Page 203 - Tunable Lasers Handbook

5 Dye Lasers 181

or CuSQ,. An alternative approach is to use an active filter that absorbs the dam-
aging ultraviolet radiation and emit at longer wavelengths compatible with the
absorption b'and of the laser dye. The use of the dye converter stilbene 420 is
reported to enhance efficiency of coumarin 504 by up to 75% [66].
A third excitation configuration uses linear flashlamps to excite the dye
region transversely (Fig, 7). This transverse excitation configuration uses two
rows of linear lamps to excite a narrow dye region channel. The dimensions of
the active region volume depicted in Fig. 7 are a 0.5-mm width, a 55-mm height.
and a 150-mm length [67]. This transverse excitation geometry allom the use of
higher dye concentrations and more importantly the rapid flow of Lhe dye solu-
tion. Also note that dye converters are also used in the cooling fluid. Using this
type of excitation geometry, with eight lamps at each side, Klimek et al. [68]
report an average power of 1.4 kW. The performance of various flashlarnp-
pumped dye lasers is listed on Table 7.

3.2 Multiple-Prism Grating Master Oscillators

Important features for master oscillators are narrow-linewidth emission,
good beam quality, and very low ASE levels. Given the geometrical and excita-
tion characteristics of flashlamp-pumped dye lasers it is a particular challenge to
achieve stable narrow-linewidth oscillation. MPL and HMPGI oscillator config-
urations coupled with thermal and fluid flow controls have been crucial to the
demonstration of stable long-term narrow-linewidth emission in this class of
master oscillators p7.721. In this subsection the physics and technological ele-
ments central to this topic are surveyed.
The first step in the design of a high-performance dispersive oscillator is to
apply the generalized interference equation [Eq. (2) in Chapter 21 to determine
the aperture necessary to yield a single-transverse mode for a given cavity
length. Then, for a given grating and grating configuration the necessary intra-
cavity beam expansion is calculated followed by an estimate of the dispersive
linewidth. I€ the dispersive linewidth is within the desired range, then the multiple-
prism beam expander is designed. In the event that the dispersive linewidth is
not appropriate, then a higher dispersion grating should be considered. In this
approach Eqs. (8) to (12) of Chapter 2 should be applied.
The multiple prism should be designed for near-orthogonal beam exit and
(a@/,/a?i), = 0. This approach reduces significantly back reflections of ASE and min-
imizes frequency detuning due to thermal change. Equations (22) and (23) sf
Chapter 2 are then utilized to determine the transmission efficiency of the multiple-
pnsrn beam expander. Here the quest for efficiency must be balanced against the
length of the prism expander and its cost. Duarte [ 1,501 provides a detailed and
explicit discussion on the design of multiple-prism beam expanders.
Further avenues in the reduction of ASE include the use of very low dye
concentrations -0.01 mM and the use of a polarizer output coupler [47,72].