[Pw_forum] negative frequency

[Mike Mehl]

On 05/19/2010 08:11 AM, partha sarathi ghosh wrote:
> Dear all,
>               I am trying to do phonon calculation of bcc-Zr from
> (0,0,0) to (1,1,1) in q-space.But I am getting negative frequencies in
> several q-points, also my curve is very different from experimental one.
> I optimized the system w.r.t ecutwf, ecutrho ,conv_thr, k-point sampling
> .I also tried the same calculation by changing potential also.I am new
> fellow in this field so I am unable to understand what is happening wrong !!

First, the ground state of Zr is HCP, not BCC, so you shouldn't expect 
being able to match the HCP phonon spectrum.  With a bcc calculation.

Second, the high-temperature hcp-bcc transition is driven by an unstable 
bcc phonon, so, yes, you'd expect "negative" (actually imaginary) 
frequencies at some k-points.  See below.  It's written about Mg, but it 
will apply to Zr as well.

@Article{wentzcovitch88:hcp_to_bcc,
author =         {Renata M. Wentzcovitch and Marvin L. Cohen},
title =          {Theoretical model for the hcp-bcc transition in Mg},
journal =        {Phys. Rev. B},
year =   1988,
volume =         37,
number =         10,
pages =  {5571-5576},
month =  {April},
url =            {http://link.aps.org/abstract/PRB/v37/p5571},
doi =            {10.1103/PhysRevB.37.5571},
keywords = {structure},
abstract =       {Using a first-principles total-energy pseudopotential 
method, we investigate the transition mechanism for a pressure-induced 
martensitic transformation hcp-->bcc which occurs in Mg at pressures 
around 50 GPa. Two internal structural degrees of freedom are selected 
and one lattice is transformed into the other by relaxing these two 
parameters continuously.  One of the parameters characterizes the 
relative displacement of the hexagonal layers and corresponds to a 
transverse phonon at the Brillouin-zone edge A in the hexagonal 
structure.  The other characterizes the distortion of the internal 
hexagonal angles and corresponds to uniform strain along one of the 
[0010][hcp] directions.  The interaction between these two distortion 
modes causes important anharmonic effects in the zone-edge phonon a
nd provides a low-energy path for the structural transition. The small 
activation barrier at the transition indicates that quantum fluctuations 
between the two structures could be taking place.}



-- 
Michael J. Mehl
Head, Center for Computational Materials Science
Naval Research Laboratory Code 6390
Washington DC