1
Introduction

1.1 Overview of Heterojunction Bipolar Transistors

Semiconductor material systems can be categorized into silicon-based and III-V compound semiconductor-based devices [1, 2]. Silicon-based semiconductor devices, with their low-cost, high-volume production, have improved frequency response significantly as the channel length is made smaller and up to 22 nm. In contrast, compound semiconductor-based devices take advantages of their intrinsic material properties and offer superior device performance in high-frequency applications such as monolithic microwave integrated circuits. Alternatively, in terms of transistor operation principles, semiconductor transistor technologies can be categorized into two major types depending on their physical carrier transportation mechanisms: field effect transistors (FETs) and bipolar transistors. The bipolar transistors include bipolar junction transistors (BJTs) and heterojunction bipolar transistors (HBTs)

Table 1.1 shows the comparison of some device parameters for both FET and bipolar transistor devices [3-5]. FETs are majority carrier devices with lateral current conduction, while bipolar transistors are the vertical devices that allow the electron and hole conduction. The speed of the bipolar transistor device is determined by the transit time through the thin vertical base-collector (B-C) layers. The maximum speed of the FET is determined by a transit time and is controlled by the gate length defined by the lithographic techniques. FET devices are also referred as unipolar devices because the majority carriers are in principle responsible for the transport characteristics. Drain current in an FET is modulated by gate voltage through channel width modulation scheme. The amplification process in FET is characterized by a transconductance to assess the controllability of the gate voltage modulation over the output drain current. On the other hand, the collector current in bipolar transistor is modulated by the minority current injection from the base. Bipolar transistor is equivalent to a current amplifier as the input base current is "amplified" by a factor of current gain through the transistor and the output current is "collected" at the collector end.

Table 1.1 Comparison of FET and bipolar transistor

There are wide varieties of the HBT device technologies available for the implementation of microwave and radio frequency integrated circuits (RFICs). The commonly used HBT devices are as follows:

1. Gallium arsenide-based heterojunction bipolar transistors (GaAs HBTs)
2. Indium phosphide-based heterojunction bipolar transistors (InP HBTs)
3. Silicon-germanium-based heterojunction bipolar transistors (SiGe HBTs)

III-V compound HBTs (GaAs HBTs and InP HBTs) largely retain the advantages of their Si predecessors but extend them to higher frequencies. Additionally, a variety of disadvantages of Si bipolar transistors can be overcome. HBTs in the GaAs/AlGaAs material system have been the first beneficiaries of the improved materials. These devices are now becoming available commercially and are poised for application in a wide variety of high-performance circuits. HBTs enjoy several advantages over their conventional silicon cousins [6]. These include:

• A thinner base and lower base resistance which yields higher gain, cutoff frequency, and maximum oscillation frequency
• Higher transconductance due to the exponential output current to input voltage variation
• High power density since the entire emitter area can carry the current because of the low emitter resistance
• High breakdown voltage
• Lower 1/f noise
• Low parasitics

A cross section of a simple HBT is shown in Figure 1.1. In a single heterojunction device, the base, collector, and subcollector will all be of the same material, such as GaAs, while in the AlGaAs system (double heterojunction device (DHBT)), for example, a small mole fraction of aluminum is added to the emitter to increase the bandgap. HBT operation involves the following three steps [7]: (i) minority carrier injection from emitter to base, (ii) carrier transport in the base region, and (iii) carrier collection at the B-C junction. In normal operation (forward bias), electrons are injected from emitter into base crossing over the heterostructure barrier. For an abrupt heterojunction barrier, the electron injection is due to thermionic emission, while for a graded base-emitter (B-E) junction, the electrons diffuse to the base. The C-B junction is reverse biased, and the high electric field present in the space charge region is responsible for collection of electrons in the collector terminal.

Figure 1.1 Cross section of a typical heterojunction bipolar transistor with a single emitter finger, two base contacts, and two collector contacts

The detailed layer structure for GaAs/AlGaAs HBTs grown by molecular organic chemical vapor deposition (MOCVD) is shown in Figure 1.2. The dopants used are silicon for n type and carbon for p type. The sequence of growth starts with the n+-GaAs subcollector layer on a (100) semi-insulating GaAs substrate, followed by an n-GaAs collector layer. The p+-GaAs base layer is then grown followed by N-AlGaAs emitter. The final layer is n+-GaAs emitter contact layer. The emitter layer consists of a high-doped (5 × 1018 cm-3) GaAs cap and a wide bandgap Al0.3Ga0.7As layer. The cap layer is used to produce low-resistivity emitter ohmic contact. The doping and thickness of Al0.3Ga0.7As layer are chosen to minimize the emitter resistance and E-B capacitance while maximizing reverse breakdown voltage.

Figure 1.2 Cross section of GaAs/AlGaAs HBT

The InP-based HBTs offer the advantages over GaAs HBTs of a low turn-on voltage, higher electron mobility, better thermal dissipation, and better microwave performance while still obtaining a high collector-to-base breakdown voltage. The InP HBTs have been used successfully to implement complex digital ICs for 40 Gb/s optical communication. The InP HBTs used in this book were grown by gas-source molecular beam epitaxy (GSMBE) on semi-insulating (100) InP substrates supplied by a commercial vendor. Be and Si are used for p- and n-type dopants, respectively. The detailed layer structure of the InP/InGaAs/InP DHBT is shown in Table 1.2 [8]. An InGaAs/InP composite collector structure with a dipole doping at the InGaAs/InP interface is employed to avoid current blocking effect. The devices were fabricated with a triple mesa process with different emitter size. Nonalloyed Ti/Pt/Au were used for emitter, base, and collector ohmic contacts. Gold-electroplated air bridges were then used to connect the emitter, base, and collector contacts to the external wire-bonding pads.

Table 1.2 Epitaxial structure of InP/InGaAs/InP DHBT

Compared to the III-V compound semiconductor devices, the silicon-based device offers the advantages of low cost, high integration, and the possibility of a single-chip solution [9]. Compared with III-V compound devices, silicon has numerous practical advantages as a semiconductor material, including the following [10]: (i) an extremely high-quality dielectric (SiO2) can be trivially grown on Si and used for isolation; (ii) ease of growth of large,...