 | This articleneeds additional citations forverification. Please helpimprove this article byadding citations to reliable sources. Unsourced material may be challenged and removed. Find sources: "Semiconductor device" – news ·newspapers ·books ·scholar ·JSTOR(July 2017) (Learn how and when to remove this message) |
Asemiconductor device is anelectronic component that relies on theelectronic properties of asemiconductor material (primarilysilicon,germanium, andgallium arsenide, as well asorganic semiconductors) for its function. Its conductivity lies between conductors and insulators. Semiconductor devices have replacedvacuum tubes in most applications. Theyconductelectric current in thesolid state, rather than as free electrons across avacuum (typically liberated bythermionic emission) or as free electrons and ions throughan ionized gas.
Semiconductor devices are manufactured both as singlediscrete devices and asintegrated circuits, which consist of two or more devices—which can number from the hundreds to the billions—manufactured and interconnected on a single semiconductorwafer (also called a substrate).
Semiconductor materials are useful because their behavior can be easily manipulated by the deliberate addition of impurities, known asdoping. Semiconductorconductivity can be controlled by the introduction of an electric or magnetic field, by exposure tolight or heat, or by the mechanical deformation of a dopedmonocrystalline silicon grid; thus, semiconductors can make excellent sensors. Current conduction in a semiconductor occurs due to mobile or "free"electrons andelectron holes, collectively known ascharge carriers. Doping a semiconductor with a small proportion of an atomic impurity, such asphosphorus orboron, greatly increases the number of free electrons or holes within the semiconductor. When a doped semiconductor contains excess holes, it is called ap-type semiconductor (p for positiveelectric charge); when it contains excess free electrons, it is called ann-type semiconductor (n for a negative electric charge). A majority of mobile charge carriers have negative charges. The manufacture of semiconductors controls precisely the location and concentration of p- and n-type dopants. The connection of n-type and p-type semiconductors formp–n junctions.
The most common semiconductor device in the world is theMOSFET (metal–oxide–semiconductorfield-effect transistor),[1] also called the MOStransistor. As of 2013, billions of MOS transistors are manufactured every day.[2] Semiconductor devices made per year have been growing by 9.1% on average since 1978, and shipments in 2018 are predicted for the first time to exceed 1 trillion,[3] meaning that well over 7 trillion have been made to date.
A semiconductor diode is a device typically made from a singlep–n junction. At the junction of a p-type and ann-type semiconductor, there forms adepletion region where current conduction is inhibited by the lack of mobile charge carriers. When the device isforward biased (connected with the p-side, having a higherelectric potential than the n-side), this depletion region is diminished, allowing for significant conduction. Contrariwise, only a very small current can be achieved when the diode isreverse biased (connected with the n-side at higher electric potential than the p-side, and thus the depletion region expanded).
Exposing a semiconductor tolight can generateelectron–hole pairs, which increases the number of free carriers and thereby the conductivity. Diodes optimized to take advantage of this phenomenon are known asphotodiodes.Compound semiconductor diodes can also produce light, as inlight-emitting diodes andlaser diode
Bipolar junction transistors (BJTs) are formed from two p–n junctions, in either n–p–n or p–n–p configuration. The middle, orbase, the region between the junctions is typically very narrow. The other regions, and their associated terminals, are known as theemitter and thecollector. A small current injected through the junction between the base and the emitter changes the properties of the base-collector junction so that it can conduct current even though it is reverse biased. This creates a much larger current between the collector and emitter, controlled by the base-emitter current.
Another type of transistor, thefield-effect transistor (FET), operates on the principle that semiconductor conductivity can be increased or decreased by the presence of anelectric field. An electric field can increase the number of free electrons and holes in a semiconductor, thereby changing its conductivity. The field may be applied by a reverse-biased p–n junction, forming ajunction field-effect transistor (JFET) or by an electrode insulated from the bulk material by an oxide layer, forming ametal–oxide–semiconductor field-effect transistor (MOSFET).
Themetal-oxide-semiconductor FET (MOSFET, or MOS transistor), asolid-state device, is by far the most used widely semiconductor device today. It accounts for at least 99.9% of all transistors, and there have been an estimated 13 sextillion MOSFETs manufactured between 1960 and 2018.[4]
Thegate electrode is charged to produce an electric field that controls theconductivity of a "channel" between two terminals, called thesource anddrain. Depending on the type of carrier in the channel, the device may be ann-channel (for electrons) or ap-channel (for holes) MOSFET. Although the MOSFET is named in part for its "metal" gate, in modern devicespolysilicon is typically used instead.
 |
Two-terminal devices:
Three-terminal devices:
Four-terminal devices:
By far,silicon (Si) is the most widely used material in semiconductor devices. Its combination of low raw material cost, relatively simple processing, and a useful temperature range makes it currently the best compromise among the various competing materials. Silicon used in semiconductor device manufacturing is currently fabricated intoboules that are large enough in diameter to allow the production of 300 mm (12 in.)wafers.
Germanium (Ge) was a widely used early semiconductor material but its thermal sensitivity makes it less useful than silicon. Today, germanium is often alloyed with silicon for use in very-high-speed SiGe devices;IBM is a major producer of such devices.
Gallium arsenide (GaAs) is also widely used in high-speed devices but so far, it has been difficult to form large-diameter boules of this material, limiting the wafer diameter to sizes significantly smaller than silicon wafers thus making mass production of GaAs devices significantly more expensive than silicon.
Gallium Nitride (GaN) is gaining popularity in high-power applications includingpower ICs,light-emitting diodes (LEDs), andRF components due to its high strength and thermal conductivity. Compared to silicon, GaN'sband gap is more than 3 times wider at 3.4eV and it conducts electrons 1,000 times more efficiently.[5][6]
Other less common materials are also in use or under investigation.
Silicon carbide (SiC) is also gaining popularity inpower ICs and has found some application as the raw material for blue LEDs and is being investigated for use in semiconductor devices that could withstand very highoperating temperatures and environments with the presence of significant levels ofionizing radiation.IMPATT diodes have also been fabricated from SiC.
Variousindium compounds (indium arsenide,indium antimonide, andindium phosphide) are also being used in LEDs and solid-statelaser diodes.Selenium sulfide is being studied in the manufacture ofphotovoltaicsolar cells.
The most common use fororganic semiconductors isorganic light-emitting diodes.
All transistor types can be used as the building blocks oflogic gates, which are fundamental in the design ofdigital circuits. In digital circuits likemicroprocessors, transistors act as on-off switches; in theMOSFET, for instance, thevoltage applied to the gate determines whether theswitch is on or off.
Transistors used foranalog circuits do not act as on-off switches; rather, they respond to a continuous range of inputs with a continuous range of outputs. Common analog circuits includeamplifiers andoscillators.
Circuits that interface or translate between digital circuits and analog circuits are known asmixed-signal circuits.
Power semiconductor devices are discrete devices or integrated circuits intended for high current or high voltage applications. Power integrated circuits combine IC technology with power semiconductor technology, these are sometimes referred to as "smart" power devices. Several companies specialize in manufacturing power semiconductors.
Thepart numbers of semiconductor devices are often manufacturer specific. Nevertheless, there have been attempts at creating standards for type codes, and a subset of devices follow those. Fordiscrete devices, for example, there are three standards:JEDEC JESD370B in the United States,Pro Electron in Europe, andJapanese Industrial Standards (JIS).
Semiconductor device fabrication is the process used to manufacturesemiconductor devices, typicallyintegrated circuits (ICs) such asmicroprocessors,microcontrollers, and memories (such asRAM andflash memory). It is a multiple-stepphotolithographic and physico-chemical process (with steps such asthermal oxidation, thin-film deposition,ion implantation, etching) during whichelectronic circuits are gradually created on awafer, typically made of pure single-crystalsemiconducting material.Silicon is almost always used, but variouscompound semiconductors are used for specialized applications. Steps such as etching and photolithography can be used to manufacture other devices, such as LCD and OLED displays.[7]
The fabrication process is performed in highly specializedsemiconductor fabrication plants, also called foundries or "fabs",[8] with the central part being the "clean room". In more advanced semiconductor devices, such as modern14/10/7 nm nodes, fabrication can take up to 15 weeks, with 11–13 weeks being the industry average.[9] Production in advanced fabrication facilities is completely automated, with automated material handling systems taking care of the transport of wafers from machine to machine.[10]
A wafer often has several integrated circuits, which are calleddies as they are pieces diced from a single wafer. Individual dies are separated from a finished wafer in a process calleddie singulation, also called wafer dicing. The dies can then undergo further assembly and packaging.[11]
Within fabrication plants, the wafers are transported inside special sealed plastic boxes calledFOUPs.[10] FOUPs in many fabs contain an internal nitrogen atmosphere[12][13] which helps prevent copper from oxidizing on the wafers. Copper is used in modern semiconductors for wiring.[14] The insides of the processing equipment and FOUPs is kept cleaner than the surrounding air in the cleanroom. This internal atmosphere is known as a mini-environment and helps improve yield, which is the number of working devices on a wafer. This mini environment is within an EFEM (equipment front end module)[15] which allows a machine to receive FOUPs, and introduces wafers from the FOUPs into the machine. Additionally, many machines also handle wafers in clean nitrogen or vacuum environments to reduce contamination and improve process control.[10] Fabrication plants need large amounts of liquid nitrogen to maintain the atmosphere inside production machinery and FOUPs, which are constantly purged with nitrogen.[12][13] There can also be an air curtain or a mesh[16] between the FOUP and the EFEM which helps reduce the amount of humidity that enters the FOUP and improves yield.[17][18]
Some of the companies that manufacture machines used in the industrial semiconductor fabrication process includeASML,Applied Materials,Tokyo Electron, andLam Research. | This sectiondoes notcite anysources. Please helpimprove this section byadding citations to reliable sources. Unsourced material may be challenged andremoved.(October 2007) (Learn how and when to remove this message) |
Semiconductors had been used in the electronics field for some time before the invention of the transistor. Around the turn of the 20th century they were quite common as detectors inradios, used in a device called a "cat's whisker" developed byJagadish Chandra Bose and others. These detectors were somewhat troublesome, however, requiring the operator to move a small tungsten filament (the whisker) around the surface of agalena (lead sulfide) orcarborundum (silicon carbide) crystal until it suddenly started working.[19] Then, over a period of a few hours or days, the cat's whisker would slowly stop working and the process would have to be repeated. At the time their operation was completely mysterious. After the introduction of the more reliable and amplifiedvacuum tube based radios, the cat's whisker systems quickly disappeared. The "cat's whisker" is a primitive example of a special type of diode still popular today, called aSchottky diode.
Another early type of semiconductor device is the metal rectifier in which the semiconductor iscopper oxide orselenium.Westinghouse Electric (1886) was a major manufacturer of these rectifiers.
During World War II,radar research quickly pushed radar receivers to operate at ever higherfrequencies about 4000 MHz and the traditional tube-based radio receivers no longer worked well. The introduction of thecavity magnetron from Britain to the United States in 1940 during theTizard Mission resulted in a pressing need for a practical high-frequency amplifier.[citation needed]
On a whim,Russell Ohl ofBell Laboratories decided to try acat's whisker. By this point, they had not been in use for a number of years, and no one at the labs had one. After hunting one down at a used radio store inManhattan, he found that it worked much better than tube-based systems.
Ohl investigated why the cat's whisker functioned so well. He spent most of 1939 trying to grow more pure versions of the crystals. He soon found that with higher-quality crystals their finicky behavior went away, but so did their ability to operate as a radio detector. One day he found one of his purest crystals nevertheless worked well, and it had a clearly visible crack near the middle. However, as he moved about the room trying to test it, the detector would mysteriously work, and then stop again. After some study he found that the behavior was controlled by the light in the room – more light caused more conductance in the crystal. He invited several other people to see this crystal, andWalter Brattain immediately realized there was some sort of junction at the crack.
Further research cleared up the remaining mystery. The crystal had cracked because either side contained very slightly different amounts of the impurities Ohl could not remove – about 0.2%. One side of the crystal had impurities that added extra electrons (the carriers of electric current) and made it a "conductor". The other had impurities that wanted to bind to these electrons, making it (what he called) an "insulator". Because the two parts of the crystal were in contact with each other, the electrons could be pushed out of the conductive side which had extra electrons (soon to be known as theemitter), and replaced by new ones being provided (from a battery, for instance) where they would flow into the insulating portion and be collected by the whisker filament (named thecollector). However, when the voltage was reversed the electrons being pushed into the collector would quickly fill up the "holes" (the electron-needy impurities), and conduction would stop almost instantly. This junction of the two crystals (or parts of one crystal) created a solid-state diode, and the concept soon became known as semiconduction. The mechanism of action when the diode off has to do with the separation ofcharge carriers around the junction. This is called a "depletion region".
Armed with the knowledge of how these new diodes worked, a vigorous effort began to learn how to build them on demand. Teams atPurdue University,Bell Labs,MIT, and theUniversity of Chicago all joined forces to build better crystals. Within a year germanium production had been perfected to the point where military-grade diodes were being used in most radar sets.
After the war,William Shockley decided to attempt the building of atriode-like semiconductor device. He secured funding and lab space, and went to work on the problem with Brattain andJohn Bardeen.
The key to the development of the transistor was the further understanding of the process of theelectron mobility in a semiconductor. It was realized that if there were some way to control the flow of the electrons from the emitter to the collector of this newly discovered diode, an amplifier could be built. For instance, if contacts are placed on both sides of a single type of crystal, current will not flow between them through the crystal. However, if a third contact could then "inject" electrons or holes into the material, the current would flow.
Actually doing this appeared to be very difficult. If the crystal were of any reasonable size, the number of electrons (or holes) required to be injected would have to be very large, making it less than useful as anamplifier because it would require a large injection current to start with. That said, the whole idea of the crystal diode was that the crystal itself could provide the electrons over a very small distance, the depletion region. The key appeared to be to place the input and output contacts very close together on the surface of the crystal on either side of this region.
Brattain started working on building such a device, and tantalizing hints of amplification continued to appear as the team worked on the problem. Sometimes the system would work but then stop working unexpectedly. In one instance a non-working system started working when placed in water. Ohl and Brattain eventually developed a new branch ofquantum mechanics, which became known assurface physics, to account for the behavior. The electrons in any one piece of the crystal would migrate about due to nearby charges. Electrons in the emitters, or the "holes" in the collectors, would cluster at the surface of the crystal where they could find their opposite charge "floating around" in the air (or water). Yet they could be pushed away from the surface with the application of a small amount of charge from any other location on the crystal. Instead of needing a large supply of injected electrons, a very small number in the right place on the crystal would accomplish the same thing.
Their understanding solved the problem of needing a very small control area to some degree. Instead of needing two separate semiconductors connected by a common, but tiny, region, a single larger surface would serve. The electron-emitting and collecting leads would both be placed very close together on the top, with the control lead placed on the base of the crystal. When current flowed through this "base" lead, the electrons or holes would be pushed out, across the block of the semiconductor, and collect on the far surface. As long as the emitter and collector were very close together, this should allow enough electrons or holes between them to allow conduction to start.
The Bell team made many attempts to build such a system with various tools but generally failed. Setups, where the contacts were close enough, were invariably as fragile as the original cat's whisker detectors had been, and would work briefly, if at all. Eventually, they had a practical breakthrough. A piece of gold foil was glued to the edge of a plastic wedge, and then the foil was sliced with a razor at the tip of the triangle. The result was two very closely spaced contacts of gold. When the wedge was pushed down onto the surface of a crystal and voltage was applied to the other side (on the base of the crystal), current started to flow from one contact to the other as the base voltage pushed the electrons away from the base towards the other side near the contacts. The point-contact transistor had been invented.
While the device was constructed a week earlier, Brattain's notes describe the first demonstration to higher-ups at Bell Labs on the afternoon of 23 December 1947, often given as the birthdate of the transistor. What is now known as the "p–n–p point-contact germanium transistor" operated as a speech amplifier with a power gain of 18 in that trial.John Bardeen,Walter Houser Brattain, andWilliam Bradford Shockley were awarded the 1956Nobel Prize in physics for their work.
Bell Telephone Laboratories needed a generic name for their new invention: "Semiconductor Triode", "Solid Triode", "Surface States Triode" [sic], "Crystal Triode" and "Iotatron" were all considered, but "transistor", coined byJohn R. Pierce, won an internal ballot. The rationale for the name is described in the following extract from the company's Technical Memoranda (May 28, 1948) [26] calling for votes:
Transistor. This is an abbreviated combination of the words "transconductance" or "transfer", and "varistor". The device logically belongs in the varistor family, and has the transconductance or transfer impedance of a device having gain, so that this combination is descriptive.
Shockley was upset about the device being credited to Brattain and Bardeen, who he felt had built it "behind his back" to take the glory. Matters became worse when Bell Labs lawyers found that some of Shockley's own writings on the transistor were close enough to those of an earlier 1925 patent byJulius Edgar Lilienfeld that they thought it best that his name be left off the patent application.
Shockley was incensed, and decided to demonstrate who was the real brains of the operation.[citation needed] A few months later he invented an entirely new, considerably more robust,bipolar junction transistor type of transistor with a layer or 'sandwich' structure, used for the vast majority of all transistors into the 1960s.
With the fragility problems solved, the remaining problem was purity. Makinggermanium of the required purity was proving to be a serious problem and limited the yield of transistors that actually worked from a given batch of material. Germanium's sensitivity to temperature also limited its usefulness. Scientists theorized that silicon would be easier to fabricate, but few investigated this possibility. Former Bell Labs scientistGordon K. Teal was the first to develop a working silicon transistor at the nascentTexas Instruments, giving it a technological edge. From the late 1950s, most transistors were silicon-based. Within a few years transistor-based products, most notably easily portable radios, were appearing on the market. "Zone melting", a technique using a band of molten material moving through the crystal, further increased crystal purity.
In 1955,Carl Frosch and Lincoln Derick accidentally grew a layer of silicon dioxide over the silicon wafer, for which they observed surface passivation effects.[20][21] By 1957 Frosch and Derick, using masking and predeposition, were able to manufacture silicon dioxide field effect transistors; the first planar transistors, in which drain and source were adjacent at the same surface.[22] They showed that silicon dioxide insulated, protected silicon wafers and prevented dopants from diffusing into the wafer.[20][22] At Bell Labs, the importance of Frosch and Derick technique and transistors was immediately realized. Results of their work circulated around Bell Labs in the form of BTL memos before being published in 1957. AtShockley Semiconductor, Shockley had circulated the preprint of their article in December 1956 to all his senior staff, includingJean Hoerni,[23][24][25][26] who would later invent theplanar process in 1959 while atFairchild Semiconductor.[27][28]
-Gate_oxide_transistor_by_Frosch_and_Derrick.png)
After this, J.R. Ligenza and W.G. Spitzer studied the mechanism of thermally grown oxides, fabricated a high quality Si/SiO2 stack and published their results in 1960.[29][30][31] Following this research,Mohamed Atalla andDawon Kahng proposed a silicon MOS transistor in 1959[32] and successfully demonstrated a working MOS device with their Bell Labs team in 1960.[33][34] Their team included E. E. LaBate and E. I. Povilonis who fabricated the device; M. O. Thurston, L. A. D’Asaro, and J. R. Ligenza who developed the diffusion processes, and H. K. Gummel and R. Lindner who characterized the device.[35][36]
With itsscalability,[37] and much lower power consumption and higher density thanbipolar junction transistors,[38] the MOSFET became the most common type of transistor in computers, electronics,[39] andcommunications technology such assmartphones.[40] TheUS Patent and Trademark Office calls the MOSFET a "groundbreaking invention that transformed life and culture around the world".[40]
Bardeen's 1948 inversion layer concept, forms the basis of CMOS technology today.[41]CMOS (complementaryMOS) was invented byChih-Tang Sah andFrank Wanlass atFairchild Semiconductor in 1963.[42] The first report of afloating-gate MOSFET was made by Dawon Kahng andSimon Sze in 1967.[43]FinFET (fin field-effect transistor), a type of 3Dmulti-gate MOSFET, was proposed by H. R. Farrah (Bendix Corporation) and R. F. Steinberg in 1967[44] and first built by Digh Hisamoto and his team of researchers atHitachi Central Research Laboratory in 1989.[45][46]
Annual semiconductor unit shipments (integrated circuits and Opto-sensor-discrete, or O-S-D, devices) are expected to grow 9% [..] For 2018, semiconductor unit shipments are forecast to climb to 1,075.1 billion, which equates to 9% growth for the year. Starting in 1978 with 32.6 billion units and going through 2018, the compound annual growth rate for semiconductor units is forecast to be 9.1%, a solid growth figure over the 40-year span. [..] In 2018, O-S-D devices are forecast to account for 70% of total semiconductor units compared to 30% for ICs.
{{cite journal}}:ISBN / Date incompatibility (help)