[Pw_forum] vc-relax

[Ngoc Linh Nguyen]

> Dear all I wanted to run some vc-relax calculations. To start with I took
> pbtio3 system and tried to relax it. for the following input file it
> predicted the c/a ratio as 1.23. The output structure (both coordinates
> and
> lattice parameters) is grossly in error. In the forum i have seen people
> reporting c/a ratio which is in good agreement with the experimental
> value.

Normally, we should start with a value that's close experiment one.

> Later I realized for different input configurations it is predicting
> different c/a ratios. At one time for pbtio3 I got cubic structure after
> relaxing!!!. As i understood it is stuck up in local minimum. Since i know
> the structure is tetragonal i could figure out the error but for an
> unknown
> system how can i relax the system with confidence.

That depends on how much your understand about the configurations that you
wanna consider. For example, under pressure structure will be induced a
structural phase transition, so you need a knowledge about the kinds of
structure, which are tetrahedral or hexagonal, at different finite
pressures.


> (if there is some literature that i can read up on this please refere me
> the
> same i will read)
Renata M. Wentzconvitch et. al. , Physical Review Letter 70, 3947 (1993).
 or you can search on google.com by keyword: "Molecular dynamics variable
cell shape "  to see more related literatures.


Good luck,

Linh


-----------------------------------------------------
 Nguyen Ngoc Linh, PhD Student
 c/o:   SISSA & CNR-INFM Democritos,
        via Beirut 2-4, 34014 Trieste (Italy)
 email: nnlinh at sissa.it
 phone: +39 04 03787 319
 skype: ngoclinh84phys
-----------------------------------------------------
"The physics is theoretical but the fun is real"


----------------------------------------------------------------
  SISSA Webmail https://webmail.sissa.it/
  Powered by SquirrelMail http://www.squirrelmail.org/