LAMMPS WWW Site - LAMMPS Documentation - LAMMPS Mailing List Archives
[lammps-users] Using compute/xrd with manual c
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[lammps-users] Using compute/xrd with manual c


From: Sebastian Hütter <sebastian.huetter@...7220...>
Date: Mon, 6 Nov 2017 15:52:45 +0100

Hi all,

I want to use the xrd compute to obtain XRD powder patterns from a structure. As a simple test, I used bcc-Mo and the following input script: <https://gist.github.com/martok/9102b3e174e960e2f8a0be1fd8c7e3c4>

In general, this works well - but only for the "default" K-space lattice with c=1 1 1:
<http://puu.sh/yg4mv/758f9eda76.png>

Changing to a coarser lattice by increasing c>1 causes some peaks to be "missed" by the sampling - this is to be expected. Changing c<1 however completely destroys the result, effectively creating fake reflections at almost every K-space lattice point sampled:
<http://puu.sh/yg4n1/23f84d512b.png>

It is my understanding (which may be lacking...), that sampling points corresponding to angles of incidence that do not cause constructive interference should just cancel out (i.e. sum up to 0 in the final binning), so the final result should always be consistent?

The main consequence of this issue is that using a manual K-space lattice to compare different results is simply impossible, but the documentation clearly states that this should work, for example writing something like "c 0.05 0.05 0.05 manual".

Am I doing something wrong (probably) and if so, what?


Kind regards,

Sebastian

--
M.Sc. Sebastian Hütter
Otto-von-Guericke University Magdeburg
Institute of Materials and Joining Technology
Universitätsplatz 2
D-39106 Magdeburg
Tel.: +49 (0)391-67-54566
Fax: +49 (0)391-67-44569