Dear Martin,
What is called "texture" in the link you've provided actually is a
"bad powder average". Texture sounds to me as a rather general term,
while "preferred orientation" is its particular case. Texture is
the distribution of crystallographic orientations in a
polycrystalline sample, it does not need to be uniaxial like
"preferred orientation", and it does not need to be only due to a
more-or-less random distribution of intensities from a few
randomly oriented particles. The latter effect should be better
called "bad powder average".
IMHO, the link you've provided pretends to revise the terminology.
I wonder, what "texture analysis" means according to it...
Best regards,
Yaroslav
http://filinchuk.com
===8<==============Original message text===============
Any data from anywhere will lead the gullible and unwary up a certain
creek without a particular implement and flat-plate is better than
transmission for doing so. Older saying still (ca. 1503-07), "thar be
Dragons".Martin PS For anyone interested in an explanation of texture
vs. PO, see:
http://pd.chem.ucl.ac.uk/pdnn/inst1/texture1.htm
http://pd.chem.ucl.ac.uk/pdnn/inst1/texture2.htm
PPS for anyone interested in the TV ad that came to mind during
this discussion, see:
http://www.nationalarchives.gov.uk/films/1964to1979/filmpage_lonely.htmMartin
Subject: RE: Preferred orientation?Date: Fri, 16 May 2008 13:05:24 -0400From:
[EMAIL PROTECTED]: [EMAIL PROTECTED]: rietveld_l@ill.fr
I'm not really arguing with you here as I prefer capillary data myself as it
gives better data in many circumstances - however it can sometimes lead you up
the garden path (another old saying!). I suppose what it boils down to is that
needles are a pain as they can orientate whatever you do them (reflection or
transmission).
The texture versus preferred orientation difference has some signficant blurry
edges from a practical point of view.
Anyway - I'm on holiday so I'm going to put my brain to sleep and go and do
some gardening!
Pam
From: Martin [mailto:[EMAIL PROTECTED]: Fri 16/05/2008 11:47 AMTo: Whitfield,
Pamela; [EMAIL PROTECTED]: RE: Preferred orientation?
Hi, happy RietveldersSome elements of confusion creeping in here. I think you
said, Pam that transmission wont help much if it's wollastonite and what I'm
saying is that it does and gave a pointer to a study that shows it. Indeed I
don't claim transmission gets rid of PO either, but it does reduce it hugely
which, if one reads back, is my claim here. The PO function in this case is
merely to illustrate the point: 0.9 and a refined model vs. 1.6 and a bad
refinement. The merits of various PO functions aren't important when it's the
data that really count (or the counts that count, if you like). Why start out
with bad data in the first place? As my old dad says, you can't make a silk
purse out of a pigs ear. regards, Martin PS Just to clear up another possible
point of confusion: large particles lead to texture effects, not preferred
orientation.
Subject: FW: Preferred orientation?Date: Fri, 16 May 2008 10:43:21 -0400From:
[EMAIL PROTECTED]: rietveld_l@ill.fr
I don't remember saying that reflection would work well with wollastonite only
that a capillary won't get rid of the orientation (or at least that was my
intention). As always this is going to vary from sample to sample, i.e. how
large the particles are, aspect ratio of the particles, the diameter of the
capillary, pure or mixture, etc. Orientation with needles is going to be more
of a problem in a 0.3mm versus a 0.8mm capillary. Particle statistics are also
a potential issue with samples like this. If they are large enough to
orientate badly then the crystallites are probably large and the capillary will
do a better job with the statistics
The MD correction doesn't work very well for alot of these systems, SH is
better as long as the correlations don't get out of hand (which they can quite
easily). However 0.9 is still significantly orientated so it doesn't get rid
of it by any means. With platey particles you can pretty much eliminate the
preferential orientation with a capillary versus flat plate with significant
impact of quantitative analysis results (paper published in Powder Diffraction
a couple of years ago).
I do have reflection data from 400 mesh wollastonite (albeit with MoKa from a
high pressure gas cell) which can be fitted quite nicely with SH (lousy with
MD). The quantitative analysis results from the carbonation are good enough to
extract a rate constant which suits me nicely. The additional penetration of
the MoKa should help with the stats in this case even if transparency becomes a
problem - you can't win eh?
Unfortunately it's quite difficult to completely decouple orientation,
statistcs and microabsorption as they are all related to size.
Pam
From: Martin [mailto:[EMAIL PROTECTED]: Fri 16/05/2008 10:04 AMTo: Whitfield,
Pamela; [EMAIL PROTECTED]: RE: Preferred orientation?
In fact I think you might find it helps quite a bit. Have a look at:
http://img.chem.ucl.ac.uk/www/vickers/po/po.htm Martin
Subject: RE: Preferred orientation?Date: Thu, 8 May 2008 10:55:12 -0400From:
[EMAIL PROTECTED]: rietveld_l@ill.fr
I do that myself but it doesnt always help much if youve got something like
wollastonite! J
From: Martin [mailto:[EMAIL PROTECTED] Sent: May 8, 2008 10:51 AMTo: [EMAIL
PROTECTED]: RE: Preferred orientation?
Forget all that long winded stuff. Just collect the data on capillary
transmission geometry and avoid all (well, most of) the fuss. Martin Vickers
Get fish-slapping on Messenger! Play Now
Get fish-slapping on Messenger! Play Now
Get fish-slapping on Messenger! Play Now
_________________________________________________________________
http://clk.atdmt.com/UKM/go/msnnkmgl0010000007ukm/direct/01/
===8<===========End of original message text===========
