Turn-on transient dynamics in a multimode, compound-cavity laser

Abstract

The turn-on-time statistics for an individual longitudinal mode and the total intensity of a multimode laser are shown to be very different. In both experiment and theory we find that the differences are due mainly to transitory modes that decay as a result of frequency-dependent losses and gains. Associated with this phenomenon is a kink in the time evolution of the mode intensity, but not in the total intensity, that represents a transition in the dynamics from domination by independent growth of all modes to domination by competition between modes with different net gains. This unequal competition increases the average and the standard deviation of the individual mode turn-on time. The experimental dye-laser system has a thin gain medium (100 m) that is strongly coupled to a short cavity (2.5 cm) and is adjacent to one of the cavity mirrors. The opposite cavity mirror serves as a relatively weak output coupler. The presence of the thin gain medium in the cavity causes the effective pump and loss rates to be frequency dependent. The result is a transient spectrum in which the cavity modes that have the highest net gain dominate the system more as the turn-on transient progresses. The gain spectrum is found to be strongly affected by the frequency dependence of the compound-cavity modes. Using realistic laser parameters, numerical simulations of a multimode laser model ͕developed in companion paper [J. Opt. Soc. Am. B 14, 191 (1997)]͖ yield turn-on dynamics and statistics that agree well with those measured experimentally.

Multimode laser model with coupled cavities and quantum noise

“…3), (f) multiplicative noise that is due to pump fluctuations, 43,44 and (g) spectral hole-burning effects including NDFWM contributions. 45,46 Despite these shortcomings, the model presented yields good agreement with the short-cavity dye-laser measurements presented in the companion paper 28 and in Refs. [25][26][27], and this comparison is the primary motivation for what follows.…”

“…45,46 Despite these shortcomings, the model presented yields good agreement with the short-cavity dye-laser measurements presented in the companion paper 28 and in Refs. [25][26][27], and this comparison is the primary motivation for what follows.…”

“…However, when ⌬ ӷ ␥ ʈ , j , as in most semiconductor laser systems, it can be seen that the highfrequency inversion components X j,k p 0 that correlate the mode noises are small (of the order of ␥ ʈ /⌬ ) compared with the average inversion density N(t). Thus, in certain class-B lasers, such as the one considered in the companion paper 28 and in Refs. 25 and 26, the correlated mode noise is small compared with the uncorrelated noise, as pointed out by Marcuse.…”

“…These treatments typically neglect the dynamical aspect of the noise 10,11,23 or borrow from single-mode theories. 14,24 The motivation for the study described here 25 is the desire to model the dynamics of individual modes in a multimode laser, including turn-on-time statistics 26 and mode hopping. 27 We have applied the theory developed here for these applications in the case of a short-cavity dye laser with good success, improving the simulation results presented in Ref.…”

Numerical modelling of lasing in microstructures

“…In order to investigate the effects related to the mode structure, interaction, and dynamics of laser radiation in microstructures, it is more efficient to develop methods that would rely on the eigenmode decomposition of the cavity field. Such theories have been developed for bulk resonators [9,10] and successfully checked against experiments [31]. Transition to microcavities involves two issues.…”