Re: [sdpd] Self-calibration (method of harmonics)

[Robin Shirley <R.Shirley...@surrey.ac.uk>]

Dear Irini (and others who may be interested)

Specimen displacement is often the major cause of apparent 2Theta zero
error (for which the correction is "Z2Theta"), especially for
diffractometer data collected in para-focusing mode.

However, specimen displacement errors and genuine zero-point errors have
different 2Theta dependencies, so that their effects become increasingly
distinct as 2Theta increases.

The differences between them remain small below 2Theta=90 degrees, but 
become visible in modern high-precision laboratory datasets.

Self-calibration (="method of harmonics"), for Z2Theta and for specimen
displacement error (but not both simultaneously), is offered in the main
CRYS front-end component of my Crysfire 2000 suite.

This is distributed free for non-profit use from the CCP14 website at
Daresbury (webmaster Lachlan Cranswick), where you will find tutorial
examples and instructions for downloading the code :

   http://www.ccp14.ac.uk/tutorial/crys/

Direct downloads (as a 2.2MB zipfile) can be made from:

   http://www.ccp14.ac.uk/ccp/web-mirrors/crys-r-shirley/
or
   http://ccp14.sims.nrc.ca/ccp/web-mirrors/crys-r-shirley/

To activate the self-calibration dialogue, use command SC after entering
your observed pattern, either by loading it from disk (LO) or typing it
in (OB).

For simplicity, only second-order relations are supported by CRYS, since
these are more useful for most non-trivial powder-indexing problems.

Higher-order relations are supported by Cheng Dong's PowderX program, 
using a different (but equivalent) algorithm for Z2Theta.  These are 
particularly useful for high-symmetry inorganic materials for which such 
higher diffraction orders are more likely to be observable.

The theory (for Z2Theta) is discussed in my 1980 paper:

Shirley, R. (1980), "Data accuracy for powder indexing", in "Accuracy in
Powder Diffraction", ed. Block, S. & Hubbard, C. R., NBS Spec. Publ. 567,
361 382.

The extension to specimen displacement errors was presented in my poster
at the ECM-19 meeting in Nancy, August 2000, and a paper on this is
currently in preparation.

Best wishes

Robin Shirley

--------------------------------

To:            <sdpd...@egroups.com>
From:          "Mastoraki Eirini" <mastorak...@iesl.forth.gr>
Date:          Thu, 11 Jan 2001 13:43:02 +0200
Reply-to:      sdpd...@egroups.com
Subject:       Re: [sdpd] more info on the method of harmonics !!!


Dear Dr.Le Bail,
I've read the FAQs regarding the method of the harmonics but we still
haven't figure out what really happens. For instance, how do you combine the
d-spacing of 12.6 with the alpha cell parameter a=12.6A?  Are these (h00)
'oriented' reflections already exist in an imaginary lets say pattern?  How
am I going to combine some other series with the b/c parameters? And what
about the space group? It is still unknown. Do I have to have exact x/2,x/3,
x/4 reflections or an error of 0.02 deg is tolerable?
Thank you for the patience of reading this mail,
best regards,
Irini