[sdpd] The things "user" beamlines can achieve?

["L. Cranswick" <L.M.D.Cranswick...@dl.ac.uk>]

> >Is this a flaw in the synchrotron method that
> >experiments and data collections are assumed to
> >work flawlessly the first time?
> 
> Recording a powder pattern at room temperature
> works generally flawlessly the first time. 
> But brillant patterns of bad samples are of little use.
> Unpublished results are also of little use for the
> community. You gave an interesting number of
> several thousands of single crystal data collected
> those last years at the 9.8 Daresbury station, and
> solved. How many were published 2 years later ?
> Can we expect to see a new winner soon in the CSD 
> list of authors http://www.cryst.chem.uu.nl/s100R ?-).

The above is correct with respect to the statement in my 
talk notes: that an estimated 1000 to 1500 had been 
"solved" on the 9.8 single crystal beam-line in its
roughly 3.5 year lifetime.

What has been published is at:
  http://srs.dl.ac.uk/XRD/9.8/Publications_index.html?subject=Publications

Beamline 9.8 run by Simon Teat is unlike any other
"user" small molecule beamline I have encounted in that 
it was seriously designed as a "user" beam line that
users could use with minimum of setup and 
hardware generated frustration!  To achieve this
it uses a standard off the shelf integrated diffraction 
system - a Bruker SMART CCD - on a "dedicated" station.  
This means it is very similar to using a laboratory system: 
arrive - put on a crystal - press the button using familiar 
software.  If your crystal is any good 
you will be collecting excellent quality 
data with the odds of solving the structure before
the data collection is finished.    If crystal is bad -
mount next crystal.
It is a "user" beamline, not a collection of
custom made equipment (potentially not well integrated)
that users are allowed to bumble along with on a 
very limited time frame - potentially spending most
of the beam line setting up/problem solving - rather
than collecting good data.

The collection and solving of a micro crystal of
tetracycline hydrochloride by Bill Clegg and Simon 
Teat to help validate the Powder Round Robin results
was a routine endeavour.  Extrapolate from the
tetracycline example (1998), and 1000 to 1500
solved crystal structures is not unreasonable.

A "user" beamline like 9.8 makes it practical for 
chemists and materials scientists to routinely 
get very productive results from synchrtron radiation.
Though as this synchrotron usage may be part of a
project with wider aims than "goto synchrotron,
collect data" - this may explain a little why
only ~82 publications have been registered?

-----

Many of the problems you hear of with old
and new "state of the art" diffraction beamlines
is they are custom monstrocities of equipment from
different vendors (or custom made) that do not
really interact that well.  The usual definition of
"state of the art user beamline" seems to be that of
getting better resolution, intensity or prettier
diffraction plots compared to a rival down the road - 
where the concept of it being "usable" by users 
and providing reliable results is not seriously 
considered.

It would be interesting to hear opinions what
might be possible for high quality "user" powder
diffraction beamlines by putting standard commercial
powder diffractometers such as a Siefert, Scintag, Bruker,
Philips, Rigaku, Stoe, etc - with their sample
automation and software control?  

I would wager that for the synchrotron applications
of micro-diffraction powder diffraction beamlines
for spatial mapping applications - Bruker or
Rigaku microdiffractometer have been totally ignored
despite good odds that stuck on a synchrotron - they
would give the equivalent of 9.8 in terms of
easy to obtain high quality data?

Thus a summary of the above rant is that a valued
"state of the art 'user' beamlines" for some
powder diffraction applications might be better
achieved by following the method used in
developing 9.8 - using standard commercial
"integrated" laboratory diffraction systems run
by a very capable station manager.  Not custom
monstrocities where major amounts of user time
is spent on "setup" and hardware problem solving  - 
with the potential for achieving little during
the few hours/days of beamtime but accelerated 
hair loss independent of the radiation hazards.

Lachlan.

PS:  An obvious rebuttal to the above would be to
point to the BM16 powder diffractometer at ESRF.

 http://www.esrf.fr/exp_facilities/BM16/handbook/handbook.html

 http://www.esrf.fr/info/science/highlights/1999/chemistry/powder.html 

Though a common comment I hear is that the BM16
strategy of multiple detectors is not affordable on
beamlines that would like to do this?  
If correct, this would be an extra advantage of 
using affordable commercial integrated diffraction 
systems on beam-lines?

-- 
Lachlan M. D. Cranswick

Collaborative Computational Project No 14 (CCP14)
    for Single Crystal and Powder Diffraction
Daresbury Laboratory, Warrington, WA4 4AD U.K
Tel: +44-1925-603703  Fax: +44-1925-603124
E-mail: l.cranswick...@dl.ac.uk  Ext: 3703  Room C14
                           http://www.ccp14.ac.uk