03/18/2024 | Press release | Distributed by Public on 03/18/2024 04:14
In nuclear magnetic resonance (NMR) analyzes of phosphorus-containing compounds, 1H and 13C signals near phosphorus are affected by heteronuclear spin couplings. Given the natural abundance of 31P at 100%, these splittings are easily visible, reducing sensitivity and increasing complexity. Additionally, some of these compounds are unstable in solvents, leading to the formation of decomposition products, which further complicates the interpretation of 1H and 13C spectra. On the other hand, when analyzing phosphorus-containing samples, 31P NMR information can be invaluable for structural analysis. To illustrate this, we will provide some examples below. In the measurements detailed in this application note, we utilized the ROYAL ProbeTM P+ [1], capable of 1H, 31P, and X triple-resonance measurements. We employed a three-channel 600 MHz NMR spectrometer, the JNM-ECZL600G, featuring one high-frequency (HF) and two low-frequency (LF) channels.
Alkyl phosphonates known as Horner-Emmons reagents are used in the olefination reaction. These include diethyl methylphosphonate 1, ethyl dimethyl phosphonoacetate 2, and diethyl (ethoxymethyl) phosphonate 3. They were mixed and dissolved in CDCl3 to create a sample containing 4.8 vol% of each compound.
A conventional 13C spectrum of the sample recorded with 1H decoupling is depicted in Fig. 1. Upon examining the expansions of this spectrum in Fig. 2, an experienced chemist may observe that several signals appear doubled, indicating splitting caused by 13C-31P coupling. However, identifying such signal doublings may prove challenging in the case of unknown samples or very complex mixtures with numerous carbon signals. To illustrate the efficacy of triple-resonance experiments, a 13C spectrum recorded with simultaneous 1H and 31P decoupling is presented in Fig. 2c) and Fig. 2d). A straightforward comparison between the double-resonance and triple-resonance spectra allows for the unequivocal identification of carbon signals split by one-bond M13C-31P couplings, which can exceed 100 Hz in alkyl phosphonates. The signals observed at 10.85 ppm (JCP = 144 Hz)*1, 33.13 ppm (JCP = 135 Hz)*2, and 64.35 ppm JCP = 167 Hz)*3 can thus be attributed to the carbon atoms directly bound to phosphorus, with the one-bond coupling constants provided in parentheses. Smaller splittings are indicative of 2-bond and 3-bond interactions. Table 1 summarizes the 13C chemical shifts and 13C-31P coupling constants.
No. | 13C/ppm | J/Hz |
1 | 10.85 | 1Jcp=144 |
2 | 13.75 | - |
3 | 14.57 | - |
4 | 16.08 | nJcp=7 |
5 | 16.15 | nJcp=6 |
6 | 33.13 | 1Jcp=135 |
7 | 52.84 | nJcp=6 |
No. | 13C/ppm | J//Hz |
8 | 61.14 | nJcp=6 |
9 | 61.35 | - |
10 | 62.05 | nJcp=7 |
11 | 64.35 | 1Jcp=167 |
12 | 68.73 | nJcp=12 |
13 | 165.29 | nJcp=6 |
Table 1: 13C chemical shifts and coupling constants
Fig. 3 shows the1H-31P HMBC spectrum. 1H-31P coupling via 2- and 3-bonds are observed, but it is difficult to distinguish each from this spectrum. On the other hands, by measuring the 1H-13C-31P double-INEPT as shows Fig. 4, only the 2-bonded 1H-31P can be selectively extracted [2]. Then we can distinguish between 2- and 3-bonds by comparing them. the 1JCH and 1JCP couplings are used to obtain correlations in double-INEPT, since 1JCP of alkyl phosphonates have a large coupling constant as mentioned above, it is possible to extract only the H-C-P coupling by setting the parameters according to that.
Correlations observed in 1H-31P HMBC
Fig. 3: ¹H-³¹P HMBC spectrumCorrelations observed in 1H-13C-31P double-INEPT
Fig. 4: ¹H-¹³C-³¹P double-INEPT spectrum31P-13C correlations can also provide significant insights into the structural analysis of phosphorus-containing compounds. Correctly chosen and optimized 31P-13C correlation experiments enable the identification of 13C coupling partners for each 31P nucleus. Fig. 5 illustrates a 31P-13C LR-HSQC (Long-Range Heteronuclear Single-Quantum Coherence) spectrum. The pulse program employed was a standard HSQC, with parameters optimized for long-range 13C-31P couplings, thereby capturing correlations from weak couplings as well. In Fig.6, we compare the 13C NMR spectrum with three slices extracted from the 31P-13C LR-HSQC spectrum for each phosphorus atom. This facilitates the assignment of 13C signals to individual phosphonates. Additionally, Fig. 6 demonstrates that 31P-13C LR-HSQC enables the detection of carbon atoms up to three bonds away from each phosphorus atom.
JEOL Application note NM220010E
Tetrahedron Letters 48 (2007) 7586-7590
JNM-ECZL series FT NMR