Semiconductor Technology from A to Z

A transistor is an electronic semiconductor device that can be used for switching or amplifying current. Current flows through two terminals (drain, source), while the third (gate) is used for control. In addition to the field-effect transistor (FET), there is another fundamental type of transistor, the bipolar transistor. In this type, the terminals are called emitter (source in the FET), base (gate), and collector (drain). The bipolar transistor operates based on charge carriers of both polarities (hence bipolar), i.e. holes and electrons. In the field-effect transistor, also referred to as a unipolar transistor, depending on the design either electrons or holes participate in the transport of current.

The transistor described below is a so-called MOSFET (metal oxide semiconductor field-effect transistor). Although highly doped polysilicon is now typically used as the gate material rather than aluminium, the term MOSFET is still commonly used for this type of transistor. In this case, the term IGFET (insulated gate FET) would technically be more accurate. In newer transistors using high-k metal gate technology, however, the term MOSFET is once again correct, provided an oxide is still used as the insulator.

The transistor is the fundamental building block of semiconductor manufacturing; modern microchips contain several hundred billion transistors. By combining multiple transistors, all logic gates can be realized in order to generate the appropriate logical output signals from input signals. Transistors therefore form the core of every microprocessor, memory chip, and so on. The transistor is the technical component that has been produced by humankind in the greatest total quantity, and it has become indispensable in modern life.

During production, the transistor is built up layer by layer. The focus here is on the basic structure of a simple MOSFET; how these layers are actually fabricated is covered in the chapter Field-effect transistors in the Devices section.

The structure described here is the planar transistor: the gate lies flat on the surface of the wafer and controls the channel from above. This design shaped semiconductor technology for four decades and remains the correct starting point for understanding how the device works. In processes for the smallest structures, however, it has since been superseded, because a gate that acts from only one side can no longer reliably switch off the increasingly shorter channel. The chapter Transistors and Memory describes what its successors look like.

Enhancement-mode FET: Without a positive voltage applied to the gate, no free charge carriers in the form of electrons are available between source and drain, since the substrate is p-doped. In the steady state, holes are the majority charge carriers here, while electrons are the minority charge carriers.

A positive voltage applied to the gate attracts electrons from the substrate through the electric field (thereby pushing holes away), thus forming a conductive n-channel between source and drain. The insulating silicon dioxide layer prevents any current flow between the substrate and the gate.

Because the transistor blocks current flow without an applied voltage, it is also referred to as normally-off. The channel does not form gradually but only once a certain gate voltage, the threshold voltage, is exceeded. This is the most important characteristic parameter of the transistor: below it, the transistor is off; above it, the transistor conducts. It can be specifically adjusted through the doping of the substrate and the choice of gate material.

Depletion-mode FET: By lightly n-doping the region between source and gate, a transistor can be made conductive even without a gate voltage (a voltage between source and drain alone is sufficient). So-called depletion FETs, also known as normally-on transistors, only switch off when a voltage more negative than that at the source terminal is applied to the gate. This pushes away the electrons located beneath the gate — the conductive electron bridge is lost.

The second important type of transistor, alongside the field-effect transistor, is the bipolar transistor. Its mode of operation relies on both types of charge carriers (hence bipolar), electrons and holes. Bipolar transistors are faster than field-effect transistors, but they require more space and can therefore not be manufactured as cost-effectively.

Bipolar transistors essentially consist of two p-n junctions connected back to back, with the layer sequence n-p-n or p-n-p. The terminals of a bipolar transistor are referred to as emitter (E), base (B), and collector (C); the emitter and collector each have the same type of doping. Between the two terminals lies the very thin base layer, which is accordingly doped differently.

Described here is an NPN transistor built using the standard buried collector (SBC) design. The mode of operation of a PNP transistor is analogous; only the signs of the applied voltages need to be reversed.

The two p-n junctions are referred to below as EB (emitter-base) and CB (collector-base). Without an external voltage, space-charge regions form at EB and CB (see The p-n junction). With a negative voltage applied to the emitter and a positive voltage applied to the collector, the space-charge region at EB is reduced, while the one at CB is enlarged. If a positive voltage is now applied to the base, EB becomes conductive — electrons enter the base layer. Since this layer is very thin, the charge carriers can be injected into the collector, where they are drawn away due to the applied positive voltage. As a result, a current flows from emitter to collector. Almost all electrons reach the collector in this way, even at a low voltage at the base (>95 %), meaning that a relatively small base current (emitter to base) enables a very large collector current (emitter to collector).

The two deep p⁺-doped regions provide lateral isolation from other components. In addition to the transistor, a resistor is also required (not shown in the graphic), since bipolar transistors cannot be driven without current.

The basic structure and mode of operation of a FinFET do not differ from those of a conventional MOS field-effect transistor. Here too there are source and drain terminals, through which the current flows. A gate electrode controls the transistor. In contrast to the classic, planar field-effect transistor, however, the channel between source and drain is formed as a three-dimensional structure on the silicon substrate, so that the gate electrode can enclose it from several sides. This enables considerably improved electrical behavior: leakage currents can be reduced and the drive current controlled more effectively.

The three-dimensional structure, however, also creates new parasitic capacitances and critical dimensions that have to be optimized. In a FinFET, the gate length is measured parallel to the channel, while the gate width corresponds to twice the fin height plus the fin width. The height of the channel limits the drive current and the gate capacitance, while the width influences the breakdown voltage and short-channel effects, and the quantities that result from them, such as power consumption.

Context

The FinFET is no longer a future technology; rather, it has defined an era that has now come to a close. After being researched for decades, it entered volume production for the first time in 2011 and remained the standard device architecture for all high-performance processors for more than a decade. At the smallest feature sizes, it has since been superseded itself: as the fin becomes ever narrower and taller, it becomes increasingly difficult to control mechanically, and the gate width can only be adjusted in discrete steps – it results from the number of fins and cannot be chosen continuously. Its successor solves both problems by laying the fin down (see From FinFET to nanosheet).

Below, the structure of a multigate transistor with three gates (tri-gate) is described for the bulk process.

The DRAM Memory Cell: One Transistor, One Capacitor

A DRAM cell (Dynamic Random Access Memory) stores a single bit as electrical charge on a capacitor. An access transistor connects this capacitor to the bitline on demand — while the transistor is off, the charge stays isolated and the bit remains stored; once switched on via the wordline, the charge can be read out or rewritten. This pairing of one transistor and one capacitor is called a 1T1C cell and is by far the most area-efficient way to store a bit — considerably more compact than the six transistors of an SRAM cell, but with one decisive drawback: the stored charge slowly leaks away and must be periodically refreshed every few milliseconds, hence the name "dynamic."

The challenge in cell design is purely geometric: a capacitor needs area to store enough charge for a reliably readable signal — typically on the order of a few femtofarads. Yet with every new technology node, the available cell area keeps shrinking while the required capacitance stays nearly constant. The manufacturing industry has solved this problem in two fundamentally different ways.

Two Paths to the Same Goal: Stack versus Trench

The stack capacitor (dominant at Samsung, SK Hynix, and Micron) builds the needed area vertically above the access transistor: cylindrical or crown-shaped electrode structures rise above the wafer surface and can, within the limits of lithography and etch technology, grow nearly arbitrarily tall.

The trench capacitor (historically developed by IBM and used at, among others, Infineon/Qimonda) instead pushes the same area downward: a deep, narrow trench is etched several micrometers into the substrate, with the capacitor electrodes forming concentric cylindrical surfaces along its walls. The key advantage of this approach lies in the process sequence: the trench — the most demanding and critical part of fabrication — is formed before the access transistor, so the transistor itself is built on top of an almost planar wafer surface, without having to overcome the tall topography of an already-completed stack capacitor.

The diagram below compares both capacitor types in cross-section; how the complete trench process is built up step by step is covered in the chapter DRAM Fabrication: The Trench Capacitor Process in the Devices section.