Re: [Phonopy-users] Fwd: Using Phonopy
Brought to you by:
atztogo
Menu
▾
▴
Re: [Phonopy-users] Fwd: Using Phonopy
|
From: Mostafa Y. <myo...@mi...> - 2015-11-10 11:45:26
|
Hi Ashish, In principle NSW should not matter but for some versions of VASP,NSW has to be set to 1. Checking EDIFF and EDIFFG is done by testing and comparing to literature or experiments. EDIFFG=-0.001 and EDIFF=1E-8 are usually sufficient. If you plan to use DFPT,you can add to INCAR: LEPSILON = .TRUE. LPEAD = .TRUE. The first to calculate the dielectric constant at a marginal cost, and the second accelerates the calculation of certain terms. Check VASP manual. Please note that your question is more suitable for VASP forum rather than PHONOPY forum. Mostafa ________________________________ From: Ashish Bhatnagar [ash...@gm...] Sent: Tuesday, November 10, 2015 2:20 AM To: pho...@li... Subject: [Phonopy-users] Fwd: Using Phonopy Hi all, I need your kind help in calculating Phonon DOS for my system I have relaxed my structure using the following tags in vasp NCORE = 8 Startparameter for this Run: NWRITE = 1 (LPETIM=F write-flag & time) ISTART = 0 !0=new, 1=wave, 2=same cutoff&latt (1&2 need WAVECAR # ICHARG = 0 !0=charge from wave, 1=charge from CHGCAR, 2=atom # ISPIN = 2 LWAVE = .FALSE. (TRUE-Write WAVECAR, FALSE-not) LCHARG = .FALSE. (TRUE-Write CHGCAR and CHG, FALSE-not) GGA = PE Electronic Relaxation ENCUT = 525.0 PREC = Accurate !Medium=default, Low, High; affects ENMAX, mesh, pspot EDIFF = 1E-8 !energy stopping-criterion for electr. iterations EDIFFG = -0.0001 !force stopping-criterion for geometry steps Ionic Relaxation NSW = 500 !max number of geometry steps NELM = 200 NELMDL = -12 IBRION = 1 !ionic relax: 0-MD 1-quasi-New (RMM-DIIS for NEB) ISIF = 2 !(0:force=y stress=n ions=y shape=n volume=n, # 1:f=y s=diag i=y s=n v=n, 4:f=y s=y i=y s=y v=n, 5:f=y s=y i=n s=y v=n) ISYM = 1 !1=use symmetry, 0 = no symmetry #POTIM = .25 !initial time step for geo-opt (increase for soft sys) DOS related values: ISMEAR = 1 ! -4-tet -1-fermi 1=Methfessel/Paxton 1.order SIGMA = .05 !broadening in eV Electronic Relaxation ALGO = Normal # IALGO = 48 !algorithm (8=CG for small, 48=RMM for big systems) LREAL = FALSE !real-space projection After that I move CONTCAR to POSCAR and want to use phonopy So I simply change IBRION=1 to IBRION=8. in my INCAR. Now at some places I found NSW=1 with IBRION=8 for calculating Phonon DOS using Phonopy. So should I set NSW=500 or should I use NSW=1? Also please tell me whether EDIFF=1E-8 and EDIFFG=-0.0001 is sufficient condition for phonopy if not then how should i check the optimum condition? Thanks -- Ashish Bhatnagar Senior Research Fellow Hydrogen Energy Centre, Dept. of Physics Banaras Hindu University, Varanasi India - 221005 ---------- Forwarded message ---------- From: Ashish Bhatnagar <ash...@gm...<mailto:ash...@gm...>> Date: Mon, Nov 9, 2015 at 3:32 PM Subject: Using Phonopy To: pho...@li...<mailto:pho...@li...> Hi all, I need your kind help in calculating Phonon DOS for my system I have relaxed my structure using the following tags in vasp NCORE = 8 Startparameter for this Run: NWRITE = 1 (LPETIM=F write-flag & time) ISTART = 0 !0=new, 1=wave, 2=same cutoff&latt (1&2 need WAVECAR # ICHARG = 0 !0=charge from wave, 1=charge from CHGCAR, 2=atom # ISPIN = 2 LWAVE = .FALSE. (TRUE-Write WAVECAR, FALSE-not) LCHARG = .FALSE. (TRUE-Write CHGCAR and CHG, FALSE-not) GGA = PE Electronic Relaxation ENCUT = 525.0 PREC = Accurate !Medium=default, Low, High; affects ENMAX, mesh, pspot EDIFF = 1E-8 !energy stopping-criterion for electr. iterations EDIFFG = -0.0001 !force stopping-criterion for geometry steps Ionic Relaxation NSW = 500 !max number of geometry steps NELM = 200 NELMDL = -12 IBRION = 1 !ionic relax: 0-MD 1-quasi-New (RMM-DIIS for NEB) ISIF = 2 !(0:force=y stress=n ions=y shape=n volume=n, # 1:f=y s=diag i=y s=n v=n, 4:f=y s=y i=y s=y v=n, 5:f=y s=y i=n s=y v=n) ISYM = 1 !1=use symmetry, 0 = no symmetry #POTIM = .25 !initial time step for geo-opt (increase for soft sys) DOS related values: ISMEAR = 1 ! -4-tet -1-fermi 1=Methfessel/Paxton 1.order SIGMA = .05 !broadening in eV Electronic Relaxation ALGO = Normal # IALGO = 48 !algorithm (8=CG for small, 48=RMM for big systems) LREAL = FALSE !real-space projection After that I move CONTCAR to POSCAR and want to use phonopy So I simply change IBRION=1 to IBRION=8. in my INCAR. Now at some places I found NSW=1 with IBRION=8 for calculating Phonon DOS using Phonopy. So should I set NSW=500 or should I use NSW=1? Also please tell me whether EDIFF=1E-8 and EDIFFG=-0.0001 is sufficient condition for phonopy if not then how should i check the optimum condition? Thanks -- Ashish Bhatnagar Senior Research Fellow Hydrogen Energy Centre, Dept. of Physics Banaras Hindu University, Varanasi India - 221005 -- Ashish Bhatnagar Senior Research Fellow Hydrogen Energy Centre, Dept. of Physics Banaras Hindu University, Varanasi India - 221005 |