The Effects of Oxide on the Spectroscopic Properties of the 1.3, 1.7, and 2.9 μm Transitions in Dysprosium-Doped Gallium-Lanthanum-Sulphide Glass

Abstract

Results of a spectroscopic investigation suggest the existence of two dysprosium sites, of which one is unsuitable for pumping a potential 1.3µm optical amplifier. Absorption, emission, and excitation spectra are presented along with fluorescent lifetime measurements. Gallium-lanthanum-sulphide (GLS) chalcogenide glass doped with dysprosium (Dy3+) shows promise as 1.3µm fibre-optic amplifier and mid-infrared laser material.[1] GLS glass has attracted great interest due to its low phonon energy of 425 cm, resulting in low non-radiative decay rates of rare-earth ions. Its high refractive index of 2.4 results in high radiative emission rates. Both properties improve the radiative quantum efficiencies for all transitions. The high quantum efficiency for the 1.3µm transition in Dy3+ -doped GLS glass and the large cross sections for this transition imply relatively short device lengths, making Dy3+ a very promising active ion for 1.3µm amplification. The optical characteristics of Dy3+ ion are also attractive because four absorption bands exist at 1.28, 1.10, 0.90 and 0.80µm, all with a large cross section allowing efficient pumping of the emitting level. All these absorption bands show a double structure in GLS, in contrast to other Dy3+- doped chalcogenide glasses.[2] We investigated the nature of the 0.80µm double structure by pumping Dy3+-doped GLS over a range of wavelengths while observing the fluorescence at 1.3µm. While pumping the short wavelength peak of the double structure, no fluorescence was observed, ie the excitation spectrum does not follow the absorption spectrum. This result suggests the existence of two spectroscopic sites in GLS glass, which may be created by different local environments of the Dy3+ ions. For example, an oxide environment with a higher phonon energy would quench the fluorescence by multiphonon decay. Oxide impurities in Dy3+:GLS are introduced by the starting material, and have proven difficult to eliminate and control. However, oxide is beneficial in stabilising the glass, increasing the region of glass formation amongst the glass constituents. We have also studied the fluorescence at 1.7 and 2.9µm from levels with a larger energy gap to the next lower level. The larger energy gap decreases the multiphonon decay rate and leads to 2.9µm emission from Dy3+ ions in the site quenched for the 1.3 and 1.7 µm transitions. This supports our suggestion of two sites with different environments and phonon energies. In addition, the emission bands exhibit a double-exponential decay profile. Absorption, emission and excitation spectra of Dy3+ will be presented at the conference, along with fluorescent lifetime measurements. Dy3+:GLS offers a rare opportunity to study the influence of the local environment on the spectroscopy of the rare-earth ion by site-selective measurements. The results also give important guidance for the choice of pump wavelengths in a 1.3µm amplifier