[Pw_forum] Ph.x and force constants calculation questions.

[Eyvaz Isaev]

Dear Alex,

--- On Sat, 12/19/09, Zhen Huang <huang87 at purdue.edu> wrote:

> 
> 1. I wonder if force constants outputs are in the unit of
> Hartree/bohr^2.

In QE all quantities are in atomic units, so IFC are in Ry/au^2 

> 2. For a 2 atoms unit cell, for example, the program (ph.x
> calcualtion) only gives the force constants between the atoms but any > other, say, second and third nearest neighbors interactions. Other 
> than increasing the number and size of unit cell generate 
> non-primitive unit cell, is there a way to obtain the force constants > for atoms beyond the atoms in unit cell?
> 

In the Linear Response all calculations are performed inside the unit cell. What you are talking about is intrinsic for the Direct Method. You can increase number of q-points for which dynamical matrices have to be calculated. 
Calculated force constants F(R) are defined for R=(i-1)*a+(j-1)*b+(k-1)*c
where a,b,c are basis vectors, and i,j,k are those from nq1,nq2,nq3.
nq1,nq2,nq3 are division numbers from ph.in file.

Calculations for a cell within N atoms and nq1,nq2,nq3 set of q-points 
mimic a supercell containing N*nq1*nq2*nq3 atoms. 

> 3. On top of the last question, is there a way to control
> the cutoff number of the nearest neighbors.For example,  if I only 
> want to consider upto fourth nearest neighbor based force constants 
> for a certain  material which perimeter should I study and adjust to 
> gives a promising dispersion relation based the selected number of 
> force constants from calculation.

If you know the nearest neighbor distance you are interested in, then just calculate for which nq1,nq2,nq3 you can get this distance and use these nq1,nq2,nq3 set to calculate dynamical matrices. 
 
Usually one starts from a reasonable number of q-set (which is different for different systems depending on the atomic bonding, complexity, symmetry) and then gradually increase nq's to see how the phonon dispersion relations are affected.  

Hope this helps.

Bests,
Eyvaz.
 
-------------------------------------------------------------------
Prof. Eyvaz Isaev, 
Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, 
Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden 
Condensed Matter Theory Group, Uppsala University, Sweden  isaev at ifm.liu.se, eyvaz_isaev at yahoo.com