Double resonance detection of 17O quadrupole resonance in potassium bicarbonate

Abstract
17 O quadrupole resonance signals have been detected from the C—17OH and C 17O…H groups in solid potassium bicarbonate by means of a double resonance technique based on adiabatic demagnetisation. The method of detection depends on the presence of a strong dipolar coupling between the 17O nucleus under study and the attached (or hydrogen-bonded) proton. With a single radiofrequency, simultaneous 17O and 1H transitions are excited, the cycle being completed by rapid quadrupole relaxation and spin diffusion to the dipolar bath of the remaining protons, whose magnetisation is thereby slowly destroyed; with two radiofrequencies whose separation lies within the dipolar multiplet, the 17O—1H system is driven round a complete cycle of induced quadrupole absorption and emission, with simultaneous spin diffusion which again slowly destroys the proton magnetisation. In both cases, the signals have well-resolved fine structure due to the 17O…1H dipolar coupling, whose appearance depends not only on the sign of the quadrupole coupling constant but also on the orientation of the quadrupole tensor with respect to the O—H vector. It is shown that in KHCO3 the observed line shapes and the behaviour of the signals under varying cycle times and r.f. intensities can be understood in terms of spin temperature theory. The calculations suggest that the C—17OH quadrupole coupling constant is negative, with e2qQ/h=–7709 kHz, η= 0.420 at 77 K and –7341.5 kHz, η= 0.237 at 291 K, the angle between q(qzz) and the OH bond being 50 ± 10 ° at both temperatures. For C 17O…H, e2qQ/h is probably positive and approximately equal to 6765 kHz, η= 0.934 at 77 K and 6772 kHz, η= 0.661 at 291 K, the angle between q and O…H being very approximately 60 °(±30 °). These values are motional averages over the vibrational modes of the nearly-planar (HCO3)2– 2 ion in the crystal; there is some evidence that ring torsional modes have a considerable effect on the quadrupolar constants.