MAGNETIC INTERACTIONS BETWEEN RARE-EARTH IONS IN INSULATORS .2. ELECTRON-PARAMAGNETIC-RESONANCE MEASUREMENTS OF GD3+ PAIR AND GD3+-EU3+ INTERACTION CONSTANTS IN EUCL3

Paramagnetic resonance has been observed from both single ions and isolated pairs of Gd3+ in EuCl3 at 77°K and 25.6 GHz. The pair spectra were analyzed in the same way as those observed previously from Gd3+ in LaCl3. The results give accurate values for the interaction parameters between both nearest (nn) and next-nearest neighbors (nnn). The dominant terms are of the form of isotropic exchange (JS1•S2), with Jnn(pair)=0.0488±0.0010 cm-1 and Jnnn(pair)=-0.0637±0.0020 cm-1. The anisotropic parts of the interactions could be accounted for completely by the calculated magnetic dipole-dipole coupling. Combining these results with earlier measurements in LaCl3, good estimates can be made of the exchange interactions in pure GdCl3: Jnn(GdCl3)=0.056±0.006 cm-1 and Jnnn(GdCl3)=-0.064±0.003 cm-1. These values are in excellent agreement with those inferred from analyses of susceptibility and specific-heat measurements, and they resolve unambiguously earlier uncertainties in the choice of possible parameters. The magnitudes and even the signs of the interactions remain unexplained. From measured g-value shifts relative to Gd3+ in LaCl3 it was also possible to estimate the isotropic part of the Gd3+-Eu3+ exchange interaction: Jnn(Gd-Eu)=0.052±0.030 cm-1 and Jnnn(Gd-Eu)=-0.061±0.030 cm-1. The similarity of these results to the corresponding Gd3+-Gd3+ interactions suggests that the average electron-electron exchange interaction may be relatively insensitive to the particular configuration of the individual ions, as proposed theoretically by Van Vleck. On the other hand, comparisons between the results for different isostructural lattices shows that some of the interactions can be extremely sensitive to even small changes in the neighbor distances, and great care is therefore required in extrapolating from one lattice to another. © 1969 The American Physical Society.