Re: Accurate/Consistent Proton Integrations

Bill C Hutton (wchutt@sigurd.monsanto.com)
Wed, 14 Jun 95 08:41:51 CDT

So many factors are involved for reasonable integration estimates
(yes... they are only estimates). Primarily the FT is very poor at
resonance amplitude estimation unless resonances are extremely well
seperated and your spectrometer is perfect... but I digress.

It would be real helpful to know what the pesky post-doc did
differently between the two spectra, how seperated are the peaks, what
their deacay rates are and the freq. domain s:n. Could this sample have
some extremely broad signals? Assuming this system has the dynamic
range to properly record these signals, I doubt this has anything to do
with a Bruker AC console, Bruker SW or anything else related to a
specific vendor.

We had a case where two peaks with different decay rates (by about a
factor of 3) were almost perfectly overlaped. The amplitude estimate
for these peaks (which we believed to be a single peak) relative to a
third standard peak (which really was a single peak) was a function of
the exponential weighting function we applied before the FT. Of course
it is highly unlikely this is what's going on here.

**********************************************************************
W.C. Hutton, Monsanto Company, Corporate Research

wchutt@monsanto.com, 314 537 6021

This message does not necessarily reflect the views of Monsanto Company
or, now that I think of it, anyone else.

**********************************************************************