Crystal-field study for Yb3+ doped KY(WO4)2 crystal – Revisited based on superposition model

Crystal-field study for the orthorhombic Yb3+ centers in Ca3Sc2Ge3O12 and Lu3Ga5O12 garnets

“…Consistent with our previous works [60,24], it is once again shown in this study that the seeming best fitted CFPs to explain optical spectroscopy of some lanthanide ion in crystal well are likely not to describe the real quantum state of central paramagnetic ion, which will decidedly have an influence on the prediction of magnetic properties such as EPR parameters and magnetic susceptibility. The solution to this problem given in the present study is phenomenological but effective and widely used.…”

“…Note that the obtained best value of f by SPM can be fixed at 2890 cm À1 in the global search in CFP space to reduce free parameters since it mainly influences the optical spectra and causes little effect on EPR g factors [25]. Though all fitting results are pretty good, here comes up one question: why the direct parametric modeling by SPM is failure for Yb 3+ -doped LuGG garnet but successful for other crystal with even lower site symmetry (e.g., KY(WO 4 ) 2 :Yb 3+ in [24])? The reasons may be complex but most could be ascribed to the deficiencies in the framework of SPM such as a set of prior exponents t k and approximate local (or defect) structure around central ion.…”

“…This local structural distortion, mainly caused by the size mismatch or charge compensation between the doped and host ion, can be obtained by ab initio [37] or molecular dynamics (MD) calculations [20]. However, for simplicity a reasonable empirical assumption is taken in this work and given as follows [38,24] …”

“…Surprisingly, our test calculations indicate that the fitting based on SPM could not simultaneously match experimental optical spectra [14] and EPR g factors [43] of LuGG:Yb 3+ crystal (please see more details in Section 3). This 'unexpected' result means that the direct parametric modeling by SPM [24] is not available for LuGG:Yb 3+ crystal, although it has been successfully utilized in several Yb 3+ -doped inorganic systems [44,45,24]. Consequently, treating all nine CFPs as adjustable parameters in fitting procedures has to be adopted and this 'last resort' has been demonstrated in many Yb 3+ (or Ce 3+ )-doped garnets [46,47,43,10,8] (note that experimental optical spectra and EPR parameters should be fitted together, otherwise the 'over-fit' problem will be hard to settle).…”

“…Therefore, several ways of tackling this problem were taken by different groups. For example, the extrapolation method based on the point charge model (PCM) was utilized by Trabelsi et al [21]; keeping ratios between CFPs fixed to reduce free variables was adopted by Kamiń ska et al [18] and more works [22][23][24][25][26] are based on effective crystal-field models such as simple overlap model (SOM) [27], exchange charge model (ECM) [28] and superposition model (SPM) [29]. In this study, the combination of complete diagonalization (of energy matrix) method (CDM) [30] and SPM are employed to explain optical spectra for CaSGG:Yb 3+ and LuGG:Yb 3+ garnets and EPR g factors of such two garnets are also investigated (see Section 2).…”