I stood up last Tuesday night at the AMMRL meeting in Orlando and
volunteered to compile a more relevant and complete list of demo tests for
NMR instrumentation. Yes, it took Bob Dudley's message to get me going,
but now I am ready.
Specifically, I am talking about:
COMPILING A NEW GROUP OF STANDARDIZED TESTS TO ALLOW MORE OBJECTIVE
ASSESSMENT OF NMR INSTRUMENTATION CAPABILITIES.
[Hit delete at this point if this does not interest you in the least...]
Here is a good example to consider:
When you are running a sample of 0.1% ethylbenzene, wouldn't it make more
sense to limit the total volume of sample as well to prevent a s/n hungry
probe manufacturer from making a coil the size of Arizona? Obviously,
this only has true value when one is solubility and not sample limited.
The much more often encountered case is that of being only able to make
up 500 or 700 ul of a 1.5 mM sample of some incredibly precious material,
and in that case people always complain about the dilution required to
obtain narrow lines (so the entire sample volume in the coil can be
properly shimmed).
Here is another example:
For a broadband probe, most everyone just quotes the 13C sensitivity, even
though the probe might tune between 15N and 31P. I understand (but have not
seen this first hand, so this can be viewed as hearsay) that certain broad
banded probes are optimized to maximize 13C sensitivity at the expense of
other nuclei far, far away. This has alarming implications when one
considers that the inorganic chemists out there often want to look at
something besides 13C...not to mention the biochemists who for some reason
might be concerned with making a direct observation of 15N...
There are other examples I could cite, but I hope I have made my point,
which is: IT WOULD BE NICE IF THERE WAS AN AGREED-UPON SET OF EXPERIMENTS
THAT PERSONS COULD REFER TO IN THE DEMOING OF CONTEMPLATED INSTRUMENTATION.
Just think of how you might be able to just refer to someone else's data
so you could save yourself the trouble of going to Billerica, Fort Collins,
Palo Alto, or Peabody (trying to be fair here...notice the presentation in
alphabetical order). Not that this is always a good thing, but I think one
can always justify a trip just so one can see a working spectrometer in
action...
Alright, time to cut to the chase. Here is what I think might be
appropriate:
Standards based upon particular concentrations in particular volumes (for
solids, I would just say particular amounts for a given rotor size).
Maybe even multiple sample volumes could be done.
Tests of line shape, (50%, 0.55%, 0.11%), spinning side bands at a particular
speed (e.g., 20 Hz), single shot sensitivity with no lb, no zero filling
(make them collect noise for 20 seconds if they want the digital resolution).
I imagine that we need this for every nucleus that people are interested
in - at the bare minimum, this is 15N, 13C, 31P, 1H.
Practical tests of rf stability (i.e., in the 2-5 mM range) would also be
nice. In contrast, I note that Varian has an incredibly elaborate procedure
worked out for testing the rf stability. When it was demo'd to me, my eyes
glazed over in about ten minutes (it went one for about an hour as I recall).
Something perhaps a wee bit more accessible, like picking some standard
tripeptides and collecting 1H-13C HMQC & HMBC, and doing an 15N-1H HSQC might
be good. All parameters need to be spelled out in detail to prevent
inappropriate massaging of the data.
T1rho experiments might be useful to test the pulse reproducibility, if a
sample doesn't have a decent exponential decay, then maybe the rf ain't so
good, right? I suppose T2 experiments would be good as well.
And how about selective excitation, and gradient reproducibility, and
NOE difference experiments (the chemists using my lab really love this
experiment for some reason).
IF THIS SOUNDS AT ALL WORTHWHILE, I WOULD LIKE TO HAVE PEOPLE SEND ME
THEIR IDEAS OF WHAT TESTS WE MIGHT AGREE UPON, AND I WILL COMPILE THIS
INFORMATION, POSSIBLY ASK THE AMMRL TO VOTE ON THE BEST OF REDUNDANT OR
OVERLAPPING SUGGESTIONS, AND THEN FREELY DISTRIBUTE THE AGREED-UPON LIST
OF DISCERNING EXPERIMENTS (I am hesitant to call these true standards) TO
ANY WHO REQUESTS IT (OR I CAN POST IT TO THE LIST SERVER, BUT IF IT IS LONG
IT WOULD BE BETTER IF I DID NOT KILL RUDI'S MACHINE BY DOING THIS.).
I will interpret silence as indifference and try not to go away hurt, so
if you care, please contribute. If you think that you might be buying
a new spectrometer or even a new probe in the near future, then maybe this
will be worth your while.
Jeff
Dr. Jeffrey H. Simpson jsimpson@mit.edu
Director of the Spectrometry Laboratory 617/253-1812 office
Department of Chemistry, 18-082 617/253-1806 lab
Massachusetts Institute of Technology 617/253-0873 fax
77 Massachusetts Avenue, Cambridge, MA 02139-4307 USA
"My curve ball's so sneaky, he might be your Daddy." -Satchel Paige