# [Pw_forum] Error in User Guide - ibrav

Paolo Giannozzi giannozz at democritos.it
Fri Mar 20 14:45:55 CET 2009

also at if.ufrj.br wrote:

> The user guide contains the following basis triplet for the BCC structure:
>
> bcc body entered cubic
> ====================
> v1 = (a/2)(1,1,1),  v2 = (a/2)(-1,1,1),  v3 = (a/2)(-1,-1,1).
>
> While the for BCT structure this is what is seen:
>
> body centered tetragonal (i)
> ================================
>    v1 = (a/2)(1,-1,c/a),  v2 = (a/2)(1,1,c/a),  v3 = (a/2)(-1,-1,c/a).
>
>
> Now, I don't think that these two are consistent with the same convention
> for the choice of basis.

there is no such thing as an "official" convention for the choice of
basis. The one used in q-e may not be the most commonly used or the
most appealing, but it is exactly and exhaustively documented.
All errors or inconsistencies are immediately corrected.

> They are not both right-handed

they are. Try v1\cdot (v2\times v3) if you don't believe your eyes.

> Also, The BCT does not have the same relative internal angle between
> two basis vectors (their scalar products are different).

so what? scalar products are different in your proposed basis as well.

> Third, it doesn't work ;)

it does, if you do things properly.

> Please, try to specify the atomic positions inside the cell using the flag
> {crystal}. Then check the results on xcrysden. Probably you will get a
> weird configuration.

if you specify atomic positions written for a different convention, yes,
you get weird results. You can specify your choice of axis, by the way,
with card CELL_PARAMETERS.

Paolo
--
Paolo Giannozzi, Democritos and University of Udine, Italy