[Pw_forum] Re: how to input crystal structure of Amm2 SrTiO3

Stefano Baroni baroni at sissa.it
Wed Apr 19 09:19:27 CEST 2006


On Apr 19, 2006, at 3:35 AM, 针叶 朱 wrote:

> Dear stefano Baroni:
> i want to caculate the properties of SrTiO3 under tensile strain. i  
> have known the crystal structure of SrTiO3 under tensile strain is  
> Amm2 space group, and atomic cartesian cooridinate : Sr(0,0,0),Ti 
> ((0.5+Δx,0,0.5),O1(0.5+Δx,0,0) O2(0.25+Δx,0.25+Δy,0.5).there is  
> several question about atomic question.
> first, in ATOMIC_POSITION{alat}, is atomic cooridinates along  
> cartesian cooridinate?

YES atomic coordinates are cartesian for every option of the  
ATOMIC_POSITION control card, but "crystal"

> second in ATOMIC_POSITION{crystal}, is atomic cooridinates along  
> crystal axis of primitive unit cell  cooridinate?for hobr and  
> angstrom,the case is same to the alat?

I do not quite understand the question, in any case the "crystal"  
atomic coordinates are defined by the convention:

x(:)  = \sum_j AT(:,j)*c(j),

where x(:) are the atomic (cartesian) coordinates, and AT(:,1), AT(:, 
2), AT(:,3) are the the basis vectors of the Bravais lattice with  
their own units (you usually specify AT giving "ibrav" and "celldm 
(:)", the latter in Bohr radii). A list of conventional basis vectors  
used by PWscf (as well as much other useful information) is found in  
the documentation files distributed with the code (INPUT_PW, in the  
present case). Beware, though, that you hardly need to pass atomic  
positions in crystal units. It is usually far easier to pass them in  
cartesian coordinates.

> from symmestry, The SrTiO3 cartesian cooridinate:
> Sr      0.000000      0.000000      0.000000
> Ti      0.500000      0.000000      0.500000
> O      0.500000      0.000000      0.000000
> O      0.250000     -0.250000      0.000000
> O      0.750000     -0.250000      0.500000
> O      0.750000     -0.250000      0.000000
> O      0.250000     -0.250000      0.500000
> but i found that the ion position in still after optimization.  
> please tell me how to input structure of AMM2 SrTiO3?

This is a more subtle question which I would invite you to think  
about carefully by yourself.

Suppose you want to find the (stable) equilibrium position of a  
particle in a double potential well, say V(x)=(x^2-1)^2, and that you  
start your numerical search from x=0. The force acting on the  
particle at this position is zero by symmetry. So, any algorithm  
aiming at finding the zeroes of the force would keep the particle  
fixed at x=0. Does this imply that this is an equilibrim position?  
Does this imply that this a _stable_ equilibrium position? How would  
you use an algorithm which finds the zeroes of the force to find a  
stable equilibrium configuration in this case? What has this all to  
do with you problem?

Please, give a thought at these questions and revert to us with or  
without a solution to your problem.

Stefano B.



> There is my inputfile of
> SrTIO3
> &CONTROL
>             title='st'
>             calculation='relax'
>             restart_mode = 'from_scratch' ,
>             outdir = 'tmp' ,
>             pseudo_dir = '/home/zzy/pwscf/pseudo/' ,
>             prefix = 'pst' ,
>             forc_conv_thr=1.0D-4
>             tstress = .true. ,
>             tprnfor = .true. ,
> /
> &SYSTEM
>            ibrav=9,
>            celldm(1)=7.5,
>            celldm(2)=1
>            celldm(3)=0.955
>            nat=7,
>            ntyp=3,
>           ecutwfc=50,
>           ecutrho=450
> /
> &ELECTRONS
>              conv_thr=1.0d-6
> /
> &IONS
>                 ion_dynamics = 'bfgs' ,
> /
> ATOMIC_SPECIES
>  Sr   87.62000  038-Sr-ca-sp-vgrp.uspp.UPF
>  Ti   47.88000  022-Ti-ca-sp-vgrp.uspp.UPF
>  O    15.99960  008-O-ca--vgrp.uspp.UPF
> ATOMIC_POSITIONS {crystal}
> Sr 0    0     0    0 0 0
> Ti 0.5  0     0.5  1 0 0
> O  0.5  0      0   1 0 0
> O  0.25 -0.25  0.0 1 1 0
> O  0.75 -0.25 0.5  1 1 0
> O  0.25 -0.25 0.5  1 1 0
> O  0.75 -0.25  0   1 1 0
> K_POINTS {automatic}
> 6 6 6 1 1 1
>
> Regards
>
>                                                  zhuzhenye
> __________________________________________________
> 赶快注册雅虎超大容量免费邮箱?
> http://cn.mail.yahoo.com

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060419/c2cc1698/attachment.htm 


More information about the Pw_forum mailing list