A comparison of iron concentration and photorefractive gain in iron doped indium phosphide

Abstract

The infrared photorefractive (PR) effect in InP can be enhanced using applied electric fields to increase gain for potential applications such as amplifying coherent optical signals. Knowledge of the defect physics of InP material is a prerequisite to understanding the PR behavior and consequently the optimization of the PR response. Previous attempts to explain the PR effect have been marked by the use of InP:Fe samples which were not fully characterized by techniques other than by PR measurement alone. Some results have been reported which lead to the conclusion that both shallow traps and electron-hole competition are dominant in semi-insulating photorefractive materials, while other workers using different crystals, have concluded that the shallow levels are relatively unimportant for InP:Fe and that the photorefractive effect is dominated by holes at room temperature. The divergent conclusions may be due in part to defect concentrations which vary from sample to sample. In this experimental investigation a series of crystals were grown using different concentrations of iron dopant and shallow donor impurities, which were then compared in PR experiments at room temperature to measure the two-wave mixing gain. The two main defects were then measured by independent means to determine a relationship between PR gain and defect concentration. Iron exists in one of two electronic states, either ionized (Fe/sup 2+/) or neutral (Fe/sup 3+/). The concentration of Fe/sup 2+/ is equal to the net shallow donor concentration. The neutral state Fe/sup 3+/ can be determined by electron paramagnetic resonance (EPR) or from an analysis of the near infrared absorption spectra in the range 0.6 to 1.3 eV. The absorption of iron doped InP in this photon range is caused by two photoionization processes, the optical excitation of electrons from Fe/sup 2+/ into the conduction band and the excitation of holes from Fe/sup 2+/ into the valence band. Both processes also give rise to the YAG laser absorption at 1.06 /spl mu/m (1.17 eV) and provide the basis for the photorefractive effect. From an analysis of these data, we conclude that the diffraction efficiency is primarily dependent on the Fe/sup 3+/ concentration, and that the PR gain can be optimized by selecting crystals within a range of Fe/sup 3+Fe/sup 2+/ ratios. >