[Pw_forum] Si diamond-cubic to beta-Sn transformation

[Mike Mehl]

Dayrl,

Looking at your output, it seems that you are very far from the minimum 
energy beta-Sn structure:

entering subroutine stress ...

           total   stress  (Ry/bohr**3)                   (kbar)     P= 
   42.95
0.00079257   0.00000000   0.00000000   116.59      0.00      0.00
0.00000000   0.00079725   0.00000001     0.00    117.28      0.00
0.00000000   0.00000001  -0.00071388     0.00      0.00   -105.01

If you find the actual minimum energy structure for beta-Sn you'll get a 
much smaller difference in energy.

I recently did this calculation using Si.pbe-n-van.UPF, rather than 
Si.pw91-n-van.UPF, and got an energy difference of 0.29 eV/atom, rather 
in better agreement with VASP and other published results.

Ideally, all ultrasoft pseudopotentials should give the same result, but 
unfortunately they often don't.

On 10/05/2010 02:36 PM, Daryl Chrzan wrote:
> Colleagues,
>
> I am working through a problem that I have assigned to my Computational
> Materials Science class - computation of the transition pressure for the
> diamond-cubic to beta-Sn structures in diamond using pw.x.

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