Abstract

The design of high power eye-safe lasers based on Er³⁺ involves choosing the optimum host, pump source, wavelength, geometry, Er concentration, length of the gain medium, etc. We are focusing on hosts such as YAG and Y₂O₃, and resonant pumping of the ⁴I_13/2 level with laser diode bars and stacks. Although the most impressive slope efficiencies so far have been obtained with narrow-line Er fiber laser pumping [1], direct resonant diode-pumping offers many important practical advantages and higher overall efficiency [2]. The choice of doping level and rod length is influenced by several factors. The relatively low absorption cross section of Er³⁺ in the 1.5 μm region favors higher doping and longer rods. Cooperative upconversion, which reduces the population of the ⁴I_13/2 upper laser level, favors lower doping. Longitudinal pumping with extended sources, e.g. diode laser bars that are difficult to collimate, favors shorter rods. One approach to the problem is to couple the bar to the rod with a lens duct, and confine the pump light within the rod using total internal reflection. Another approach is to use micro-optics to collimate the emission from the bar. The use of shorter rods and higher doping can benefit either approach, but reintroduces the problem of upconversion. It is also important to consider the effect of temperature on laser performance, due to the striking results obtained recently for cryogenic operation of diode-pumped Yb:YAG lasers. Due to the similar quasi-four-level structures of Er³⁺ and Yb³⁺, similar advantages occur upon cooling Er-doped lasers. But gain medium cooling is even more advantageous for Er³⁺ ion, which at room temperature is more nearly three-level than Yb³⁺, due to its smaller ground state splitting. This paper describes a quantitative study of these tradeoffs over a wide temperature range, based on a laser model anchored to experimental laser results.

© 2007 IEEE