Theory of Soliton-Modelocked Femtosecond Solid-State Lasers

Abstract

Listen

Translate article

Since the early work of Martinez et al., it is well known that modelocked lasers will yield pulses that are both shorter and more stable if solitonlike pulse shaping is used, i.e. self-phase modulation (SPM) and negative group velocity dispersion (GVD) is properly adjusted. This work extends these results. Based on soliton perturbation theory we derive stability ranges for actively and passively modelocked femtosecond solid-state lasers. The theory shows that the response time of the saturable absorber in a passively modelocked laser that operates in the soliton regime can be much longer than the pulse width of the generated solitonlike pulses. This is in contrast to the traditional concepts of ultrashort pulse generation that rely either on an artificial fast saturable absorber as is the case for APM or KLM systems or on the interplay between a slow saturable absorber and gain saturation as is the case with dye lasers. From this finding we can conclude that semiconductor absorbers with typical absorption recovery times of 100 fs due to intraband thermalization processes can be used directly to generate 10 fs pulses. The theoretical results are compared with experiments. Thus far we achieved pulses as short as 16 fs from a Ti:sapphire laser modelocked only with a semiconductor absorber2.