[xcrysden] Multiple 3D-grids?

Tone Kokalj tone.kokalj at ijs.si
Fri Dec 18 12:39:54 CET 2009

On Fri, 2009-12-18 at 12:18 +0100, Lorenzo Paulatto wrote:
> Replying to message "Re: [xcrysden] Multiple 3D-grids?" from Henrik Löfås  
> (18/12/09):
> > Yes, I would to like have them superimposed. What I want to plot is the
> > transmission eigenchannels, which can be seen as travelling waves hence,
> > both real and imag part separately are interesting, see for example
> > http://link.aps.org/doi/10.1103/PhysRevB.76.115117
> Dear Henrik,
> I could never find a software that does both multiple grids and  
> non-simple-cubic cells. It may still exist, but I could not find it. On  
> ther other hand, if your unit cell *is* simple cubic you can use gOpenMol  
> (which I think was used for the pictures in that paper) and/or vmd.  
> Otherwise, you can try with XCrysDen + gimp, plot the different datagrids  
> and then compose them by hand; its painful, but it works.

Lorenzo is right. You cannot do plot like the Fig.3 in that paper. But
you can use the gimp for the assembly, and in same cases it may not be
that painful. However if the two surfaces "penetrate one into the other"
like a chain the result may not be very good.

Nevertheless, here is the recipe.

1. Plot first, say the real part, (you can do two positive and negative
simultaneous plot.

2. save he xcrysden script, i.e., menu File-->Save Current State and
Structure (this is important for having the two pictures displayed in a
completely the same way)

3. edit the so-saved script, i.e., replace the first structure part of
the file with something like:

::scripting::open -r my_dim --xsf my_second_grid.xsf

(where my_dim is the dimension of your structure: 0=molecule, 1=polymer,
2=slab,3=crystal, i.e. it's a periodic dimensionality)

4. load so edited script with: xcrysden -s my_script.xcrysden and plot
it accordingly

Then you need some image manipulation program. This is the most tedious
part. I've seen people doing "miracles" with gimp ... You may make both
plots (layers) a bit transparent for better superposition.

Good luck!


Anton Kokalj
J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia 
(tel: +386-1-477-3523 // fax:+386-1-477-3822)

Please, if possible, avoid sending me Word or PowerPoint attachments.
See:  http://www.gnu.org/philosophy/no-word-attachments.html

More information about the XCrySDen mailing list