Many thanks for all of the replies to my query concerning temperature
calibration some time ago. I have taken a couple of minutes to summarize
them for those who are interested ...
----------------------------------------------------------------------
It's my understanding that you can change the temperature calibration on
most Bruker probes by adjusting the thermocouple position. You might not
have the new thermocouple in the same position as the old one?
Your Methanol capillary tube sounds like a good idea, but I haven't seen it
done here.
Regards,
Jerry
-----------------------------------------------------------------------
We have in new probe heads the "T"-thermocouple (what ever this mean)
and big differences between the temp display and the shift measurements
with the BRUKER sampels (methanole, glycol , but not neat, there is any
lock sovent within).
The difference is about 10 degree at 100 grd Celsius, and similar at
-100 grd Celsius.
If we plot the "shift"-temperature vs. "display"-temperature we got a
straight line (what means: linear increasing differnce between the to
temperatures, or, other spooken: the slope of the built-in thermocouple
is different from that of the shift-measurements).
I rely on the shift measurements.
A Bruker engenier sad to me that the thermocouple is possilble worse
located in the probe head (to big or short distance from the sample
buttom) - but I never tried to proof.
----------------------------------------------------------------------
The methanol or glycol calibration curves given in the VTU manual gives
the theoretical chemical shift differences for methanol and glycol as a
function of temperature as could be obtained if you calculate your self the
chemical shift differences against temperature.
With some thermocouples you may obtain a nearly same slope and bias as
found on the theoretical curve. Nevertheless, with some other thermocouples
the slope and the bias can be different from the theoretical curve. Thus you
may use the obaserved chemical shifts to obtain your own calibration curve
for this thermocouple.
As far as the slope and bias of the temperature calibration curve may be
function of the chosen maximum power, the manner in which the PID has been
done as well as function of the gas flow, if you wan't to know exacly your
sample temperature the best thing to as you suggest is to use a capilary
containing a substance whose chemical shift or chemical shift difference is
overlapping of the calibration substance peaks with the peaks of your
compound and that no reaction takes place between them.
Sincerely yours,
Dr. Philippe LUX
Bruker R&D.
----------------------------------------------------------------------
> On a related note, has anyone ever used a methanol-filled capillary
>tube as an internal temperature standard? It seems to me that if the
>methanol lines don't interfere with the observed spectrum that it might be
>a neat way to measure the actual temperature of the sample.
Yes, from personal visits I know that Kuchel's group in Sydney
uses both methanol and ethylene glycol capillaries, depending
on the temperature range to be covered, with great success.
See, e.g., Potts and Kuchel. 1992. Biochem. J. 281:753-759.
Siegfried Schoberth
----------------------------------------------------------------------
I read your BUM posting and have a few comments:
First if you got the methanol calibration curve to agree well with the
original thermocouple you were lucky. Second, was the second thermocouple
of the same type as the first one (if it is not you will have to change the
programming of the temperature controller as well as perhaps using
different jacks at the back of the controller in connecting the
thermocouple). Third, the positioning of the thermocouple within a mm or so
can have a very large difference on the calibration. I put a note into the
NMR newsletter about this in 1996 (see 453-27). I would trust the methanol.
----------------------------------------------------------------------
The only thing I can think of is that the thermocouple is not inserted
into the probe to exactly the same length as the old one; but it seems
like this is something you would've checked....
Tom Pratum
Dept of Chemistry
Box 351700
Univ of Washington
pratum@u.washington.edu
http://weber.u.washington.edu/~pratum
----------------------------------------------------------------------
I am surprised that you you have reproduced (at least once) the Bruker
curve, because a different slope is a rule.
If you are using the same hardware (probe, VT-unit, precooling unit(?)) as
in August, check the the following:
The main problem is the position of the thermocouple. Is the lenght of the
new thermocouple excactly the same as the old one? Is it inserted properly?
Otherwise you are measuring the temperature of the surrounding teflon.
If the air/N2 flow rate is different than before, the slope will be
different. Check also the air flow connection in the probe.
Did you wait the same time in every point as in your previous calibration
measurement? Different stabilization time will also produce a different
slope.
Small effects (usually below 1K): sample position, spinning-nonspinning,
heating-cooling
Petri Ingman
University of Oulu
Oulu, Finland
----------------------------------------------------------------------
I don't think the thermocouple should change things via voltage changes.
Look to make sure the thermocouple is in the identical placement in the
probe; 1mm shorter or longer can make a difference! Also check that the
gas flow is the same.
I would be interested in the comments you get back about methanol
capillaries. I suspect that susceptibility shifts from the tube may
cause trouble, but perhaps someone's worked this out. If you could send
a summary, that would be great.
Good luck,
Charlie Fry
----------------------------------------------------------------------
If the tip of the new TC is not in the same exact position (ie. in the air
stream)
then the slope could change. Also I would think that the MeOH capillary
would work but give it a longer equilibration time??
regards
Nick
--------------
Dr Nick Burlinson
Chemistry Dept
University of British Columbia
Vancouver BC Canada
V6T 1Z1
Email nick@unixg.ubc.ca
---------------------------------------------------------------------------
Keith Brown
University of Saskatchewan
Saskatoon, Saskatchewan
---------------------------------------------------------------------------
Pinky: What shall we do tonight, Brain?
Brain: The same thing we do every night, Pinky ... try to take over the
world!