[Pw_forum] pressure from pw calculation-further

[Katalin Gaal-Nagy]

Dear Yu,

the Pulay-stress is called after Peter Pulay, so you can search for this
name. During my PhD I was reading specially the papers Chem. Phys. Let.
Vol 73, pp 393 (1980) and Mol. Phys. Vol 17, pp 197 (1969). A nice
explanation I found in some lecture notes of Paolo Giannozzi, but I don't
find the link right now.

In the past I was doing some test calculations for different
pseudopotentials and it also turned out, that some of them converge very
badly with respect to the pressure. The pressure values can be positive or
negative, depending on the fact if you are overestimating or
underestimating the equilibrium volume due to the error of the Pulay
stress.


Best wishes and good luck,
Katalin


>
> Dear Paolo and pw users,
>
> The following is the relationship between lattice
> parameter a0 and pressure from the results of pw
> calculation.The fitted zero pressure lattice parameter
> is about 7.74au. If the pressure is the derivative of
> the energy with respect to the volume (given by -dE/dV
> ?) at fixed basis set,the absolute value of pressures
> might be different from the fitted ones, but the sign
> of the pressure should be at least positive with a0
> less than about 7.7au. But the output pressures from
> pw are all negtive. Why?
>
> By the way, where can I find the literature explain
> this "Pulay stress". I didn't find it in the mannual
> or the input pw.
>
> best regards!
>
> W. YU
> ======================================================
>
> a0(au)      P(ressure,kbar)
>
> 7.4     -2966.80
> 7.5     -3074.92
> 7.6     -3148.36
> 7.7     -3186.48
> 7.8     -3243.19
> 7.9     -3286.29
> 8.0     -3315.97
> 8.1     -3327.72
> 8.2     -3336.04
> 8.3     -3336.26
> 8.4     -3309.90
> 8.5     -3227.79
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>