The ABINIT Users Mailing List ( CLOSED )

["matthieu verstraete" <matthieu.jean.verstraete@gmail.com>]

On Sat, Jan 10, 2009 at 7:02 AM, Julie Smart <smart.julie@rocketmail.com> wrote:

Hi all,

Hello
 

Working on bulk GaAs optimization, I have imposed natfix and iatfix inorder to fix "Ga" position. But after the optimization I have the output xred like:

0.3333333333333  0.6666666666667  1.0658143578E-15
0.6666666666667  0.3333333333333  0.500000000000
0.3333333333333  0.6666666666667  0.374956338791
0.6666666666667  0.3333333333333  0.874956338791

you mean the 1.e-15? That is machine precision: simple numerical roundoff errors will always give you this kind of thing. You can reset it to 0 if you like but it won't change anything. You can also see that the last 2 atoms are precisely 0.5 apart in z, down to 12 digits - this is amply enough precision. Finally, the details of the numerical precision are down to which platform and compiler you use, so if it's not the same as the reference machine for tnlo you can get these tiny differences. Abinit is one of the most portable and precise codes around, for these matters.

For symmetry purposes, abinit examines down to 1.e-10 in reduced coordinates.
 
Further, by symmetry, you do not need to fix the Ga here. However if you didn't you might end up with a shift of both atoms, e.g. z_Ga=0.2    and   z_As=0.2+0.374956338791, which is physically the same but the numbers are not as aesthetic.

Why do I have the slight displace in the first atom z parameter (following exactly tnlo-1.in) ? Also, please let me know how we can set "strfact" in the optimization in order to get  to the level of  zero stress in optimization?

strfact is 0 by default, which relaxes to 0 stress (see the definition of strfact for units and components).


Matthieu

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Dr. Matthieu Verstraete

European Theoretical Spectroscopy Facility (ETSF) 
Dpto. Fisica de Materiales, 
U. del Pais Vasco,            
Centro Joxe Mari Korta, Av. de Tolosa, 72,   Phone: +34-943018393
E-20018 Donostia-San Sebastian, Spain        Fax  : +34-943018390

Mail : matthieu.jean.verstraete@gmail.com
http://www-users.york.ac.uk/~mjv500