1 total cpu time : 34535.4 secs av.it.: 5.0 thresh= 0.100E-01 alpha_mix = 0.700 |ddv_scf|^2 = 0.288E-04 (STOP HERE WITHOUT ANY ERROR PROMPT) ------------------------tmp files---------------------------------- [hzh at linux1 03]$ ll tmp total 2478664 drwxrwxr-x 2 hzh hzh 4096 Jul 12 04:29 ./ drwxr-xr-x 7 hzh hzh 4096 Jul 11 18:53 ../ -rw-rw-r-- 1 hzh hzh 46656000 Jul 12 04:28 flmixdpot1 -rw-rw-r-- 1 hzh hzh 68984832 Jul 12 04:24 Ni.bar1 -rw-rw-r-- 1 hzh hzh 68984832 Jul 12 04:26 Ni.dwf1 -rw-rw-r-- 1 hzh hzh 89872 Jul 11 18:52 Ni.igk -rw-rw-r-- 1 hzh hzh 11664008 Jul 11 18:29 Ni.pot -rw-rw-r-- 1 hzh hzh 2239488000 Jul 12 04:13 Ni.prd1 -rw-rw-r-- 1 hzh hzh 11664008 Jul 11 18:35 Ni.rho -rw-rw-r-- 1 hzh hzh 524487 Jul 11 18:35 Ni.save -rw-rw-r-- 1 hzh hzh 13 Jul 11 18:51 Ni.stat -rw-rw-r-- 1 hzh hzh 68984832 Jul 11 18:26 Ni.wfc1 -rw-rw-r-- 1 hzh hzh 18557428 Jul 12 04:29 recover From mousumi at jncasr.ac.in Tue Jul 12 20:28:47 2005 From: mousumi at jncasr.ac.in (Mousumi Upadhyay Kahaly) Date: Tue, 12 Jul 2005 23:58:47 +0530 (IST) Subject: [Pw_forum] negative charge density! Message-ID: <46254.202.41.111.151.1121192927.squirrel@202.41.111.151> Dear all, I am trying to do charge analysis for Cd-di-acetate system. With the following files for pp.x and then chdens.x, as I am plotting the isosurfaces of charge densities with XCRYSDEN, I am getting some negative charge; Minimum grid value= -0.498760, and maximum grid value= 1.09220. How come I get such negative values(mainly around the Cd atom, as found) which is not negligible? What is the meaning of this NEGATIVE charge density? Can anyone please tell me, what is the mistake that I have made & what should I do to get rid of this? Best regards, Mousumi. 1. Input file for SCF pw.x run: ========================================================================= &CONTROL calculation = 'scf' , restart_mode = 'from_scratch' , tstress =.true., outdir = '/export/mousumi/PWSCF_2.0.1/Cd_2CH3COO/ELF/tmp//' , pseudo_dir = '/export/mousumi/PWscf_GUI/pseudo/' , prefix = 'Cd_di_acetate' , / &SYSTEM ibrav = 8, celldm(1) = 50.000000, celldm(2) = 0.30, celldm(3) = 0.30, nat = 15, ntyp = 4, ecutwfc = 25.00000 , ecutrho = 150.00000 , nosym = .false. , nbnd = 37, occupations = 'smearing' , degauss = 0.003 , smearing = 'methfessel-paxton' , nspin = 1 , lda_plus_u = .false. , / &ELECTRONS conv_thr = 1.0d-12 , mixing_beta = 0.3 , / ATOMIC_SPECIES C 12.01000 C.pbe-van_bm.UPF H 1.00000 H.pbe-van_bm.UPF O 15.99990 O.pbe-van_bm.UPF Cd 112.4000 Cd.pbe-van.UPF ATOMIC_POSITIONS {angstrom} O -2.181887203 -0.288129610 -0.491107761 O -1.981730329 -0.473233788 1.733566845 C -2.708815067 -0.432156029 0.673205720 C -4.208743648 -0.532432174 0.801074150 H -4.477821741 -1.131831132 1.680695448 H -4.617325903 0.482417907 0.942246198 H -4.646292897 -0.952999527 -0.113856718 Cd -0.137714936 -0.219807184 0.461048531 C 2.434799121 -0.031519760 0.246012557 O 1.729314564 1.038606335 0.354394821 O 1.885568682 -1.194338232 0.239827517 C 3.936349099 0.076394885 0.148653197 H 4.228550980 1.056061785 -0.250634315 H 4.360345258 -0.025882832 1.161749300 H 4.338742013 -0.738423646 -0.467966988 K_POINTS {automatic} 1 1 1 0 0 0 ========================================================================== 2. After the pw.x run with above file, with files in "tmp" dir, file for pp.x =========================================================================== &inputpp prefix = 'Cd_di_acetate' outdir = '/export/mousumi/PWSCF_2.0.1/Cd_2CH3COO/ELF/tmp//' filplot = 'Cd_2CH3COO_chden' plot_num = 0 / ========================================================================== 3. Then, input file for chdens.x ========================================================================== &input nfile = 1 filepp(1)='Cd_2CH3COO_chden' iflag = 3 output_format = 5 / ========================================================================== From dchrobak at us.edu.pl Wed Jul 13 14:48:30 2005 From: dchrobak at us.edu.pl (Dariusz Chrobak) Date: Wed, 13 Jul 2005 14:48:30 +0200 Subject: [Pw_forum] spherical avarage of the charge density Message-ID: <1121258910.7385.5.camel@kometk.ifichm.us.edu.pl> Dear PWscf users, Could anybody tell mi how the spherical avarage of the charge density (plot_out=0 in the chdens.x input file) is calculated? Best regards Dariusz Chrobak From lsl0426 at yahoo.com Thu Jul 14 04:14:45 2005 From: lsl0426 at yahoo.com (Shihn Lun) Date: Wed, 13 Jul 2005 19:14:45 -0700 (PDT) Subject: [Pw_forum] Psuedopotential generator of José-Luís Martins Message-ID: <20050714021445.81304.qmail@web31107.mail.mud.yahoo.com> Hello pwscf users, I want to use pseudopotential made by Jos?-Lu?s Martins' code in pwscf, but it doesn't seem to have a tool to convert to UPF. Anyone konw how to convert or where the converting tool is? Thanks in advance, Shihn-Lun. -- __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From giannozz at nest.sns.it Thu Jul 14 09:11:52 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 14 Jul 2005 09:11:52 +0200 Subject: [Pw_forum] Psuedopotential generator of =?iso-8859-1?q?Jos=E9-Lu=EDs?= Martins In-Reply-To: <20050714021445.81304.qmail@web31107.mail.mud.yahoo.com> References: <20050714021445.81304.qmail@web31107.mail.mud.yahoo.com> Message-ID: <200507140911.52569.giannozz@nest.sns.it> On Thursday 14 July 2005 04:14, Shihn Lun wrote: > I want to use pseudopotential made by Jos?-Lu?s > Martins' code in pwscf, but it doesn't seem to have > a tool to convert to UPF. > > Anyone konw how to convert or where the converting > tool is? there is a template (any2upf.f90) in the upftools/ directory. If you know the format of JLM's pseudopotentials, it should be quite easy to write a converter. The UPF format is fully documented in upftools/UPF 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 Thu Jul 14 15:35:35 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 14 Jul 2005 15:35:35 +0200 Subject: [Pw_forum] spherical avarage of the charge density In-Reply-To: <1121258910.7385.5.camel@kometk.ifichm.us.edu.pl> References: <1121258910.7385.5.camel@kometk.ifichm.us.edu.pl> Message-ID: <200507141535.35899.giannozz@nest.sns.it> On Wednesday 13 July 2005 14:48, Dariusz Chrobak wrote: > Could anybody tell mi how the spherical avarage of the charge density > (plot_out=0 in the chdens.x input file) is calculated? it is calculated as explained in subroutine plot_1d of chdens.f90: ! spherically averaged charge: rho0(|r|) = int rho(r) dOmega ! rho0(r) = 4pi \sum_G rho(G) j_0(|G||r|) with j_0(x) = sin(x)/x; or as in the attached TeX file (which actually differs for a 4pi factor from the above) 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 -------------- next part -------------- A non-text attachment was scrubbed... Name: spherical.tex Type: text/x-tex Size: 705 bytes Desc: not available Url : /pipermail/attachments/20050714/d41ad714/attachment.tex From patricia at fcq.unc.edu.ar Thu Jul 14 12:05:19 2005 From: patricia at fcq.unc.edu.ar (Patricia Paredes) Date: Thu, 14 Jul 2005 12:05:19 Subject: [Pw_forum] espresso 2.1.4 and ifc8.1.024 In-Reply-To: <42D1311C.2020507@mit.edu> References: <3.0.1.32.20050707175957.011fa2c0@mail.fcq.unc.edu.ar> <3.0.1.32.20050701190023.012f7aa0@mail.fcq.unc.edu.ar> <3.0.1.32.20050707175957.011fa2c0@mail.fcq.unc.edu.ar> Message-ID: <3.0.1.32.20050714120519.01a420d0@mail.fcq.unc.edu.ar> A non-text attachment was scrubbed... Name: not available Type: text/enriched Size: 5677 bytes Desc: not available Url : /pipermail/attachments/20050714/be771edf/attachment.bin From marzari at MIT.EDU Thu Jul 14 17:15:00 2005 From: marzari at MIT.EDU (Nicola Marzari) Date: Thu, 14 Jul 2005 11:15:00 -0400 Subject: [Pw_forum] espresso 2.1.4 and ifc8.1.024 In-Reply-To: <3.0.1.32.20050714120519.01a420d0@mail.fcq.unc.edu.ar> References: <3.0.1.32.20050707175957.011fa2c0@mail.fcq.unc.edu.ar> <3.0.1.32.20050701190023.012f7aa0@mail.fcq.unc.edu.ar> <3.0.1.32.20050707175957.011fa2c0@mail.fcq.unc.edu.ar> <3.0.1.32.20050714120519.01a420d0@mail.fcq.unc.edu.ar> Message-ID: <42D68174.3000104@mit.edu> Thanks Patricia ! There is some mistery going around here - do make sure hat the three calculations with pwscf 2.0.4 are all taking the same number of bfgs steps, and the same number of iterations on the electrons (by the way, that could explain the difference coming just from different smearings). Ideally, you want to have tests that always have the same number of iterations - i.e. I would avoid ionic relaxations, and would do a fixed number of electronic iterations (i.e., use a very tight electronic threshold, but limit the number of iterations to, say, 10, so that the code doesn't take less than 10 to converge). The most relevant issue seem to be the huge difference between 2.0.4 and 2.1.1 - could you redo that on one single node, making sure noone else run on the machine ? In a parallel job, and with a shared machine, all timings can be really thrown off. Thanks, nicola Patricia Paredes wrote: > Hi Nicola - Paolo: > OK, now I made all the test (including marzari-vanderbilt smearing...) > and changing the pseupotentials from the old format > to the *.UPF format currently on the pwscf web page. > here are the results for a cluster of PIV-3.0GHZ, compiled with ifc > 8.0.034 (both pw and the mpi) running on 5 nodes. > > > pwscf 2.0.4: > bfgs converged in 10 scf cycles and 8 bfgs steps > final energy: -1259.5907420114 ryd > > > electrons : 8813.48s CPU ( 10 calls, 881.348 s avg) > first ionic iteration > scf - last iteration: 1527.75 secs > scf + forces : 1674.88 secs > > PWSCF : 2h35m CPU time > > using smearing='marzari-vanderbilt' > PWSCF : 2h 0m CPU time > > using pseudopotential in UPF format+ marzari-vanderbilt smearing > S 32.06 S.pbe-van_bm.UPF > C 12.01 C.pbe-van_bm.UPF > H 1 H.pbe-van_bm.UPF > Au 196.96 Au.pbe-nd-van.UPF > > PWSCF : 1h18m cpu time!!!! > > > pwscf 2.1.1 > bfgs converged in 12 scf cycles and 11 bfgs steps > Final energy = -1259.5907145126 ryd > > electrons : 11685.19s CPU ( 12 calls, 973.766 s avg) > fist ionic iteration > scf - last iteration: 1539.16 secs > scf + forces : 1685.91 secs > > > PWSCF : 3h24m CPU time > > using smearing='marzari-vanderbilt' > PWSCF : 3h33m CPU time > > using pseudopotential in UPF format+ marzari-vanderbilt smearing > S 32.06 S.pbe-van_bm.UPF > C 12.01 C.pbe-van_bm.UPF > H 1 H.pbe-van_bm.UPF > Au 196.96 Au.pbe-nd-van.UPF > > PWSCF : 3h30m CPU time > > pwscf 2.1.2 > bfgs converged in 12 scf cycles and 11 bfgs steps > Final energy = -1259.5907141926 ryd > > electrons : 13977.43s CPU ( 12 calls,1164.786 s avg) > fist ionic iteration > scf - last iteration:2028.56 secs > scf + forces : 2181.71 secs > > > > > > A 10:30 AM 7/10/2005 -0400, ha escrito: > > > > > >Thanks a lot Patricia - these are *exceedingly* useful tests. > > > >We have even a standard test for CP: > >http://www.democritos.it/pipermail/pw_forum/2004-October/001483.html > > > >Axel had kindly provided some results, still in the repository, and > >that I should add to the report. > > > >If you'd like, we can add yours as a standard test for PWSCF. Of course, > >you should switch to a "marzari-vanderbilt" smearing, then :-) . > > > > nicola > > > > > >Patricia Paredes wrote: > > > >> Hi Axel, > >> I made some experiments compiling all the versions of pwscf since 2.0.3 > >> using ifc8.1.024, cluster mkl 7.1 and compiling mpi with the same > compiler. > >> first iteration > >> 2.0.3 117.11 secs > >> 2.0.4 109.89 secs > >> 2.1.1 110.03 secs > >> 2.1.2 164.93 secs > >> 2.1.3 152.52 secs > >> 2.1.4 154.43 secs > >> > >> So, the breaking point was version 2.1.1 (btw, v2.1.0 didn?t work). In > >> the web page it is stated that it was a modification in soubroutine > >> electrons.f90 to make a double check... (By carlo sbraccia), perhaps > >> there is the point: the double check. I was taking a look to the final > >> statistics and the major differences were at electrons.f90 (Sorry, I > >> don?t have the numbers now since I overwrite the files :). > >> > >> The input file was: > >> > >> cuatricapa de Au > >> Au > >> &control > >> calculation='relax' > >> restart_mode='from_scratch', > >> prefix='prueba', > >> pseudo_dir = '/home/pwscf/pseudo/', > >> outdir='/home/patricia/plata/temp/' > >> / > >> &system > >> ibrav=0, > >> nat=17, > >> ntyp=4, > >> ecutwfc = 22.0, > >> ecutrho = 180.0 > >> celldm(1)=7.89304630 > >> degauss=0.03 > >> occupations='smearing' > >> smearing='methfessel-paxton' > >> / > >> &electrons > >> mixing_mode = 'local-TF' > >> conv_thr = 1.0d-6 > >> mixing_beta = 0.5 > >> diagonalization = 'cg' > >> / > >> &ions > >> ion_dynamics = 'bfgs' > >> / > >> ATOMIC_SPECIES > >> S 32.06 S.gpbe.vdb > >> C 12.01 C.gpbe.vdb > >> H 1 H.gpbe.vdb > >> Au 196.96 Au.pbe-nd-van.UPF > >> ATOMIC_POSITIONS {alat} > >> H -0.187034332 0.221830025 4.921552258 > >> H -0.013559323 0.486812870 4.628026988 > >> H 0.242342701 0.256878798 4.888036613 > >> C 0.014538967 0.257636565 4.754725119 > >> S 0.016846606 -0.082426513 4.477818864 > >> Au 0.383806249 0.011800463 4.017799873 > >> Au 1.076105787 -0.011642255 3.950466565 > >> Au 1.768205765 0.011736229 4.018611550 > >> Au 0.706660429 -0.205310574 3.426597457 > >> Au 1.418423038 -0.205689922 3.424408560 > >> Au 1.062125184 0.398360581 3.405678032 > >> Au 0.701778921 0.206802972 2.843998759 > >> Au 1.435470249 0.207226266 2.845135398 > >> Au 1.067569501 -0.420576720 2.851998036 > >> Au 0.353553300 0.000000000 2.267949200 0 0 0 > >> Au 1.060660200 0.000000000 2.267949200 0 0 0 > >> Au 1.767767000 0.000000000 2.267949200 0 0 0 > >> CELL_PARAMETERS {cubic} > >> 1.0606602 0.6123724 0.0 > >> 1.0606602 -0.6123724 0.0 > >> 0.0000000 0.0 7.0 > >> K_POINTS {automatic} > >> 4 4 1 0 0 0 > >> > >> > >> *_Prof. Dr. Patricia Paredes Olivera > >> _*Facultad de Ciencias Qu?micas > >> Universidad Nacional de C?rdoba > >> ph/fax: 054-0351-434-4972 > >> _______________________________________________ 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 617.2586534 marzari at mit.edu http://nnn.mit.edu > >_______________________________________________ > >Pw_forum mailing list > >Pw_forum at pwscf.org > >http://www.democritos.it/mailman/listinfo/pw_forum > > > *_Prof. Dr. Patricia Paredes Olivera > _*Facultad de Ciencias Qu?micas > Universidad Nacional de C?rdoba > ph/fax: 054-0351-434-4972 > _______________________________________________ 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 617.2586534 marzari at mit.edu http://nnn.mit.edu From giannozz at nest.sns.it Thu Jul 14 18:02:00 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 14 Jul 2005 18:02:00 +0200 Subject: [Pw_forum] why phonon calc stops at the first mode In-Reply-To: <1121175554.42d3c802bc4de@webmail.asu.edu> References: <1121175554.42d3c802bc4de@webmail.asu.edu> Message-ID: <200507141802.00087.giannozz@nest.sns.it> On Tuesday 12 July 2005 15:39, Yong Jiang wrote: > the phonon calc stops at the first mode without any error/reason. in such a large calculation, there can be 1001 reasons for the code to stop, with lack of memory or of disk space high in the list 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 Thu Jul 14 18:34:46 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 14 Jul 2005 18:34:46 +0200 Subject: [Pw_forum] ??? marks in ELF analysis In-Reply-To: <48467.202.41.111.151.1121100121.squirrel@202.41.111.151> References: <48467.202.41.111.151.1121100121.squirrel@202.41.111.151> Message-ID: <200507141834.46430.giannozz@nest.sns.it> On Monday 11 July 2005 18:42, Mousumi Upadhyay Kahaly wrote: > I want to do elf analysis for Cd-di-acetate system. For that, I am > using ULTRASOFT PSEUDOPOTENTIALS in recent versions of the code, the calculation of elf with ultrasoft pseudopotentials is only partially implemented (i.e. the elf is calculated without a contribution to the kinetic energy) > Program POST-PROC v.2.0.1 starts ... in such old versions, the calculation of elf with ultrasoft pseudopotentials is likely to be very wrong. The ???????????? are used by some compilers to signal that what is printed is not a number, I think 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 Thu Jul 14 18:53:03 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 14 Jul 2005 18:53:03 +0200 Subject: [Pw_forum] About phonon calculation and IR In-Reply-To: <32810.10.41.38.18.1121069398.squirrel@webmail3.ictp.trieste.it> References: <20050710222020.7684.qmail@web60312.mail.yahoo.com> <200507110855.08193.giannozz@nest.sns.it> <32810.10.41.38.18.1121069398.squirrel@webmail3.ictp.trieste.it> Message-ID: <200507141853.03550.giannozz@nest.sns.it> On Monday 11 July 2005 10:09, Diep Vinh Quang wrote: > 1. Why the program phcg.x cannot use with ultra-solf pseudopotentials ? because it takes time to implement US-PP and nobody ever did it > what is the different between phcg.x and ph.x the former uses a "global minimization" approach: the linear response functions dpsi/du, where u is the atomic displacement from equilibrium (or dpsi/dE in the case of an electric field) are calculated from the minimisation of a quadratic functional, that is, the solution of a linear system involving all dpsi/du . See for instance Rev. Mod. Phys. 73, 515 (2001), Sec. II C 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 mousumi at jncasr.ac.in Thu Jul 14 22:05:28 2005 From: mousumi at jncasr.ac.in (Mousumi Upadhyay Kahaly) Date: Fri, 15 Jul 2005 01:35:28 +0530 (IST) Subject: [Pw_forum] negative charge density! Message-ID: <47591.202.41.111.151.1121371528.squirrel@202.41.111.151> Hi all, This is a resubmission of my mail regarding charge density calculation. -------------------------------------------------------------------------- Dear all, I am trying to do charge analysis for Cd-di-acetate system. With the following files for pp.x and then chdens.x, as I am plotting the isosurfaces of charge densities with XCRYSDEN, I am getting some negative charge; Minimum grid value= -0.498760, and maximum grid value= 1.09220. How come I get such negative values(mainly around the Cd atom, as found) which is not negligible? What is the meaning of this NEGATIVE charge density? Can anyone please tell me, what is the mistake that I have made & what should I do to get rid of this? Best regards, Mousumi. 1. Input file for SCF pw.x run: ========================================================================= &CONTROL calculation = 'scf' , restart_mode = 'from_scratch' , tstress =.true., outdir = '/export/mousumi/PWSCF_2.0.1/Cd_2CH3COO/ELF/tmp//' , pseudo_dir = '/export/mousumi/PWscf_GUI/pseudo/' , prefix = 'Cd_di_acetate' , / &SYSTEM ibrav = 8, celldm(1) = 50.000000, celldm(2) = 0.30, celldm(3) = 0.30, nat = 15, ntyp = 4, ecutwfc = 25.00000 , ecutrho = 150.00000 , nosym = .false. , nbnd = 37, occupations = 'smearing' , degauss = 0.003 , smearing = 'methfessel-paxton' , nspin = 1 , lda_plus_u = .false. , / &ELECTRONS conv_thr = 1.0d-12 , mixing_beta = 0.3 , / ATOMIC_SPECIES C 12.01000 C.pbe-van_bm.UPF H 1.00000 H.pbe-van_bm.UPF O 15.99990 O.pbe-van_bm.UPF Cd 112.4000 Cd.pbe-van.UPF ATOMIC_POSITIONS {angstrom} O -2.181887203 -0.288129610 -0.491107761 O -1.981730329 -0.473233788 1.733566845 C -2.708815067 -0.432156029 0.673205720 C -4.208743648 -0.532432174 0.801074150 H -4.477821741 -1.131831132 1.680695448 H -4.617325903 0.482417907 0.942246198 H -4.646292897 -0.952999527 -0.113856718 Cd -0.137714936 -0.219807184 0.461048531 C 2.434799121 -0.031519760 0.246012557 O 1.729314564 1.038606335 0.354394821 O 1.885568682 -1.194338232 0.239827517 C 3.936349099 0.076394885 0.148653197 H 4.228550980 1.056061785 -0.250634315 H 4.360345258 -0.025882832 1.161749300 H 4.338742013 -0.738423646 -0.467966988 K_POINTS {automatic} 1 1 1 0 0 0 ========================================================================== 2. After the pw.x run with above file, with files in "tmp" dir, file for pp.x =========================================================================== &inputpp prefix = 'Cd_di_acetate' outdir = '/export/mousumi/PWSCF_2.0.1/Cd_2CH3COO/ELF/tmp//' filplot = 'Cd_2CH3COO_chden' plot_num = 0 / ========================================================================== 3. Then, input file for chdens.x ========================================================================== &input nfile = 1 filepp(1)='Cd_2CH3COO_chden' iflag = 3 output_format = 5 / ========================================================================== From yjiang at asu.edu Thu Jul 14 22:14:04 2005 From: yjiang at asu.edu (Yong Jiang) Date: Thu, 14 Jul 2005 13:14:04 -0700 (MST) Subject: [Pw_forum] why phonon calc stops at the first mode In-Reply-To: <200507141802.00087.giannozz@nest.sns.it> References: <1121175554.42d3c802bc4de@webmail.asu.edu> <200507141802.00087.giannozz@nest.sns.it> Message-ID: <1121372044.42d6c78c6c218@webmail.asu.edu> Dear Dr. Paolo Giannozzi, Thank you very much for your reply. If it is the case, could the code be updated in later versions to be able to prompt such error message whenever it encounts, so that users could take action accordingly? By the way, is there any unreasonable setting in my input files can be downscaled/modified to avoid such problem? Would a parallel run be helpful? Thanks very much! Yong Quoting Paolo Giannozzi : > On Tuesday 12 July 2005 15:39, Yong Jiang wrote: > > > the phonon calc stops at the first mode without any error/reason. > > in such a large calculation, there can be 1001 reasons for the > code to stop, with lack of memory or of disk space high in the > list > > 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 eyvaz_isaev at yahoo.com Thu Jul 14 22:42:24 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Thu, 14 Jul 2005 13:42:24 -0700 (PDT) Subject: [Pw_forum] negative charge density! In-Reply-To: <47591.202.41.111.151.1121371528.squirrel@202.41.111.151> Message-ID: <20050714204225.10648.qmail@web60320.mail.yahoo.com> Hi, Did you try increase k-points instead of using only Gamma-point? Besides, test also a little bit larger ecutrho, say 30Ry. Bests, Eyvaz. --- Mousumi Upadhyay Kahaly wrote: > Hi all, > > This is a resubmission of my mail regarding > charge density calculation. > > -------------------------------------------------------------------------- > > Dear all, > > I am trying to do charge analysis for > Cd-di-acetate system. With > the following files for pp.x and then chdens.x, as I > am plotting > the isosurfaces of charge densities with XCRYSDEN, I > am getting > some negative charge; > Minimum grid value= -0.498760, and maximum > grid value= 1.09220. > How come I get such negative values(mainly around > the Cd atom, as found) > which is not negligible? What is the meaning of this > NEGATIVE charge > density? Can anyone please tell me, what is the > mistake that I have made & > what should I do to get rid of this? > > Best > regards, Mousumi. > > 1. Input file for SCF pw.x run: > ========================================================================= > &CONTROL > calculation = 'scf' , > restart_mode = 'from_scratch' , > tstress =.true., > outdir = > '/export/mousumi/PWSCF_2.0.1/Cd_2CH3COO/ELF/tmp//' , > pseudo_dir = '/export/mousumi/PWscf_GUI/pseudo/' , > prefix = 'Cd_di_acetate' , > / > &SYSTEM > ibrav = 8, > celldm(1) = 50.000000, > celldm(2) = 0.30, > celldm(3) = 0.30, > nat = 15, > ntyp = 4, > ecutwfc = 25.00000 , > ecutrho = 150.00000 , > nosym = .false. , > nbnd = 37, > occupations = 'smearing' , > degauss = 0.003 , > smearing = 'methfessel-paxton' , > nspin = 1 , > lda_plus_u = .false. , > / > &ELECTRONS > conv_thr = 1.0d-12 , > mixing_beta = 0.3 , > / > ATOMIC_SPECIES > C 12.01000 C.pbe-van_bm.UPF > H 1.00000 H.pbe-van_bm.UPF > O 15.99990 O.pbe-van_bm.UPF > Cd 112.4000 Cd.pbe-van.UPF > ATOMIC_POSITIONS {angstrom} > O -2.181887203 -0.288129610 -0.491107761 > O -1.981730329 -0.473233788 1.733566845 > C -2.708815067 -0.432156029 0.673205720 > C -4.208743648 -0.532432174 0.801074150 > H -4.477821741 -1.131831132 1.680695448 > H -4.617325903 0.482417907 0.942246198 > H -4.646292897 -0.952999527 -0.113856718 > Cd -0.137714936 -0.219807184 0.461048531 > C 2.434799121 -0.031519760 0.246012557 > O 1.729314564 1.038606335 0.354394821 > O 1.885568682 -1.194338232 0.239827517 > C 3.936349099 0.076394885 0.148653197 > H 4.228550980 1.056061785 -0.250634315 > H 4.360345258 -0.025882832 1.161749300 > H 4.338742013 -0.738423646 -0.467966988 > > K_POINTS {automatic} > 1 1 1 0 0 0 > ========================================================================== > > 2. After the pw.x run with above file, with files in > "tmp" dir, file for pp.x > =========================================================================== > &inputpp > prefix = 'Cd_di_acetate' > outdir = > '/export/mousumi/PWSCF_2.0.1/Cd_2CH3COO/ELF/tmp//' > filplot = 'Cd_2CH3COO_chden' > plot_num = 0 > / > ========================================================================== > > 3. Then, input file for chdens.x > ========================================================================== > &input > nfile = 1 > filepp(1)='Cd_2CH3COO_chden' > iflag = 3 > output_format = 5 > / > ========================================================================== > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > __________________________________ Yahoo! Mail Stay connected, organized, and protected. Take the tour: http://tour.mail.yahoo.com/mailtour.html From mousumi at jncasr.ac.in Thu Jul 14 23:28:12 2005 From: mousumi at jncasr.ac.in (Mousumi Upadhyay Kahaly) Date: Fri, 15 Jul 2005 02:58:12 +0530 (IST) Subject: [Pw_forum] negative charge density! In-Reply-To: <20050714204225.10648.qmail@web60320.mail.yahoo.com> References: <47591.202.41.111.151.1121371528.squirrel@202.41.111.151> <20050714204225.10648.qmail@web60320.mail.yahoo.com> Message-ID: <59008.202.41.111.151.1121376492.squirrel@202.41.111.151> Respected Sir, I am trying to do calculations for a molecule(Cd-2CH3COO); so is there much meaning of taking more k-points instead only Gamma-pt? Secondly, I have tested pp.x and the chdens.x for Cd crystal system(hexagonal) with K_POINTS automatic 7 7 6 -6 -6 -5 and ecutwfc = 30.00000 Ry , ecutrho = 240.00000 Ry. Then also I get a negative charge of -0.70243E+00 at the output .xsf file & hence in xcrysden plot. All these negative charges come particularly on the metal atom. I have done these testing for Pb crystal system also with both LDA and GGA pseudopotentials. For LDA pseudopotential ( Pb.vdb.UPF) no negative charges appear whereas for GGA psp( pb_ps.uspp.UPF ) negative charges appear. So, what is the explanation for this negative charges? Can you please suggest some way out to get rid of this problem while using GGA pseudopotential? Best regards, mousumi > Hi, > > Did you try increase k-points instead of using only > Gamma-point? Besides, test also a little bit larger > ecutrho, say 30Ry. > > Bests, > Eyvaz. > > > --- Mousumi Upadhyay Kahaly > wrote: > >> Hi all, >> >> This is a resubmission of my mail regarding >> charge density calculation. >> >> > -------------------------------------------------------------------------- >> >> Dear all, >> >> I am trying to do charge analysis for >> Cd-di-acetate system. With >> the following files for pp.x and then chdens.x, as I >> am plotting >> the isosurfaces of charge densities with XCRYSDEN, I >> am getting >> some negative charge; >> Minimum grid value= -0.498760, and maximum >> grid value= 1.09220. >> How come I get such negative values(mainly around >> the Cd atom, as found) >> which is not negligible? What is the meaning of this >> NEGATIVE charge >> density? Can anyone please tell me, what is the >> mistake that I have made & >> what should I do to get rid of this? >> >> Best >> regards, Mousumi. >> >> 1. Input file for SCF pw.x run: >> > ========================================================================= >> &CONTROL >> calculation = 'scf' , >> restart_mode = 'from_scratch' , >> tstress =.true., >> outdir = >> '/export/mousumi/PWSCF_2.0.1/Cd_2CH3COO/ELF/tmp//' , >> pseudo_dir = '/export/mousumi/PWscf_GUI/pseudo/' , >> prefix = 'Cd_di_acetate' , >> / >> &SYSTEM >> ibrav = 8, >> celldm(1) = 50.000000, >> celldm(2) = 0.30, >> celldm(3) = 0.30, >> nat = 15, >> ntyp = 4, >> ecutwfc = 25.00000 , >> ecutrho = 150.00000 , >> nosym = .false. , >> nbnd = 37, >> occupations = 'smearing' , >> degauss = 0.003 , >> smearing = 'methfessel-paxton' , >> nspin = 1 , >> lda_plus_u = .false. , >> / >> &ELECTRONS >> conv_thr = 1.0d-12 , >> mixing_beta = 0.3 , >> / >> ATOMIC_SPECIES >> C 12.01000 C.pbe-van_bm.UPF >> H 1.00000 H.pbe-van_bm.UPF >> O 15.99990 O.pbe-van_bm.UPF >> Cd 112.4000 Cd.pbe-van.UPF >> ATOMIC_POSITIONS {angstrom} >> O -2.181887203 -0.288129610 -0.491107761 >> O -1.981730329 -0.473233788 1.733566845 >> C -2.708815067 -0.432156029 0.673205720 >> C -4.208743648 -0.532432174 0.801074150 >> H -4.477821741 -1.131831132 1.680695448 >> H -4.617325903 0.482417907 0.942246198 >> H -4.646292897 -0.952999527 -0.113856718 >> Cd -0.137714936 -0.219807184 0.461048531 >> C 2.434799121 -0.031519760 0.246012557 >> O 1.729314564 1.038606335 0.354394821 >> O 1.885568682 -1.194338232 0.239827517 >> C 3.936349099 0.076394885 0.148653197 >> H 4.228550980 1.056061785 -0.250634315 >> H 4.360345258 -0.025882832 1.161749300 >> H 4.338742013 -0.738423646 -0.467966988 >> >> K_POINTS {automatic} >> 1 1 1 0 0 0 >> > ========================================================================== >> >> 2. After the pw.x run with above file, with files in >> "tmp" dir, file for pp.x >> > =========================================================================== >> &inputpp >> prefix = 'Cd_di_acetate' >> outdir = >> '/export/mousumi/PWSCF_2.0.1/Cd_2CH3COO/ELF/tmp//' >> filplot = 'Cd_2CH3COO_chden' >> plot_num = 0 >> / >> > ========================================================================== >> >> 3. Then, input file for chdens.x >> > ========================================================================== >> &input >> nfile = 1 >> filepp(1)='Cd_2CH3COO_chden' >> iflag = 3 >> output_format = 5 >> / >> > ========================================================================== >> >> >> >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> > > > > > __________________________________ > Yahoo! Mail > Stay connected, organized, and protected. Take the tour: > http://tour.mail.yahoo.com/mailtour.html > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From eyvaz_isaev at yahoo.com Fri Jul 15 00:18:09 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Thu, 14 Jul 2005 15:18:09 -0700 (PDT) Subject: [Pw_forum] negative charge density! In-Reply-To: <59008.202.41.111.151.1121376492.squirrel@202.41.111.151> Message-ID: <20050714221810.81811.qmail@web60314.mail.yahoo.com> Hi, At this time I would like just to note that you use plane waves which means you use periodic boundary condition, i.e. a supercell. Bests, Eyvaz. --- Mousumi Upadhyay Kahaly wrote: > Respected Sir, > > I am trying to do calculations for a > molecule(Cd-2CH3COO); so > is there much meaning of taking more k-points > instead only > Gamma-pt? > > Secondly, I have tested pp.x and the > chdens.x for Cd crystal > system(hexagonal) with K_POINTS automatic > 7 7 6 -6 -6 -5 > and > ecutwfc = 30.00000 Ry , > ecutrho = 240.00000 Ry. > > Then also I get a negative charge of > -0.70243E+00 at the > output .xsf file & hence in xcrysden plot. > All these negative charges come > particularly on the metal > atom. I have done these testing for Pb crystal > system also > with both LDA and GGA pseudopotentials. For LDA > pseudopotential ( Pb.vdb.UPF) no negative charges > appear > whereas for GGA psp( pb_ps.uspp.UPF ) negative > charges appear. > So, what is the explanation for this negative > charges? > > Can you please suggest some way out to > get rid of this > problem while using GGA pseudopotential? > > Best regards, > mousumi > > > Hi, > > > > Did you try increase k-points instead of using > only > > Gamma-point? Besides, test also a little bit > larger > > ecutrho, say 30Ry. > > > > Bests, > > Eyvaz. > > > > > > --- Mousumi Upadhyay Kahaly > > wrote: > > > >> Hi all, > >> > >> This is a resubmission of my mail regarding > >> charge density calculation. > >> > >> > > > -------------------------------------------------------------------------- > >> > >> Dear all, > >> > >> I am trying to do charge analysis for > >> Cd-di-acetate system. With > >> the following files for pp.x and then chdens.x, > as I > >> am plotting > >> the isosurfaces of charge densities with > XCRYSDEN, I > >> am getting > >> some negative charge; > >> Minimum grid value= -0.498760, and maximum > >> grid value= 1.09220. > >> How come I get such negative values(mainly around > >> the Cd atom, as found) > >> which is not negligible? What is the meaning of > this > >> NEGATIVE charge > >> density? Can anyone please tell me, what is the > >> mistake that I have made & > >> what should I do to get rid of this? > >> > >> Best > >> regards, Mousumi. > >> > >> 1. Input file for SCF pw.x run: > >> > > > ========================================================================= > >> &CONTROL > >> calculation = 'scf' , > >> restart_mode = 'from_scratch' , > >> tstress =.true., > >> outdir = > >> > '/export/mousumi/PWSCF_2.0.1/Cd_2CH3COO/ELF/tmp//' , > >> pseudo_dir = '/export/mousumi/PWscf_GUI/pseudo/' > , > >> prefix = 'Cd_di_acetate' , > >> / > >> &SYSTEM > >> ibrav = 8, > >> celldm(1) = 50.000000, > >> celldm(2) = 0.30, > >> celldm(3) = 0.30, > >> nat = 15, > >> ntyp = 4, > >> ecutwfc = 25.00000 , > >> ecutrho = 150.00000 , > >> nosym = .false. , > >> nbnd = 37, > >> occupations = 'smearing' , > >> degauss = 0.003 , > >> smearing = > 'methfessel-paxton' , > >> nspin = 1 , > >> lda_plus_u = .false. , > >> / > >> &ELECTRONS > >> conv_thr = 1.0d-12 , > >> mixing_beta = 0.3 , > >> / > >> ATOMIC_SPECIES > >> C 12.01000 C.pbe-van_bm.UPF > >> H 1.00000 H.pbe-van_bm.UPF > >> O 15.99990 O.pbe-van_bm.UPF > >> Cd 112.4000 Cd.pbe-van.UPF > >> ATOMIC_POSITIONS {angstrom} > >> O -2.181887203 -0.288129610 -0.491107761 > >> O -1.981730329 -0.473233788 1.733566845 > >> C -2.708815067 -0.432156029 0.673205720 > >> C -4.208743648 -0.532432174 0.801074150 > >> H -4.477821741 -1.131831132 1.680695448 > >> H -4.617325903 0.482417907 0.942246198 > >> H -4.646292897 -0.952999527 -0.113856718 > >> Cd -0.137714936 -0.219807184 0.461048531 > >> C 2.434799121 -0.031519760 0.246012557 > >> O 1.729314564 1.038606335 0.354394821 > >> O 1.885568682 -1.194338232 0.239827517 > >> C 3.936349099 0.076394885 0.148653197 > >> H 4.228550980 1.056061785 -0.250634315 > >> H 4.360345258 -0.025882832 1.161749300 > >> H 4.338742013 -0.738423646 -0.467966988 > >> > >> K_POINTS {automatic} > >> 1 1 1 0 0 0 > >> > > > ========================================================================== > >> > >> 2. After the pw.x run with above file, with files > in > >> "tmp" dir, file for pp.x > >> > > > =========================================================================== > >> &inputpp > >> prefix = 'Cd_di_acetate' > >> outdir = > >> > '/export/mousumi/PWSCF_2.0.1/Cd_2CH3COO/ELF/tmp//' > >> filplot = 'Cd_2CH3COO_chden' > >> plot_num = 0 > >> / > >> > > > ========================================================================== > >> > >> 3. Then, input file for chdens.x > >> > > > ========================================================================== > >> &input > >> nfile = 1 > >> filepp(1)='Cd_2CH3COO_chden' > >> iflag = 3 > >> output_format = 5 > >> / > >> > > > ========================================================================== > >> > >> > >> > >> > >> _______________________________________________ > >> Pw_forum mailing list > >> Pw_forum at pwscf.org > >> > http://www.democritos.it/mailman/listinfo/pw_forum > >> > > > > > > > > > > __________________________________ > > Yahoo! Mail > > Stay connected, organized, and protected. Take the > tour: > === message truncated === ____________________________________________________ Start your day with Yahoo! - make it your home page http://www.yahoo.com/r/hs From degironc at sissa.it Fri Jul 15 09:19:03 2005 From: degironc at sissa.it (Stefano de Gironcoli) Date: Fri, 15 Jul 2005 09:19:03 +0200 Subject: [Pw_forum] negative charge density! References: <47591.202.41.111.151.1121371528.squirrel@202.41.111.151> <20050714204225.10648.qmail@web60320.mail.yahoo.com> <59008.202.41.111.151.1121376492.squirrel@202.41.111.151> Message-ID: <42D76367.7000103@sissa.it> Never use Methfessel-Paxton smearing in molecules. It is not positive definite. Your smearing width is in my opinion very small and could be the source of the problem also in the solid Cd case. Use Marzari-Vanderbilt or even simple gaussian broadening instead what do you mean by K_POINTS automatic 7 7 6 -6 -6 -5 ???????? check the syntax of automatic grid definition in INPUT_PW !!! stefano From mousumi at jncasr.ac.in Fri Jul 15 09:26:20 2005 From: mousumi at jncasr.ac.in (Mousumi Upadhyay Kahaly) Date: Fri, 15 Jul 2005 12:56:20 +0530 (IST) Subject: [Pw_forum] negative charge density! In-Reply-To: <20050714221810.81811.qmail@web60314.mail.yahoo.com> References: <59008.202.41.111.151.1121376492.squirrel@202.41.111.151> <20050714221810.81811.qmail@web60314.mail.yahoo.com> Message-ID: <39606.202.41.111.151.1121412380.squirrel@202.41.111.151> Dear Sir, Yes, i use supercell. So what should I do now? regards, mousumi > Hi, > > At this time I would like just to note that you use > plane waves which means you use periodic boundary > condition, i.e. a supercell. > > Bests, > Eyvaz. > --- Mousumi Upadhyay Kahaly > wrote: > >> Respected Sir, >> >> I am trying to do calculations for a >> molecule(Cd-2CH3COO); so >> is there much meaning of taking more k-points >> instead only >> Gamma-pt? >> >> Secondly, I have tested pp.x and the >> chdens.x for Cd crystal >> system(hexagonal) with K_POINTS automatic >> 7 7 6 -6 -6 -5 >> and >> ecutwfc = 30.00000 Ry , >> ecutrho = 240.00000 Ry. >> >> Then also I get a negative charge of >> -0.70243E+00 at the >> output .xsf file & hence in xcrysden plot. >> All these negative charges come >> particularly on the metal >> atom. I have done these testing for Pb crystal >> system also >> with both LDA and GGA pseudopotentials. For LDA >> pseudopotential ( Pb.vdb.UPF) no negative charges >> appear >> whereas for GGA psp( pb_ps.uspp.UPF ) negative >> charges appear. >> So, what is the explanation for this negative >> charges? >> >> Can you please suggest some way out to >> get rid of this >> problem while using GGA pseudopotential? >> >> Best regards, >> mousumi >> >> > Hi, >> > >> > Did you try increase k-points instead of using >> only >> > Gamma-point? Besides, test also a little bit >> larger >> > ecutrho, say 30Ry. >> > >> > Bests, >> > Eyvaz. >> > >> > >> > --- Mousumi Upadhyay Kahaly >> > wrote: >> > >> >> Hi all, >> >> >> >> This is a resubmission of my mail regarding >> >> charge density calculation. >> >> >> >> >> > >> > -------------------------------------------------------------------------- >> >> >> >> Dear all, >> >> >> >> I am trying to do charge analysis for >> >> Cd-di-acetate system. With >> >> the following files for pp.x and then chdens.x, >> as I >> >> am plotting >> >> the isosurfaces of charge densities with >> XCRYSDEN, I >> >> am getting >> >> some negative charge; >> >> Minimum grid value= -0.498760, and maximum >> >> grid value= 1.09220. >> >> How come I get such negative values(mainly around >> >> the Cd atom, as found) >> >> which is not negligible? What is the meaning of >> this >> >> NEGATIVE charge >> >> density? Can anyone please tell me, what is the >> >> mistake that I have made & >> >> what should I do to get rid of this? >> >> >> >> Best >> >> regards, Mousumi. >> >> >> >> 1. Input file for SCF pw.x run: >> >> >> > >> > ========================================================================= >> >> &CONTROL >> >> calculation = 'scf' , >> >> restart_mode = 'from_scratch' , >> >> tstress =.true., >> >> outdir = >> >> >> '/export/mousumi/PWSCF_2.0.1/Cd_2CH3COO/ELF/tmp//' , >> >> pseudo_dir = '/export/mousumi/PWscf_GUI/pseudo/' >> , >> >> prefix = 'Cd_di_acetate' , >> >> / >> >> &SYSTEM >> >> ibrav = 8, >> >> celldm(1) = 50.000000, >> >> celldm(2) = 0.30, >> >> celldm(3) = 0.30, >> >> nat = 15, >> >> ntyp = 4, >> >> ecutwfc = 25.00000 , >> >> ecutrho = 150.00000 , >> >> nosym = .false. , >> >> nbnd = 37, >> >> occupations = 'smearing' , >> >> degauss = 0.003 , >> >> smearing = >> 'methfessel-paxton' , >> >> nspin = 1 , >> >> lda_plus_u = .false. , >> >> / >> >> &ELECTRONS >> >> conv_thr = 1.0d-12 , >> >> mixing_beta = 0.3 , >> >> / >> >> ATOMIC_SPECIES >> >> C 12.01000 C.pbe-van_bm.UPF >> >> H 1.00000 H.pbe-van_bm.UPF >> >> O 15.99990 O.pbe-van_bm.UPF >> >> Cd 112.4000 Cd.pbe-van.UPF >> >> ATOMIC_POSITIONS {angstrom} >> >> O -2.181887203 -0.288129610 -0.491107761 >> >> O -1.981730329 -0.473233788 1.733566845 >> >> C -2.708815067 -0.432156029 0.673205720 >> >> C -4.208743648 -0.532432174 0.801074150 >> >> H -4.477821741 -1.131831132 1.680695448 >> >> H -4.617325903 0.482417907 0.942246198 >> >> H -4.646292897 -0.952999527 -0.113856718 >> >> Cd -0.137714936 -0.219807184 0.461048531 >> >> C 2.434799121 -0.031519760 0.246012557 >> >> O 1.729314564 1.038606335 0.354394821 >> >> O 1.885568682 -1.194338232 0.239827517 >> >> C 3.936349099 0.076394885 0.148653197 >> >> H 4.228550980 1.056061785 -0.250634315 >> >> H 4.360345258 -0.025882832 1.161749300 >> >> H 4.338742013 -0.738423646 -0.467966988 >> >> >> >> K_POINTS {automatic} >> >> 1 1 1 0 0 0 >> >> >> > >> > ========================================================================== >> >> >> >> 2. After the pw.x run with above file, with files >> in >> >> "tmp" dir, file for pp.x >> >> >> > >> > =========================================================================== >> >> &inputpp >> >> prefix = 'Cd_di_acetate' >> >> outdir = >> >> >> '/export/mousumi/PWSCF_2.0.1/Cd_2CH3COO/ELF/tmp//' >> >> filplot = 'Cd_2CH3COO_chden' >> >> plot_num = 0 >> >> / >> >> >> > >> > ========================================================================== >> >> >> >> 3. Then, input file for chdens.x >> >> >> > >> > ========================================================================== >> >> &input >> >> nfile = 1 >> >> filepp(1)='Cd_2CH3COO_chden' >> >> iflag = 3 >> >> output_format = 5 >> >> / >> >> >> > >> > ========================================================================== >> >> >> >> >> >> >> >> >> >> _______________________________________________ >> >> Pw_forum mailing list >> >> Pw_forum at pwscf.org >> >> >> http://www.democritos.it/mailman/listinfo/pw_forum >> >> >> > >> > >> > >> > >> > __________________________________ >> > Yahoo! Mail >> > Stay connected, organized, and protected. Take the >> tour: >> > === message truncated === > > > > > ____________________________________________________ > Start your day with Yahoo! - make it your home page > http://www.yahoo.com/r/hs > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From proffess at yandex.ru Fri Jul 15 14:28:29 2005 From: proffess at yandex.ru (Sergey Lisenkov) Date: Fri, 15 Jul 2005 16:28:29 +0400 (MSD) Subject: [Pw_forum] Distribution of the memory when use -npool option Message-ID: <42D7ABED.000004.08210@tide.yandex.ru> Dear PWscf users and authors, I am interested how are memory distributed when use -npool option. In my case, I perform calculations with 64 k-points. When I use : mpirun -np 64 ./pw.x -npool 64 -in my.inp (with k-point parallelization) I see that a lot of memory (~30%) are used on each node (measured by 'top' command). But when I use: mpirun -np 64 ./pw.x -in my.inp (without k-point parallelization) only ~5% of memory of each node is used. I believe, that using k-point parallelization is highly efficient, but it follows to out of memory, killing network mappers and etc. So, is there a compromise? Best wishes, Sergey From sbraccia at sissa.it Fri Jul 15 14:54:24 2005 From: sbraccia at sissa.it (sbraccia carlo) Date: Fri, 15 Jul 2005 08:54:24 -0400 Subject: [Pw_forum] Distribution of the memory when use -npool option In-Reply-To: <42D7ABED.000004.08210@tide.yandex.ru> References: <42D7ABED.000004.08210@tide.yandex.ru> Message-ID: <42D7B200.10309@sissa.it> Dear Sergey, the compromise is using the two schemes together: mpirun -np 64 ./pw.x -npool 32 -in my.inp (to have 32 pools each one of 2 cpus, the memory is ~1/2 the memory for a serial run) or mpirun -np 64 ./pw.x -npool 16 -in my.inp (to have 16 pools each one of 4 cpus, the memory is ~1/4 the memory for a serial run) and so on. carlo Sergey Lisenkov wrote: > Dear PWscf users and authors, > > > I am interested how are memory distributed when use -npool option. In my case, I perform calculations with 64 k-points. When I use : > mpirun -np 64 ./pw.x -npool 64 -in my.inp (with k-point parallelization) > > I see that a lot of memory (~30%) are used on each node (measured by 'top' command). But when I use: > > mpirun -np 64 ./pw.x -in my.inp (without k-point parallelization) > > only ~5% of memory of each node is used. I believe, that using k-point parallelization is highly efficient, but it follows to out of memory, killing network mappers and etc. > > So, is there a compromise? > > Best wishes, > Sergey > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From proffess at yandex.ru Fri Jul 15 15:15:25 2005 From: proffess at yandex.ru (Sergey Lisenkov) Date: Fri, 15 Jul 2005 17:15:25 +0400 (MSD) Subject: [Pw_forum] Distribution of the memory when use -npool option In-Reply-To: <42D7B200.10309@sissa.it> References: <42D7ABED.000004.08210@tide.yandex.ru> <42D7B200.10309@sissa.it> Message-ID: <42D7B6ED.000002.31277@mfront7.yandex.ru> Dear Carlo, Thank you very much for your suggestions. Best wishes, Sergey From yjiang at asu.edu Fri Jul 15 15:40:27 2005 From: yjiang at asu.edu (Yong Jiang) Date: Fri, 15 Jul 2005 06:40:27 -0700 (MST) Subject: [Pw_forum] why phonon calc stops at the first mode In-Reply-To: <1121372044.42d6c78c6c218@webmail.asu.edu> References: <1121175554.42d3c802bc4de@webmail.asu.edu> <200507141802.00087.giannozz@nest.sns.it> <1121372044.42d6c78c6c218@webmail.asu.edu> Message-ID: <1121434827.42d7bccb2bca8@webmail.asu.edu> Dear Dr. Paolo Giannozzi, I find a parallel run helps! The parallel run can proceed within the first phonon mode calc while the serial run always stop at the very beginning. But this time it seems I lose the convergency during the self-consistent calc of the first mode. Could you make some comment on my input? Thanks very much for your kind help. Best regards, Yong ---------------------scf.in----------------------------- &control calculation='scf', restart_mode='from_scratch', tprnfor = .true., prefix='Ni', pseudo_dir = '/home/hzhwc8/PWscf/espresso-2.1.4/pseudo/', outdir='/home/hzhwc8/PWscf/espresso-2.1.4/examples/example03/tmp/' / &system ibrav=1, celldm(1) =13.31444, nat= 32, ntyp= 1, nspin=2, starting_magnetization(1)=0.5, degauss=0.003, smearing='mp', occupations='smearing', ecutwfc =27.0, ecutrho =350.0 / &electrons diagonalization='cg', mixing_beta = 0.3, / ATOMIC_SPECIES Ni 58.6934 Ni.pbe-nd-rrkjus.UPF ATOMIC_POSITIONS {angstrom} Ni 0.000 0.000 0.000 Ni 0.000 1.762 1.762 Ni 1.762 0.000 1.762 Ni 1.762 1.762 0.000 Ni 0.000 0.000 3.523 Ni 0.000 1.762 5.285 Ni 1.762 0.000 5.285 Ni 1.762 1.762 3.523 Ni 0.000 3.523 0.000 Ni 0.000 5.285 1.762 Ni 1.762 3.523 1.762 Ni 1.762 5.285 0.000 Ni 0.000 3.523 3.523 Ni 0.000 5.285 5.285 Ni 1.762 3.523 5.285 Ni 1.762 5.285 3.523 Ni 3.523 0.000 0.000 Ni 3.523 1.762 1.762 Ni 5.285 0.000 1.762 Ni 5.285 1.762 0.000 Ni 3.523 0.000 3.523 Ni 3.523 1.762 5.285 Ni 5.285 0.000 5.285 Ni 5.285 1.762 3.523 Ni 3.523 3.523 0.000 Ni 3.523 5.285 1.762 Ni 5.285 3.523 1.762 Ni 5.285 5.285 0.000 Ni 3.523 3.523 3.523 Ni 3.523 5.285 5.285 Ni 5.285 3.523 5.285 Ni 5.285 5.285 3.523 K_POINTS {automatic} 2 2 2 1 1 1 -------------------phG.in------------------------- phonons of Ni at Gamma &inputph tr2_ph=1.0d-14, prefix='Ni', amass(1)=58.6934, outdir='/home/hzhwc8/PWscf/espresso-2.1.4/examples/example03/tmp/', fildyn='Ni.dynG', / 0.0 0.0 0.0 -------------------phG.out------------------------- Representation 62 1 modes - To be done Representation 63 2 modes - To be done Representation 64 1 modes - To be done PHONON : 6h46m CPU time Alpha used in Ewald sum = 2.8000 Representation # 1 mode # 1 Self-consistent Calculation Pert. # 1: Fermi energy shift (Ryd) = 0.0000 0.0000 iter # 1 total cpu time : 25160.0 secs av.it.: 5.0 thresh= 0.100E-01 alpha_mix = 0.700 |ddv_scf|^2 = 0.410E-04 Pert. # 1: Fermi energy shift (Ryd) = -0.0035 0.0000 iter # 2 total cpu time : 26199.2 secs av.it.: 11.5 thresh= 0.640E-03 alpha_mix = 0.700 |ddv_scf|^2 = 0.498E-01 Pert. # 1: Fermi energy shift (Ryd) = 0.0002 0.0000 iter # 3 total cpu time : 27074.1 secs av.it.: 8.0 thresh= 0.100E-01 alpha_mix = 0.700 |ddv_scf|^2 = 0.282E-0 ... Pert. # 1: Fermi energy shift (Ryd) = -0.3054 0.0000 iter # 48 total cpu time : 68304.7 secs av.it.: 13.8 thresh= 0.100E-01 alpha_mix = 0.700 |ddv_scf|^2 = 0.132E+03 Pert. # 1: Fermi energy shift (Ryd) = 0.2026 0.0000 iter # 49 total cpu time : 69515.5 secs av.it.: 15.0 thresh= 0.100E-01 alpha_mix = 0.700 |ddv_scf|^2 = 0.191E+03 Pert. # 1: Fermi energy shift (Ryd) = 0.0478 0.0000 iter # 50 total cpu time : 70733.4 secs av.it.: 15.0 thresh= 0.100E-01 alpha_mix = 0.700 |ddv_scf|^2 = 0.412E+02 Pert. # 1: Fermi energy shift (Ryd) = -0.0590 0.0000 iter # 51 total cpu time : 71859.1 secs av.it.: 13.5 thresh= 0.100E-01 alpha_mix = 0.700 |ddv_scf|^2 = 0.217E+02 -------------------------------------------------------------------------- Quoting Yong Jiang : > Dear Dr. Paolo Giannozzi, > > Thank you very much for your reply. > > If it is the case, could the code be updated in later versions to be > able to > prompt such error message whenever it encounts, so that users could take > action > accordingly? > > By the way, is there any unreasonable setting in my input files can be > > downscaled/modified to avoid such problem? Would a parallel run be > helpful? > > Thanks very much! > > Yong > > Quoting Paolo Giannozzi : > > > On Tuesday 12 July 2005 15:39, Yong Jiang wrote: > > > > > the phonon calc stops at the first mode without any error/reason. > > > > in such a large calculation, there can be 1001 reasons for the > > code to stop, with lack of memory or of disk space high in the > > list > > > > 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 Fri Jul 15 18:18:58 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Fri, 15 Jul 2005 09:18:58 -0700 (PDT) Subject: [Pw_forum] why phonon calc stops at the first mode In-Reply-To: <1121434827.42d7bccb2bca8@webmail.asu.edu> Message-ID: <20050715161859.12506.qmail@web60320.mail.yahoo.com> Hi, > > Pert. # 1: Fermi energy shift (Ryd) = > -0.0590 0.0000 > iter # 51 total cpu time : 71859.1 secs > av.it.: 13.5 > thresh= 0.100E-01 alpha_mix = 0.700 > |ddv_scf|^2 = 0.217E+02 > Just decrease alpha_mix for phonon calculations, i.e. in *.phG.in file add alpha_mix(1)=0.1 By default alpha_mix=0.7 Bests, Eyvaz. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From lsl0426 at yahoo.com Sat Jul 16 10:39:23 2005 From: lsl0426 at yahoo.com (Shihn Lun) Date: Sat, 16 Jul 2005 01:39:23 -0700 (PDT) Subject: [Pw_forum] Convert FHI Pseudopotential to UPF Message-ID: <20050716083923.24396.qmail@web31109.mail.mud.yahoo.com> Hi everybody, I try to convert FHI pseudopotential to UPF by fhi2upf.f90, when I perform this converting program, this program tell me enter "Wavefunction # 1: label, occupancy >", but I dont konw how to enter suitable 'Wavefunction', it seems very important. And I read a file 'upftool/UPF', I still not understand what 'Wavefunction' means. could anybody give me some hints or suggestions ? Thanks in advance, Shinn-Lun. -----Here is FHI input-------- 40.00 8 2 8 0.00 : z nc nv 1 0 2.00 : n l f 2 0 2.00 2 1 6.00 3 0 2.00 3 1 6.00 3 2 10.00 4 0 2.00 4 1 6.00 4 2 2.00 5 0 2.00 3 "t" : lmax s_pp_def 1 0.0 15.00 "t" 0 0.0 0 "t" 2 0.0 0 "t" 3 0.0 0.01 "t" -- __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From leoant21 at hotmail.com Mon Jul 18 05:17:52 2005 From: leoant21 at hotmail.com (=?gb2312?B?1cUguumx8g==?=) Date: Mon, 18 Jul 2005 03:17:52 +0000 Subject: [Pw_forum] Some simple questions on phonon calculation results. In-Reply-To: <20050717053631.13415.17400.Mailman@democritos.sissa.it> Message-ID: Dear all: Here are some questions I have about the phonon calculation and how to process its results.plz help me. 1. the units of the output file in .modes file, is it THz? and if I want to get the energy, is it right to multiply omegas by h-bar? and why I got modes with omega( 1) = 0.000000 [THz] = -0.000010 [cm-1] ~~~~~is this negative walue meaningful? 2. From which output file can I get the dispersion curves plotted? using plotband.x? As to the special k-lines for the dispersion curve plotting, are they included in matdyn.x's input file with dos!=.true? and in example06, I wonder, why there are K-points input in matdyn.x's input file while dos=.true, since in the matdyn.f90 files it is said that there's no need for k-points when dos=.true? 3. what is the .modes and .freq file used for in further analysis? It's seems to be silly to ask these questions, but please help me. Best regards _________________________________________________________________ ???? MSN Explorer: http://explorer.msn.com/lccn From adeagbo at thp.Uni-Duisburg.DE Mon Jul 18 10:30:55 2005 From: adeagbo at thp.Uni-Duisburg.DE (Adeagbo Waheed Adeniyi) Date: Mon, 18 Jul 2005 10:30:55 +0200 (MESZ) Subject: [Pw_forum] Electron-phonon coupling In-Reply-To: <20050718053632.17970.26492.Mailman@democritos.sissa.it> References: <20050718053632.17970.26492.Mailman@democritos.sissa.it> Message-ID: Dear all, Is there is anyway to calculate el-ph coupling for spin-polarized system?. I got the message that el-ph coupling is not implemented for spin system (when ISPIN =2) while the code runs for non-spin systems. Best regards Adeagbo =================================================== ADEAGBO WAHEED ADENIYI Institute of Physics, MC 328 University Duisburg-Essen Lotharstr. 1 47048 Duisburg Germany ==================================================== Phone: +49 203 379 1606 Fax: +49 203 379 3665 Present e-mail address: adeagbo at thp.uni-duisburg.de Permanent e-mail address: wadeagbo at mail.skannet.com http://www.thp.uni-duisburg.de ==================================================== "The intellect has little to do on the road to discovery. There comes a leap in consciousness, call it intuition or what you will, and the solution comes to you and you don't know how or why." -- AlbertEinstein From giannozz at nest.sns.it Mon Jul 18 15:56:00 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 18 Jul 2005 15:56:00 +0200 Subject: [Pw_forum] Some simple questions on phonon calculation results. In-Reply-To: References: Message-ID: <200507181556.00775.giannozz@nest.sns.it> On Monday 18 July 2005 05:17, ? ?? wrote: > 1. the units of the output file in .modes file, is it THz? units for frequencies are explicitly specified; normal modes are normalized to 1, thus they are dimensionless > As to the special k-lines for the dispersion curve plotting, > are they included in matdyn.x's input file with dos!=.true? no, you have to supply the list of wavevectors for which you want to plot them > and in example06, I wonder, why there are K-points input in > matdyn.x's input file while dos=.true, since in the matdyn.f90 > files it is said that there's no need for k-points when dos=.true? they shouldn't be there, thank you for bringing this to our attention 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 Mon Jul 18 17:13:27 2005 From: burde1jb at cmich.edu (John B Burdette) Date: Mon, 18 Jul 2005 11:13:27 -0400 Subject: [PW_Forum] Phonon Calculation Problem Message-ID: <750ce52.1dd9af54.81f5b00@student-b.csv.cmich.edu> Dear all, I am trying to run ph.x for phonon calculations of TiO2 in rutile form. However, when I have created my input file, along with already running the TiO2 through the pw.x program, I receive the following error: Ultrasoft (Vanderbilt) Pseudopotentials Reading file TiO2.save ... read complete Reading file TiO2.save ... %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%% from read_restart_pseudo : error # 1 wrong size %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%% Now, the only other situation on this forum in which this occurred was solved by increasing the number of points in the atomic mesh in Modules/parameters.f90 to beyond the number grid points in the pseudpotentials being used. My ndmx in parameters.f90 is set to 2500 while the pseudopotential grid points are at 2199 at most. So this is not the problem. Can anyone help me? Best regards, John Burdette From giannozz at nest.sns.it Mon Jul 18 17:16:05 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 18 Jul 2005 17:16:05 +0200 Subject: [Pw_forum] Convert FHI Pseudopotential to UPF In-Reply-To: <20050716083923.24396.qmail@web31109.mail.mud.yahoo.com> References: <20050716083923.24396.qmail@web31109.mail.mud.yahoo.com> Message-ID: <200507181716.05714.giannozz@nest.sns.it> On Saturday 16 July 2005 10:39, Shihn Lun wrote: > I try to convert FHI pseudopotential to UPF by fhi2upf.f90, > when I perform this converting program, this program tell > me enter "Wavefunction # 1: label, occupancy >", > but I dont know how to enter suitable 'Wavefunction' like all pseudopotentials, the one you are trying to convert was generated starting from atomic valence orbitals for some reference electronic configuration. The label and occupancy of these orbitals is what is required. Actually the only REALLY important thing is the letter in the label part (i.e. "s" in "5s") 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 Jul 18 18:31:12 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 18 Jul 2005 18:31:12 +0200 Subject: [PW_Forum] Phonon Calculation Problem In-Reply-To: <750ce52.1dd9af54.81f5b00@student-b.csv.cmich.edu> References: <750ce52.1dd9af54.81f5b00@student-b.csv.cmich.edu> Message-ID: <200507181831.12259.giannozz@nest.sns.it> On Monday 18 July 2005 17:13, John B Burdette wrote: > I am trying to run ph.x for phonon calculations of TiO2 > in rutile form. However, when I have created my input file, > along with already running the TiO2 through the pw.x program, > I receive the following error: [...] > from read_restart_pseudo : error # 1 > wrong size if this happens reproducibly, please provide a test that can be rerun (i.e. complete with pseudopotentials or pointers to where they are) and a brief description of your "experimental setup" (i.e. code version, compiler, ...) 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 Jul 18 18:38:37 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 18 Jul 2005 18:38:37 +0200 Subject: [Pw_forum] pseudopotential of Ba and Ti In-Reply-To: <8d38992a05070600395a94d698@mail.gmail.com> References: <8d38992a05070600395a94d698@mail.gmail.com> Message-ID: <200507181838.37802.giannozz@nest.sns.it> On Wednesday 06 July 2005 09:39, Jian ZHOU wrote: > What is the difference between the choice ibrav=6 or ibrav=7? it is explained towards the end of the INPUT_PW file ibrav=6: Simple Tetragonal ibrav=7: Body-Centred Tetragonal 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 Jul 18 19:16:49 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 18 Jul 2005 19:16:49 +0200 Subject: [Pw_forum] Re: FHI pseudopotential In-Reply-To: <20050628112756.67653.qmail@web31108.mail.mud.yahoo.com> References: <20050628112756.67653.qmail@web31108.mail.mud.yahoo.com> Message-ID: <200507181916.49705.giannozz@nest.sns.it> On Tuesday 28 June 2005 13:27, Shihn Lun wrote: > Recently, I tried to generate norm-conserving pseudo- > potential with FHI tool, and compared its bandstrure > with pwscf's(Zr.pbe-nsp-van.UPF).But they are not > very agreement, I am sure that pwscf's result is right. the UPF pseudopotential is PBE, the FHI one is LDA; the UPF pseudopotential has the nonlinear core correction, the FHI maybe not 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 Jul 18 19:18:05 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 18 Jul 2005 19:18:05 +0200 Subject: [Pw_forum] Fwd: Re: Segmentation fault with Setting 2 --- follow-up of Re. ATOMIC_POSITIONS In-Reply-To: <1120075597.42c2ff4d914de@webmail.asu.edu> References: <1120075597.42c2ff4d914de@webmail.asu.edu> Message-ID: <200507181918.05761.giannozz@nest.sns.it> On Wednesday 29 June 2005 22:06, Yong Jiang wrote: > running the phonon calculation at Gamma..../test_RUN_atGnK: line 88: > 7992 Segmentation fault $PH_COMMAND Ni.phG.out > done it works for me , or more exactly, it doesn't crash. v.2.1.4, Intel PC with ifc 7.1 and mkl > running the phonon calculation at K...forrtl: severe (24): end-of-file > during read, unit 99, file /home/hzhwc8/PWscf/tmp/recover this is something different: the restart file left by the previously interrupted calculation has confused the following calculation. This is fixed in the cvs version. > (ENDS HERE FOR WHAT REASON ?!) many possible reasons, including a buggy compiler 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 patricia at fcq.unc.edu.ar Mon Jul 18 15:13:02 2005 From: patricia at fcq.unc.edu.ar (Patricia Paredes) Date: Mon, 18 Jul 2005 15:13:02 Subject: [Pw_forum] espresso 2.1.4 and ifc8.1.024 In-Reply-To: <42D68174.3000104@mit.edu> References: <3.0.1.32.20050714120519.01a420d0@mail.fcq.unc.edu.ar> <3.0.1.32.20050707175957.011fa2c0@mail.fcq.unc.edu.ar> <3.0.1.32.20050701190023.012f7aa0@mail.fcq.unc.edu.ar> <3.0.1.32.20050707175957.011fa2c0@mail.fcq.unc.edu.ar> <3.0.1.32.20050714120519.01a420d0@mail.fcq.unc.edu.ar> Message-ID: <3.0.1.32.20050718151302.01a420d0@mail.fcq.unc.edu.ar> A non-text attachment was scrubbed... Name: not available Type: text/enriched Size: 8053 bytes Desc: not available Url : /pipermail/attachments/20050718/41febf86/attachment.bin From leoant21 at hotmail.com Tue Jul 19 08:28:16 2005 From: leoant21 at hotmail.com (=?gb2312?B?1cUguumx8g==?=) Date: Tue, 19 Jul 2005 06:28:16 +0000 Subject: [Pw_forum] RE: Some simple questions on phonon calculation results. In-Reply-To: <20050719053624.2677.52644.Mailman@democritos.sissa.it> Message-ID: Dear Paolo: Thank you very much for your reply. However, I want to know exactly which program should be used to get the dispersion curves plot? And using .freq file generated by matdyn.x? Best Regards. Hongbin Zhang _________________________________________________________________ ???? MSN Explorer: http://explorer.msn.com/lccn/ From adeagbo at thp.Uni-Duisburg.DE Tue Jul 19 11:27:54 2005 From: adeagbo at thp.Uni-Duisburg.DE (Adeagbo Waheed Adeniyi) Date: Tue, 19 Jul 2005 11:27:54 +0200 (MESZ) Subject: [Pw_forum] [PW_Forum]Electron-phonon coupling Message-ID: Dear all, Is there is a way to calculate el-ph coupling for spin-polarized system such as Ni2MnGa?. I got the message that el-ph coupling is not implemented for spin polirazied system (when ISPIN =2) %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from phq_readin : error # 1 El-ph and spin not implemented %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% while the code runs for non-spin polarized system. Best regards Adeagbo From giannozz at nest.sns.it Tue Jul 19 13:34:34 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 19 Jul 2005 13:34:34 +0200 Subject: [Pw_forum] negative charge density! In-Reply-To: <46254.202.41.111.151.1121192927.squirrel@202.41.111.151> References: <46254.202.41.111.151.1121192927.squirrel@202.41.111.151> Message-ID: <200507191334.35064.giannozz@nest.sns.it> On Tuesday 12 July 2005 20:28, Mousumi Upadhyay Kahaly wrote: > How come I get such negative values (mainly around the Cd atom, > as found) which is not negligible? What is the meaning of this > NEGATIVE charge density? Can anyone please tell me, what is the > mistake that I have made what should I do to get rid of this? there is no mistake. Ultrasoft pseudopotentials may give raise to a small negative charge, for various reasons that are explained in the manual. In most but not all cases, it disappears with increasing cutoff (see for instance: http://www.democritos.it/pipermail/pw_forum/2003-April/000207.html ). In your case the pseudopotential of Cd is clearly the culprit. I am a little bit puzzled by the size of the negative charge, which does not seem to be reduced by increasing cutoff, so please check carefully in simple systems the correct behavior of the Cd pseudopotentials. 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 katawut at excc.ex.ac.uk Tue Jul 19 17:19:13 2005 From: katawut at excc.ex.ac.uk (Katawut Chuasiripattana) Date: Tue, 19 Jul 2005 16:19:13 +0100 Subject: [Pw_forum] Slab calculation. Message-ID: <42DD19F1.4010704@excc.ex.ac.uk> Dear all, Please answer about slab surface calculation with ESPRESSO/pwscf. How to do the slab calculation by fixing several layers of the bottom of the slab and let the rest relax e.g. Si(001) surface with 8 layers with H passivated at the bottom and fixed the last bottom Si layer and let the rest relax? Thank you. Katawut -- ========================================================= Katawut Chuasiripattana 2nd Floor School of Physics Exeter University Stocker road Exeter, EX4 4QL UK http://newton.ex.ac.uk/research/qsystems/people/katawut/ ========================================================= From proffess at yandex.ru Wed Jul 20 08:39:33 2005 From: proffess at yandex.ru (Sergey Lisenkov) Date: Wed, 20 Jul 2005 10:39:33 +0400 (MSD) Subject: [Pw_forum] Pt pseudopotential from atomic code Message-ID: <42DDF1A5.000009.07784@pantene.yandex.ru> Dear PWscf users, I try to generate Pt pseudopotential using ld1.x code and example which is given in atomic_doc as test. This example with rel = 2 option (fully relativistic). I switch to rel=1 (scalar relativistic) and got the error: 'strange occupations'. What should I change in the input file else? Thanks, Best wishes, Sergey From giannozz at nest.sns.it Wed Jul 20 10:33:46 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Wed, 20 Jul 2005 10:33:46 +0200 Subject: [Pw_forum] RE: Some simple questions on phonon calculation results. In-Reply-To: References: Message-ID: <200507201033.46519.giannozz@nest.sns.it> On Tuesday 19 July 2005 08:28, ? ?? wrote: > which program should be used to get the dispersion curves plot? the one you prefer. Presently the plot of phonon dispersions is not implemented. It will be done sooner or later. In the meantime you may easily modify the format of file 'flfrq' to fit your needs. 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 dalcorso at sissa.it Wed Jul 20 13:14:02 2005 From: dalcorso at sissa.it (Andrea Dal Corso) Date: Wed, 20 Jul 2005 13:14:02 +0200 Subject: [Pw_forum] Pt pseudopotential from atomic code In-Reply-To: <42DDF1A5.000009.07784@pantene.yandex.ru> References: <42DDF1A5.000009.07784@pantene.yandex.ru> Message-ID: <1121858043.3755.16.camel@dhpc-5-33.sissa.it> The occupations must be those of the scalar relativistic case. For Pt they become: 4 5D 3 2 8.00 0.00 2.10 2.40 1.50 5D 3 2 0.00 -0.20 2.10 2.40 1.50 6P 2 1 0.00 -0.00 3.30 3.30 0.50 6S 1 0 2.00 0.00 2.60 2.60 0.50 or you can also try the configuration: 4 5D 3 2 9.00 0.00 2.10 2.40 1.50 5D 3 2 0.00 -0.20 2.10 2.40 1.50 6P 2 1 0.00 -0.00 3.30 3.30 0.50 6S 1 0 1.00 0.00 2.60 2.60 0.50 Best Regards, Andrea On Wed, 2005-07-20 at 10:39 +0400, Sergey Lisenkov wrote: > Dear PWscf users, > > I try to generate Pt pseudopotential using ld1.x code and example which is given in atomic_doc as test. This example with rel = 2 option (fully relativistic). I switch to rel=1 (scalar relativistic) and got the error: 'strange occupations'. What should I change in the input file else? > > Thanks, > Best wishes, > Sergey > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- Andrea Dal Corso Tel. 0039-040-3787428 SISSA, Via Beirut 2/4 Fax. 0039-040-3787528 34014 Trieste (Italy) e-mail: dalcorso at sissa.it From proffess at yandex.ru Wed Jul 20 13:36:03 2005 From: proffess at yandex.ru (Sergey Lisenkov) Date: Wed, 20 Jul 2005 15:36:03 +0400 (MSD) Subject: [Pw_forum] Pt pseudopotential from atomic code In-Reply-To: <1121858043.3755.16.camel@dhpc-5-33.sissa.it> References: <42DDF1A5.000009.07784@pantene.yandex.ru> <1121858043.3755.16.camel@dhpc-5-33.sissa.it> Message-ID: <42DE3723.000001.23337@pantene.yandex.ru> Dear Andrea, Thank you very much for your help. It helps me a lot! Best wishes, Sergey From eyvaz_isaev at yahoo.com Wed Jul 20 14:00:29 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Wed, 20 Jul 2005 05:00:29 -0700 (PDT) Subject: [Pw_forum] RE: Some simple questions on phonon calculation results. In-Reply-To: <200507201033.46519.giannozz@nest.sns.it> Message-ID: <20050720120029.42430.qmail@web60314.mail.yahoo.com> Dear Paolo, Dear users, I can add this feature into my small programs. Besides, my apologies for not submitting the FS program, I would like just test it on a parallel computer. Hopefully, I will do that in a few days. Bests, Eyvaz. --- Paolo Giannozzi wrote: > On Tuesday 19 July 2005 08:28, ?? ???? wrote: > > > which program should be used to get the dispersion > curves plot? > > the one you prefer. Presently the plot of phonon > dispersions is > not implemented. It will be done sooner or later. In > the meantime > you may easily modify the format of file 'flfrq' to > fit your needs. > > 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 From proffess at yandex.ru Wed Jul 20 19:42:48 2005 From: proffess at yandex.ru (Sergey Lisenkov) Date: Wed, 20 Jul 2005 21:42:48 +0400 (MSD) Subject: [Pw_forum] Error during CP calculations Message-ID: <42DE8D18.000001.12807@ariel.yandex.ru> Dear Espresso users and authors, I try to run cp.x code but I got the next message: .... iprsta = 1 unit vectors of full simulation cell in real space: in reciprocal space: 58.5815 0.0000 0.0000 1.0000 0.0000 0.0000 0.0000 33.0891 0.0000 0.0000 1.7704 0.0000 0.0000 0.0000 33.0891 0.0000 0.0000 1.7704 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from set_fft_para : error # 64512 ncplanex too small %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... It seems I have to increase the parameter 'ncplanex'. What is preffered value? Thanks, Sergey From axel.kohlmeyer at theochem.ruhr-uni-bochum.de Wed Jul 20 19:57:57 2005 From: axel.kohlmeyer at theochem.ruhr-uni-bochum.de (Axel Kohlmeyer) Date: Wed, 20 Jul 2005 19:57:57 +0200 (CEST) Subject: [Pw_forum] Error during CP calculations In-Reply-To: <42DE8D18.000001.12807@ariel.yandex.ru> Message-ID: On Wed, 20 Jul 2005, Sergey Lisenkov wrote: SL> SL> Dear Espresso users and authors, dear sergey, please always state which version of espresso you are using. SL> SL> I try to run cp.x code but I got the next message: SL> SL> .... SL> iprsta = 1 SL> SL> SL> unit vectors of full simulation cell SL> in real space: in reciprocal space: SL> 58.5815 0.0000 0.0000 1.0000 0.0000 0.0000 SL> 0.0000 33.0891 0.0000 0.0000 1.7704 0.0000 SL> 0.0000 0.0000 33.0891 0.0000 0.0000 1.7704 SL> SL> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% SL> from set_fft_para : error # 64512 SL> ncplanex too small SL> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% SL> SL> stopping ... SL> SL> It seems I have to increase the parameter 'ncplanex'. What is preffered value? how about 65000? seems like you need at least 64512... ;-) hope, you have a machine able to handle such a big job and don't run into other internal limitations. good luck, axel. SL> SL> Thanks, SL> Sergey SL> _______________________________________________ SL> Pw_forum mailing list SL> Pw_forum at pwscf.org SL> http://www.democritos.it/mailman/listinfo/pw_forum SL> SL> -- ======================================================================= Dr. Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/ ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From yi.kong at gmail.com Thu Jul 21 09:12:26 2005 From: yi.kong at gmail.com (Yi Kong) Date: Thu, 21 Jul 2005 15:12:26 +0800 Subject: [Pw_forum] How to relax the HCP type structures? Message-ID: <8c9dc66c05072100126826f5ac@mail.gmail.com> Dear All, I want to find the ground state of D019 (HCP type) structure of Pd-Ru alloys in the PWSCF, I follow the input of the EXAMPLE 3, and the structure parameters I input are: ********************************************* ibrav= 4, celldm(1)= $acons,celldm(3)=$ccons, nat=8, ntyp=2, ... ... ATOMIC_POSITIONS (crystal) Ru 0.50 0.0 0.00 Ru 0.50 0.50 0.00 Ru 0.00 0.50 0.00 Ru 0.333333 0.1666667 0.50 Ru 0.833333 0.1666667 0.50 Ru 0.833333 0.6666667 0.50 Pd 0.00 0.00 0.00 Pd 0.333333 0.6666667 0.50 ************************************************ The $acons and $ccons are varied so as to find the lattice constants a and c through comparing the obtained energies. I just want to know that are these input correct, and are there more efficient method to find the ground state of the HCP type structures. Thanks a lot! -- ============================= Yi Kong Department of MSE Tsinghua University PR China, 100084 ============================= From mousumi at jncasr.ac.in Thu Jul 21 20:48:25 2005 From: mousumi at jncasr.ac.in (Mousumi Upadhyay Kahaly) Date: Fri, 22 Jul 2005 00:18:25 +0530 (IST) Subject: [Pw_forum] problem getting electronuc dipole moment Message-ID: <57252.202.41.111.151.1121971705.squirrel@202.41.111.151> Dear All, I am trying to compute electronic and ionic dipole moments of an isolated H2 molecule in a big box. For this, after relaxation, I am running one scf step followed by a pp.x run and then chdens.x run(with the option idpol = 1). The problem is that, irrespective of the positions of the H2 molecule in the big box, always it comes that charge density inside the Wigner-Seitz cell = 0.00000000 & hence, electronic dipole moment is sero. Have I done something wrong may be, while setting up the input files??? Because this is the case for any molecule in any position within a box of any size. Can anyone please help me to get rid of this problem? All the inputs files i am using(one case: when the molecule centre is in the centre of the box) are written below. Best regards, mousumi. ========================================================================== 1. Input file for scf run: pw.x input file &CONTROL calculation = 'scf' , tstress=.true., restart_mode = 'from_scratch' , outdir = '/export/mousumi/PWSCF_2.0.1/H2/PostProcess/dipole_mom/tmp2//' , pseudo_dir = '/export/mousumi/PWscf_GUI/pseudo/' , prefix = 'H2' , / &SYSTEM ibrav = 1, celldm(1) = 30.00000, nat = 2, ntyp = 1, ecutwfc = 30.00000 , ecutrho = 180.00000 , nosym = .false. , nbnd = 5, nspin = 1 , lda_plus_u = .false. , / &ELECTRONS conv_thr = 1.0d-8 , / &IONS upscale = 10.0 , / ATOMIC_SPECIES H 1.00000 H.pbe-van_bm.UPF ATOMIC_POSITIONS {angstrom} H 7.70055287150 7.70055287150 7.70055287150 H 8.15472682850 8.15472682850 8.15472682850 K_POINTS {automatic} 1 1 1 0 0 0 mixing_beta = 0.6 , ===================================================================== 2. With the files created in temp2/ directory from the above run, input file for pp.x &inputpp prefix = 'H2' outdir = '/export/mousumi/PWSCF_2.0.1/H2/PostProcess/dipole_mom/tmp2//' filplot = 'H2_chdens2' plot_num = 0 / ======================================================================= 3. With the 'H2_chdens2' file created in the above run, file for running chdens.x &input nfile = 1 filepp(1)='H2_chdens2' iflag = 3 plot_out = 1 idpol = 1 output_format = 5 / ========================================================================= From axel.kohlmeyer at theochem.ruhr-uni-bochum.de Thu Jul 21 21:17:27 2005 From: axel.kohlmeyer at theochem.ruhr-uni-bochum.de (Axel Kohlmeyer) Date: Thu, 21 Jul 2005 21:17:27 +0200 (CEST) Subject: [Pw_forum] problem getting electronuc dipole moment In-Reply-To: <57252.202.41.111.151.1121971705.squirrel@202.41.111.151> Message-ID: On Fri, 22 Jul 2005, Mousumi Upadhyay Kahaly wrote: MK> Dear All, MK> MK> I am trying to compute electronic and ionic dipole moments of an MK> isolated H2 molecule in a big box. For this, after relaxation, I MK> am running one scf step followed by a pp.x run and then chdens.x MK> run(with the option idpol = 1). MK> MK> The problem is that, irrespective of the positions of the H2 MK> molecule in the big box, always it comes that charge density MK> inside the Wigner-Seitz cell = 0.00000000 & hence, electronic MK> dipole moment is sero. Have I done something wrong may be, while i may be missing something here, but the h2 molecule has no dipole moment (unless you polarize it). axel. MK> setting up the input files??? Because this is the case for any MK> molecule in any position within a box of any size. Can anyone MK> please help me to get rid of this problem? MK> MK> All the inputs files i am using(one case: when the molecule MK> centre is in the centre of the box) are written below. MK> MK> Best regards, mousumi. MK> ========================================================================== MK> 1. Input file for scf run: pw.x input file MK> MK> &CONTROL MK> calculation = 'scf' , MK> tstress=.true., MK> restart_mode = 'from_scratch' , MK> outdir = MK> '/export/mousumi/PWSCF_2.0.1/H2/PostProcess/dipole_mom/tmp2//' , MK> pseudo_dir = '/export/mousumi/PWscf_GUI/pseudo/' , MK> prefix = 'H2' , MK> / MK> &SYSTEM MK> ibrav = 1, MK> celldm(1) = 30.00000, MK> nat = 2, MK> ntyp = 1, MK> ecutwfc = 30.00000 , MK> ecutrho = 180.00000 , MK> nosym = .false. , MK> nbnd = 5, MK> nspin = 1 , MK> lda_plus_u = .false. , MK> / MK> &ELECTRONS MK> conv_thr = 1.0d-8 , MK> / MK> &IONS MK> upscale = 10.0 , MK> / MK> ATOMIC_SPECIES MK> H 1.00000 H.pbe-van_bm.UPF MK> ATOMIC_POSITIONS {angstrom} MK> H 7.70055287150 7.70055287150 7.70055287150 MK> H 8.15472682850 8.15472682850 8.15472682850 MK> K_POINTS {automatic} MK> 1 1 1 0 0 0 MK> mixing_beta = 0.6 , MK> MK> ===================================================================== MK> 2. With the files created in temp2/ directory from the above run, MK> input file for pp.x MK> &inputpp MK> prefix = 'H2' MK> outdir = MK> '/export/mousumi/PWSCF_2.0.1/H2/PostProcess/dipole_mom/tmp2//' MK> filplot = 'H2_chdens2' MK> plot_num = 0 MK> / MK> MK> ======================================================================= MK> 3. With the 'H2_chdens2' file created in the above run, file for MK> running chdens.x MK> &input MK> nfile = 1 MK> filepp(1)='H2_chdens2' MK> iflag = 3 MK> plot_out = 1 MK> idpol = 1 MK> output_format = 5 MK> / MK> MK> ========================================================================= MK> MK> MK> _______________________________________________ MK> Pw_forum mailing list MK> Pw_forum at pwscf.org MK> http://www.democritos.it/mailman/listinfo/pw_forum MK> MK> -- ======================================================================= Dr. Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/ ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From burde1jb at cmich.edu Thu Jul 21 22:49:20 2005 From: burde1jb at cmich.edu (John B Burdette) Date: Thu, 21 Jul 2005 16:49:20 -0400 Subject: [Pw_forum] Errors in PW.x calculation Message-ID: <1194ed35.1f83ee83.81d5a00@student-b.csv.cmich.edu> Dear Users, I am performing a pw.x calculation for potassium oxide. When I am running the pw.x program for certain values of celldm(1), since ibrav=2, I am receiving the following error: iteration # 2 ecut= 70.00 ryd beta=0.70 Conjugate-gradient style diagonalization %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%% from cdiaghg : error # 12 info =/= 0 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%% But only for celldm(1) < 5.00. For values greater than 5.00, I receive a steep curve and not the parabola I expect. The energy curve drops steeply down as celldm(1) approaches < 5.00. Here is the input file for pw.x: K2O Potassium Oxide &control calculation = 'relax' restart_mode='from_scratch', prefix='Potassium Oxide', tstress = .true., tprnfor = .true., pseudo_dir = './', outdir='./', / &system ibrav=2, celldm(1) = 7.000, nat=3, ntyp= 2, ecutwfc =70.0, / &electrons diagonalization='cg' mixing_mode = 'plain' mixing_beta = 0.7 conv_thr = 1.0d-8 / &ions ion_dynamics = 'bfgs' / ATOMIC_SPECIES K 39.0983 K+.US.Burdette.UPF O 15.9994 O.US.LDA.RRKJ3.UPF ATOMIC_POSITIONS crystal O 0.00 0.00 0.00 K 0.25 0.25 0.25 K -.25 -.25 -.25 K_POINTS automatic 4 4 4 1 1 1 The pseudopotentials are of neutral Oxygen and K with one extra electron. Each was used in other pw.x experiments and worked fine. This is why I am so confused. Hopefully, one or more of you can help me. Best regards, John Burdette From mousumi at jncasr.ac.in Fri Jul 22 08:24:12 2005 From: mousumi at jncasr.ac.in (Mousumi Upadhyay Kahaly) Date: Fri, 22 Jul 2005 11:54:12 +0530 (IST) Subject: [Pw_forum] problem getting electronuc dipole moment In-Reply-To: References: <57252.202.41.111.151.1121971705.squirrel@202.41.111.151> Message-ID: <58750.202.41.111.151.1122013452.squirrel@202.41.111.151> Dear Sir, 1. I was actually trying to do these calculations for different di-acetates molecules, where I am expecting electronic dipole moment. As it failed, I was trying to test the cause with H2 molecule. And, as I told, this is the case for any molecule in any position within a box of any size. So, I want to know if any mistake is there in my input files and any extra flag is to be set for getting correct result. 2. I have even polarised the H2 molecule and found similar result(charge density inside the WS cell = 0.00000000). By changing the position of the H2 molecule in the box, I wanted to see How WS cell is contructed and whether any charge come within the WS cell. How is this possible that always the same 0.0000000 comes? Sir, will you please see if I made something wrong(or may be I need to add some extra lines) in the input files? Best regards, mousumi. > On Fri, 22 Jul 2005, Mousumi Upadhyay Kahaly wrote: > > MK> Dear All, > MK> > MK> I am trying to compute electronic and ionic dipole moments of an > MK> isolated H2 molecule in a big box. For this, after relaxation, I MK> am running one scf step followed by a pp.x run and then chdens.x MK> run(with the option idpol = 1). > > MK> > MK> The problem is that, irrespective of the positions of the H2 > MK> molecule in the big box, always it comes that charge density MK> inside the Wigner-Seitz cell = 0.00000000 & hence, electronic MK> dipole moment is sero. Have I done something wrong may be, while > > i may be missing something here, but the h2 molecule has > no dipole moment (unless you polarize it). > > axel. > > MK> setting up the input files??? Because this is the case for any MK> molecule in any position within a box of any size. Can anyone MK> please help me to get rid of this problem? > MK> > MK> All the inputs files i am using(one case: when the molecule > MK> centre is in the centre of the box) are written below. > MK> > MK> Best regards, mousumi. MK> > ========================================================================== MK> 1. Input file for scf run: pw.x input file > MK> > MK> &CONTROL > MK> calculation = 'scf' , > MK> tstress=.true., > MK> restart_mode = 'from_scratch' , > MK> outdir = > MK> '/export/mousumi/PWSCF_2.0.1/H2/PostProcess/dipole_mom/tmp2//' , MK> pseudo_dir = '/export/mousumi/PWscf_GUI/pseudo/' , MK> prefix = 'H2' , > MK> / > MK> &SYSTEM > MK> ibrav = 1, > MK> celldm(1) = 30.00000, > MK> nat = 2, > MK> ntyp = 1, > MK> ecutwfc = 30.00000 , > MK> ecutrho = 180.00000 , > MK> nosym = .false. , > MK> nbnd = 5, > MK> nspin = 1 , > MK> lda_plus_u = .false. , > MK> / > MK> &ELECTRONS > MK> conv_thr = 1.0d-8 , > MK> / > MK> &IONS > MK> upscale = 10.0 , > MK> / > MK> ATOMIC_SPECIES > MK> H 1.00000 H.pbe-van_bm.UPF > MK> ATOMIC_POSITIONS {angstrom} > MK> H 7.70055287150 7.70055287150 7.70055287150 > MK> H 8.15472682850 8.15472682850 8.15472682850 > MK> K_POINTS {automatic} > MK> 1 1 1 0 0 0 > MK> mixing_beta = 0.6 , > MK> > MK> ===================================================================== > MK> 2. With the files created in temp2/ directory from the above run, MK> input file for pp.x > MK> &inputpp > MK> prefix = 'H2' > MK> outdir = > MK> '/export/mousumi/PWSCF_2.0.1/H2/PostProcess/dipole_mom/tmp2//' MK> filplot = 'H2_chdens2' > MK> plot_num = 0 > MK> / > MK> > MK> > ======================================================================= MK> 3. With the 'H2_chdens2' file created in the above run, file for MK> running chdens.x > MK> &input > MK> nfile = 1 > MK> filepp(1)='H2_chdens2' > MK> iflag = 3 > MK> plot_out = 1 > MK> idpol = 1 > MK> output_format = 5 > MK> / > MK> > MK> > ========================================================================= MK> > MK> > MK> _______________________________________________ > MK> Pw_forum mailing list > MK> Pw_forum at pwscf.org > MK> http://www.democritos.it/mailman/listinfo/pw_forum > MK> > MK> > > -- > > ======================================================================= Dr. Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/ ======================================================================= 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 > From giannozz at nest.sns.it Fri Jul 22 10:11:09 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 22 Jul 2005 10:11:09 +0200 Subject: [Pw_forum] Errors in PW.x calculation In-Reply-To: <1194ed35.1f83ee83.81d5a00@student-b.csv.cmich.edu> References: <1194ed35.1f83ee83.81d5a00@student-b.csv.cmich.edu> Message-ID: <200507221011.09579.giannozz@nest.sns.it> On Thursday 21 July 2005 22:49, John B Burdette wrote: > When I am running the pw.x program for certain values of > celldm(1), since ibrav=2, I am receiving the following error: > [...] > from cdiaghg : error # 12 > info =/= 0 > [...] > But only for celldm(1) < 5.00 i.e. d(K-O) < sqrt(3)*5.00/2*0.529177 = 2.3 A. Note that this distance should be larger than the sum of the "core radii" of the pseudopotentials. Is it? > The pseudopotentials are of neutral Oxygen and K with one > extra electron strange choice: K- might be not bound with LDA or GGA, like most negative ions 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 dalcorso at sissa.it Fri Jul 22 15:13:16 2005 From: dalcorso at sissa.it (Andrea Dal Corso) Date: Fri, 22 Jul 2005 15:13:16 +0200 Subject: [Pw_forum] problem getting electronuc dipole moment In-Reply-To: <57252.202.41.111.151.1121971705.squirrel@202.41.111.151> References: <57252.202.41.111.151.1121971705.squirrel@202.41.111.151> Message-ID: <1122037997.3735.8.camel@dhpc-5-33.sissa.it> On Fri, 2005-07-22 at 00:18 +0530, Mousumi Upadhyay Kahaly wrote: > Dear All, > > I am trying to compute electronic and ionic dipole moments of an > isolated H2 molecule in a big box. For this, after relaxation, I > am running one scf step followed by a pp.x run and then chdens.x > run(with the option idpol = 1). > > The problem is that, irrespective of the positions of the H2 > molecule in the big box, always it comes that charge density > inside the Wigner-Seitz cell = 0.00000000 & hence, electronic > dipole moment is sero. Have I done something wrong may be, while > setting up the input files??? Because this is the case for any > molecule in any position within a box of any size. Can anyone > please help me to get rid of this problem? > > All the inputs files i am using(one case: when the molecule > centre is in the centre of the box) are written below. > > Best regards, mousumi. > ========================================================================== > 1. Input file for scf run: pw.x input file > > &CONTROL > calculation = 'scf' , > tstress=.true., > restart_mode = 'from_scratch' , > outdir = > '/export/mousumi/PWSCF_2.0.1/H2/PostProcess/dipole_mom/tmp2//' , > pseudo_dir = '/export/mousumi/PWscf_GUI/pseudo/' , > prefix = 'H2' , > / > &SYSTEM > ibrav = 1, > celldm(1) = 30.00000, > nat = 2, > ntyp = 1, > ecutwfc = 30.00000 , > ecutrho = 180.00000 , > nosym = .false. , > nbnd = 5, > nspin = 1 , > lda_plus_u = .false. , > / > &ELECTRONS > conv_thr = 1.0d-8 , > / > &IONS > upscale = 10.0 , > / > ATOMIC_SPECIES > H 1.00000 H.pbe-van_bm.UPF > ATOMIC_POSITIONS {angstrom} > H 7.70055287150 7.70055287150 7.70055287150 > H 8.15472682850 8.15472682850 8.15472682850 > K_POINTS {automatic} > 1 1 1 0 0 0 > mixing_beta = 0.6 , > > ===================================================================== > 2. With the files created in temp2/ directory from the above run, > input file for pp.x > &inputpp > prefix = 'H2' > outdir = > '/export/mousumi/PWSCF_2.0.1/H2/PostProcess/dipole_mom/tmp2//' > filplot = 'H2_chdens2' > plot_num = 0 > / > > ======================================================================= > 3. With the 'H2_chdens2' file created in the above run, file for > running chdens.x > &input > nfile = 1 > filepp(1)='H2_chdens2' > iflag = 3 > plot_out = 1 > idpol = 1 > output_format = 5 > / > The dipole moment is not calculated with all values of output_format. A warning message appears in the output if the dipole is not calculated. Best Regards, Andrea > ========================================================================= > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- Andrea Dal Corso Tel. 0039-040-3787428 SISSA, Via Beirut 2/4 Fax. 0039-040-3787528 34014 Trieste (Italy) e-mail: dalcorso at sissa.it From proffess at yandex.ru Fri Jul 22 15:42:12 2005 From: proffess at yandex.ru (Sergey Lisenkov) Date: Fri, 22 Jul 2005 17:42:12 +0400 (MSD) Subject: [Pw_forum] Pt pseudopotential from atomic code In-Reply-To: <1121858043.3755.16.camel@dhpc-5-33.sissa.it> References: <42DDF1A5.000009.07784@pantene.yandex.ru> <1121858043.3755.16.camel@dhpc-5-33.sissa.it> Message-ID: <42E0F7B4.000002.12751@colgate.yandex.ru> Dear Andrea and other Espresso authors and users, I generated a Pt PsP using the first configuration (the second one was failed with error 'strange occupations'). After that I tried to test this pseudopotential, but got the next error: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from read_pseudo_header : error # 1 unknown pseudo type %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... STOP 2 The header of PsP is printed: Generated using Andrea Dal Corso code (rrkj3) Author: Andrea Dal Corso Generation date: unknown Info:Pt 1 The Pseudo was generated with a Scalar-Relativistic Calculation 2.60000000000E+00 Local Potential cutoff radius nl pn l occ Rcut Rcut US E pseu 5D 3 2 8.00 2.10000000000 2.40000000000 0.00000000000 5D 3 2 0.00 2.10000000000 2.40000000000 0.00000000000 6P 2 1 0.00 3.30000000000 3.30000000000 0.00000000000 6S 1 0 2.00 2.60000000000 2.60000000000 0.00000000000 0 Version Number Element US Ultrasoft pseudopotential T Nonlinear Core Correction SLA PW PBE PBE PBE Exchange-Correlation functional 10.00000000000 Z valence -52.93790096200 Total energy 0.0000000 0.0000000 Suggested cutoff for wfc and rho 2 Max angular momentum component 1277 Number of points in mesh 2 3 Number of Wavefunctions, Number of Projectors Wavefunctions nl l occ 5D 2 8.00 6S 0 2.00 .... What is a problem? Thanks, Sergey From giannozz at nest.sns.it Fri Jul 22 15:54:15 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 22 Jul 2005 15:54:15 +0200 Subject: [Pw_forum] Pt pseudopotential from atomic code In-Reply-To: <42E0F7B4.000002.12751@colgate.yandex.ru> References: <42DDF1A5.000009.07784@pantene.yandex.ru> <1121858043.3755.16.camel@dhpc-5-33.sissa.it> <42E0F7B4.000002.12751@colgate.yandex.ru> Message-ID: <200507221554.15654.giannozz@nest.sns.it> On Friday 22 July 2005 15:42, Sergey Lisenkov wrote: > [...] > I tried to test this pseudopotential, but got the next error: > [...] > from read_pseudo_header : error # 1 > unknown pseudo type > [...] > The header of PsP is printed: > [...] > > 0 Version Number > Element > US Ultrasoft pseudopotential Element ? nothing ! read the following line, get obscure error ... fixed in the cvs version. In the meantime, provide the atom name when you generate a PP (or simply add it to the file) P. -- 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 proffess at yandex.ru Fri Jul 22 16:04:50 2005 From: proffess at yandex.ru (Sergey Lisenkov) Date: Fri, 22 Jul 2005 18:04:50 +0400 (MSD) Subject: [Pw_forum] Pt pseudopotential from atomic code In-Reply-To: <200507221554.15654.giannozz@nest.sns.it> References: <42DDF1A5.000009.07784@pantene.yandex.ru> <1121858043.3755.16.camel@dhpc-5-33.sissa.it> <42E0F7B4.000002.12751@colgate.yandex.ru> <200507221554.15654.giannozz@nest.sns.it> Message-ID: <42E0FD02.000001.30424@ariel.yandex.ru> Dear Paolo, Thank you very much for the solution! Sergey From burde1jb at cmich.edu Fri Jul 22 16:06:04 2005 From: burde1jb at cmich.edu (John B Burdette) Date: Fri, 22 Jul 2005 10:06:04 -0400 Subject: [Pw_forum] Errors in PW.x calculation Message-ID: <8d1656c4.1fe2d805.81dcb00@student-b.csv.cmich.edu> i.e. d(K-O) < sqrt(3)*5.00/2*0.529177 = 2.3 A. Note that this distance should be larger than the sum of the "core radii" of the pseudopotentials. Is it? The sum of the "core radii" equals 0.237 A. Also, by K with one extra electron, I meant K with one less electron. My mistake on that one. Even so, any idea why I might not be getting a parabola for the energy curve? I can send you the pseudopotentials Paolo if you so wish. Best regards, John Burdette From dalcorso at sissa.it Fri Jul 22 16:13:02 2005 From: dalcorso at sissa.it (Andrea Dal Corso) Date: Fri, 22 Jul 2005 16:13:02 +0200 Subject: [Pw_forum] Pt pseudopotential from atomic code In-Reply-To: <42E0F7B4.000002.12751@colgate.yandex.ru> References: <42DDF1A5.000009.07784@pantene.yandex.ru> <1121858043.3755.16.camel@dhpc-5-33.sissa.it> <42E0F7B4.000002.12751@colgate.yandex.ru> Message-ID: <1122041582.3735.18.camel@dhpc-5-33.sissa.it> On Fri, 2005-07-22 at 17:42 +0400, Sergey Lisenkov wrote: > Dear Andrea and other Espresso authors and users, > > I generated a Pt PsP using the first configuration (the second one was failed with error 'strange occupations'). Did you changed also the configuration in the all electron calculation? With the second configuration you should also put config='[Xe] 4f14. 5d9 6s1 6p0' in the input namelist. Best regards, Andrea > After that I tried to test this pseudopotential, but got the next error: > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > from read_pseudo_header : error # 1 > unknown pseudo type > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > stopping ... > STOP 2 > > The header of PsP is printed: > > > Generated using Andrea Dal Corso code (rrkj3) > Author: Andrea Dal Corso Generation date: unknown > Info:Pt > 1 The Pseudo was generated with a Scalar-Relativistic Calculation > 2.60000000000E+00 Local Potential cutoff radius > nl pn l occ Rcut Rcut US E pseu > 5D 3 2 8.00 2.10000000000 2.40000000000 0.00000000000 > 5D 3 2 0.00 2.10000000000 2.40000000000 0.00000000000 > 6P 2 1 0.00 3.30000000000 3.30000000000 0.00000000000 > 6S 1 0 2.00 2.60000000000 2.60000000000 0.00000000000 > > > > > 0 Version Number > Element > US Ultrasoft pseudopotential > T Nonlinear Core Correction > SLA PW PBE PBE PBE Exchange-Correlation functional > 10.00000000000 Z valence > -52.93790096200 Total energy > 0.0000000 0.0000000 Suggested cutoff for wfc and rho > 2 Max angular momentum component > 1277 Number of points in mesh > 2 3 Number of Wavefunctions, Number of Projectors > Wavefunctions nl l occ > 5D 2 8.00 > 6S 0 2.00 > > > > > .... > > > What is a problem? > > Thanks, > Sergey > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- Andrea Dal Corso Tel. 0039-040-3787428 SISSA, Via Beirut 2/4 Fax. 0039-040-3787528 34014 Trieste (Italy) e-mail: dalcorso at sissa.it From mousumi at jncasr.ac.in Fri Jul 22 20:40:46 2005 From: mousumi at jncasr.ac.in (Mousumi Upadhyay Kahaly) Date: Sat, 23 Jul 2005 00:10:46 +0530 (IST) Subject: [Pw_forum] problem getting electronuc dipole moment In-Reply-To: <1122037997.3735.8.camel@dhpc-5-33.sissa.it> References: <57252.202.41.111.151.1121971705.squirrel@202.41.111.151> <1122037997.3735.8.camel@dhpc-5-33.sissa.it> Message-ID: <41935.202.41.111.151.1122057646.squirrel@202.41.111.151> Dear Sir, >> running chdens.x >> &input >> nfile = 1 >> filepp(1)='H2_chdens2' >> iflag = 3 >> plot_out = 1 >> idpol = 1 >> output_format = 5 >> / > > > The dipole moment is not calculated with all values of output_format. A warning message appears in the output if the dipole is not calculated. Thank you for your suggestion. With output_format 4, for H2 molecule system, it works. When I used the output_format 5, then no warning msg was printed. But with option 3, the warning msg is printed. In my case, possible other output format option is output_format 4. For Cd-di-acetate, using "output_format = 4" I get the warning msg as ======================================================================== Writing coordinates on file coord.xyz Requested parallelepiped sides : 1.0000 1.0000 1.0000 Redefined parallelepiped sides : 1.0000 1.0000 1.0000 Requested parallelepiped origin: 0.0000 0.0000 0.0000 Redefined parallelepiped origin: 0.0000 0.0000 0.0000 Warning: the box is too small to calculate dipole ========================================================================= Then anyway, it prints out some electronic and ionic dipole moments. My questions are- 1. With the above warning msg, does the code take into account the whole charge densities of the system and prints the electronic dipole moment correctly? 2. Why for the output format 5, neither any warning nor correct electronic dipole moment is shown? Best regards, mousumi From jasonsun98 at hotmail.com Sat Jul 23 17:07:28 2005 From: jasonsun98 at hotmail.com (sun jason) Date: Sat, 23 Jul 2005 15:07:28 +0000 Subject: [Pw_forum] about energy gap In-Reply-To: <20050708053634.10888.36235.Mailman@democritos.sissa.it> Message-ID: Dear all, I send it again for more discussions. Much appreciation to Mr. Stefano de Gironcoli for your kind answers! but I still have some doubts. 1) we mentioned that LDA underestimate band gaps, the experimental gaps for c-BN and diamond are 6.4 eV and 5.6 eV, and the LDA results are about 4.5 and 4.1, respectively. so why the LDA gap of B-C-N system is not in the range of 4.1~4.5 eV? 2)N doping may couse gap reduction, but should I looks my B-C-N system as N doping in boron carbide? doping may couse impurity (donor or acceptor), but my systems are perfect single crystals.I have not consider the amorphous systems they are in the case of doping? 3)as above one, are there "disorders" in perfect single crystals( in computational model)? 4)yes, the top of the valence band or the bottom of conduction band is a little localized, but very small, within 0.5 eV. the gap is still small compared with the parents materials even we add the lineup value. Thank you very much! ==================================================== Message: 2 Date: Thu, 07 Jul 2005 11:27:01 +0200 From: Stefano de Gironcoli To: pw_forum at pwscf.org Subject: Re: [Pw_forum] about energy gap Reply-To: pw_forum at pwscf.org 1) LDA strongly underestimate band gaps how large are your gaps for the parents materials compared to literature values (Exp and Theory) 2) N doping is known to produce large reduction (giant negative bowing) in GaAs. A few % of N actually reduces the gap. 3) disorder tends to close the gaps 4) are your valence-band top and conduction -band bottom states localized on different atoms species. is your cell a kind of nano-heterojunction ? in that case you might have some band-lineup effects that could reduce the gap Dig better in the literature and in your calculations. Compare with known results in similar systems. best regards, Stefano de Gironcoli sun jason wrote: > Dear all, > > I'm a freshman in computational materials, I'm puzzled by a simple > question about energy gap, when I calculated the C2-BN system which > was construtured by mixing cubic BN and diamond. > Well, the computed LDA band gap of c-BN and diamond is around 4.5 and > 4.1 eV, respectively. Therefore, by mixing two compounds with such a > wide band gap (roughly of the order of 4 > eV) one might expect to get a BC2N compound with an Eg between 4.5 and > 4.1eV. Hence, why is our calculated Eg (about 1.0 to 2.0 eV) is much > lower than expected for a C2-BN system? > the unit cell I calculated is crystallined in diamond-like structure > and the atoms are sp3 hybridization. > > Thank you very much! From astroppa at ts.infn.it Sun Jul 24 15:31:08 2005 From: astroppa at ts.infn.it (Alessandro Stroppa) Date: Sun, 24 Jul 2005 15:31:08 +0200 (CEST) Subject: [Pw_forum] Diropn Message-ID: Dear Pwscf users, I had the following error message while running a LDA+U calculation, with Espresso.2.1.4: task # 6 from diropn : error # 0 can't open a connected unit Has anybody faced this problem? A. From zucco at dipteris.unige.it Mon Jul 25 08:49:34 2005 From: zucco at dipteris.unige.it (Marino Vetuschi Zuccolini) Date: Mon, 25 Jul 2005 08:49:34 +0200 Subject: [Pw_forum] non-diagonalized stress on Chromium spinel Message-ID: Hello to all, I need to approach a study on chromium-spinel mineral and I would like to use a planewave code as pwscf. My experiences was only with localized atomic basis set and then the change of "language" is for me a source of questions. My main question about the results of a run of the reported input file is: - how can I detect the possible sources of errors or simply a bad description of system in the case as is, if the stress on the primitive cell is not diagonalized (that for me is a sign of structural unstability)? This is the last one input file. In the previous I changed diagonalization, smearing, broadening, ecut*, number of k-points but nothing really has changed. Then vibrations are completely wrong.. Many thanks to all for any help, marino ************************************************************************ **** &CONTROL title = 'Chromium spinel ICSD #75623' , calculation = 'scf' , restart_mode = 'restart' , outdir = '/tmp/' , wf_collect = .true., pseudo_dir = '/apps/espresso-2.1.3/pseudo/', prefix = 'MgCrspinel' , verbosity = 'high' , disk_io = 'minimal', tstress = .true., / &SYSTEM ibrav = 1, celldm(1) = 15.75, nat = 3, ntyp = 3, ecutwfc = 35 , ecutrho = 300 , nbnd = 20 , nspin = 2, starting_magnetization(2) = 0.2, occupations = 'smearing' , degauss = 0.05 , smearing = 'm-v' / &ELECTRONS electron_maxstep = 200, mixing_beta = 0.1, diagonalization = 'diis', mixing_ndim = 3 / ATOMIC_SPECIES Mg 24.30500 Mg.pw91-np-van.UPF Cr 51.99610 Cr.pw91-sp-van.UPF O 15.99940 O.pw91-van_ak.UPF ATOMIC_POSITIONS crystal Mg 0.125 0.125 0.125 Cr 0.500 0.500 0.500 O 0.2615 0.2615 0.2615 K_POINTS automatic 4 4 4 1 1 1 ************************** output ********************************* ... npt with |zeta| > 1: 91168, npt tot 729000, 12.51 % npt with rhoup < 0: 14443, npt tot 729000, 1.98 % npt with rhodw < 0: 105467, npt tot 729000, 14.47 % total stress (ryd/bohr**3) (kbar) P= -10.24 -0.00006959 -0.00003860 -0.00003860 -10.24 -5.68 -5.68 -0.00003860 -0.00006959 -0.00003860 -5.68 -10.24 -5.68 -0.00003860 -0.00003860 -0.00006959 -5.68 -5.68 -10.24 kinetic stress (kbar) 1319.27 -14.50 -14.50 -14.50 1319.27 -14.50 -14.50 -14.50 1319.27 local stress (kbar) -3106.47 -1019.77 -1019.77 -1019.77 -3106.47 -1019.77 -1019.77 -1019.77 -3106.47 nonloc. stress (kbar) 663.92 1.49 1.49 1.49 663.92 1.49 1.49 1.49 663.92 hartree stress (kbar) 2283.49 458.71 458.71 458.71 2283.49 458.71 458.71 458.71 2283.49 exc-cor stress (kbar) -362.41 0.38 0.38 0.38 -362.41 0.38 0.38 0.38 -362.41 corecor stress (kbar) -114.70 -0.09 -0.09 -0.09 -114.70 -0.09 -0.09 -0.09 -114.70 ewald stress (kbar) -693.32 568.11 568.11 568.11 -693.32 568.11 568.11 568.11 -693.32 hubbard stress (kbar) 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Writing file MgCrspinel.sav for program phonon PWSCF : 2h 4m CPU time ******************************************************* Marino Vetuschi Zuccolini zucco at dipteris.unige.it PhD / Geochemist Laboratory of Geochemistry http://ugo.dipteris.unige.it/pesto Visit our BAXEICO computing cluster homepage http://qed.dipteris.unige.it/ganglia/?c=Baxeico%20Cluster DIPartimento per lo studio della TErra e delle sue RISorse - Universit? di Genova Tel. ++39 010 3538136 Fax. ++39 010 352169 Corso Europa 26, 16132 - Genova - Italy -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/enriched Size: 4966 bytes Desc: not available Url : /pipermail/attachments/20050725/3eca0473/attachment.bin From giannozz at nest.sns.it Mon Jul 25 09:16:22 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 25 Jul 2005 09:16:22 +0200 Subject: [Pw_forum] non-diagonalized stress on Chromium spinel In-Reply-To: References: Message-ID: <200507250916.22199.giannozz@nest.sns.it> On Monday 25 July 2005 08:49, Marino Vetuschi Zuccolini wrote: > This is the last one input file. In the previous I changed > diagonalization, smearing, broadening, ecut*, number of > k-points but nothing really has changed. > Then vibrations are completely wrong.. you mean that you calculated phonons and they are wrong, or what else? I don't see anything wrong in your results. The stress you get, both diagonal and non-diagonal, is really very small Poolo -- 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 Jul 25 09:18:49 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 25 Jul 2005 09:18:49 +0200 Subject: [Pw_forum] Diropn In-Reply-To: References: Message-ID: <200507250918.50033.giannozz@nest.sns.it> On Sunday 24 July 2005 15:31, Alessandro Stroppa wrote: > I had the following error message while running a LDA+U > calculation, with Espresso.2.1.4: > task # 6 > from diropn : error # 0 > can't open a connected unit this refers to unit 0, which of course shouldn't be used to write anything to file. How do you manage to get this? 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 Jul 25 09:28:26 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 25 Jul 2005 09:28:26 +0200 Subject: [Pw_forum] Errors in PW.x calculation In-Reply-To: <8d1656c4.1fe2d805.81dcb00@student-b.csv.cmich.edu> References: <8d1656c4.1fe2d805.81dcb00@student-b.csv.cmich.edu> Message-ID: <200507250928.26252.giannozz@nest.sns.it> On Friday 22 July 2005 16:06, John B Burdette wrote: > The sum of the "core radii" equals 0.237 A. this is very unlikely, since O alone has rc=1.6 a.u=0.85A > Also, by K with one extra electron, I meant K with one less > electron. My mistake on that one. Even so, any idea why I > might not be getting a parabola for the energy curve? it is not easy to generate pseudopotentials for alkali atoms, especially for ionic compounds. If you have a one-electron PP, you need the nonlinear core correction for sure. If you generate 9-electron PP, beware ghosts. 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 Jul 25 11:48:54 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 25 Jul 2005 11:48:54 +0200 Subject: [Pw_forum] problem getting electronuc dipole moment In-Reply-To: <41935.202.41.111.151.1122057646.squirrel@202.41.111.151> References: <57252.202.41.111.151.1121971705.squirrel@202.41.111.151> <1122037997.3735.8.camel@dhpc-5-33.sissa.it> <41935.202.41.111.151.1122057646.squirrel@202.41.111.151> Message-ID: <200507251148.54052.giannozz@nest.sns.it> On Friday 22 July 2005 20:40, Mousumi Upadhyay Kahaly wrote: > 2. Why for the output format 5, neither any warning nor > correct electronic dipole moment is shown? because computers do only what they are instructed to do. Nobody had considered what to do in that particular case. 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 chaohao2002 at 163.com Mon Jul 25 11:59:53 2005 From: chaohao2002 at 163.com (=?gb2312?B?uvqzr7rG?=) Date: Mon, 25 Jul 2005 17:59:53 +0800 (CST) Subject: [Pw_forum] =?gb2312?B?YSBxdWVzdGlvbiBhYm91dCBwaG9ub24gY2FsY3VsYXRpb24=?= Message-ID: <42E4B819.00004A.19226@m0201.mail.163.com> Dear pwscf users, I want to ask you a question about phonon calculation.In my calculations, if I add lines described as below, I can get the phonon frequences at the wave vector q (0.5,0.5,0.5) ======= &phonon xqq(1) = 0.50, xqq(2) = 0.50, xqq(3) = 0.50 / ======= But I want to get the phonon disperson curve over entire phonon wave vectors q, how do I set these wave vectors in my input files? Please help me kindly. Thanks in advance. Best regards, Chaohao -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20050725/c125592f/attachment.htm From eyvaz_isaev at yahoo.com Mon Jul 25 12:56:36 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Mon, 25 Jul 2005 03:56:36 -0700 (PDT) Subject: [Pw_forum] a question about phonon calculation In-Reply-To: <42E4B819.00004A.19226@m0201.mail.163.com> Message-ID: <20050725105636.89689.qmail@web60313.mail.yahoo.com> Hi, To do it first you have to calculate force constants for a number if q-points, and then transform C(q), force constants in the reciprocal space, to C(R) using q2r.x. Finally, using these real-space force constants, C(R), you can calculate \omega(q) for a given q-point using matdyn.x by means of the inverse Fourier transformation C(q)=\int C(R)exp(-iqR). These q-vectors could be defined manually or you can wait some time. I am going to include my small programs to PWSCF distribution. Please have a look also at example06. Best regards, Eyvaz. --- ?????? wrote: > Dear pwscf users, > I want to ask you a question about phonon > calculation.In my calculations, if I add lines > described as below, I can get the phonon frequences > at the wave vector q (0.5,0.5,0.5) > ======= > &phonon > xqq(1) = 0.50, xqq(2) = 0.50, xqq(3) = 0.50 > / > ======= > But I want to get the phonon disperson curve over > entire phonon wave vectors q, how do I set these > wave vectors in my input files? Please help me > kindly. > Thanks in advance. > > Best regards, > Chaohao __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From konstantin_kudin at yahoo.com Mon Jul 25 18:35:31 2005 From: konstantin_kudin at yahoo.com (Konstantin Kudin) Date: Mon, 25 Jul 2005 09:35:31 -0700 (PDT) Subject: [Pw_forum] CP & the cpu time limit Message-ID: <20050725163532.24210.qmail@web52007.mail.yahoo.com> Hi all, While running the CP code from the CVS I encountered the following: Maximum CPU time exceeded max_seconds = 1000000.00 elapsed seconds = 1000010.00 And the job killed itself ... I recomputed these seconds into days, it comes out to be 11.6 days. Our queue allows 15 days, so the code was not constrained by that. Is there any reason for this 1000000.00 second limit? Kostya ____________________________________________________ Start your day with Yahoo! - make it your home page http://www.yahoo.com/r/hs From giannozz at nest.sns.it Mon Jul 25 18:56:07 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 25 Jul 2005 18:56:07 +0200 Subject: [Pw_forum] CP & the cpu time limit In-Reply-To: <20050725163532.24210.qmail@web52007.mail.yahoo.com> References: <20050725163532.24210.qmail@web52007.mail.yahoo.com> Message-ID: <200507251856.07963.giannozz@nest.sns.it> On Monday 25 July 2005 18:35, Konstantin Kudin wrote: > Is there any reason for this 1000000.00 second limit? it is the default value, which was supposed to be large enough to be equivalent to "no time limit", 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 chaohao2002 at 163.com Tue Jul 26 03:46:30 2005 From: chaohao2002 at 163.com (=?gb2312?B?uvqzr7rG?=) Date: Tue, 26 Jul 2005 09:46:30 +0800 (CST) Subject: [Pw_forum] =?gb2312?B?YSBxdWVzdGlvbiBhYm91dCB6ZXJvIHBvaW50IGVuZXJneQ==?= Message-ID: <42E595F6.0000A0.10598@m0201.mail.163.com> Dear PWscf users, In PWscf codes, how can I obtain the results of zero point energy by phonon calculation? Thanks in advance. Best regards, Chaohao -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20050726/90f13256/attachment.htm From baroni at sissa.it Tue Jul 26 04:29:54 2005 From: baroni at sissa.it (Stefano Baroni) Date: Mon, 25 Jul 2005 19:29:54 -0700 Subject: [Pw_forum] a question about zero point energy In-Reply-To: <42E595F6.0000A0.10598@m0201.mail.163.com> References: <42E595F6.0000A0.10598@m0201.mail.163.com> Message-ID: <23886D71-91EC-4C9D-BE6C-943125166BEF@sissa.it> As in textbooks: by calculating phonons all over the Brillouin zone (in practice, on a regular grid), and them by calculating {\hbar \over 2} \sum_{{\bf q},\nu} \omega_{{\bf q},\nu}. SB On Jul 25, 2005, at 6:46 PM, ??? wrote: > Dear PWscf users, > In PWscf codes, how can I obtain the results of zero point energy > by phonon calculation? > Thanks in advance. > > Best regards, > Chaohao > > > > > > ????2000???????? > ???????????????????? > --- 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/20050725/4c2e3fda/attachment.htm From nitya at iitk.ac.in Tue Jul 26 07:00:31 2005 From: nitya at iitk.ac.in (Nitya Nath) Date: Tue, 26 Jul 2005 10:30:31 +0530 (IST) Subject: [Pw_forum] Bands plot. Message-ID: Dear PWScf Users, I want to plot bulk bands of bcc Fe for a low symmetry line, like (1,0,0) to (1,0,1), which corresponds to a point on symmetry line in the 2D BZ. We also want to look odd and even symmetries with respect to reflections in the [010] plane. How can I do it using PwScf. Any help is welcome in this regard. Thanking you in advance. Best regards. Nitya From vdiep at ictp.it Tue Jul 26 16:46:46 2005 From: vdiep at ictp.it (Diep Vinh Quang) Date: Tue, 26 Jul 2005 16:46:46 +0200 (CEST) Subject: [Pw_forum] Simulated annealing In-Reply-To: <23886D71-91EC-4C9D-BE6C-943125166BEF@sissa.it> References: <42E595F6.0000A0.10598@m0201.mail.163.com> <23886D71-91EC-4C9D-BE6C-943125166BEF@sissa.it> Message-ID: <40896.10.41.38.19.1122389206.squirrel@webmail1.ictp.trieste.it> Dear pwscf users, Does anyone know how to include the simulated annealing technique into the pwscf code in order to find the best configuration of atoms in a unit cell (lowest energy)? My system is polyethylene and at very high pressure it has a lot of local minimums. Therefore we need to find the global one. Thank you very much in advance. Best regards Vinh From yonasb at yahoo.com Tue Jul 26 17:26:22 2005 From: yonasb at yahoo.com (Yonas Abraham) Date: Tue, 26 Jul 2005 08:26:22 -0700 (PDT) Subject: [Pw_forum] Simulated annealing In-Reply-To: <40896.10.41.38.19.1122389206.squirrel@webmail1.ictp.trieste.it> Message-ID: <20050726152622.60124.qmail@web33504.mail.mud.yahoo.com> One way is to use cp.x (Car-Parrinello MD) that comes with the espresso package. the simple step would be -minimize the wavefinction -minimize the geometry -heat the system -tern the temperature off and let the geometry relax But I don't know if there is such thing on the pw.x. /yonas --- Diep Vinh Quang wrote: > Dear pwscf users, > > Does anyone know how to include the simulated > annealing technique into the > pwscf code in order to find the best configuration > of atoms in a unit cell > (lowest energy)? > > My system is polyethylene and at very high pressure > it has a lot of local > minimums. Therefore we need to find the global one. > > Thank you very much in advance. > > Best regards > Vinh > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From giannozz at nest.sns.it Tue Jul 26 17:57:21 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 26 Jul 2005 17:57:21 +0200 Subject: [Pw_forum] Simulated annealing In-Reply-To: <20050726152622.60124.qmail@web33504.mail.mud.yahoo.com> References: <20050726152622.60124.qmail@web33504.mail.mud.yahoo.com> Message-ID: <200507261757.21209.giannozz@nest.sns.it> On Tuesday 26 July 2005 17:26, Yonas Abraham wrote: > One way is to use cp.x (Car-Parrinello MD) that comes > with the espresso package. The simple step would be > > -minimize the wavefunction > -minimize the geometry > -heat the system > -turn the temperature off and let the geometry relax > > But I don't know if there is such thing on the pw.x. just a clarification: simulated annealing is something distinct from Car-Parrinello dynamics. The above steps can be performed either with "Born-Oppenheimer" dynamics (i.e. with electrons always on the ground state) or with Car-Parrinello dynamics (i.e. with an additional dynamics on electrons that keeps them close to the groung state). The latter case is not implemented in PWscf: one has to use either CP or FPMD, both part of the quantum-espresso distribution. Also note that temperature control in the current release of PWscf is not very sophisticated. 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 baroni at sissa.it Tue Jul 26 18:02:38 2005 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 26 Jul 2005 09:02:38 -0700 Subject: [Pw_forum] Simulated annealing In-Reply-To: <200507261757.21209.giannozz@nest.sns.it> References: <20050726152622.60124.qmail@web33504.mail.mud.yahoo.com> <200507261757.21209.giannozz@nest.sns.it> Message-ID: Yet another bit of clarification: On Jul 26, 2005, at 8:57 AM, Paolo Giannozzi wrote: > On Tuesday 26 July 2005 17:26, Yonas Abraham wrote: > > >> One way is to use cp.x (Car-Parrinello MD) that comes >> with the espresso package. The simple step would be >> >> -minimize the wavefunction >> -minimize the geometry >> -heat the system >> -turn the temperature off and let the geometry relax This is not quite "simulated annealing", but rather "simulated quenching". Simulated annealing requires a real-time control of temperature (not just "on/off") which I do not know if it is available in either code, but which should be not so difficult to implement, should anyone be interested. SB >> >> But I don't know if there is such thing on the pw.x. >> > > just a clarification: simulated annealing is something distinct > from Car-Parrinello dynamics. The above steps can be performed > either with "Born-Oppenheimer" dynamics (i.e. with electrons > always on the ground state) or with Car-Parrinello dynamics > (i.e. with an additional dynamics on electrons that keeps them > close to the groung state). The latter case is not implemented > in PWscf: one has to use either CP or FPMD, both part of the > quantum-espresso distribution. > > Also note that temperature control in the current release of > PWscf is not very sophisticated. > > 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 > --- 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/20050726/fdddebf8/attachment.htm From yonasb at yahoo.com Tue Jul 26 19:39:38 2005 From: yonasb at yahoo.com (Yonas Abraham) Date: Tue, 26 Jul 2005 10:39:38 -0700 (PDT) Subject: [Pw_forum] Simulated annealing In-Reply-To: Message-ID: <20050726173938.72187.qmail@web33501.mail.mud.yahoo.com> --- Stefano Baroni ssissa wrote: > This is not quite "simulated annealing", but rather > "simulated > quenching". Simulated annealing requires a real-time > control of > temperature (not just "on/off") which I do not know > if it is > available in either code, but which should be not so > difficult to > implement, should anyone be interested. > > SB I think this could be a real use of the "steering" functionality of espresso. by steering i mean, change input parameters on the fly. I am pretty sure that I have used to change the temperature while the simulation is running for cp.x Yonas From chaohao2002 at 163.com Wed Jul 27 07:03:23 2005 From: chaohao2002 at 163.com (=?gb2312?B?Q2hhb2hhbyBIdQ==?=) Date: Wed, 27 Jul 2005 13:03:23 +0800 (CST) Subject: [Pw_forum] =?gb2312?B?YSBxdWVzdGlvbiBhYm91dCBjYWxjdWxhdGluZyBmcmVlIGVuZXJneQ==?= Message-ID: <42E7159B.0000C2.05788@m0211.mail.163.com> Dear PWscf users, I did some phonon calculations using pw.x and ph.x and obtained the results of phonon density of states (1.phdos). But when I calculated the free energy using fqha.x, some wrong informations always appear. In the fqha.f90 file, I deleted the following line "if (abs(nu(i)-de*(i-0.5d0)).gt.1.0d-6) stop ' wrong grid'" (I think that the limitative condition is too rigorous). the executing processing: fqha file containing the dos >>>1.phdos 1368 0.2048 1.0 file for the Free energy >>>1.txt temperature >>>500.0 the error information: "run-time error M6202: MATH - log: SING error" I also know the negative value appears in log function described as the following line, " if (T.gt.0.d0) Ftot=Ftot+dos(i)*a2*T*log(1.0d0-exp(-a3*nu(i)/T)) " but I do not know how to resolve this problem. Who can help me? Thanks in advance. Best regards, Chaohao -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20050727/20762092/attachment.htm From eyvaz_isaev at yahoo.com Wed Jul 27 07:37:36 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Tue, 26 Jul 2005 22:37:36 -0700 (PDT) Subject: [Pw_forum] a question about calculating free energy In-Reply-To: <42E7159B.0000C2.05788@m0211.mail.163.com> Message-ID: <20050727053736.52654.qmail@web60319.mail.yahoo.com> Hi, > the error information: > "run-time error M6202: MATH > - log: SING error" > > I also know the negative value appears in log > function described as the following line, > " if (T.gt.0.d0) > Ftot=Ftot+dos(i)*a2*T*log(1.0d0-exp(-a3*nu(i)/T)) > " > but I do not know how to resolve this problem. If you have imaginary (negative in your files) frequences, your system is dynamically unstable, so, it is meaningless to calculate the free energy for such a kind unstable system. Bests, Eyvaz. ____________________________________________________ Start your day with Yahoo! - make it your home page http://www.yahoo.com/r/hs From chaohao2002 at 163.com Wed Jul 27 08:14:48 2005 From: chaohao2002 at 163.com (=?gb2312?B?Q2hhb2hhbyBIdQ==?=) Date: Wed, 27 Jul 2005 14:14:48 +0800 (CST) Subject: [Pw_forum] =?gb2312?B?UmU6IFJlOiBbUHdfZm9ydW1dIGEgcXVlc3Rpb24gYWJvdXQgY2FsY3U=?= =?gb2312?B?bGF0aW5nIGZyZWUgZW5lcmd5?= Message-ID: <42E72658.000028.13153@m0211.mail.163.com> Hi,thank you for your great help. In my pdos file, the first frequence value is indeed negative. When deleting the first line, it can run very well. Best regards, Chaohao > Hi, > > > the error information: > > "run-time error M6202: MATH > > - log: SING error" > > > > I also know the negative value appears in log > > function described as the following line, > > " if (T.gt.0.d0) > > Ftot=Ftot+dos(i)*a2*T*log(1.0d0-exp(-a3*nu(i)/T)) > > " > > but I do not know how to resolve this problem. > > If you have imaginary (negative in your files) > frequences, your system is dynamically unstable, so, > it is meaningless to calculate the free energy for > such a kind unstable system. > > Bests, > Eyvaz. > > > > ____________________________________________________ > Start your day with Yahoo! - make it your home page > http://www.yahoo.com/r/hs > > _______________________________________________ > 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/20050727/5b13a437/attachment.htm From konstantin_kudin at yahoo.com Thu Jul 28 01:53:58 2005 From: konstantin_kudin at yahoo.com (Konstantin Kudin) Date: Wed, 27 Jul 2005 16:53:58 -0700 (PDT) Subject: [Pw_forum] ibrav=0 vs. other choices in CP In-Reply-To: <42E72658.000028.13153@m0211.mail.163.com> Message-ID: <20050727235358.51065.qmail@web52001.mail.yahoo.com> Hi there, Is there any performance benefit in specifying for example bcc (body centered cubic) cell for water in CP via ibrav=3 with a celldm as opposed to just using ibrav=0 and then giving proper lattice vectors in CELL_PARAMETERS ? Or will the code just convert whatever ibrav into the lattice vectors, and proceed from there? Thanks! Kostya __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From giannozz at nest.sns.it Thu Jul 28 10:21:44 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 28 Jul 2005 10:21:44 +0200 Subject: [Pw_forum] ibrav=0 vs. other choices in CP In-Reply-To: <20050727235358.51065.qmail@web52001.mail.yahoo.com> References: <20050727235358.51065.qmail@web52001.mail.yahoo.com> Message-ID: <200507281021.44992.giannozz@nest.sns.it> On Thursday 28 July 2005 01:53, Konstantin Kudin wrote: > Is there any performance benefit in specifying for example bcc > (body centered cubic) cell for water in CP via ibrav=3 with a celldm > as opposed to just using ibrav=0 and then giving proper lattice > vectors in CELL_PARAMETERS ? there shouldn't be any difference at all in performance > Or will the code just convert whatever ibrav into the lattice vectors, > and proceed from there? correct 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 leoant21 at hotmail.com Thu Jul 28 10:57:16 2005 From: leoant21 at hotmail.com (=?gb2312?B?1cUguumx8g==?=) Date: Thu, 28 Jul 2005 08:57:16 +0000 Subject: [Pw_forum] Question on phonon calculation of MgB2. Message-ID: Dear all: Now I am doing some calculation on the phonon dispersion curves of MgB2.To get a more detailed and correct picture, I have tried several different values of ecutwfc ecutrho and degauss, however there are some problems I want an answer: let me show the results first: (ecutrho=300 for the parameters below) * ecutwfc ***** degauss *** E1u ******* A2u ****** E2g ******** B1g ************** 60 0.005 309 383 585 693 60 0.01 309 383 551 692 60 0.02 309 383 541 692 70 0.005 *******as above**** 566 as above 70 0.01 * * 532 * 70 0.02 * * 522 * 80 0.005 * * 570 * 80 0.01 * * 536 * 80 0.02 * * 527 * ***************************************************************************************** so it seems like that E1u, A2u, and B1g have reached the convergence, while E2g varies as degauss value, since electron-phonon coupling for this modes is very strong, so is this the reason why it varies? and which value should I choose? Or is it related to the values of structure vectors? Or pseudopotentials? Further more, for ecutwfc=50, I didn't get the converged value for A2u mode(692), but I think ecutwfc=50 is so large a value, is there anything else to improve? PP? my input file as follows: **omitted some here** ecutws="60\ 70\ 80" gausss="0.005\ 0.01\ 0.02" for ecutw in $ecutws;do for gauss in $gausss;do # clean TMP_DIR $ECHO " cleaning $TMP_DIR...\c" rm -rf $TMP_DIR/* $ECHO " done" # self-consistent calculation cat > mgb.scf.in << EOF &control calculation='scf' restart_mode='from_scratch', tstress = .true. tprnfor = .true. prefix='mgb', pseudo_dir = '$PSEUDO_DIR/', outdir='$TMP_DIR/' / &SYSTEM ibrav = 4, celldm(1) = 5.803350584, celldm(3) = 1.165092804, nat = 3, ntyp = 2, ecutwfc = $ecutw , ecutrho = 300, occupations = 'smearing' , degauss = $gauss , smearing = 'gaussian' , / &ELECTRONS conv_thr = 1.0d-8 , mixing_beta = 0.7 , diagonalization = 'cg' , / ATOMIC_SPECIES Mg 24.30500 Mg.pz-bhs.UPF B 10.81100 B.pz-bhs.UPF ATOMIC_POSITIONS alat Mg 0.000000000 0.000000000 0.000000000 B 0.500000000 0.288680000 0.570500000 B 0.000000000 0.577350000 0.570500000 K_POINTS automatic 30 30 30 0 0 0 EOF $ECHO " running the scf calculation...\c" $PW_COMMAND < mgb.scf.in > mgb.scf.out $ECHO " done" # phonons cat > mgb.phG.in << EOF phonons of mgb at Gamma &inputph tr2_ph=1.0d-12, prefix='mgb', amass(1)=24.305, amass(2)=10.811, outdir='$TMP_DIR/', fildyn='mgb.$ecutw.$ecutr.$gauss.dynG', / 0.0 0.0 0.0 EOF $ECHO " running the phonon calculation at Gamma...\c" $PH_COMMAND < mgb.phG.in > mgb.phG.out $ECHO " done" #using dynmat.x to recalculate phonon at /gamma $ECHO " running dynmat.x calculation at Gamma...\c" cat > mgb.$ecutw.$ecutr.$gauss.dynmat.in << EOF &input fildyn='mgb.$ecutw.$ecutr.$gauss.dynG' asr=.true. amass(1)=24.305 amass(2)=10.811 filout='mgb.$ecutw.$ecutr.$gauss.dynmat' filmol='mgb.$ecutw.$ecutr.$gauss.modes' / EOF $DYNMAT_COMMAND < mgb.$ecutw.$ecutr.$gauss.dynmat.in >mgb.$ecutw.$ecutr.$gauss.dynmat.out $ECHO "done for $ecutw $ecutr.$gauss" done done _________________________________________________________________ ?????????????? MSN Messenger: http://messenger.msn.com/cn From giannozz at nest.sns.it Thu Jul 28 11:19:53 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 28 Jul 2005 11:19:53 +0200 Subject: [Pw_forum] Question on phonon calculation of MgB2. In-Reply-To: References: Message-ID: <200507281119.53537.giannozz@nest.sns.it> On Thursday 28 July 2005 10:57, ? ?? wrote: > so it seems like that E1u, A2u, and B1g have reached the convergence, > while E2g varies as degauss value, since electron-phonon coupling for > this modes is very strong, so is this the reason why it varies? this is the reason, and it is well known to everybody who has tried to do calculations on MgB2. It is very hard to achieve converged values for the frequency of the E2g mode, and even harder for electron-phonon coefficients. > Further more, for ecutwfc=50, I didn't get the converged value for A2u > mode(692), but I think ecutwfc=50 is so large a value, is there anything > else to improve? PP? convergence with the cutoff depends mostly on pseudopotentials 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 chesrunners at yahoo.com Thu Jul 28 17:30:46 2005 From: chesrunners at yahoo.com (Matt Small) Date: Thu, 28 Jul 2005 08:30:46 -0700 (PDT) Subject: [Pw_forum] fpmd.x questions... Message-ID: <20050728153046.2536.qmail@web30514.mail.mud.yahoo.com> Dear pwscf users, I'm relatively new to pwscf, and would like to ask a few general questions which I haven't really been able to find the answers to in either the input files, manual, or on the forum. This may be because I want to use fpmd.x for my calculations and most of the material in these sources involves pw.x. 1) In the input file it says to use ibrav=14 for variable cell cp type calculations. Is this a typo? (both cp and pw use ibrav=0 for vc calculations) 2) How do you choose the size of your supercell? I got the impression that this is what the cell parameters are for, but have not found any confirming information. 3) The runs that I've been doing aren't stabilizing at the temp.s or the pressures that I want. In part, I realize this will be due to the simulation length and cell size; but for some reason my system does not start at the given starting temperature. I'm not starting with an scf relaxtion (want to make sure I know how to do the md calculations first) but that really shouldn't matter. 4) How do I get the Entropy, Enthalpy, and Heat Capacity? And are k-points or gamma points better for over all bulk properties? - In fpmd.x I realize the Enthalpy is given, but there also appears to be a fortran file "enthropy" (??) which I presume refers to entropy - how do I activate it? - I know it's possible to get the heat capacity via the relat- ion: entropy = integration of (heat capacity)/(temperature) with respect to temperature. I am just wondering if fpmd outputs an equation for entropy or the heat capacity is actually calculated somewhere. 5) Does anyone have any suggestions on optimizing dt, emass, ecutwfc, ecutrho and (if needed) emass_cutoff for sodium containing ionic solids? I'm using NCPP with NLCCs (PZ LDAs available on pwscf.org). Other information: I am using espresso 2.1.3 and below is a sample of one of my input files. A big thank you in advance to everyone! Matt Small &CONTROL calculation ='vc-md' restart_mode='from_scratch' nstep=10 iprint=10 isave=10 tstress=.TRUE. tprnfor=.TRUE. dt = 10.0 ndr=52 ndw=52 outdir ='/home/shorej/espresso/espresso-2.1.3/Frank/NaCl/' prefix = 'fpmd1' pseudo_dir = '/home/shorej/espresso/espresso-2.1.3/pseudo/' ekin_conv_thr = 1.D-5 etot_conv_thr = 1.D-5 forc_conv_thr = 1.D-4 / &SYSTEM ibrav= 0 celldm(1) = 10.6591128 nat=2, ntyp=2, nbnd = 6 nelec= 12 ecutwfc = 40.0 occupations = 'fixed' xc_type = 'PZ' / &ELECTRONS ortho_max = 100 emass = 400.d0 emass_cutoff = 2.5 electron_temperature = 'not_controlled' / &IONS ion_temperature ='not_controlled' ion_dynamics = 'verlet' tempw =298.15 tolp = 10 / &CELL cell_dynamics ='pr' press = .000101325 temph = 300 wmass = 2921.28 / ATOMIC_SPECIES Na 22.99 Na.pz-n-vbc.UPF Cl 35.45 Cl.pz-bhs.UPF ATOMIC_POSITIONS (crystal) Cl 0.5 0.5 0.5 Na 0 0 0 CELL_PARAMETERS (cubic) 10.6591128 0.00000000 0.0000000 0.00000000 10.6591128 0.00000000 0.00000000 0.00000000 10.6591128 __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com