Recently, I have become interested in trying to observe 1H signals
expected to be quite weak and broad in typical liquids probes. Our probes
turn out to have a degree of weak background signal that I had never
noticed before, as I had never looked. This signal would be easily
baseline corrected into oblivion in the presence of normal sharp peaks,
but could be a deal-breaker interference with the signals we would like to
see.
What I get, when I open the sweep width up to a couple of hundred ppm and
turn the receiver gain up high, is a broad, asymmetric peak with a width
at half height of around 8,000 Hz.
Normally, we operate our spectrometers with compressed air that goes
through a desiccant dryer expected to produce a dew point of -40 C, so
it's reasonably dry, but not bone dry. When we do either very high or
very low temperature VT, we switch to boiloff from liquid nitrogen, which
should be extremely dry, but some of the piping/tubing is shared between
the two gas systems. The nitrogen is fed both through the VT and the
spin/air bearing systems (these are Varian instruments).
As a function of temperature, the appearance of this background signal
changed little between ambient temperature and -80 C, using dry nitrogen
gas the whole way. When the temperature was raised to 120 C, the
background diminished visibly over a period of minutes, perhaps to 5% of
its original intensity, but did not completely disappear. Coming back
down to room temperature, though, the signal came back, even though the
probe was still under nitrogen gas.
I would be delighted to receive any suggestions about how to get rid of
whatever is causing the background. It completely goes away with a short
spin echo or diffusion sequence, but these would be expected to attenuate
our desired signals just as effectively. Barring that, I'd be glad to
hear from anyone who has seen this before and knows what is causing it.
I'm just guessing, but I would think 8 kHz is too narrow of a line to come
from e.g. glue holding the coil onto the insert. If it were from adsorbed
water, I might expect more of a change going down in temperature, and if
anything, the signal might be sharper at high temperature, broader at low
temperature. These probes are fairly old, so I suppose they could contain
a very little bit of oil from the air compressor. But there could be
various other possibilities too.
David VanderVelde
Manager, High Resolution NMR Facility, Caltech
davidv_at_caltech.edu
Received on Fri Mar 26 2010 - 13:37:00 MST