Hi everyone! I would like to know if there are any quidelines or literature on how to set up and run a quantitative 1H-NMR experiment on a 400MHz Bruker NMR. What commands do I use and how do I process the results?
Thank you!!
Enter the Manuals section in topspin (or type "docs" in the command line). In the "Analysis and Simulation" section, you will find a manual (or rather a presentation) called "Quantitative NMR". I have, in any case, attached the file to this answer.
In principle, you just have to set up a normal 1H spectrum. You add an internal standard (some known amount of substance), and can then calculate the contents of the analyte either on your own from the integrals or use the "nmrq"-command to do that for you (which is the better way to do this).
Christian Adam and Jean-Marie R Peron, thank you so much for your help! I will try your suggestions and see how it goes!
I'll talk generically, because I do not have much practice with Bruker software. Complementing Jean M. Peron comments, it is important to adjust the acquisition parameters to the type of molecule to be analyzed, and one of the parameters and interval between pulses (delay). For molecules of low molar mass, it is good to use a delay of 15 to 20 seconds. In the case of macromolecules (eg polymers) a delay of 10 seconds works well. It is important to keep the temperature of the probe about five degrees Celsius above the laboratory room temperature. Do not forget to make the optimum shimming. Get a good signal-to-noise ratio and in the process use a fourier number in the 64K range, exponential (or line broadening) equal to zero, phase adjustment and baseline correction. Best regards
The most foolproof, but longer way is to first measure your T1 relaxation time. Then use the longest T1 in your sample (except the solvent signal) to set the relaxation delay D1 (D1 = 5 or 7 x longest T1).
Dear Ioanna,
what I do (if I have enough substance in the NMR tube) to overcome the T1 problem is to acquire a spectrum with a 90° pulse (manually determined as 360° pulse) with a single scan and no dummy scan.
Note: between this scan used for quantification and the previous scan on 1H (for example from receiver gain adjustment, pulse determination...) I wait 60s. Which seems to be long, but is adequate for low-MW solvents.
I hope this helps
Alfred
Perhaps the first thing to decide is how accurate you need the result. It is very easy and quick to get a +/- 10% result. If you need greater accuracy then you will need to be very careful about how you achieve it. The use of a 90 degree pulse is good if you are using an external reference (e.g. ERETIC 2 or QUANTAS). If you are using an internal standard I would be tempted to use a 30 degree pulse and 30 second recycle delay (depending on the nature of your sample and choice of internal standard).
If you are using an internal standard (as suggested by Christian) then you need to select a suitable material. Do not use TMS - it has a long relaxation time and is also volatile so is inaccurate.
Lastly, if you are trying to get to a high level of accuracy, you will need to consider how you integrate the peaks. In my experience, peak-fitting is the only reliable way. Conventional integration is too prone to operator influence (in one experiment we noted as high as 20% difference between independent operators working on the same data set!).
QUANTAS reference: https://doi.org/10.1002/mrc.2647
https://doi.org/10.1002/mrc.2647