dkriegner
commented
4 years ago I am afraid the behavior you describe is how the code works (at the moment). The names of reflections in PowderDiffraction can change when the lattice parameters are different. I, however, want to point out that always an equivalent set of Miller indices is used. For spacegroup 160, 212 and 3-12 are equivalent. See Crystal.equivalent_hkls((2,1,2)).
To get what you want I see two options: 1) define a series of PowderDiffraction objects as you do and then check every equivalent of the Miller indices you look for. This should work but might be cumbersome. 2) define only one PowderDiffraction object and change the lattice parameters in a loop after the definition and recalculate the powder lines after every change. In this way the names of reflections stay the same since the naming is only done once (during initialization of the object)
To change for example 'a' do
p.mat.a = 5 p.set_sample_parameters() p.update_powder_lines()
'p' must be your PowderDiffraction object.
I would not recommend to try to access the reflections via q-range. This will be confusing if you change lattice parameters.
On Mon, 10 Aug 2020, 19:43 Pej-ecp, notifications@github.com wrote:
I am interested in the effect of the rhombohedral angle on the intensity of some super-structure peak. I have therefore defined my crystal (sp.group R3m, #160) in hexagonal setting, and made a loop to define crystals (with xu.materials.SGlattice(160, a_H, c_H), atoms=[...], pos=[...], occ=[...])) with the values of the lattice parameters a_H and c_H calculated from a_R (held constant) and alpha (varying between 90 and 89 degrees). I then generate the PowderDiffraction pattern for each with xu.simpack.Powder(...) and create a list of [alpha, xu.simpack.PowderDiffraction(...)] So far so good.
When I try to plot the evolution of the intensity of a given reflection vs alpha, say (212) I get the error message: KeyError: (2, 1, 2) because they key is absent from some patterns (but not all, it does exist in some as verified with individual prints of each PowderDiffraction element).
On further inspection, it turns out that the chosen Miller indices for (2 1 2) are (3 -1 2) in some patterns. The Miller-Bravais indices corresponding to (2 1 2) being (2 1 -3 2), the equivalent directions should be (-3 2 2) and (1 -3 2), not (3 -1 2). This happens for several reflections, e.g. (0 3 0) becomes (3 0 0), (2 1 8) becomes (3 -1 8), etc. but, strangely some reflections remain unchanged ex: (4 -1 4), (4 3 2), (6 1 4)... as if the reference frame was changing (from (a_H, b_H, c_H) to (-b_H, -a_H, c_H) for some reflections.
I therefore have two questions: 1/ is this a bug or I am missing something ? 2/ is there a way for me to access these reflections other than by hand (e.g. through a selection of q or 2theta range) ?
Thanks a lot for your help (I can pm a python script or minimal working example if needed)
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/dkriegner/xrayutilities/issues/104, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACKZJFKWNM3PENKV6F27KE3SAAWVBANCNFSM4P2GDZFQ .
Pej-ecp
I am interested in the effect of the rhombohedral angle on the intensity of some super-structure peak. I have therefore defined my crystal (sp.group R3m, #160) in hexagonal setting, and made a loop to define crystals (with
xu.materials.SGlattice(160, a_H, c_H), atoms=[...], pos=[...], occ=[...])) with the values of the lattice parameters a_H and c_H calculated from a_R (held constant) and alpha (varying between 90 and 89 degrees). I then generate the PowderDiffraction pattern for each withxu.simpack.Powder(...)and create a list of[alpha, xu.simpack.PowderDiffraction(...)]So far so good.When I try to plot the evolution of the intensity of a given reflection vs alpha, say (212) I get the error message:
KeyError: (2, 1, 2)because they key is absent from some patterns (but not all, it does exist in some as verified with individual prints of each PowderDiffraction element).On further inspection, it turns out that the chosen Miller indices for (2 1 2) are (3 -1 2) in some patterns. The Miller-Bravais indices corresponding to (2 1 2) being (2 1 -3 2), the equivalent directions should be (-3 2 2) and (1 -3 2), not (3 -1 2). This happens for several reflections, e.g. (0 3 0) becomes (3 0 0), (2 1 8) becomes (3 -1 8), etc. but, strangely some reflections remain unchanged ex: (4 -1 4), (4 3 2), (6 1 4)... as if the reference frame was changing (from (a_H, b_H, c_H) to (-b_H, -a_H, c_H) for some reflections.
I therefore have two questions: 1/ is this a bug or I am missing something ? 2/ is there a way for me to access these reflections other than by hand (e.g. through a selection of q or 2theta range) ?
Thanks a lot for your help (I can pm a python script or minimal working example if needed)