Analytical modeling of pnp InP/InGaAs heterojunction bipolar transistors

Abstract

Pnp InP/InGaAs heterojunction bipolar transistors have been modeled using a modified Gummel–Poon model, and the results are compared with experimental measurements and results from a commercial simulator. The model provides a good description of the transistor’s high frequency performance and describes the falloff in device performance at high current densities. The model overestimates the current gain at low current densities by neglecting hole recombination in the base side of the emitter–base space charge region.

Introduction

Recently, Pnp heterojunction bipolar transistors (HBTs) have been demonstrated in InP-based materials operating at microwave frequencies [1], [2], [3], [4], [5], [6]. Lunardi et al. [1] have reported InP/InGaAs Pnp HBTs with a current gain as high as 420, a cutoff frequency f_T of 10.5 GHz and a maximum frequency of oscillation f_max of 25 GHz. Stanchina et al. [2], [3] have reported comparable results for InAlAs/InGaAs Pnp HBTs. The devices are of interest for integration with Npn HBTs in complementary HBT (CHBT) based circuits [2], [3], [4], [5], [6], [7], [8] and for power applications [9]. Previously, we reported the development of a modified Gummel–Poon model for Pnp HBTs and compared the model’s results with experimental measurements for InAlAs/InGaAs HBTs [10]. In this work, we compare the results of the analytical model for InP/InGaAs Pnp HBTs with the experimental reports [1], and the results from a commercial numerical simulator for Ref. [11]. In Section 2, we briefly summarize the device physics included in the analytical model. A comparison of the experimental and simulation results is presented in Section 3. In Section 4, we draw conclusions regarding the limitations of the analytical model and means for improving it, and factors limiting the device’s performance.