[gpaw-users] Problem simulating STM images

[Susi Toma]

Great, thanks a lot! Probably an optional structural overlay is then not necessary.

Perhaps Jussi can update the svn on Taito when the fix is submitted? :)

"Jens Jørgen Mortensen" <jensj at fysik.dtu.dk<mailto:jensj at fysik.dtu.dk>> kirjoitti 25.9.2013 kello 13.16:

Den 25-09-2013 09:01, Jens Jørgen Mortensen skrev:
Den 24-09-2013 12:31, Susi Toma skrev:
Hello again,

I have a few further questions.

First of all, my graphene unit cell is not a square, but orthorhombic, and furthermore, it is 8 unit cells in one direction and 6 in the other (made with the attached script). The stm image simulation, however, yields square images, resulting in a significant and non-trivial distortion compared to the actual real-space structure. This makes interpretation of the images rather hard.

Can the stm functionality somehow create non-square images? Or if not, is there a way to automatically overlay the structure on the images?

You could repeat your calculation in a rectangular cell which should be possible for your 8*6 cell (since 6 is even).

See also:

  http://matplotlib.org/api/axes_api.html#matplotlib.axes.Axes.set_aspect

Alternatively, you will have to teach yourself some matplotlib magic - I'm sure there is an elegant way to do it, but I don't know it.  Any matplotlib masters here?

OK, I figured it out.  Matplotlib's contour() function can do it - it just needs arrays of x and y coordinates.  Let me fix ASE's STM.scan() method so that it delivers that.

Jens Jørgen


You could also do the interpolation to a rectangular grid yourself just using some ndarray tricks - we should have such an example on the ASE web-page.

Jens Jørgen


Many thanks,
Toma


On 5.9.2013, at 15.28, Toma Susi <toma.susi at aalto.fi<mailto:toma.susi at aalto.fi>> wrote:

Hi,

After Jussi kindly updated the gpaw-env module on the CSC Taito cluster, the simulations now run correctly following Jens' instructions. I'm now getting reasonable looking STM image simulations for my system, with no problems with either of the previous errors regardless of the values I use for z, bias or z0.

With any luck we'll have a chance to compare the simulations to experimental measurements that we are hoping to do soon.

Thanks a lot for all your help!

Best,
Toma





_______________________________________________
gpaw-users mailing list
gpaw-users at listserv.fysik.dtu.dk<mailto:gpaw-users at listserv.fysik.dtu.dk>
https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.fysik.dtu.dk/pipermail/gpaw-users/attachments/20130925/3c65e634/attachment.html>