Infra-red spectra from MD

[alex]

Hello all,

I'd like to calculate spectra from MD simulations, because I'm also interested in anharmonic effects (e.g. solute in water).
I also figured out that I'll need the velocity autocorrelation function (VACF) and p4vasp to do that. But so far I failed to convince p4v to plot the VACF. I'm also not sure how to continue after that.

I'd be happy if somebody is out there who could provide some kind of recipie with, e.g. working INCARs and which steps to do.

Thanks a lot!

alex

[admin]

Dear Alex,
a reason for the not-working p4v can be the setting of your local computer, cf.
http://cms.mpi.univie.ac.at/vasp-forum/ ... p4v#p17024
IR spectra were simulated by several users, check e.g.
http://homepage.univie.ac.at/david.karh ... loads.html

[alex]

Dear admin and all,

sorry for bothering you again.
p4v _is_ working, I just don't get any VACF or FFT output after pressing the button (or pressing twice as outlined in the cited post below). I assume I'm missing some details in my INCAR (attached below). And yes, I also know David's work on intensities, which is working fine for a normal, harmonic frequency analysis.

However, I'm after anharmonic frequcies and intensities.

Thanks

alex


PS: my INCAR


SYSTEM = water in a box

Electronic minimisation
GGA = PE
PREC = Low
EDIFFG = -1.e-2
ENCUT = 400.
NELMIN = 4
NELM = 200
LREAL = A
# ISPIN = 2
# LVDW = .T.

DOS related values
SIGMA = 0.1
ISMEAR = 0 ! Gaussian smearing for big cells

Ionic relaxation
IBRION = 0 ionic relax: 0-MD 1-quasi-New 2-CG
NSW = 1000 ! number of steps for IOM
NBLOCK = 1 ; KBLOCK = 10
SMASS = -3.0 ! Nose mass-parameter (am)
POTIM = 0.50 ! time-step for ion-motion
TEBEG = 498.15 ! temperature
TEEND = 498.15

LEPSILON = .TRUE. # enables to calculate and to print the BEC tensors
IDIPOL = 3
# LDIPOL = .TRUE.
DIPOL = 0.5 0.5 0.5
NWRITE = 3 # affects OUTCAR verbosity: explicitly forces SQRT(mass)-divided eigenvectors to be printed

PC-function
APACO = 10.0 ! distance for P.C.

Tomas Bucko MD package stuff
MDALGO = 1
ANDERSEN_PROB = 0.1

Do not write huge files
LCHARG = .FALSE.
LWAVE = .FALSE.

Performance options
NPAR = 4 ! adjust to number of processors
ALGO = V

[admin]

There is nothing special in INCAR. p4v produces VAC and FFT when MD makes the number of steps
defined in INCAR by NSW and vasprun.xml file is closed properly.