INTERPRETATION OF THE TEMPERATURE-DEPENDENCE OF THE EPR-SPECTRUM OF CU2-DOPED (NH4)(2)[CD(NH3)(2)(CRO4)(2)] AND CRYSTAL-STRUCTURES OF THE HIGH-TEMPERATURE AND LOW-TEMPERATURE FORMS OF THE HOST LATTICE()

The crystal structure of (NH4)(2)[Cd(NH3)(2)(CrO4)(2)] is reported. Be low about 300 K the compound changes from a monoclinic cell (space gro up C2/m, Z = 2, a = 12.8380(11) Angstrom, b = 6.0308(6) Angstrom, c = 7.5890(6) Angstrom, beta = 110.154(14)degrees) in which all four Cd-O bonds of the trans-Cd(NH3)(2)O-4 coordination sphere are crystallograp hically equivalent to a triclinic cell (space group P ($) over bar 1, Z = 1, a = 6.0210(4) Angstrom, b = 7.0363(4) Angstrom, c = 7.5714(8) A ngstrom, alpha = 106.802(18); beta = 93.032(12); gamma = 114.079(11)de grees) in which the only symmetry element of the Cd complex is an inve rsion center. It is shown that the previously reported temperature dep endence of the EPR spectrum of similar to 0.3% Cu2+ doped into this co mpound is consistent with the change in crystal structure. The spectra may be explained using a model of dynamic vibronic coupling in which the effects of Jahn-Teller coupling and a ''strain'' due to the inequi valence of the ligands are applied to the e(g) vibrational and E(g) el ectronic wave functions of the Cu2+ ion. The balance between the ligan d field asymmetry and the natural tendency of Cu2+ to adopt a tetragon ally elongated octahedral coordination geometry results in a complex w ith an orthorhombic coordination geometry having short bonds to the am mine groups and intermediate and long bonds to the chromate oxygen ato ms. However, the long Cu-O bonds may occur to either pair of trans chr omate oxygen atoms. In the high-temperature monoclinic unit cell, the EPR spectrum confirms that these two conformations are energetically e quivalent, but in the low-temperature triclinic cell this is no longer the case, and the EPR spectrum is consistent with a temperature-depen dent equilibrium between the two possible structural isomers. However, the model suggests that significant delocalization of the vibronic wa ve functions may occur, so that it is difficult to define precisely th e bond lengths and electronic wave function parameters of the guest co pper(II) complexes.