Stable, Chirped, Ultrashort Pulses in Lasers Using the Optical Kerr Effect

Abstract

efi 0 and gA = 1132/50.1 = 22.6 for eJ 1 by comparing our data with those of Fairfield and Gokhale. Apart from experimental errors of about±lO% in our data and±15% in c~0 and±30% in cfi 1 in Fairfield and Gokhale's data, the remaining difference of the two values of gA may in part be due to the electric field dependence of the emission and capture rates.The field dependence of the measured emission rates is further illustrated by the data of the hole emission rate at the gold-donor level, efi_ 1 in Table I, which were taken at 10 5 V/cm. A comparison of our result with the thermal equilibrium value of Fairfield and Gokhale given in the table shows that 8D = 3.88 x 10 7 /7.67 x 10 5 = 51, which reflects further the electric-field enhancement of the thermal emission rate. A self-consistent analysis of pulse propagation inside a laser cavity containing, in addition to an amplifying medium, some material displaying the optical Kerr effect, yields ultrashort pulses. These pulses become stable in the limit of high line-center excess gain with the stabilization provided by an interplay between strong chirping in the Kerr medium and bandlimiting in the laser medium.This paper presents an analysis of pulse propagation in a laser cavity containing some material displaying the optical Kerr effect, in addition to an amplifying medium. The question of mode locking a laser by means of the Kerr effect has been considered previously using a coupled mode approach\ in this paper we consider the same problem using the circulating pulse techniques developed by Cutler 2 and applied recently to laser systems. 3 ' 4 We find that this approach enables us to predict, in addition to the existence of ultrashort' pulses, their amplitude, pulse width, and chirp. The analysis also reveals the presence of strong pulse stability occurring in the limit of high line-center excess gain in the laser medium and strong chirping in the Kerr cell.We consider a Gaussian pulse with chirp circulating in the laser cavity. Simplified models for the laser and Kerr media are adopted, allowing the pulse to remain Gaussian, so that closed-form ana-*Work supported jointly by the U.S. Army Research Office-Durham, and by the U.S. Air Force of Scientifrc Research. 148 lytical solutions can be obtained. Also, all constant phase terms and group delays are ignored, since we are not calculating the loop transit time. The physical interpretation of the results suggest, however, that qualitatively they do not depend on the simple model chosen.The laser, shown in Fig. 1, is presumed to have a length Lo of amplifying medium near one end, and a cell of length L 1 containing the liquid (or solid) Kerr medium near the other end. An optical pulse in the form ofis assumed to be traveling from the Kerr cell toward the laser medium. If we neglect the overlap of a pulse with itself during reflection, we may represent one round trip as a traversal of a laser medium of length 2Lo followed by a passage through a Kerr medium! of length 2L1· The laser medium is characterized by a ...

Passive Q-switching and mode-locking for the generation of nanosecond to femtosecond pulses

“…A theoretical analysis of this mode coupling has been given in [124][125][126]. Mode-locking experiments were carried out with different optical Kerr media in ruby lasers [128] and Nd: glass lasers [127,129,130].…”

“…Instead of absorption changes the intensity dependent refractive index may be applied to passive mode-locking [121,[124][125][126][127][128][129][130]. …”