Homonuclear Decoupling

This traditional 1D NMR experiment involves selectively decoupling a proton from its J coupling partners. This is accomplished by low-power irradiation at the frequency of the peak of interest in the !H spectrum during the acquisition of the FID. The power level required is a bit higher because the presaturation and NOE experiments need only enough power to equalize populations, whereas the decoupling experiment requires enough power to rapidly (relative to the "NMR timescale" 1/(2.2 J)) flip each spin at the selected frequency back and forth between the upper and lower energy levels. As with heteronuclear decoupling, the rapid flipping on the NMR timescale means that other protons that are J coupled to this spin see an average magnetic field that is unaffected by the orientation (a or p state) of the spin being irradiated. The selected spin is thus removed from the coupling network, and any peak in the spectrum that is coupled to it will be simplified by removal of that one splitting. For example, a doublet will become a singlet if the J coupling is from the peak being irradiated, and a double doublet will become a doublet. In the latter case, the coupling that is removed can be unequivocally assigned to the proton being irradiated and the remaining coupling must be due to some other proton in the molecule. In some cases the decoupling power is insufficient to completely remove the coupling and the apparent J value is reduced rather than being eliminated. This is the same as the heteronuclear case where the reduced J value (JR) is a function of decoupler field strength (Chapter 4, Section 4.4.2). For example, a triplet might be changed into a double doublet with one small coupling due to the proton being irradiated. In general, any change in a peak's coupling pattern can be interpreted as a J coupling to the peak being irradiated.

The results of a homonuclear decoupling experiment on sucrose are shown in Fig. 5.18. The experiment is set up by acquiring a normal 1H spectrum and determining the exact RF frequency of each peak (each resonance) we wish to "test" by CW irradiation during the acquisition of the FID. The desired frequencies are actually offsets from a fundamental decoupler frequency; for example, an offset (Bruker: o2 for channel 2 offset; Varian: dof for decoupler offset) of 132.6 Hz is added to the fundamental frequency (Bruker: BF2

for base frequency, channel 2; Varian: dfrq for decoupler frequency) of 299.956 MHz to get the exact frequency of the decoupler, 299.9561326 MHz (299.956 + 0.0001326). The experiment is run as a series of 1H acquisitions yielding a series of FIDs, each using a different decoupler offset. In Fig. 5.18, the top spectrum is a control spectrum using a decoupler frequency away from any of the peaks, and the other spectra (a)-(f) are acquired with different peaks selected. On the Varian, the dof values are loaded into an array by typing in the values: dof = 1536.7, 467.8, 258.4, and so on. On the Bruker, a separate text file is created, a frequency list, in which the o2 values are entered one line at a time.

From a hardware standpoint, homonuclear decoupling is more challenging than presaturation or NOE difference because we need to do the CW irradiation at the same time that we are acquiring FID data. This is accomplished by shutting off ("gating") the decoupler RF for a brief period while each FID data point is being observed and digitized. Thus the send/receiver (or T/R) switch is very busy going back and forth between transmitting the decoupling CW RF signal and "listening" to the FID.

For sucrose, the experiment allows us to assign all of the peaks in the 1H spectrum. As always, we have to start with some prior knowledge, based on a unique chemical shift or coupling pattern. For the glucose part, we have H-g1, the only anomeric proton, which is farthest downfield because it is bonded to a carbon with two bonds to oxygen. For the fructose part, we have H-f3, which is the only doublet peak (besides H-g1) because it has a quaternary carbon on one side (C-f2). From these two pieces of information we begin the process of assignment. Irradiation of H-g1 (Fig. 5.18(f)) converts the double doublet at 3.5 ppm into a doublet. Thus the double doublet represents H-g2, and the remaining large doublet coupling is the J coupling between H-g2 and H-g3. The smaller coupling of the H-g2 double doublet is the small (axial-equatorial) coupling observed in the H-g1 doublet (3.8 Hz). Irradiation of H-g2 (Fig. 5.18(b)) causes the triplet at 3.7 ppm to "collapse" into a broad doublet, as well as converting the H-g1 doublet into a broad singlet. Thus the triplet at 3.7 ppm is H-g3. Irradiation at this position in turn causes (Fig. 5.18(c)) the triplet at 3.4 ppm to collapse to a sort of ugly doublet, as well as simplifying the H-g2 double doublet into a narrow doublet (glucose Ji-2 remains). Now we select H-g4 (Fig. 5.18(a)) and see the H-g3 triplet simplify to a broad doublet (3.7 ppm) and a very subtle change in the overlapped region at 3.8 ppm, which is evident if you compare to the control spectrum directly above it. So we can guess that the H-g5 resonance is buried in overlap at about 3.8 ppm. This is the end of the trail for the glucose part. Now pick the H-f3 peak to irradiate at 4.16 ppm (Fig. 5.18(e)). The triplet at 3.99 ppm collapses very neatly to a doublet, proving that this resonance is H-f4. No other peak is changed, confirming that H-f3 is at the end of the "spin system," next to a quaternary carbon. Irradiating H-f4 (Fig. 5.18(d)) we see a subtle change in the messy region at 3.84 ppm as well as the collapse of the H-f3 doublet to a narrow pattern resembling the superposition of a doublet and a singlet. The H-f5 proton can be assigned to an overlapped peak at 3.84 ppm. The only protons that remain to be assigned are the CH2OH protons H-f6, H-f1, and H-g6. The singlet at 3.62 ppm integrates to 2 protons, so it must be H-f1. H-f6 and H-g6 should have couplings to the proton at position 5. In addition, the NOE difference spectrum (see Fig. 5.30, below) clearly shows an NOE from this singlet peak to H-g1 (across the glycosidic linkage) and to H-f3 within the fructose unit. By process of elimination, the large overlapped peak at 3.76 ppm must be both H-f6 and H-g6.

Homonuclear decoupling is pretty much a historical experiment, as newer experiments such as 1D-TOCSY (Chapter 8) and 2D-COSY (Chapter 9) have replaced it.

0 0

Post a comment