Abstract
It is shown that the previous parametric-instability explanation of the jump in the Stokes intensity IS as a function of laser intensity IL in stimulated Raman scattering experiments is valid for many solids, liquids, and gases (with high optical dispersion dndλ and high Raman frequency ωf), while a few materials (low dndλ and ωf) should show an enhanced gain, ISexpIL2, which is greater than the usual stimulated-Raman gain but not as great as the jump result. The previously anomalous experimental results of Grun, McQuillan, and Stoicheff and of others, which show the jump in IS, and of Hagenlocker, Minck, and Rado, which show both types of behavior, are explained. The instability is expected to be important in laser damage of Raman-active crystals and possibly in determining the limiting diameter of self-focused beams. The transient solution for the case of high dispersion indicates that the steady state is not reached until a time much greater than the Raman phonon-relaxation time. It is also shown that a phonon parametric instability studied previously in another context can reduce the Stokes intensity. Experiments to detect the phonon instability and measure the magnitude and temperature dependence of "half-frequency" phonons Q (with frequencies ωQ+ωQ=ωf) are suggested.