Can a NMR tube de-resolve a portion of a sample's spectrum?

Recently we received spectra from a customer who was questioning a loss in resolution between two different NMR tube brands using an identical sample for the test.

spectrum1

spectrum2

The data was acquired on a Bruker DRX 500 taken at a temperature of 303 K and spin rate of 20Hz. The compound was from a stock solution containing a chiral, aliphatic, small molecular weight organic compound. The customer also mentioned that a shim file derived from gradient shimming was used and then further hand shimmed on Z-Z5, 8 scans with no line broadening.

Looking at the aliphatic region between 1 and 2.2 ppm, the two spectra seemed identical. However, between 7.35 and 7.4 ppm there was a noticeable difference between the two tubes in resolving the long range H-H coupling of a imine CH peak. It was assumed by the customer that these differences were caused by quality differences between the tube brands.

The customer's concerns were very convincing until they realized that when a NMR spectrum is acquired, the raw data is in the form of a free-induction decay which is carried over the entire length of the spectrum, and converted into a frequency domain function by Fourier transform. Therefore, the entire spectrum is subjected to one resolution constant, not many at various points of the spectrum. Since NMR sample tube quality plays only a partial role in resolution, other effects such as slight changes in the sample's water content, pH effects from the glass, or others, can lead to chemical exchange broadening, which could have caused this customer's problem.

It is important to note that strict quality control standards are used in NMR tube manufacturing. Since variation in concentricity and camber is at its worst ± 0.025 mm and ± 0.075mm in a disposable tube, it is critical not to rule out chemical influences that will effect spectrum resolution.

Click here for useful articles on shimming magnets.