Cl Nuclear Magnetic Resonance in CuCl2·2H2O

The ${}_{}{}^{35}{}_{}{}^{}\mathrm{Cl}$ NMR in single crystals of Cu${\mathrm{Cl}}_2$·2${\mathrm{H}}_2$O has been studied at 5.0 and 76°K and in the antiferromagnetic state at zero field between 1.3 and 4.2°K. The orientation dependence of the NMR was measured at 5.0°K for fields lying in the $\mathrm{ac}$ and $\mathrm{bc}$ crystallographic planes. In addition, the dependence of the NMR frequencies upon applied field strength was measured for particular field orientations. Two possible sets of principal axis systems for the Cl nuclear-site electric field gradient (EFG) and hyperfine field tensors result. This ambiguity is a consequence of the presence of two Cl sites in the unit cell which are distinguishable for general field orientations. However, one set requires that the anisotropic part of the hyperfine field should be dominated by $p_{\pi }$ interactions. This is contrary to the expected dominant $p_{\sigma }$ contribution for ${\mathrm{Cu}}^{++}$ in Cu${\mathrm{Cl}}_2$·2${\mathrm{H}}_2$O. The second and correct set leads to the hyperfine field values (in units of ${10}^{-4\mathrm{}}$ ${\mathrm{cm}}^{-1\mathrm{}}$): $A_s=7.8\mathrm{}\pm 0.6$, $A_{\sigma }=5.0\mathrm{}\pm 0.7$, and $A_{\pi }=0.0\mathrm{}\pm 0.7$, where we have neglected the second-nearest-neighbor contribution in determining these values. The magnitude of the internal field at a Cl nucleus in Cu${\mathrm{Cl}}_2$·2${\mathrm{H}}_2$O is 47.5±3.0 kOe at 0°K. The orientation of the EFG is in much better agreement with point-charge-model predictions than with the results of the calculation of Rao and Narasimhamurty in which induced dipole effects were included. The NMR were observed with a superregenerative detector. Since it was not possible to separate the sidebands from the broad central resonance, the resonance patterns consisted of the central resonance with sidebands superimposed. The uncertainty in the location of the central resonance frequencies (>±15 kc/sec) proved to be the major source of experimental error. Attempts to observe the resonance with marginal oscillator detectors proved unsuccessful.