Hello,
I have a sample that contains a certain number of organophosphorus compounds, each one of them has an alkyl (P-R bond) or alkoxy side chain (P-OR bond). I would like to perform a rapid screening to know the exact number of those organophosphorus compounds (OPs). Would it be quicker to run 31P{1H} or 2D [1H-31P] HMQC ?
In general, 2D NMR experiments require more time than 1D experiments but in the 2D [1H-31P] HMQC there is polarization transfer from a sensitive (1H) to a less sensitive (31P) nucleus, and knowing that all the OPs that I am looking for contain P-R or P-OR bond I thought that it might be quicker to run the 2D experiment.
Thank you for your comments.
There is no general answer, it depends on what you want to see. For setting up the 2D experiment you need some knowlegde about your compounds (I doubt whether it will provide results since coupling constants in P-OR groups may be rather small). If you have reasonable amounts of substance and if you can achieve sufficient concentration, it's safer to record 1D spectra: it is less likely that you will miss any of your compounds.
Wolfgang Baumann
Thank you for your answer sir.
The typical J coupling for P-OR is 3JPH= 7-11 Hz and 2JPH= 15-17 Hz for P-R. The samples have a concentration of 5-10 ppm. Taking this information into consideration, do you maintain that it would be safer to run 31P{1H} NMR ?
... I never worked with concentrations in the ppm range so it is really difficult to predict which experiment will deliver results more quickly. The problem remains: With 2D experiment (indirect detection) you might miss a compound due to inappropriate parameter settings, with 1D experiment (direct detection) due to restricted sensitivity. Only a real experiment and a comparison of both methods might deliver the answer for your specific system, I'm afraid... However, phosphorus is not a really "insensitive" nucleus.
What field strength? At higher field strengths P-31 line widths increase. Chemical shift anisotropy contributes to line width. The DFT signal-to-noise ratio depends on peak height. So, any gain in signal strength due to increased field could be compromised by broader lines.
Salem try the parameterset PROP31DEC this is 1H{31P} ,1H is the most sensititive an you might easily see which 1H signals "loose" a splitting when decoupling 31P. for PO(Or)3 you can set the 31P frequency near zero ppm. For other species 31P shifts are typically well documented. Clemens
Hi Clemens,
The problem is that, the OPs I'm trying to screen and Identify are within complex matrices and at low concentrations (below 10 ppm), it is then initially difficult to see their 1H signals. However, 2D 1H-31P HMQC directly targets those compounds and extracts their signals. But, I would definitely try the HX COSY you told me about, just to evaluate the necessary time to get exploitable results.
Dear Salem,
what was not yet mentioned by the other contributers is the following:
- If you have an inverse probe the 31P coil is remote of the sample and offers only moderate sensitvity - here 1H detection is most sensitve. In the other case (real broadband probe) the 31P coil is next to the sample and offers high sensitivity.
- If you need a large spectral width in 31P the 180 deg pulse in the HMQC has only limited inversion performance (can be quite substantial dependent on the B0 field and probehead in use).
- If I know the spectral range and the resolution required along 31P I probably would use 2D 1H-31P. You might also have a look on the Hxy sequence published some time back which works excellent if you only need a small 31P spectral range.
- T1 of 31P also tends to be longer compared to that of 1H - often (but definitely not always) faster pulsing is possible in 2D.
I know that this comments do not answer your question, but moreover add more "dimensions" to your problem.
Anyways I hope this helps
Alfred
Dear Alfred,
Thank you very much for this valuable addition. I now realize that, there won't be a direct answer. The reason is that there are multiple factors influencing the final result (sample nature, type of probe/spectrometer, spectral parameters...). I think the best approach would be trial and "error", taking into account all the factors you have cited and those in the previous answers.
Thank you.
In my experience the inverse 1H-31P HMBC correlation is extremely sensitive and provides a lot of information about the intermediates '(and of course also impurities) that you will never see in the 1D spectra. But it is always your decision, where to make a cut.
- Badges
- Science topic
- Similar topics
- Engineering
- Chemical Engineering
- Membranes