From:   GORDON::COCKCROFT    "Jeremy Karl Cockcroft (Tel. 0171-631-6800)" 17-OCT-2000 18:15:08.08
To:     SMTP%"sprokic@yahoo.com"
CC:     COCKCROFT
Subj:   RE: Project, step 4

Dear Sinisa,

>It is strictly said in the step 4 of our project that we should NOT try
>to refine in the group Pbnm the structure obtained from previous three
>steps with the whole range of data set. In that case I find I have a
>problem to claim that the best possible result obtained through
>refinement with one of two lower symmetry groups is better then the one
>that could have been obtained from a refinement with Pbnm. For example,
>it is obvious that one of the main points of refinements is improvement
>of the peak shapes at higher angles, and this applies both to Pbnm-case
>and the remaining two groups.

>Should we neglect that?

The point about not refining the parameters when using the whole data set
is to visually see the difference between the fit based on just 8 to
62.5 degrees versus the fit between 62.5 and 143 degrees.
I hope it is obvious that the while the low angle data fits reasonably well,
the high angle data does not.

There can be several reasons for this which you are expected to comment on.
[You could test out some of these reasons (e.g. resolution function/peak
shape) by then refining the model in Pbnm using the whole data set.]

The idea is that you then test out one of the possible reasons, which is that
the structure may not be orthorhombic still at low temperature.
This is step 5, etc.

I am not sure whether this answers your question or not. If not, please
come back to me with further questions.

With best regards,
Jeremy Karl.

Back