MOSFET Guide, Meaning , Facts, Information and Description
The MOSFET, or Metal-Oxide-Semiconductor Field-Effect Transistor, is by far the most common Field effect transistor in both digital and analog circuits. The MOSFET is composed of a channel of n-type or p-type semiconductor material (see article on semiconductor devices), and is accordingly called an NMOSFET or a PMOSFET. Usually the semiconductor of choice is silicon, but some chip manufacturers, most notably IBM, have begun to use a mixture of silicon and germanium (SiGe) in MOSFET channels. Unfortunately, many semiconductors with better electrical properties than silicon, such as gallium arsenide, do not form good gate oxides and thus are not suitable for MOSFETs.The gate terminal is a layer of polysilicon (polycrystalline silicon; why polysilicon is used will be explained below) placed over the channel, but separated from the channel by a thin layer of insulating silicon dioxide. When a voltage is applied between the gate and source terminals, the electric field generated penetrates through the oxide and creates a so-called "inversion channel" in the channel underneath. The inversion channel is of the same type--p-type or n-type--as the source and drain, so it provides a conduit through which current can pass. Varying the voltage between the gate and body modulates the conductivity of this layer and makes it possible to control the current flow between drain and source.
| Table of contents |
|
2 MOSFET scaling 3 Other MOSFET design issues 4 Depletion mode MOSFETs and non-CMOS logic 5 DMOS |
The growth of digital technologies like the microprocessors has provided the motivation to advance MOSFET technology faster than any other type of silicon-based transistor. The principal reason for the success of the MOSFET was the development of CMOS digital logic, which uses MOSFETs as building blocks. The great advantage of CMOS circuits is that they allow no current to flow (ideally), and thus no power to be consumed, except when the inputs to logic gates are being switched. CMOS accomplishes this by complementing every NMOSFET with a PMOSFET and wiring the same input to both in such a way that whenever one is conducting, the other is not (see article on CMOS). Not only does this arrangement conserve energy, but perhaps more importantly it prevents overheating that would cause chips to fail. Overheating is a major concern in integrated circuits, since millions of transistors are packed into small chips.
Another advantage of MOSFETs for digital switching is that the oxide layer between the gate and the channel prevents any DC current from flowing through the gate, reducing power consumption. Even more importantly, this isolation between the gate and channel effectively isolates a MOSFET in one logic state from earlier and consequent stages, since the gate of one MOSFET is usually driven by the output from a previous logic stage. This isolation makes it easier for designers to design logic stages independently.
The MOSFET's strengths as the workhorse transistor in most digital circuits do not translate into supremacy in analog circuits, in which the bipolar junction transistor (BJT) has traditionally been seen as the transistor of choice, due largely to its high transconductance. Nevertheless, since it is both economically and operationally advantageous to incorporate digital and analog circuits onto the same chip, and since it is technologically difficult to fabricate BJTs and MOSFETs on the same chip, MOSFETs are widely relied upon for analog purposes as well. Ironically, the BJT has some advantages over the MOSFET in digital circuits, and some complex digital circuit designs incorporate BJTs to speed things up in critical locations. These mixed-transistor digital circuits are called BiCMOS (bipolar-CMOS) circuits.
For more on the BJT, which is not considered a field-effect transistor, see bipolar junction transistor.
Over the past decades, the MOSFET has continually been scaled down in size; typical MOSFET channel lengths were once several micrometres, but today's integrated circuits are incorporating MOSFETs with channel lengths of about a tenth of a micrometre. Until the late 1990s, this size reduction resulted in great improvement to MOSFET operation with no deleterious consequences. Historically, the difficulties with decreasing the size of the MOSFET have been associated with the semiconductor device fabrication process.
Smaller MOSFETs are desirable for two main reasons. First, smaller MOSFETs allow more current to pass. Conceptually, MOSFETs are like resistors in the on-state, and shorter resistors have less resistance. Second, smaller MOSFETs have smaller gates, and thus lower gate capacitance. These two factors contribute to lower switching times, and thus higher processing speeds.
To see why, note that the "switching" of a CMOS logic gate corresponds physically to the toggling of output voltages from high to low or vice versa by charging or discharging the gates of MOSFETs in the next logic stage. The time it takes a logic gate to accomplish this charging or discharging is directly proportional to the amount of current the MOSFETs in the logic gate are producing, since current is defined as the passage of charge per unit of time. Furthermore, since smaller MOSFETs have lower gate capacitance, and since the amount of charge on a gate is proportional to its capacitance, logic gates incorporating smaller MOSFETs have less charge to move. Indeed, these two factors combined traditionally resulted in a switching times proportional to the squared length of the MOSFET channel. In other words, integrated circuits using 1 micrometre MOSFETs would be roughly 100 times faster than those using 10 micrometre MOSFETs.
There is a third reason why MOSFETs have been scaled down in size: smaller MOSFETs can obviously be packed more densely, resulting in either smaller chips or chips with more computing power in the same area. Since the cost of producing integrated circuits is highly related to the number of chips that can be produced per wafer, this third reason for MOSFET scaling is perhaps as important as the first two.
Producing MOSFETs with channel lengths smaller than a micrometre is a tremendous challenge, and the difficulties of semiconductor device fabrication are always a limiting factor in advancing integrated circuit technology. Recently though, the MOSFET has shrunk to the point where its small size creates operational problems. The biggest problem with "deep submicrometre" MOSFETs--MOSFETs with channel lengths much smaller than a micrometre--is that they have a tendency to be always on. In more precise terms, such tiny MOSFETs tend to form inversion layers without the presence of a gate voltage. Of course, if MOSFETs are always conductive, they cannot perform their roles in digital switching, so semiconductor device engineers in industry and academia devote considerable energy to fighting this "leakage current" problem of submicrometre MOSFETs.
Integrated circuit designers have encountered many other scaling difficulties as they push for smaller transistors and faster switching speeds. Whereas traditionally switching time was roughly proportional to gate capacitance, as described above, MOSFET gate capacitances have decreased to the point where they are beginning to be overwhelmed by other capacitances, most notably interconnect capacitances, which result from the metal wires that connect inputs to outputs. The ever-increasing density of MOSFETs in a circuit, always considered to be a plus, is creating problems of substantial localized heat generation that can impair or even destroy circuit operation. Thus some of the motivations for scaling have become less compelling in recent IC generations, and switching time is no longer considered proportional to the square of gate length.
The gate oxide is a critical part of the MOSFET. It is desirable for the gate oxide to be as thin as possible, because a thin gate oxide results in higher on-currents, and also improves the leakage current problem. Unfortunately, most experts believe that today's gate oxides--around 2 nanometers (5 atoms!) thick--cannot be made thinner, because thinner oxides break down and create a conductive path from gate to channel. Insulators besides silicon dioxide, such as silicon nitride, are now being aggressively researched in hopes that their higher dielectric constants will have the same effect as decreasing gate-oxide thickness without the disastrous consequences of dielectric breakdown.
The primary criterion for the gate material is that it is a good conductor. Highly-doped polysilicon is an acceptable, but certainly not ideal conductor, and it also suffers from some more technical deficiences in its role as the standard gate material. So why use polysilicon instead of a metal like aluminum? The reason is simple: in the MOSFET IC fabrication process, the gate material must be deposited prior to high-temperature steps that would melt metals. To improve the performance of the gate, some manufacturers form a silicide by blending a metal into the polysilicon. Such a silicide has better electrical properties than polysilicon but doesn't melt in subsequent processing.
There are also depletion mode MOSFET devices, which are less commonly used than the standard "enhancement mode" devices already described. These are MOSFET devices which are doped so that a channel exists even without any voltage applied to the gate. When one then applies a voltage to the gate, the channel is depleted, which reduces the current flow through the device. In essence the depletion mode device is equivalent to a normally closed switch, while the enhancement mode device is equivalent to a normally open switch.
Historically, n-channel MOSFETs tended to be smaller and therefore cheaper to produce.
These were the driving principles in the design of NMOS logic which uses n-channel MOSFETs exclusively. However, NMOS logic consumes power even when no switching is taking place,
unlike CMOS logic which combines n-channel and p-channel MOSFETs on a single chip. With advances in technology, CMOS logic displaced NMOS in the 1980s to become the preferred choice for digital chips.The primacy of MOSFETs
MOSFET scaling
Reasons for MOSFET scaling
Scaling difficulties in state-of-the-art integrated circuits
Other MOSFET design issues
Gate oxide
Gate material
Depletion mode MOSFETs and non-CMOS logic