Surface effects on voltage- and current-source bias phototransistors grown by low-pressure MOCVD

We report on low-pressure MOCVD grown InGaP/GaAs heterojunction phototransistors (HPT) with and without passivated base surface. The typical structure consists of a 0.5-/spl mu/m n/sup +/-GaAs sub-collector (n/sup +/=5/spl times/10/sup 18/cm/sup -3/), a 0.6-/spl mu/m n/sup -/-GaAs collector (n/sup -/=5/spl times/10/sup 16-3/), a 0.14-/spl mu/m p/sup +/-GaAs base (p/sup +/=4/spl times/10/sup 19/cm/sup -3/), a 0.07-/spl mu/m n-InGaP emitter (n=5/spl times/10/sup 17/cm/sup -3/), and a 0.3-/spl mu/m n/sup +/-GaAs cap layer (n/sup +/=5/spl times/10/sup 18/cm/sup -3/). Both fabricated HBTs and HPTs have their emitter areas of 150/spl times/150 /spl mu/m/sup 2/ with an absorption area of 4/spl times/10 cm. Optical/electrical experiments and measurements include effects of base surface on optical responsivity, performances of HPTs with a voltage-source and current-source bias base. Following are our important results and conclusions. 1. Effects of base surface: (1) wide-gap InGaP as a base surface passivation layer is transparent to 850-nm incident light, (2) a higher responsivity is available for such a passivated base layer. The measured responsivity is 0.64/spl plusmn/0.01 (0.57/spl plusmn/0.01) A/W for passivated (nonpassivated) devices. 2. Voltage-source bias HPT: (1) both collector and reverse base currents saturate at their p-i-n photocurrents for both kinds of HPTs at a small voltage level; (2) there is no amplification of photocurrent generated with base-collector region if B-E junction is off conductive due to small voltage; (3) even if the applied voltage pushes a HBT to the high current gain level, the magnitude of amplification of photocurrent is only 3, which is much smaller than that from a HPT with a floating base. 3. Current-source bias HPT: both collector photocurrent and optical gain are increased with increasing input current for both kinds of HPTs. The measured collector photocurrents equal -32 and -48 times of their photocurrent. So, we conclude that pushing HBTs operating point to a high current level using a current source is able to enhance optical performances. The enhanced trend with input current is more obvious, indicating that passivated HPTs are more suitable for present applications in integrated optoelectronic circuits. Furthermore, a voltage-source and a current-source bias HPT-circuit model using extended Ebers-Moll configuration and related results calculated therein are presented in this work. It is found that calculated results are in good agreement with experimental data.