Friday, August 29, 2008

The NMR Time Scale

Many people use statements like, "It is fast on the NMR time scale" without really having an understanding of what the "NMR time scale" means. Assume we have a compound with two inter-convertible conformations, A and B, each of which gives a different NMR spectrum. If the interconversion between A and B is very slow then we expect to get a spectrum with both conformations A and B resolved. If the interconversion is very fast then we would get an unresolved spectrum which represents the average conformation of A and B. The interconversion between A and B is fast on the NMR time scale if it occurs at a rate much greater than the difference in frequency between A and B. Since the difference in frequency between A and B depends on the Larmor frequency of the nucleus being observed and the strength of the magnetic field, the "NMR time scale" depends on the particular experiment and the field strength. Therefore, when one uses the expression, "NMR time scale" one must qualify it with details of the measurement. An example of different NMR time scales is shown in the figure below. Here the 1H (300 MHz) and 13C (75 MHz) NMR spectra of the N-methyl groups of N,N-dimethylacetamide are shown at various temperatures. The spectra were plotted on the same scale in Hz. In this molecule there is a hindered rotation about the central C-N bond. One can see that the rotation is fast on the 1H NMR time scale before it is fast on the 13C NMR time scale because the difference in frequency between the methyl groups is smaller in the 1H spectrum compared to the 13C spectrum.

18 comments:

Anonymous said...

Thanks! You cleared that up for me

Anonymous said...

I have an examination in NMR spectroscopy tomorrow. Your explanation was very useful and clarified the meaning of the term 'NMR time scale'! I'm so greatful!!!

Anonymous said...

Still I did not understand.In general it is okey. Which parameter indicates the the NMR time scale? AQ:Aacquisition time, pulse duration (P1) or what??
please help...
ozguralver@anadolu.edu.tr

Glenn Facey said...

Anonymous,

It is whether or not the frequency difference between the reonances is comparable to the frequency of interconversion between A and B.

Glenn

Anonymous said...

Dear Facey,
Lets say there are 2 peaks for one carbon because it goes some reorientataion and the difference between these 2 peaks is 20Hz. Can we say it jumps one from to another structure 20 times per second? If say "yes", I want to ask this where the problem begins:

if it is fast enough we probably observe one peak so we say that intercoversion is fast compared to NMR time scale so we observe one peak. Then it is meaningles to say "yes" for my first statement because there will be no seperation for the second case even it changes its structure more faster than 20Hz....


ozguralver@anadolu.edu.tr

Glenn Facey said...

Dear Anonymous,

No you cannot say that the jump rate is 20 times per second. You can however say that it is much slower than 20 times per second. You can however get an estimate of the jump rate at the coalescence temperature.

Glenn

Unknown said...

Dear Glenn,
The problem with anonymous on March 5, 2010 is that he/she probably did not understand why we see a single peak in an exchange situation. The single peak represents neither of the conformations A and B, rather it is an weighted average of the resonance frequencies of the two (this should be clear from the figure shown). Actually a nuclei attached to any anisotropic molecule feels different extent of the applied magnetic field at different orientations relative to the field direction and hence resonate at different frequencies. In the liquid (and gaseous) state where the rate of tumbling of the the molecule is very high (of the order of THz) we see a single sharp line, which represent an average resonance frequency from all possible orientations.
M. Habibur Rahman

Glenn Facey said...

Dear Habibur

thank you for your comment.

Anonymous said...

Thanks for this post.

There are still some things that are unclear to me.

What is the time scale if you see only a single peak (say in metal)?

I have people refer to the timescale being the inverse of the larmor frequency. Is that true?

Anonymous said...

To follow up on what I asked before, I want to clarify that I was asking what the NMR time scale is if there is no interconversion.

Sorry for the confusion

Glenn Facey said...

Anonymous,

When one refers to a time scale relative to the inverse of the Larmor frequency, one is usually talking about relaxation processes as T1 and T2 depend on spectral density functions which have minima close to the reciprocal of the product of the Larmor frequency and a motional correlation time.

Glenn Facey said...

Anonymous,

When there is no interconversion the concept of an NMR time scale as described in this post has no real meaning except that the conversion is infinitely slow on the NMR time scale.

Anonymous said...

Dear Glen,
Say, the 7Li NMR spectra has FWHM of 10kHz. can I estimate jump rate from spectral width ? if yes, can you explain me how ?

Glenn Facey said...

Anonymous,

7Li is quadrupolar and the line width may be broadened due to fast relaxation in addition to any effects due to exchange. As a result, I don't think you can extract an exchange rate directly from the width of the signal.

Glenn

Anonymous said...

Dear Glenn,
In my NMR course it is stated that the NMR timescale relates to the reciproke of the frequency of the spectrometer. I.e. for a 250 MHz spectrometer this would be 4 ns. So it is stated that rotation must be faster than this to result in equivalence. I was confused by this since later coalescence of OH protons is explained by fast proton exchange. But I had a hard time believing that this exchange occurs on nanosecond timescales, though I don't know for sure. Here a different statement is made, ie the relevant factor is the difference between the peaks of the alternate states, which brings it to timescales in the micro/-millisec regime. Is what I've learned than flat out wrong?

And second question:
How do you relate these timescales to the actual NMR measurement?

Thanks in advance!

Glenn Facey said...

Anonymous,
The NMR time scale is "related" to the field strength only in that the strength of the field affects the frequency difference between peaks. The NMR time scale is NOT equal to the reciprocal of the proton resonance frequency. I'm not sure what you mean by the question "How do you relate these timescales to the actual NMR measurement?"

Glenn

Anonymous said...

Thanks for clearing that out! I guess what I mean to ask is, why is it related to the frequency difference between the peaks? Why isn't it possible to measure/ resolve states when the system alternates faster than this?

Glenn Facey said...

Anonymous,
The answer lies in the lifetimes of the states compared to the duration of the FID and the frequency difference between the states. When an exchange event occurs during the measurement, the frequency changes in the FID. The best explanation for this is given in James Keeler's book "Understanding NMR spectroscopy" second edition pp. 286 - 293.

Glenn