Integrated Circuits

views updated

Integrated Circuits

Integrated circuits are electronic devices that contain many transistors , resistors , capacitors , and connecting wires in one package. Integrated circuits, also called ICs or chips, were invented in 1959 and have become critical components in virtually all electronic devices, including computers, radios, televisions, and videocassette recorders. The microprocessor is the most complex integrated circuit, and also the most complex single device of any kind, ever produced.

Integrated circuits are produced on a piece of semiconductor crystal. As its name implies, a semiconductor is a material that conducts electricity better than an insulator, but not as well as a conductor. The conductivity can be adjusted by introducing other elements into the crystal during manufacture. Conductivity is also changed by applying an electric field to the crystal.

Elements Used as Semiconductors

Silicon is the most commonly used semiconductor. It is the second most common element in the Earth's crust, after oxygen. To be used in an integrated circuit, silicon must be processed to extreme purity. The silicon is heated until it melts, and is allowed to cool very slowly. If cooled too quickly, silicon has a tendency to crack, which would render it unusable for use in an integrated circuit. Once cooled, the silicon ingot is sliced into thin sheets. Many identical integrated circuits are formed simultaneously on each sheet.

In the periodic table of the elements, silicon is one of the group IV elements. This means each atom has four electrons in its outer shell available for combination with other atoms. It also means each atom would like to have four more electrons from adjacent atoms to fill up the outer shell. These characteristics cause pure silicon to form a crystalline structure. Germanium , the second most popular semiconductor material, is also in group IV, directly below silicon in the periodic table.

For a semiconductor to be useful, trace amounts of other elements must be carefully added to the crystal. This process is called doping . Phosphorous and boron are common doping elements. Phosphorous is a group V element that has five electrons in its outer shell. The extra electrons from the phosphorous atoms, each with a negative charge, are available to conduct electricity. Silicon doped with phosphorous is called N-type (for negative) silicon.

In a similar way, boron, a group III element, is used to produce P-type silicon. Other elements from groups III and V may be used as well. N-type and P-type silicon together can form transistors, the most important elements in an integrated circuit.

Transistors in Integrated Circuits

The transistor was invented at Bell Labs in 1947 by Walter Brattain (1902–1987), John Bardeen (1908–1991), and William Shockley (1910–1989). They were interested in developing a solid state amplifier that would replace the vacuum tube . Vacuum tubes are large, consume a lot of power to heat the filament, and are subject to filament burn out.

Transistors are much smaller, less costly, and more efficient than vacuum tubes. The transistor is a three-terminal solid state electronic device. The terminals are called the collector, the emitter, and the base. Electricity is conducted between the collector and emitter to a greater or lesser extent depending on the current of the base. This allows a small base current to control a much larger current in the collector.

As transistors came into widespread use, circuit designs using large numbers of them were limited by a problem: connecting them together. The solution to this problem occurred independently to Jack Kilby (1923–) at Texas Instruments and Robert Noyce (1927–1990) at Fairchild Semiconductor in 1959. Many transistors could be produced on one semiconductor base simultaneously. Noyce also determined how to produce resistors, diodes, capacitors, and connecting wires on the silicon chip. The integrated circuit was born.

Types of Integrated Circuits

A transistor can be operated either as an amplifier or as a switch. When used as an amplifier, the transistor collector current replicates the base current but at a much larger scale. This amplification enables a small signal, such as that produced by a microphone, to be reproduced with enough power to drive a large device, such as loud speakers. A transistor used as an amplifier is said to be operating in its linear range, neither all the way on nor all the way off. Linear integrated circuits are used in signal processing analog signals, such as in radios and communications systems and also in analog computers.

In digital integrated circuits, transistors are used as switches. The transistor is not operating in its linear range; there is no attempt to make the output replicate the input. Instead it is either fully on or fully off, representing either a logic one or zero. Digital integrated circuits are the heart and brains of digital computers.

The first digital integrated circuits implemented logic gates and simple memories using flip-flops. A gate is a logic device with no memory; the output at any time is determined by the inputs at that time. The simplest gate is a one input device called a "not" gate. The output is simply the opposite of the input. In a two-input "and" gate, the output is one (true) if both inputs are one, and zero otherwise. In a two-input "or" gate, the output is one (true) if either input is one, and zero otherwise. Other gates include "nand" (an "and" gate followed by a "not" gate), and "nor" gates. A flip-flop is a one-bit memory. It can be made by interconnecting the inputs and outputs of two "nand" gates. A flip-flop can be flipped to a one state or flopped to a zero state. It remains in a state until switched to the other. Gates and flip-flops are the most basic computer building blocks.

Smaller and More Powerful

Over time, the complexity of integrated circuits has increased, while the size of features continues to shrink. The feature size of an integrated circuit is indicated by the width of a "wire," measured in microns (one micron is one millionth of a meter).

In 1971 Intel produced the first microprocessor, designated 4004. It contained 2,300 transistors, used 10 micron technology, and powered an electronic calculator. The original IBM-PC was based on the Intel 8088 processor, which was first produced in 1979 and contained 29,000 transistors (3 micron technology). In 1989 Intel introduced the 80486 processor, containing 1.2 million transistors (0.8 micron technology). In 1999 the company began production of the Pentium III Xeon processor, with 28 million transistors on a silicon chip (0.18 micron technology). A year later, it debuted its Pentium IV processor, which featured 42 million transistors on a silicon chip (0.13 micron technology). The same trend toward smaller and more powerful devices is evident in memory chips, and in microprocessors from other manufacturers, such as Texas Instruments, Motorola, Advanced Micro Devices, and International Business Machines.

see also Generations, Computers; Microcomputers; Minicomputers; Transistors.

Donald M. McIver