[CPMD-list] about vibrational analysis

[Juerg Hutter]

Hi

it seems that what you see is an effect of the anharmonicity
of the C=O bond (and environment). The force field
is much more harmonic (iposed by its analytic form) then
DFT. You could do the following test:
Calculate the C=O mode for H2CO for different C-O
bond lengths or map out the potential along the C-O
bond and then calculate what the distribution of
frequencies would be using your method (instantaneous
harmonic approximation).

regards

Juerg Hutter

----------------------------------------------------------
Juerg Hutter                   Phone : ++41 44 635 4491
Physical Chemistry Institute   FAX   : ++41 44 635 6838
University of Zurich           E-mail: hutter at pci.unizh.ch
Winterthurerstrasse 190
CH-8057 Zurich, Switzerland
----------------------------------------------------------


On Mon, 13 Nov 2006, Yong Zhang wrote:

> Hi CPMDers,
>
> I have a question about vibrational analysis in cpmd.
> My system is one N-methylacetamide molecule (with N-H deuterated to N-D) in
> vacuum. The situation is like this: I generated a trajectory (10ps)
> following general procedure (with Nose Hoover thermostat at 300K); then 80
> snapshots were taken from this traj (one every 100fs for the 8ps); for each
> of these snapshots, a wavefunction ompimization was carried out, and based
> on this optimized wavefunction (by "restart wavefunction..."), the freq was
> calculated using FD method; Then I checked the C=O stretching frequency (so
> called amide I mode). Surprisingly, the freq has a distribution over the
> range as wide as 400cm-1 (1500cm-1 ~ 1900cm-1). The experimental observation
> of this mode in gas phase is 1717cm-1. I did the similar calculation using
> charmm force field, the freq distribution is much narrower (1685+/-15 cm-1),
> which is more reasonable under harmonic approximation. So my question is,
> why the freq dirtribution in CPMD calaulation is so wide?
>
> If quench the snapshots to a minimized structure and do the freq ana, their
> freq are close to each other, which might indicate a "single" local minimum.
> I also tried LR for freq calculation, the resultant freq is about 10cm-1
> lower than the corresponding FD one for all cases; I also tried BO dynamics,
> similar results.
>
> Any comments or suggestions are appreciated.
>
> Best wishes
> Yong
>
>
> The input file for dynamics run:
> ==================
> &CPMD
> MOLECULAR DYNAMICS CP
> RESTART WAVEFUNCTION COORDINATES VELOCITIES ACCUMULATORS LATEST
>
> TEMPERATURE
> 300.0
> NOSE IONS MASSIVE
> 300.0 1700.0
> NOSE ELECTRONS
> 0.002 10000.0
>
> TIMESTEP
> 4.0
>
> MAXSTEP
> 10000
>
> EMASS
> 500.0
>
> TRAJECTORY XYZ SAMPLE
> 5
>
> DIPOLE DYNAMICS SAMPLE WANNIER
> 10
> WANNIER REFERENCE
> 0.0 0.0 0.0
>
> &END
>
> &SYSTEM
> ANGSTROM
> SYMMETRY
> 1
> CELL
> 12.14  1.0  1.0  0  0  0
> CUTOFF
> 70.0
> &END
>
> &DFT
> NEWCODE
> FUNCTIONAL PBE
> GC-CUTOFF
> 5.0e-6
> &END
>
> &ATOMS
> ISOTOPE
>  1.0
>  2.0
>  12.0
>  14.0
>  16.0
>
> *H_MT_PBE.psp   KLEINMAN-BYLANDER
>  LMAX=S
>  6
>  1.69080    .05583    .88430
>  3.28199   -.14332    .06344
>  2.16794  -1.45255    .00207
> -1.58359    .17630  -2.35128
>  -.19701    .20889  -3.48962
>  -.49100   1.60271  -2.37391
>
> *H_MT_PBE.psp   KLEINMAN-BYLANDER
>  LMAX=S
>  1
>  -.22859   -.88824   -.93883
>
> *C_MT_PBE.psp   KLEINMAN-BYLANDER
>  LMAX=P
>  3
>  2.18368   -.34882    .01350
>  1.56123    .24388  -1.15711
>  -.53752    .48601  -2.44731
>
> *N_MT_PBE.psp   KLEINMAN-BYLANDER
>  LMAX=P
>  1
>   .32592   -.14815  -1.44467
>
>
> *O_MT_PBE.psp   KLEINMAN-BYLANDER
>  LMAX=P
>  1
>  2.16429   1.09559  -1.77796
>
> &END
>
> ======================
>
>
>
> input file for vib ana:
> =====================
> &CPMD
> VIBRATIONAL ANALYSIS FD GAUSS SAMPLE
> 2
> RESTART WAVEFUNCTION COORDINATES LATEST
> &END
>
> &SYSTEM
> ANGSTROM
> SYMMETRY
> 1
> CELL
> 12.14  1.0  1.0  0  0  0
> CUTOFF
> 70.0
> &END
>
> &DFT
> NEWCODE
> FUNCTIONAL PBE
> GC-CUTOFF
> 5.0e-6
> &END
>
> &ATOMS
>
> ISOTOPE
>  1.0
>  2.0
>  12.0
>  14.0
>  16.0
>
> *H_MT_PBE.psp   KLEINMAN-BYLANDER
>  LMAX=S
>  6
>  1.69080    .05583    .88430
>  3.28199   -.14332    .06344
>  2.16794  -1.45255    .00207
> -1.58359    .17630  -2.35128
>  -.19701    .20889  -3.48962
>  -.49100   1.60271  -2.37391
>
> *H_MT_PBE.psp   KLEINMAN-BYLANDER
>  LMAX=S
>  1
>  -.22859   -.88824   -.93883
>
> *C_MT_PBE.psp   KLEINMAN-BYLANDER
>  LMAX=P
>  3
>  2.18368   -.34882    .01350
>  1.56123    .24388  -1.15711
>  -.53752    .48601  -2.44731
>
> *N_MT_PBE.psp   KLEINMAN-BYLANDER
>  LMAX=P
>  1
>   .32592   -.14815  -1.44467
>
>
> *O_MT_PBE.psp   KLEINMAN-BYLANDER
>  LMAX=P
>  1
>  2.16429   1.09559  -1.77796
>
> &END
> ======================
>