A recently predicted resonant effect for the enhancement of two-wave mixing in photorefractive materials is investigated. The resonance occurs when the frequency of the applied ac field agrees with the eigenfrequency of the excited space-charge wave. Experimentally a clear resonance is found, as predicted by the theory, for high dc electric fields, but the resonance is smeared out for lower fields. A modified theory, taking into account the second temporal harmonic of the space-charge wave, shows good agreement with the experimental results. © 1997 Optical Society of America It was shown by Huignard and Marrakchi 1 and Réfrégier et al.2 that the interaction between two incident waves can be significantly enhanced by application of dc voltage and detuning of one of the waves. The enhancement was maximum when the moving interference wave was in resonance with the space-charge wave, which is an eigenwave of the system.
3A nonresonant excitation employing an applied ac voltage was shown by Stepanov and Petrov 4 to lead to similar enhancement. More recently, a new kind of resonant interaction 5 -7 was proposed that, instead of detuning, relies on applied dc and ac f ields. It was shown in Refs. 5 -7 that the resonance took place when the frequency of the applied ac f ield agreed with the eigenfrequency of the space-charge wave. The purpose of this Letter is to investigate experimentally this new resonant mechanism and to introduce corrections to the model of Ref. 7 that will lead to good agreement between the experimental and the theoretical results.The experimental arrangement for two-wave mixing is shown schematically in Fig. 1. A frequency-doubled diode-pumped Nd:YAG laser ͑l 532 nm; TEM 00 power, 500 mW) was used as the light source. A variable beam splitter controlled the intensity ratio of the two beams, which were incident upon the (110) face of a bismuth silicate (BSO) crystal with an interbeam angle of 1.4 ± , thus producing fringes with approximately 20-mm spacing. Both incident beams were linearly polarized normal to the (110) face of the crystal. The external fields were applied across the (001) faces. The crystal dimensions were all 10 mm, mirrors M3 and M4 were mounted upon piezoelectric translators, and, for frequency detuning, sawtooth voltages of opposite polarity were applied to these translators with an amplitude sufficient to cause an optical phase change of 2p between the beams. The modulation depth of the interference pattern, m, was equal to 0.1 (corresponding to an input intensity ratio of ϳ100), and a photodetector was used for monitoring the intensity of the weak signal beam. The total intensity of the beams in all the experiments was I 0 5 mW cm 22 . According to the predictions of the simple model referred to above, the maximum of the new resonance occurs at the same frequency as that of the resonance obtained when the optical beams are detuned. Hence in our f irst set of experiments we measured the gain as a function of detuning. The results are shown in Fig. 2