In this paper we present the fluorescence decay profiles of vanadium and titanium doped gallium lanthanum sulphide (GLS) glass at various doping concentrations between 0.01 and 1% (molar). We demonstrate that below a critical doping concentration the fluorescence decay profile can be fitted with the stretched exponential function: exp[-(t/{ \tau})\b{ eta}], where { \tau} is the fluorescence lifetime and \b{ eta} is the stretch factor. At low concentrations the lifetime for vanadium and titanium doped GLS was 30 { \mu}s and 67 { \mu}s respectively. We validate the use of the stretched exponential model and discuss the possible microscopic phenomenon it arises from. We also demonstrate that above a critical doping concentration of around 0.1% (molar) the fluorescence decay profile can be fitted with the double exponential function: a*exp-(t/{ \tau}1)+ b*exp-(t/{ \tau}2), where { \tau}1and { \tau}2 are characteristic fast and slow components of the fluorescence decay profile, for vanadium the fast and slow components are 5 { \mu}s and 30 { \mu}s respectively and for titanium they are 15 { \mu}s and 67 { \mu}s respectively. We also show that the fluorescence lifetime of vanadium and titanium at low concentrations in the oxide rich host gallium lanthanum oxy-sulphide (GLSO) is 43 { \mu}s and 97 { \mu}s respectively, which is longer than that in GLS. From this we deduce that vanadium and titanium fluorescing ions preferentially substitute into high efficiency oxide sites until at a critical concentration they become saturated and low efficiency sulphide sites start to be filled.
Categories: cond-mat.mtrl-sci