Semiconductor Hub

Semiconductors overview

A material’s ability to conduct electrical charge is usually defined by measuring its resistance to the flow of electrons. Metals offer little in the way of resistance, whereas wood or rubber are known as insulators rather than conductors. A semiconductor sits in the middle of these two extremes and provides scientists and engineers with a key tool for building complex circuits and devices. Semiconductors’ capacity to control how power reaches a device can be affected by heat, light, pressure or the induction of a magnetic field – all of which leads to complex switching and amplification options within circuits.

Semiconductors are right at the heart of modern industry and continue to help shape how we work, rest and play. For example, microprocessors use semiconductors (and thus make personal computing

possible) in the form of transistors, as do mobile phones and many other devices. The key element that differentiates a semiconductor is its ability to be “doped” or “gated” in order to become more conductive – thus shaping the electrical current that runs through it.

 

History of semiconductors

The early history of semiconductors relates to the results of observational experiments conducted by some of the great nineteenth century thinkers. Michael Faraday (1791-1867) heated up silver sulphide and noticed that there was a decrease in the substance’s resistance. In the same way, A.E Becqurel (1820-1891), a French physicist noted for his work in optics and electricity observed a voltage when light was shone between solid and liquid electrolytes. This “photovoltaic effect” as it is known was also independently corroborated in 1876 by William Grylls Adams (1836-1915) and Richard Evans Day, whose experiment with selenium and platinum paved the way for the development of modern technologies such as solar power.

The advent of the twentieth century saw several key milestones in the refinement and use of semiconductors. In 1901, the last year of what is generally referred to as the Victorian period, Jagadis Chadra Bose invented a semiconductor device he dubbed “cat whiskers”. It was designed to detect radio waves, and represented the first patented device to use semiconductors in any capacity.

Perhaps the most pivotal discoveries were made by John Bardeen, Walter Brattain and William Shockley in 1947. Building on the work of previous scholars and taking advantage of new theories of conduction, these three men invented the transistor at Bell laboratories and changed the world. Now it was possible to switch or amplify electronic signals and electrical power from a compact semiconductor device, where valves and other large-scale apparatus had been necessary before. Transistors – and their constituent semiconductors became a fundamental part of how we live today, and paved the way for the radios, calculators and – ultimately – the computing revolution.

 

Overview of different semiconductor types

Proximity Sensor

A proximity sensor is a device designed to detect nearby objects. Unlike a pressure sensor, it does not need to physically touch an object in order to detect it. It works by emitting a beam of electromagnetic radiation (infrared), then analysing anomalies in the field as the signal returns to the device. Depending on what materials the sensor is required to target, different sensors may be used. For example, where the target is composed of metal, an inductive proximity sensor is required. If the target is plastic a photoelectric proximity sensor with a photodetector is required.

Often used in industry as part of the machine monitoring process for measuring the variations between shafts and support bearings, the proximity sensor can be found in the energy sector in steam turbines.

 

Semiconductors

Designed to offer control and switching capabilities in electronic circuits, semiconductors superseded valve technology and have greatly reduced the size – at the same time as greatly increasing the range – of electronic devices available today. Semiconductors are at the root of transistor (referred to as solid state) technology and are also integral to computer memory, audio processing and amplification circuits.

 

Transistors

Transistors are semiconductor devices that control and switch current depending on its function within a circuit. In essence they can control output of a signal relative to its input and in this way they can be used as an amplifier. They can also be used to switch current on and off as a consequence of the actions of other elements of the circuit.

Transistors are manufactured in two forms. The first is known as a bipolar transistor and features three terminals – base, collector and emitter. When current from the base terminal flows to the emitter, the bipolar transmitter controls or switches larger current in. The second type, the field-effect transistor, has gate, source and drain labels. Voltage arriving at the gate can be used to control current running between the gate and the drain and thus is useful in lower-power applications such as logic gates.

 

Accelerometers

Accelerometers are electromechanical devices that inhabit many of the communications and computing devices in common use today. Smartphones, for example, use accelerometers to detect movement and switch screen orientation relative to the phone’s position.

As the name suggests, accelerometers measure forces of acceleration. By taking a measure of the amount of static acceleration provided by gravitational forces, information about the angle of a device – for example a mobile phone or laptop – relative to the earth’s surface becomes possible. Factoring in the capacity to measure dynamic acceleration means that devices can become aware of their angle of tilt and the rate at which they are moving.

Recent applications of the technology have seen major manufacturers incorporate accelerometers into laptops. As soon as sudden, violent acceleration is detected, the laptop powers down the hard drive in order to try and minimise damage.

Accelerometers often use crystal structures that are incredibly responsive to changes in acceleration or orientation. This piezoelectric effect creates voltage which is then rendered into visible outputs by the device’s software.

 

Microprocessor

At the forefront of the computer revolution, the microprocessor is the “brain” of any modern computer system. Essentially a collection of transistors arranged on one chip, the microprocessor handles all data processing and logic functions in the device in which it is housed.

In effect, the chip is an integrated circuit comprised of a small amount of silicon (the semiconducting base material) onto which transistors are etched. In case of relatively “simple” devices, this might mean just a few thousand transistors. More complex and powerful chips are likely to contain millions of transistors.

Chips are generally classified by their ability to perform a number of instructions per second. In computing terms this is described as MIPS – or millions of instructions per second. Related to this measure is clock speed – or the maximum rate at which the chip will run. In modern PCs, this is measured in GHz and describes the number of operations the chip can complete in a given period of time. For example, a chip with a clock speed of 3.5 GHz should be capable of achieving 3.5 billion operations a second.

Microprocessors are also commonly found in electrical appliances and industry, where they control and regulate the outputs of numerous devices.

 

Zener Diode

Unlike standard semiconductor diodes that only allow current to flow in one direction, a Zener Diode will allow current to flow both ways if subjected to a sufficiently high voltage.

This voltage threshold is described as the breakdown voltage – also known as the Zener voltage.

Standard diodes have a Zener voltage (or resistance to reverse current), but if reverse current above that threshold moves through the circuit then, the diode is irreparably damaged. A Zener diode has the Zener voltage threshold set at a much lower point (often not much more than 2.4 volts). When reverse current is fed back through the diode a “controlled breakdown” occurs in which the diode remains undamaged. This is true regardless of the value of reverse voltage relative to the Zener voltage. The voltage drop is always equal to the Zener voltage of the diode - for this reason, Zener diodes are often used to regulate circuits.

 

Important technical elements of a semiconductor.

As stated at the beginning of this article, a semiconductor is a material that partially conducts current. Physically it can be said to be half way between an insulator – essentially a material with little or no conductivity and a conductor – its polar opposite. In order to better understand this concept, consider the characteristics and organisation of electrons in an atom. Electrons are layered in such a way that they form shells. The outer layer of these shells are known as “valence shells”. It is these electrons on the outer layer that form bonds with neighbouring atoms. In the case of materials commonly used as semiconductors, such as silicon, the valence shell has four electrons. Where the atoms on either side of the electrons are of the same type, they form into a structure known as “crystals”. The best way to imagine this is to think of a “lattice” occupied by atoms bound together in a pattern resembling chambered wire mesh. Each of the atom’s four electrons are shared with a neighbouring atom – or to put it another way, each atom in the silicon is bonded with four other atoms through the electrons in the outer shell.

The physical structure of pure silicon crystals means they are poor conductors in and of themselves. The process of turning them into something useful from a technological and manufacturing perspective is known as “doping”. This process sees a “dopant” added to a crystal in order to effect the differences in behaviour that make modern electronics possible.

Dopants create two types of semiconductor – the N-type and the P-type. The N-type semiconductor is created when the dopant has five electrons in its valence shell. These electrons bond with the surrounding atoms in the structure of the crystal, but the “surplus” electron is left without bonds to the adjacent atoms. It is these “free electrons” that behave like electrons found in highly conductive material, such as copper.

P-type conductors occur when a dopant material  - such as boron - possesses three electrons in its valence shell. Adding minute amounts to the silicon crystal, the atom bonds with four silicon atoms. The difference between this and P-type semiconductors is, whilst the three electrons of the boron atom bond with the silicon, there is no fourth electron – and thus a “hole” is created. The physical properties of the hole means that electrons are attracted to fill the vacant space, but whilst they may occupy that space they leave a vacancy behind them. For this reason, a feature of P-type semiconductors is that electrons are constantly moving within the crystal as they try to occupy
these spaces.

Once either N or P-type semiconductor have been doped, the application of voltage sees current flow for the same fundamental physical reasons as it does in materials that are “natural” conductors. The simultaneous pull and push of the voltage on the electrons organises them into directional current which can then be used for many different applications.

For scientists and engineers working on electrical and electronic design, the advantages of having compact materials that give varying results depending on the stimulus, and offer a switching capability, is a massive advantage. Take computers as one example. The “brain” of a computer is made of chips – essentially electronic components (transistors) etched onto a semiconducting silicon layer. The capability to “switch” current through logic gates (a part of a digital circuit that defines a sequence of events that will take place given certain actions. These logic gates are known as “Boolean gates”) combined with the ability to store and retrieve information gives computers the capability of processing data. Without this “solid state technology” - these functions would be confined to the expensive and cumbersome valve technology that predated the semiconductor revolution.

 

Semiconductors from a manufacturing perspective

Most semiconductors are manufactured as silicon wafers. The process involves heating up purified polycrystalline until it changes to a molten liquid, then adding a small amount of solid silicon to the molten mix. As the solid silicon is drawn out of the liquid, it cools down and forms a single crystal. This crystal – now described as an ingot – is subsequently ground down until the diameter measurements are uniform, after which it is cut into very small wafers.

At this point the wafers are transferred to a facility that has clean room status (an area with regular checks to make sure that the wafers are not subjected to even slight impurities from environmental factors) and prepared to be turned into integrated circuits via etching – the process by which the pattern applied using photoresist techniques is baked into the semiconductor, and doping techniques - the addition of additional materials that alter the conductive properties of the silicon.