Dependence of Upconversion on Direct Luminescence Decay in Energy-transfer Upconversion

For several decades, energy-transfer upconversion (ETU) in rare-earth-ion doped systems [1,2] has attracted much attention, firstly, because of the fundamental interest in the physical nature of this process and, secondly, because of very practical considerations, namely the demonstration of near-infrared pumped visible light sources and, in reverse, the detrimental influence of ETU on the efficiency of infrared emitting systems. We investigate fundamentally the behavior of and interaction between infrared luminescence emitted directly from a metastable level and visible luminescence emitted after ETU from this metastable level to higher-lying levels. Although these two luminescences are connected by the same metastable level and influenced by the same ETU process, they probe different classes of ions. Whereas the infrared luminescence probes all ions, the visible luminescence probes only the class of ions susceptible to ETU [3]. A simple analytical model [4] predicts that such luminescence decay curves exhibit a super-quadratic dependence of upconversion on direct luminescence decay. The fraction of ions susceptible to ETU can be derived from this model. The Nd ion can serve as a model system for such investigations. It exhibits strong ETU from the metastable F3/2 level. When doped into oxide matrices, the F3/2 level is the only metastable level within the 4f subshell. The Nd energy levels excited by ETU decay by fast multiphonon relaxation and, hence, the weak visible fluorescence emitted from these levels represents a quasi instantaneous reaction on the dynamics of the F3/2 metastable level. Experimental results obtained after pulsed laser excitation of Nd-doped oxide host materials show indeed a super-quadratic behavior of upconversion versus direct luminescence decay, in accordance with the model predictions [4].