Presentation is loading. Please wait.

Presentation is loading. Please wait.

Cross-sectional view of single wafer capacitive microphone

Similar presentations

Presentation on theme: "Cross-sectional view of single wafer capacitive microphone"— Presentation transcript:

1 Cross-sectional view of single wafer capacitive microphone

2 One of the earlier microphones was designed by Scheeper et al. in 1991
One of the earlier microphones was designed by Scheeper et al. in The design is based on using a single silicon wafer with a 1 µm-thick layer of LPCVD nitride grown on each side. The nitride was then patterned and etched in KOH. On the top side of the wafer, aluminum was evaporated as a sacrificial layer which was between 1 and 3 microns thick. Another micron-thick layer of nitride was grown on top of the aluminum layer using PECVD. Then an adhesion layer of 30 nm titanium was evaporated on the surface, followed by the addition of the gold electrode. After patterning, the acoustic holes were etched in the Ti/Au layers and in the PECVD nitride. Using the acoustic holes, the aluminum sacrificial layer was etched. Finally, the last step of the fabrication was to evaporate 100 nm of aluminum on the bottom side of the wafer.

3 microphone fabrication process

4 (a) Growth of 1 pm LPCVD silicon nitride
(a) Growth of 1 pm LPCVD silicon nitride. patterning and anisotropic etching of V grooves and diaphragm in 33wt.% KOH solution (73'C). (b) Evaporation of 3pm aluminium, growth of 1 um PECVD silicon nitride, evaporation and patterning of a 30nm titanium adhesion layer and a 30nm gold electrode, etching of accesslacoustic holes in TilAu and PECVD silicon nitride. (c) Etching of aluminium sacrificial layer, drying and evaporation of lOOnm aluminium on siliconlsilicon nitride on the reverse side (counter electrode)

5 Cross-sectional view of single wafer capacitive microphone

6 This microphone consists of a rigid 20 µm thick monocrystalline silicon back plate with acoustic holes, a 0.3 µm thick metallized low stress LPCVD silicon nitride membrane, and an air gap of 1.3 µm produced by a sacrificial layer of porous silicon and SiO2.

7 This microphone was made from a double-sided polished p-doped silicon wafer. The first fabrication step was to p+ doped both sides of the wafer. A thin nitride layer was deposited by LPCVD for use as a mask for the porous silicon formation. The porous silicon was created in a portion of the heavily p-doped silicon by etching in 25% HF and 50% ethanol in water. An electrical current was then sent through the porous silicon formation. The next step was to remove the masking layer and then sputter a 0.8 micron oxide layer onto the top side of the wafer. The oxide and the porous silicon would later be used as the sacrificial layer to create the air gap. Then a 300 nm layer of nitride was deposited with LPCVD on both sides. The nitride was used as a masking layer to create contact holes on the front side and windows on the back side of the wafer. After patterning the nitride by a mask, the holes and backside were etched in KOH. Aluminum was deposited on the front and back of the wafer to act as an RIE etch mask and a conductive layer (Linder et al. 1996). Acoustic holes were etched into the backplate to reach the porous silicon by RIE in SF6 plasma. The porous silicon was etched with KOH, and then the oxide layer was etched with HF. This last etching step releases the nitride diaphragm.


9 Cross-sectional view of a condenser microphone with polyimide

10 In this microphone, the diaphragm and back plate are made of a polymer (polyimide). In this microphone, the silicon substrates used were 3”, <100>, p-type, 5 Ω cm wafers. The thickness of the substrates was 400 μm. First, a 1.5 μm thick layer of SiO2 was grown by wet oxidation at 1150 oC. The SiO2 was subsequently etched in BHF to define the active sensor area on the backside of the substrate. During the etching, the front was protected with photoresist. The substrate was then etched in a KOH solusion (33 wt.%, 73 oC) until the thickness in the sensor area was reduced to approximately 50 μm. After rinsing, a thin diaphragm electrode was evaporated and patterned on the front, using a standard lift-off process with photoresist. The electrode was a Cr/Au/Cr multilayer, with a thickness of 4/20/4 nm.

11 The first layer of polyimide, which is later to be the diaphragm, was then spun on the front. The polyimide was then cured for 1 h at 300 oC in an N2 atmosphere, and the thickness of the polyimide diaphragm was measured to be μm. An Al sacrificial layer was then deposited and patterned, using electron-beem evaporation and a standard Al etch solution. The thickness of the sacrificial layer was 1.5 μm. Subsequently, the backplate electrode was deposited and patterned using resistive evaporation and lift-off. The thickness of the CR/Au/Cr multilayer electrode was 10/200/10 nm. After rinsing the substrate, the polyimide backplate layer (~18 μm) was spun on to the front. The sacrificial Al layer was then etched for 3 h using the standard Al etch solution. The final step was to release the polyimide diaphragm by etching the remaining silicon under the diaphragm


13 Cross section of the polysilicon diaphragm condenser microphone

14 A condenser microphone using a polysilicon diaphragm has been presented by Hsu et al. in The device consists of an n-type silicon substrate, a thin low-stress phosphorus doped polysilicon diaphragm, a P+ perforated back plate, and the metal contacts. To fabrication of this microphone, on (100) n-type silicon wafers, a 1 μm thick wet oxide is first grown at 1100 oC for 3 h. This oxide layer is patterned and etched in the buffered HF (5:1 BHF) for 12 minutes serving as a mask for the deep boron diffusion. A deep P+ boron diffusion is next introduced into the silicon from a solid source at 1175 oC for 15 h. followed by 20 minute wet oxidation at 1000 oC. The thick boron diffusion forms the stationary back electrode and the measured thickness is about 13 μm. The oxide was then stripped in a 1:1 HF: H2O solution for 4 minutes.

15 A 2 μm thick layer of LPCVD low-temperature oxide (LTO) is deposited at 420 oC for 4 h and patterned in 5:1 BHF for 23 minutes. This oxide provides isolation for the two electrodes. A 0.3 μm thick layer of low stress LPCVD SiN is deposited at 875 oC. This layer is patterned and etched in hot phosphoric acid for 3 h using a 0.5 μm layer of LTO as a mask. This nitride layer protects the passivation oxide from a subsequent the sacrificial etch. A 4 μm thick LTO sacrificial layer is next deposited defining the air gap electrode spacing. This oxide is patterned and etched 5:1 BHF for about 20 minutes.

16 Next, a 2 μm thick layer of LPCVD low-stress polysilicon is deposited at 588 oC. This material showed an unannealed tensile residual stress of about 100 MPa. The deposition is followed by a phosphorus ion implantation of 7x1015 cm-2 at 50 KeV. The remaining 1 μm thick layer of polysilicon is next deposited. The polysilicon is next annealed at 1050 oC for 1 h to redistribute the diaphragm dopants and remove as much residual stress as possible. The poly layer is next patterned and etched first using RIE with 20:5 SF6:O2 sccm, at 40 mT, and 60 W for 15 minutes, followed by a wet etch in 950:500:50 NHO3:H2O: NH4F for 25 minutes.

17 A 0.6 μm thick LTO mask is deposited and patterned in BHF for 7 minutes to define the contact area of the back plate. The nitride over the contact area is then etched in hot phosphoric acid for 3 h. A second 0.5 um thick LTO layer is deposited followed by a 0.2 um Al evaporation. The LTO protects the front side of the wafer during the backside etch and the metal is used to pattern the back to front alignment key. The backside oxide is patterned and etched in 5:1 BHF for about 8 minutes. The wafer is then anisotropically etched in EDP for 8 h at 110 oC. After striping the protective LTO in 5:1 BHF for 20 min., the wafers are dried. Cr and Au are next evaporated forming the contact pads with thickness of 50 and 400 nm. The metal is next patterned and wet Au and Cr etchants for 4 and 1 min, respectively. Finally, the device is released in concentrated HF for 1 h. in this operation, the HF removes the sacrificial LTO from the backside while the wafer front is protected by the SiN layer

18 Simplified fabrication process of the microphone

19 Cross-section of the microphone chip

20 The component consists of a low-stress polysilocon bending diaphragm and a high-stress polysilcon perforated back plate that are separated by an air gap. The lower membrane bends due to a pressure difference, while the perforated membrane remains steady and the capacitance between the membranes changes.

21 The components were fabricated by combining the bulk and surface-micromachining techniques. First, double-sided polished 4-in. p-type 100 silicon wafers were thermally oxidised. The oxide serves as an electrical insulator for the electrodes and as a masking and stopper layer in TMAH etching. A polysilicon layer for the bending membrane was then deposited using the standard LPCVD and patterned with RIE plasma etching. A sacrificial TEOS oxide layer was deposited and small holes for anti-sticking spikes were etched. The idea behind the anti-sticking spikes is to reduce the contact area between the membranes. Next, a second polysilicon layer for the perforated membrane was deposited and patterned. An LPCVD silicon nitride layer was deposited for top insulation, and contact holes for the electrodes and the substrate were etched through the nitride and oxide layers. Before the etching in TMAH, the silicon substrate, a protective LTO oxide layer was deposited on the front side of the wafers, and the thermal oxide at the backside was patterned using a double-sided lithography. After the TMAH etching, the sacrificial oxide layers were etched in a PSG etch. PSG etch was used instead of HF due to its better selectivity with respect to aluminium. The wafers were rinsed in water and tertiary butanol and dried using freezing and sublimation. Finally, the wafers were diced into chips by cleaving along the v-shaped grooves formed in the TMAH etching step.

22 Fabrication process of the microphone

23 Cross-section of the micro-machined capacitive microphone

24 The MEMS Structure in the IntelliSuite Design Environment

25 The device used a polysilicon Germanium diaphragm, an air gap and a gold-coated nitride back plate. The process starts with a <100> oriented n-type silicon wafer. A 1.0 micron thick silicon nitride, Si3N4, layer is deposited by LPCVD process. Over the nitride layer, a 30 nm thick layer of titanium is sputtered, which is followed by the evaporation-deposition of a 30 nm thick layer of gold. Next, a four micron thick layer of thermal oxide is deposited and then patterned. This is next followed by the deposition of a 300 nm thick layer of low-stress PECVD SiN layer. This nitride layer will act as an isolator between the poly SiGe diaphragm and the backplate electrode (Chowdhury et al. 2002).

26 The PECVD nitride layer is then patterned and etched in hot phosphoric acid using a 0.5 μm thick layer of low temperature oxide. In the next step a 40 nanometer thick layer of polysilicon is deposited by LPCVD process. This polysilicon layer will act as the nucleation layer for the subsequent deposition of polysilicon germanium (poly SiGe). The poly SiGe (30% Ge) is deposited by LPCVD process using SiH4 and GeH4 gases. The poly SiGe layer is then ion-implanted with phosphorous and annealed at 925 oC to reduce the residual stress. The poly SiGe layer is then patterned and etched using reactive ion etching with SF6/CH4 plasma. After that a thin layer of low-temperature oxide is deposited and patterned in Buffered Hydrofluoric acid (BHF) to define the contact hole for the backplate (gold electrode).

27 The nitride over the contact area is then etched by hot phosphoric acid. Next, the backside of the wafer is patterned using a 0.5 μm thick layer of low temperature oxide (LTO). The silicon wafer is then etched using SF6/O2 plasma. Next, the silicon nitride backplate is patterned using LTO and the silicon nitride layer and titanium layer are plasma etched using CF4/O2. The gold layer is then RIE etched using Cl2. The protective LTO is then stripped. Chromium and Gold are next evaporated to form contact pads with a thickness of 50 and 400 nm, respectively. The contact pads are then patterned and wet etched. Finally, the sacrificial thermal oxide layer is etched by BOE to release the diaphragm and then the wafer is freeze-dried using cyclohexane at atmospheric pressure under a continuous flow of nitrogen

28 Significant Microphone Fabrication Steps

29 Fabrication Process of the Capacitive Structure
The process uses five masks and starts with a double side polished p-type (100) silicon wafer as a substrate. First a 4-inch p-type (100) silicon wafer is cut into several pieces to be used as substrate. Each of the samples should be cleaned using standard cleaning procedure to remove organic contaminants such as dust particles, grease or silica gel and then remove any oxide layer from the wafer surface prior to processing. The first step in the cleaning process is to clean the wafer using ultrasonic in the acetone solution for 5 minutes. The second step is to put the samples into the methanol solution using ultrasonic for 5 minutes. Final step is to dip the samples in a 10:1 DI water-HF solution (10% HF) until hydrophobic (i.e. no water can stick to wafer). This will remove native oxide film.

30 Fabrication process for DRIE of silicon wafer
A 300-μm-thick (100)-oriented silicon wafer is used for the substrate. In step (1), a 300 nm-thick Al, which is a masking material for DRIE, is evaporated on surface of the silicon wafer. In step (2), approximately 1.5 μm of thin positive photoresist (AZ-1500) was spin-coated on the Al surface. In step (3), it was exposed for 10 sec using a contact mask aligner with a UV light source. In step (4), it was developed for 60 seconds in AZ 300 MIF. In step (5), Al was wet etched for 17 seconds in 10% hydrofluoric acid (HF) solution. In step (6), the Photoresist layer was removed in acetone. In step (7), a 260 μm deep cavity was fabricated in the silicon wafer by the DRIE system. In step (8), the Al thin film is removed by 10%HF solution.

31 A 0.5 µm thick wet oxide is grown on silicon wafer.
This oxide layer is patterned and etched in the buffered HF (6:1 BHF) serving as a mask for the phosphorus diffusion. A n+ phosophorous diffusion is next introduced into the silicon wafer as a back plate electrode. The doped area acts as a back plate electrode. A constant source diffusion process was used to obtain the n+-silicon layer with a depth of 0.4 µm and surface concentration as high as 1020 atoms/cm3. It was performed at 1000 oC for 30 minutes in N2 environment. After diffusion step, the wafer was dipped in a buffered oxide etch (BOE) to remove any silicon dioxide grown during the diffusion.

32 Next, a Silicon-On-Glass (SOG) of 1
Next, a Silicon-On-Glass (SOG) of 1.5 μm thick was spun with spin coater in order to form a sacrificial and insulating layer between diaphragm and back plate electrode. The layer was gradually hot plate baked from 20 oC to 180 oC for 5 minutes and after 4 minutes gradually cooled down.

33 The isolating layer is then patterned and etched in PAD etch solution for 3.5 minutes to define the contact area on the back plate.

34 After that, a 3 m thick layer of Al is evaporated on SOG as a material of diaphragm.
The Al layer is then patterned using photoresist mask (AZ1500) to define the geometry of the diaphragm (0.5 mm x 0.5 mm), contact pad (250 µm x 250 µm), anchors and to form the circular acoustic holes (radius = 10 µm).


36 The holes in the diaphragm are etched by Al etchant for 33 minutes.
Finally, the sacrificial oxide layer is etched through the acoustic holes in PAD etch for 70 minutes to release the diaphragm, The fabrication process is completed by the release of the diaphragm by immersing it in deionized water (DI) and then acetone.

37 Then the whole structure is dried on hot plate at 60 oC for 90 seconds to protect the diaphragm from sticking to the back plate.

38 Fabrication Process

39 Structure under Test

40 Gallium arsenide (GaAs)
Gallium arsenide (GaAs) is a compound of the elements gallium and arsenic. It is a III/V semiconductor, and is used in the manufacture of devices such: infrared light-emitting diodes, laser diodes, solar cells HEMT transistors

41 Gallium arsenide

42 Properties Molecular formula GaAs Molar mass g/mol Appearance Very dark red, vitreous crystals Odor garlic-like when moistened Density g/cm3 Melting point 1238 °C, 1511 K, 2260 °F Solubility in water insoluble Solubility soluble in HCL insoluble in ethanol, methanol, acetone Band gap 1.424 eV (at 300 K) Electron mobility 8500 cm2/(V·s) (at 300 K) Thermal conductivity 0.55 W/(cm·K) (at 300 K) Refractive index (nD) 3.8

43 Preparation Gallium arsenide can be prepared by direct reaction from the elements: Crystal growth using a horizontal zone furnace in the Bridgman-Stockbarger technique, in which gallium and arsenic vapors react and free molecules deposit on a seed crystal at the cooler end of the furnace. Liquid encapsulated Czochralski (LEC) growth is used for producing high purity single crystals that exhibit semi-insulating characteristics.

44 Alternative methods for producing films of GaAs
VPE reaction of gaseous gallium metal and arsenic trichloride: 2 Ga + 2 AsCl3 → 2 GaAs + 3 Cl2 MOCVD reaction of trimethylgallium and arsine: Ga(CH3)3 + AsH3 → GaAs + 3 CH4 Molecular beam epitaxy (MBE) of gallium and arsenic: 4 Ga + As4 → 4 GaAs or 2 Ga + As2 → 2 GaAs

45 Etching Wet etching of GaAs industrially uses an oxidizing agent such as hydrogen peroxide or bromine water.

46 Comparison with silicon GaAs advantages
It has a higher saturated electron velocity and higher electron mobility, allowing gallium arsenide transistors to function at frequencies in excess of 250 GHz. GaAs devices are relatively insensitive to heat owing to their wider bandgap. GaAs devices tend to have less noise than silicon devices, especially at high frequencies. This is a result of higher carrier mobilities and lower resistive device parasitics.

47 it has a direct band gap, which means that it can be used to emit light efficiently. Silicon has an indirect bandgap and so is very poor at emitting light. recent advances may make silicon LEDs and lasers possible. The direct gap of GaAs results in efficient emission of infrared light at eV (~870 nm). GaAs has been used to produce (near-infrared) laser diodes.

48 Silicon advantages silicon is abundant and cheap to process.
existence of a native oxide (silicon dioxide), which is used as an insulator in electronic devices. Silicon dioxide can easily be incorporated onto silicon circuits, and such layers are adherent to the underlying Si. GaAs does not have a native oxide and does not easily support a stable adherent insulating layer.

49 it possesses a much higher hole mobility
it possesses a much higher hole mobility. This high mobility allows the fabrication of higher-speed P-channel field effect transistors, which are required for CMOS logic. Silicon has a nearly perfect lattice, impurity density is very low and allows to build very small structures (currently down to 25 nm). GaAs in contrast has a very high impurity density, which makes it difficult to build ICs with small structures, so the 500 nm process is a common process for GaAs.

50 Aluminium gallium arsenide
The crystal structure of aluminium gallium arsenide

51 Aluminium gallium arsenide
Aluminium gallium arsenide (AlxGa1-xAs) is a semiconductor material with very nearly the same lattice constant as GaAs, but a larger bandgap. The x in the formula above is a number between 0 and 1 - this indicates an arbitrary alloy between GaAs and AlAs. The bandgap varies between 1.42 eV (GaAs) and 2.16 eV (AlAs). For x < 0.4, the bandgap is direct. It can also be used in 1064 nm (Infra-red) laser diodes

52 Red, pure green and blue LEDs of the 5mm diffused type
Light-emitting diode Red, pure green and blue LEDs of the 5mm diffused type Type Passive, optoelectronic Working principle Electroluminescence First production 1968 Electronic symbol Pin configuration anode and cathode

53 Parts of an LED

54 LED A light-emitting diode (LED) is a semiconductor light source.
LEDs are used as indicator lamps in many devices and are increasingly used for other lighting. Early LEDs emitted low-intensity red light, but modern versions are available across the visible, ultraviolet, and infrared wavelengths, with very high brightness.

55 Applications Light-emitting diodes are used in applications as diverse as aviation lighting, automotive lighting, advertising, general lighting, and traffic signals. LEDs have allowed new text, video displays, and sensors to be developed, while their high switching rates are also useful in advanced communications technology. Infrared LEDs are also used in the remote control units of many commercial products including televisions, DVD players and other domestic appliances. LEDs are also used in Seven-segment display.

56 Applications LEDs used on a train Automotive applications
Red and green traffic signals LED illumination LED digital display LED destination signs on buses

57 When a light-emitting diode is forward-biased (switched on), electrons are able to recombine with electron holes within the device, releasing energy in the form of photons. This effect is called electroluminescence and the color of the light (corresponding to the energy of the photon) is determined by the energy gap of the semiconductor. LEDs present many advantages over incandescent light sources including lower energy consumption, longer lifetime, improved physical robustness, smaller size, and faster switching.

58 Technology I-V diagram for a diode The inner workings of an LED
An LED will begin to emit light when the on-voltage is exceeded. Typical on voltages are 2–3 volts.

59 Physics The LED consists of a chip of semiconducting material doped with impurities to create a p-n junction. As in other diodes, current flows easily from the p-side, or anode, to the n-side, or cathode, but not in the reverse direction. Charge-carriers—electrons and holes—flow into the junction from electrodes with different voltages. When an electron meets a hole, it falls into a lower energy level, and releases energy in the form of a photon.

60 The wavelength of the light emitted, and thus its color depends on the band gap energy of the materials forming the p-n junction. In silicon or germanium diodes, the electrons and holes recombine by a non-radiative transition, which produces no optical emission, because these are indirect band gap materials. The materials used for the LED have a direct band gap with energies corresponding to near-infrared, visible, or near-ultraviolet light.

61 LED development began with infrared and red devices made with gallium arsenide.
Advances in materials science have enabled making devices with ever-shorter wavelengths, emitting light in a variety of colors. LEDs are usually built on an n-type substrate, with an electrode attached to the p-type layer deposited on its surface. P-type substrates, while less common, occur as well. Many commercial LEDs, especially GaN/InGaN, also use sapphire substrate.

62 Colors and materials Conventional LEDs are made from a variety of inorganic semiconductor materials. The following table shows the available colors with wavelength range, voltage drop and material.




66 Typical efficacy (lm/W)
As of 2012, the Lumiled catalog gives the following as the best efficacy for each color Color Wavelength range (nm) Typical efficacy (lm/W) Red 620 < λ < 645 72 Red-orange 610 < λ < 620 98 Green 520 < λ < 550 93 Cyan 490 < λ < 520 75 Blue 460 < λ < 490 37

67 In February 2008, a luminous efficacy of 300 lumens of visible light per watt of radiation (not per electrical watt) and warm-light emission was achieved by using nanocrystals.

68 Types LEDs are produced in a variety of shapes and sizes. The color of the plastic lens is often the same as the actual color of light emitted, but not always. For instance, purple plastic is often used for infrared LEDs, and most blue devices have colour less housings. Modern high power LEDs such as those used for lighting and backlighting are generally found in surface-mount technology (SMT) packages, (not shown). The main types of LEDs are miniature, high power devices and custom designs such as alphanumeric or multi-color.

69 Laser diode a packaged laser diode shown with a penny for scale.
the laser diode chip is removed from the above package and placed on the eye of a needle for scale A laser diode with the case cut away The actual laser diode chip is the small black chip at the front; a photodiode at the back is used to control output power

70 laser diode A laser diode is an electrically pumped semiconductor laser in which the active medium is formed by a p-n junction of a semiconductor diode similar to that found in a light-emitting diode. The laser diode is distinct from the optically pumped semiconductor laser, in which, while also semiconductor based, the medium is pumped by a light beam rather than electric current.

71 Theory of operation A laser diode is formed by doping a very thin layer on the surface of a crystal wafer. The crystal is doped to produce an n-type region and a p-type region, one above the other, resulting in a p-n junction, or diode.

72 Laser diodes form a subset of the larger classification of semiconductor p-n junction diodes.
Forward electrical bias across the laser diode causes the two species of charge carrier – holes and electrons – to be "injected" from opposite sides of the p-n junction into the depletion region. Holes are injected from the p-doped, and electrons from the n-doped, semiconductor. (A depletion region, devoid of any charge carriers, forms as a result of the difference in electrical potential between n- and p-type semiconductors wherever they are in physical contact.) Due to the use of charge injection in powering most diode lasers, this class of lasers is sometimes termed "injection lasers," or "injection laser diode" (ILD). As diode lasers are semiconductor devices, they may also be classified as semiconductor lasers. Either designation distinguishes diode lasers from solid-state lasers.

73 When an electron and a hole are present in the same region, they may recombine or "annihilate" with the result being spontaneous emission — i.e., the electron may re-occupy the energy state of the hole, emitting a photon with energy equal to the difference between the electron and hole states involved. (In a conventional semiconductor junction diode, the energy released from the recombination of electrons and holes is carried away as phonons, i.e., lattice vibrations, rather than as photons.) Spontaneous emission gives the laser diode below lasing threshold similar properties to an LED. Spontaneous emission is necessary to initiate laser oscillation, but it is one among several sources of inefficiency once the laser is oscillating.

74 The difference between the photon-emitting semiconductor laser and conventional phonon-emitting (non-light-emitting) semiconductor junction diodes lies in the use of a different type of semiconductor, one whose physical and atomic structure confers the possibility for photon emission. These photon-emitting semiconductors are the so-called "direct bandgap" semiconductors. The properties of silicon and germanium, which are single-element semiconductors, have bandgaps that do not align in the way needed to allow photon emission and are not considered "direct." Other materials, the so-called compound semiconductors, have virtually identical crystalline structures as silicon or germanium but use alternating arrangements of two different atomic species in a checkerboard-like pattern to break the symmetry. The transition between the materials in the alternating pattern creates the critical "direct bandgap" property. Gallium arsenide, indium phosphide, gallium antimonide, and gallium nitride are all examples of compound semiconductor materials that can be used to create junction diodes that emit light.


76 Diagram of a simple laser diode

77 Types The simple laser diode structure, described above, is extremely inefficient. Such devices require so much power that they can only achieve pulsed operation without damage. Although historically important and easy to explain, such devices are not practical.

78 Double heterostructure lasers
In these devices, a layer of low bandgap material is sandwiched between two high bandgap layers. One commonly-used pair of materials is gallium arsenide (GaAs) with aluminium gallium arsenide (AlxGa(1-x)As). Each of the junctions between different bandgap materials is called a heterostructure, hence the name "double heterostructure laser" or DH laser. The kind of laser diode described in the first part of the article may be referred to as a homojunction laser, for contrast with these more popular devices. The advantage of a DH laser is that the region where free electrons and holes exist simultaneously—the active region—is confined to the thin middle layer. This means that many more of the electron-hole pairs can contribute to amplification—not so many are left out in the poorly amplifying periphery. In addition, light is reflected from the heterojunction; hence, the light is confined to the region where the amplification takes place. Diagram of front view of a double heterostructure laser diode

79 Quantum well lasers If the middle layer is made thin enough, it acts as a quantum well. This means that the vertical variation of the electron's wavefunction, and thus a component of its energy, is quantized. The efficiency of a quantum well laser is greater than that of a bulk laser because the density of states function of electrons in the quantum well system has an abrupt edge that concentrates electrons in energy states that contribute to laser action. Lasers containing more than one quantum well layer are known as multiple quantum well lasers. Multiple quantum wells improve the overlap of the gain region with the optical waveguide mode. Further improvements in the laser efficiency have also been demonstrated by reducing the quantum well layer to a quantum wire or to a "sea" of quantum dots. Diagram of front view of a simple quantum well laser diode

80 Separate confinement heterostructure lasers
The problem with the simple quantum well diode described above is that the thin layer is simply too small to effectively confine the light. To compensate, another two layers are added on, outside the first three. These layers have a lower refractive index than the centre layers, and hence confine the light effectively. Such a design is called a separate confinement heterostructure (SCH) laser diode. Almost all commercial laser diodes since the 1990s have been SCH quantum well diodes. Diagram of front view of a separate confinement heterostructure quantum well laser diode

81 Uses Laser diodes can be arrayed to produce very high power outputs, continuous wave or pulsed. Such arrays may be used to efficiently pump solid-state lasers for high average power drilling, burning or for inertial confinement fusion

82 Laser diodes find wide use in telecommunication as easily modulated and easily coupled light sources for fiber optics communication. They are used in various measuring instruments, such as rangefinders. Another common use is in barcode readers. Visible lasers, typically red but later also green, are common as laser pointers. Both low and high-power diodes are used extensively in the printing industry both as light sources for scanning (input) of images and for very high-speed and high-resolution printing plate (output) manufacturing. Infrared and red laser diodes are common in CD players, CD-ROMs and DVD technology. Violet lasers are used in HD DVD and Blu-ray technology. Diode lasers have also found many applications in laser absorption spectrometry (LAS) for high-speed, low-cost assessment or monitoring of the concentration of various species in gas phase.

83 Diode lasers can be used in a great many fields; since light has many different properties (power, wavelength, spectral and beam quality, polarization, etc.) it is useful to classify applications by these basic properties.

84 Common wavelengths 375 nm – excitation of Hoechst stain, Calcium Blue, and other fluorescent dyes in fluorescence microscopy 405 nm – InGaN blue-violet laser, in Blu-ray Disc and HD DVD drives 445 nm – InGaN Deep blue laser multimode diode recently introduced (2010) for use in mercury free high brightness data projectors 473 nm – Sky blue laser pointers, still very expensive, output of DPSS systems 485 nm – excitation of GFP and other fluorescent dyes 510 nm – (to ~525 nm) Green diodes recently (2010) developed by Nichia and OSRAM for laser projectors. 532 nm – DPSS Green laser, frequency doubled from 1064 nm by means of a KTP-crystal, for use in light shows and laser pointers. 635 nm – AlGaInP better red laser pointers, same power subjectively 5 times as bright as 670 nm one 640 nm – High brightness red DPSS laser pointers 657 nm – AlGaInP DVD drives, laser pointers 670 nm – AlGaInP cheap red laser pointers 760 nm – AlGaInP gas sensing: O2 785 nm – GaAlAs Compact Disc drives 808 nm – GaAlAs pumps in DPSS Nd:YAG lasers (e.g. in green laser pointers or as arrays in higher-powered lasers) 848 nm – laser mice 980 nm – InGaAs pump for optical amplifiers, for Yb:YAG DPSS lasers 1064 nm – AlGaAs fiber-optic communication, DPSS laser pump frequency 1310 nm – InGaAsP, InGaAsN fiber-optic communication 1480 nm – InGaAsP pump for optical amplifiers 1512 nm – InGaAsP gas sensing: NH3 1550 nm – InGaAsP, InGaAsNSb fiber-optic communication 1625 nm – InGaAsP fiber-optic communication, service channel 1654 nm – InGaAsP gas sensing: CH4 1877 nm – GaSbAs gas sensing: H2O 2004 nm – GaSbAs gas sensing: CO2 2330 nm – GaSbAs gas sensing: CO 2680 nm – GaSbAs gas sensing: CO2 3030 nm – GaSbAs gas sensing: C2H2 3330 nm – GaSbAs gas sensing: CH4

85 Micromachined optical gas sensor
Aim: environmental monitoring of gases for health, safety and air-pollution control. For Example: Detection of important gases of smokes, such as CO2, CO, NO2, NO, O3, SO2, SO3, CH4 for air pollution control.

86 Techniques for gas detection
1- optical techniques 2- chemical techniques Infrared (IR) technology Long recognized for accurate identification of gases and reliable concentration measurements. Infrared radiation has wavelengths ranging from 0.8 to 100 microns.



89 Infrared (IR) technology
Different gas molecules have characteristic absorption wavelength bands in the infrared spectrum because of rotational and vibrational excitation modes. Fore example: CO2 (4.26 µm) CO (4.65 µm) CH4 (3.3 µm) When exposed to a beam of infrared radiation, most gas molecules will absorb IR-radiation at their vibration/rotation frequencies.

90 Micromacined optical gas sensor


92 Operating principle of the micromachined optical gas sensor
An infrared light beam of an IR-source passes a gas cell. Depending on the gases present, a part of the spectrum is absorbed. An optical tunable filter then analyzes the beam. It passes just one wavelength and blocks other wavelengths, while an IR-detector detects the transmitted intensity. The wavelength of the filter can be tuned to absorption bands of different gasses. Ggggght hjjh, 92

93 What is LASER? Light amplification by stimulated emission of radiation, or laser in short, is a device that creates and amplifies electromagnetic radiation of specific frequency through process of stimulated emission. In laser, all the light rays have the same wavelength and they are coherent; they can travel long distances without diffusing.

94 How an atom gives out light.
To understand how lasers work, we must understand how an atom gives out light. An atom is the smallest particle in the world, and it contains electrons. By introducing extra photon into the atom, the electrons are forced to move into a higher energy level, and now the atom is at an excited state. However, the excited atom is unstable and the electrons always tries to get back to its ground state, therefore releasing the excess energy it originally gained, as a photon of light radiation. This process is called spontaneous emission, as shown below in Figure. 1.


96 Laser The laser contains a chamber in which atoms of a medium are excited, bringing their electrons into higher orbits with higher energy states. When one of these electrons jumps down to a lower energy state, it gives off its extra energy as a photon with a specific frequency. By introducing more photons into the system, the photons will eventually encounter another atom with an excited electron, which will stimulate that electron to jump back to its original state, emitting two or more photons with the same frequency as the first and in phase with it. This effect cascades through the chamber, constantly stimulating other atoms to emit yet more coherent photons, and this process is called stimulated emissions. In other words, the light has been amplified, as shown below in Figure 2


98 Furthermore, mirrors at both ends of the chamber cause the light to bounce back and forth across the medium. One of the mirrors is partially transparent, allowing the laser beam to exit from that end of the chamber. By maintaining a sufficient number of atoms in the medium by external energy source in the higher energy state, the emissions are continuously stimulated, and this process is called population inversion. Ultimately, it creates a stream of coherent photons which is a very concentrated beam of powerful laser light. Lasers have many industrial, military, and scientific uses, including welding, target detection, microscopic photography, fiber optics, surgery, and etc.

99 A closer look at laser diodes
The laser diode converts electrical energy into energy in form of light.

100 Laser diode Laser diode is a laser where the medium is a semiconductor, formed by a p-n junction, as shown in Fig. 3, and powered by electric current. The semiconductor materials that are used to create p-n junction diodes that emit light today are: Gallium arsenide, indium phosphide, gallium antimonide, and gallium nitride. The reason that these are being used is because of the three-five compound properties on chemical periodic table. The materials have to be heavily doped to create P – N regions, which rules out others, leaving groups three-five the ideal options.


102 wavelengths Their wavelengths can be adjusted by changing the ratio of composition. For instance, the wavelength of the laser beam produced by InP substrate can be increased by increasing the Indium content or lowering the Phosphate content percentage. Longer wavelength usually indicates a longer travel distance.

103 Laser diodes Laser diodes (also called .injection lasers.) are in effect a specialised form of LED. Just like a LED, a form of P-N junction diode with a thin depletion layer where electrons and holes collide to create light photons, when the diode is forward biased.

104 The difference is that in this case the. active
The difference is that in this case the .active. part of the depletion layer (i.e., where most of the current flows) is made quite narrow, to concentrate the carriers. The ends of this narrow active region are also highly polished, or coated with multiple very thin reflective layers to act as mirrors, so it forms a resonant optical cavity.


106 The forward current level is also increased, to the point where the current density reaches a critical level where .carrier population inversion. occurs. This means there are more holes than electrons in the conduction band, and more electrons than holes in the valence band . or in other words, a very large excess population of electrons and holes which can potentially combine to release photons. And when this happens, the creation of new photons can be triggered not just by random collisions of electrons and holes, but also by the influence of passing photons.

107 Passing photons are then able to stimulate the production of more photons, without themselves being absorbed. So laser action is able to occur: Light Amplification by Stimulated Emission of Radiation. And the important thing to realize is that the photons that are triggered by other passing photons have the same wavelength, and are also in phase with them. In other words, they end up .in sync. and forming continuous-wave coherent radiation. Because of the resonant cavity, photons are thus able to travel back and forth from one end of the active region to the other, triggering the production of more and more photons in sync with themselves. So quite a lot of coherent light energy is generated.

108 And as the ends of the cavity are not totally reflective (typically about 90-95%), some of this coherent light can leave the laser chip . to form its output beam. Because a laser.s light output is coherent, it is very low in noise and also more suitable for use as a .carrier. for data communications. The bandwidth also tends to be narrower and better defined than LEDs, making them more suitable for optical systems where light beams need to be separated or manipulated on the basis of wavelength.

109 There are two main forms of laser diode:
The very compact size of laser diodes makes them very suitable for use in equipment like CD, DVD and MiniDisc players and recorders. As their light is reasonably well collimated (although not as well as gas lasers) and easily focussed, also used in optical levels, compact handheld laser pointers, barcode scanners etc. There are two main forms of laser diode: the horizontal type, which emits light from the polished ends of the chip, and the vertical or .surface emitting. type. They both operate in the way just described, differing mainly in terms of the way the active light generating region and resonant cavity are formed inside the chip. (See Fig.5) Because laser diodes have to be operated at such a high current density, and have a very low forward resistance when lasing action occurs, they are at risk of destroying themselves due to thermal runaway. Their operating light density can also rise to a level where the end mirrors can begin melting. As a result their electrical operation must be much more carefully controlled than a LED.

110 LEDS & LASER DIODES Light-emitting diodes (LEDs) - In a diode formed from a direct band gap semiconductor, such as gallium arsenide, carriers that cross the junction emit photons when they recombine with the majority carrier on the other side. Depending on the material, wavelengths (or colors) from the infrared to the near ultraviolet may be produced. All LEDs produce incoherent, narrow-spectrum light. LEDs can also be used as low-efficiency photodiodes in signal applications. An LED may be paired with a photodiode or phototransistor in the same package, to form an opto-isolator. Laser diodes - When an LED-like structure is contained in a resonant cavity formed by polishing the parallel end faces, a laser can be formed. Laser diodes are commonly used in optical storage devices and for high speed optical communication.

111 سلول خورشیدی سلول خورشیدی (به انگلیسی: solar cell یا photovoltaic cell یا photoelectric cell)‏ یک قطعه الکترونیکی حالت جامد است که انرژی نور خورشید را مستقیماً توسط اثر فوتوولتاییک به الکتریسیته تبدیل می‌کند.

112 سلول خورشیدی کاربرد سلول خورشیدی ساخته شده از ویفر سیلیکون، کاربرد بسیاری دارند. سلول‌های تکی برای فراهم کردن توان لازم دستگاه‌های کوچک‌تر مانند ماشین حساب الکترونیکی به کار می‌روند. آرایه‌های فوتوولتاییک الکتریسیتهٔ بازیافت‌شدنی‌ای را تولید می‌کنند که عمدتاً در موارد عدم وجود سیستم انتقال و توزیع الکتریکی کاربرد دارد. برای مثال می‌توان به محل‌های دور از دسترس، ماهواره‌های مدارگرد، کاوشگرهای فضایی و ساختمان‌های مخابراتی دور از دسترس اشاره کرد. علاوه بر این استفاده از این نوع انرژی امروزه در محل‌هایی که شبکهٔ توزیع هم موجود است، مرسوم شده‌است.

113 ساختار باتری خورشیدی باتری‌های خورشیدی معمولاً از مواد نیمه‌رسانا، مخصوصاً سیلیسیم، تشکیل شده‌است. هر اتم سیلیسیم با چهار اتم دیگر پیوند تشکیل می‌دهد و بدین صورت، شکل کریستالی آن پدید می‌آید. در باتری‌های خورشیدی به سیلیسیم مقداری جزئی ناخالصی اضافه می‌کنند. اگر اتم ناخالصی ۵ ظرفیتی باشد (اتم سیلیسیم ۴ ظرفیتی است)، آنگاه در ارتباط با چهار اتم سیلیسیم یک لایهٔ آن بدون پیوند باقی می‌ماند (یک تک الکترون). به همین دلیل چون بار نسبی منفی پیدا می‌کند به آن سیلیسیم نوع N) Negative) می‌گویند. درصورتی که اتم ناخالصی دارای ظرفیت ۳ باشد، آنگاه یک حفرهٔ اضافی ایجاد می‌شود. حفره را به گونه‌ای می‌توان گفت که جای خالی الکترون است، با بار مثبت (به اندازهٔ الکترون) و جرمی برابر با جرم الکترون. که این امر هم باعث مثبت شدن نسبی ماده می‌شود و به آن سیلیسیم نوع P) Positive) می‌گویند.

114 عملکرد باتری خورشیدی با اتصال یک نیمه هادی نوع p به یک نیمه هادی نوع n، الکترون‌ها از ناحیه n به ناحیه p و حفره‌ها از ناحیه p به ناحیه n منتقل می‌شوند. با انتقال هر الکترون به ناحیه p، یک یون مثبت در ناحیه n و با انتقال هر حفره به ناحیه n، یک یون منفی در ناحیه p باقی می‌ماند. یون‌های مثبت و منفی میدان الکتریکی داخلی ایجاد می‌کنند که جهت آن از ناحیه n به ناحیه p است. این میدان با انتقال بیشتر باربرها (الکترون‌ها و حفره‌ها)، قوی‌تر و قویتر شده تا جایی که انتقال خالص باربرها به صفر می‌رسد. در این شرایط ترازهای فرمی دو ناحیه با یکدیگر هم سطح شده‌اند و یک میدان الکتریکی داخلی نیز شکل گرفته‌است.

115 اگر در چنین شرایطی، نور خورشید به پیوند بتابد، فوتون‌هایی که انرژی آنها از انرژی شکاف نیمه هادی بیشتر است، زوج الکترون-حفره تولید کرده و زوج‌هایی که در ناحیه تهی یا حوالی آن تولید شده‌اند، شانس زیادی دارند که قبل از بازترکیب، توسط میدان داخلی پیوند از هم جدا شوند.

116 میدان الکتریکی، الکترون‌ها را به ناحیه n و حفره‌ها را به ناحیه p سوق می‌دهد.
این تراکم بار، به شکل ولتاژی در دو سر پیوند قابل اندازه گیری است. اگر دو سر پیوند با یک سیم، به یکدیگر اتصال کوتاه شود، الکترون‌های اضافی ناحیه n، از طریق سیم به ناحیه p رفته و جریان اتصال کوتاهی را شکل می‌دهند. اگر به جای سیم از یک مصرف کننده استفاده شود، عبور جریان از مصرف کننده، به آن انرژی می‌دهد. به این ترتیب انرژی فوتون‌های نور خورشید به انرژی الکتریکی تبدیل می‌شود.

117 هر چه میدان الکتریکی درون پیوند قوی‌تر باشد، ولتاژ مدار باز بزرگتری بدست می‌آید.
برای دست یافتن به یک میدان الکتریکی بزرگ، باید اختلاف ترازهای فرمی دو ماده p و n از یکدیگر زیاد باشد. برای این منظور باید انرژی شکاف نیمه هادی بزرگ انتخاب شود. بنابراین ولتاژ مدار باز یک سلول خورشیدی با انرژی شکاف آن افزایش می‌یابد. اما افزایش انرژی شکاف سبب می‌شود، فوتون‌های کمتری توانایی تولید زوج الکترون-حفره داشته باشند و بنابراین جریان اتصال کوتاه کمتری نیز تولید شود. افزایش انرژی شکاف، روی ولتاژ مدار باز و جریان اتصال کوتاه سلول دو اثر متفاوت دارد.

118 سلول خورشیدی ساخته شده از monocrystalline silicon wafer

119 Fundamentals of photovoltaics
The solar spectrum in outer space resembles the theoretical radiation provided by a black body of 5900 K [20]. As the light passes through the atmosphere, some of the light is absorbed or reflected by gasses such as water vapour and the ozone. The spectrum of the sun’s light that reaches Earth’s upper atmosphere ranges from the ultraviolet to the near infrared radiation, with peak region (48%) from 400 to 700 nm, which is the visible diapason (Figure 1).

120 Figure 1. Terrestrial and space solar spectrum [14].

121 Photovoltaic cells can be defined as p-i-n photodiodes, which are operated under forward
bias. They are designed to capture photons from the solar spectrum by exciting electrons across the bandgap of a semiconductor, which creates electron-hole pairs that are then charge separated, typically by p-n junctions introduced by doping. The space charge at the p-n junction interface drives electrons and holes in opposite directions, creating at the external electrodes a potential difference equal to the bandgap (Figure 2) [20-22]. A semiconductor can only convert photons with the energy of the bandgap with good efficiency. Photons with lower energy are not absorbed and those with higher energy are reduced to gap energy by thermalization of the photogenerated carriers.


123 Behaviour of a solar cell is represented by current versus voltage curves on Figure 3 [23].
The point at which a curve intersects the vertical axis is known as the short circuit condition, and it defines how the cell operates if a wire is connected between its terminals, shorting it out. The current flow here is known as short-circuit current, sc I . For an ideal solar cell at most moderate resistive loss mechanisms, the short-circuit current and the light-generated current are identical. Therefore, the short-circuit current is the largest current which may be drawn from the solar cell [22].

124 The point at which a curve intersects the horizontal axis is known as the open circuit
condition. The open-circuit voltage, oc V , is the maximum voltage available from a solar cell, and this occurs at zero current. The open-circuit voltage corresponds to the amount of forward bias on the solar cell due to the bias of the solar cell junction with the light-generated current. Voc depends on the saturation current of the solar cell and the light-generated current. Open-circuit voltage is then a measure of the amount of recombination in the device [22].

125 For each point on the graph, the voltage and current can be multiplied to calculate power.
Maximum power point is the point on the I-V curve of a solar cell corresponding to the maximum output electrical power,


127 Maximizing total power is the goal of solar cell’s design
Maximizing total power is the goal of solar cell’s design. Multi-junction photovoltaics, as compared to single-junction cells, have reduced currents, because fixed total number of photons is distributed over increasing number of cell layers, so that the amount available for electron promotion in any one layer is decreased. At the same time, the electrons excited are more energetic and have a greater electric potential, so the reduction of currents is compensated for by increase in voltages, and the overall power of the cell is greater. Moreover, multi-junction design is advantageous, because resistive losses, which are proportional to the square of the current, can be significantly reduced [6]. Another defining term in the overall behaviour of a solar cell is the fill factor, FF . This is the ratio that describes how close the I-V curve of a solar cell resembles a perfect rectangle, which represents the ideal solar cell:

128 Quantum efficiency is a term intrinsic to the light absorbing material and not the cell as a
whole; it refers to the percentage of absorbed photons that produce electron-hole pairs. Whereas energy conversion efficiency, is the percentage of incident electromagnetic radiation that is converted to electrical power, when a solar cell is connected to an electrical circuit. This overall efficiency depends on many factors including the temperature, amount of incident radiation and the surface area of the solar cell [21].

129 How Solar Cells Work You've probably seen calculators that have solar cells -- calculators that never need batteries, and in some cases don't even have an off button. As long as you have enough light, they seem to work forever. You may have seen larger solar panels -- on emergency road signs or call boxes, on buoys, even in parking lots to power lights. Although these larger panels aren't as common as solar powered calculators, they're out there, and not that hard to spot if you know where to look. There are solar cell arrays on satellites, where they are used to power the electrical systems. You have probably also been hearing about the "solar revolution" for the last 20 years -- the idea that one day we will all use free electricity from the sun. This is a seductive promise: On a bright, sunny day, the sun shines approximately 1,000 watts of energy per square meter of the planet's surface, and if we could collect all of that energy we could easily power our homes and offices for free. In this article, we will examine solar cells to learn how they convert the sun's energy directly into electricity. In the process, you will learn why we are getting closer to using the sun's energy on a daily basis, and why we still have more research to do before the process becomes cost effective.

130 Solar panels absorb energy to produce hydrogen at SunLine Transit Agency.

131 N-type Plus P-type Silicon
The interesting part starts when you put N-type silicon together with P-type silicon. Remember that every PV cell has at least one electric field. Without an electric field, the cell wouldn't work, and this field forms when the N-type and P-type silicon are in contact. Suddenly, the free electrons in the N side, which have been looking all over for holes to fall into, see all the free holes on the P side, and there's a mad rush to fill them in. Before now, our silicon was all electrically neutral. Our extra electrons were balanced out by the extra protons in the phosphorous. Our missing electrons (holes) were balanced out by the missing protons in the boron. When the holes and electrons mix at the junction between N-type and P-type silicon, however, that neutrality is disrupted. Do all the free electrons fill all the free holes? No. If they did, then the whole arrangement wouldn't be very useful. Right at the junction, however, they do mix and form a barrier, making it harder and harder for electrons on the N side to cross to the P side. Eventually, equilibrium is reached, and we have an electric field separating the two sides.


133 This electric field acts as a diode, allowing (and even pushing) electrons to flow from the P side to the N side, but not the other way around. It's like a hill -- electrons can easily go down the hill (to the N side), but can't climb it (to the P side). So we've got an electric field acting as a diode in which electrons can only move in one direction. Let's see what happens when light hits the cell.

134 When Light Hits the Cell
When light, in the form of photons, hits our solar cell, its energy frees electron-hole pairs. Each photon with enough energy will normally free exactly one electron, and result in a free hole aswell. If this happens close enough to the electric field, or if free electron and free hole happen to wander into its range of influence, the field will send the electron to the N side and the hole to the P side. This causes further disruption of electrical neutrality, and if we provide an external current path, electrons will flow through the path to their original side (the P side) to unite with holes that the electric field sent there, doing work for us along the way. The electron flow provides the current, and the cell's electric field causes a voltage. With both current and voltage, we have power, which is the product of the two.


136 How much sunlight energy does our PV cell absorb
How much sunlight energy does our PV cell absorb? Unfortunately, the most that our simple cell could absorb is around 25 percent, and more likely is 15 percent or less. Why so little?

137 Powering a House

138 Field-effect transistor
The field-effect transistor (FET) is a transistor that uses an electric field to control the shape and hence the conductivity of a channel of one type of charge carrier in a semiconductor material. FETs are unipolar transistors as they involve single-carrier-type operation.

139 Basic information The device consists of an active channel through which charge carriers, electrons or holes, flow from the source to the drain. Source and drain terminal conductors are connected to the semiconductor through ohmic contacts. The conductivity of the channel is a function of potential applied across the gate and source terminals.

140 The FET's three terminals are:[3]
Source (S), through which the carriers enter the channel. Conventionally, current entering the channel at S is designated by IS. Drain (D), through which the carriers leave the channel. Conventionally, current entering the channel at D is designated by ID. Drain to Source voltage is VDS. Gate (G), the terminal that modulates the channel conductivity. By applying voltage to G, one can control ID

141 Cross section of an n-type MOSFET

142 FET operation I–V characteristics and output plot of a JFET n-channel transistor.

143 The FET controls the flow of electrons (or electron holes) from the source to drain by affecting the size and shape of a "conductive channel" created and influenced by voltage (or lack of voltage) applied across the gate and source terminals.

144 In an n-channel depletion-mode device, a negative gate-to-source voltage causes a depletion region to expand in width and encroach on the channel from the sides, narrowing the channel. If the depletion region expands to completely close the channel, the resistance of the channel from source to drain becomes large, and the FET is effectively turned off like a switch. This is pinch-off, and the voltage to achieve it is the pinch-off voltage. Likewise a positive gate-to-source voltage increases the channel size and allows electrons to flow easily.

145 Conversely, in an n-channel enhancement-mode device, a positive gate-to-source voltage is necessary to create a conductive channel, since one does not exist naturally within the transistor. The positive voltage attracts free-floating electrons within the body towards the gate, forming a conductive channel. But first, enough electrons must be attracted near the gate to counter the dopant ions added to the body of the FET; this forms a region free of mobile carriers called a depletion region, and the phenomenon is referred to as the threshold voltage of the FET. Further gate-to-source voltage increase will attract even more electrons towards the gate which are able to create a conductive channel from source to drain; this process is called inversion.

146 In a p-channel depletion-mode device, a positive voltage from gate to body creates a depletion layer by forcing the positively charged holes away from the gate-insulator/semiconductor interface, leaving exposed a carrier-free region of immobile, negatively charged acceptor ions.

147 For either enhancement- or depletion-mode devices, at drain-to-source voltages much less than gate-to-source voltages, changing the gate voltage will alter the channel resistance, and drain current will be proportional to drain voltage (referenced to source voltage). In this mode the FET operates like a variable resistor and the FET is said to be operating in a linear mode or ohmic mode.[4][5]

148 If drain-to-source voltage is increased, this creates a significant asymmetrical change in the shape of the channel due to a gradient of voltage potential from source to drain. The shape of the inversion region becomes "pinched-off" near the drain end of the channel. If drain-to-source voltage is increased further, the pinch-off point of the channel begins to move away from the drain towards the source. The FET is said to be in saturation mode;[6] some authors refer to it as active mode, for a better analogy with bipolar transistor operating regions.[7][8] The saturation mode, or the region between ohmic and saturation, is used when amplification is needed. The in-between region is sometimes considered to be part of the ohmic or linear region, even where drain current is not approximately linear with drain voltage.

149 Even though the conductive channel formed by gate-to-source voltage no longer connects source to drain during saturation mode, carriers are not blocked from flowing. Considering again an n-channel device, a depletion region exists in the p-type body, surrounding the conductive channel and drain and source regions. The electrons which comprise the channel are free to move out of the channel through the depletion region if attracted to the drain by drain-to-source voltage. The depletion region is free of carriers and has a resistance similar to silicon. Any increase of the drain-to-source voltage will increase the distance from drain to the pinch-off point, increasing resistance due to the depletion region proportionally to the applied drain-to-source voltage. This proportional change causes the drain-to-source current to remain relatively fixed independent of changes to the drain-to-source voltage and quite unlike the linear mode operation. Thus in saturation mode, the FET behaves as a constant-current source rather than as a resistor and can be used most effectively as a voltage amplifier. In this case, the gate-to-source voltage determines the level of constant current through the channel.

150 Composition The FET can be constructed from a number of semiconductors, silicon being by far the most common. Most FETs are made with conventional bulk semiconductor processing techniques, using a single crystal semiconductor wafer as the active region, or channel. Among the more unusual body materials are amorphous silicon, polycrystalline silicon or other amorphous semiconductors in thin-film transistors or organic field effect transistors that are based on organic semiconductors; often, OFET gate insulators and electrodes are made of organic materials, as well. Such FETs are manufactured using variety of materials such as silicon carbide(SiC), gallium arsenide(GaAs), gallium nitride(GaN), indium gallium arsenide(InGaAs). In June 2011, IBM announced that it had successfully used graphene-based FETs in an integrated circuit. These transistors are capable of about 2.23 GHz cutoff frequency, much higher than standard silicon FETs.

151 Types of field-effect transistors
The channel of a FET is doped to produce either an N-type semiconductor or a P-type semiconductor. The drain and source may be: doped of opposite type to the channel,in the case of depletion mode FETs, or doped of similar type to the channel as in enhancement mode FETs. Field-effect transistors are also distinguished by the method of insulation between channel and gate.

152 Types of FETs CNTFET (Carbon nanotube field-effect transistor)
The DEPFET is a FET formed in a fully depleted substrate and acts as a sensor, amplifier and memory node at the same time. It can be used as an image (photon) sensor. The DGMOSFET is a MOSFET with dual gates. The DNAFET is a specialized FET that acts as a biosensor. The FREDFET (Fast Reverse or Fast Recovery Epitaxial Diode FET) is a specialized FET designed to provide a very fast recovery (turn-off) of the body diode.

153 The HEMT (high electron mobility transistor), also called a HFET (heterostructure FET), can be made using bandgap engineering in a ternary semiconductor such as AlGaAs. The fully depleted wide-band-gap material forms the isolation between gate and body. The HIGFET (heterostructure insulated gate field effect transisitor), is used mainly in research now. The IGBT (insulated-gate bipolar transistor) is a device for power control. It has a structure akin to a MOSFET coupled with a bipolar-like main conduction channel. These are commonly used for the V drain-to-source voltage range of operation. Power MOSFETs are still the device of choice for drain-to-source voltages of 1 to 200 V.

154 The ISFET (ion-sensitive field-effect transistor) used to measure ion concentrations in a solution;
when the ion concentration (such as H+, see pH electrode) changes, the current through the transistor will change accordingly. The JFET (junction field-effect transistor) uses a reverse biased p-n junction to separate the gate from the body. The MESFET (Metal–Semiconductor Field-Effect Transistor) substitutes the p-n junction of the JFET with a Schottky barrier; used in GaAs and other III-V semiconductor materials. The MODFET (Modulation-Doped Field Effect Transistor) uses a quantum well structure formed by graded doping of the active region.

155 The NOMFET is a Nanoparticle Organic Memory Field-Effect Transistor.
The MOSFET (Metal–Oxide–Semiconductor Field-Effect Transistor) utilizes an insulator (typically SiO2) between the gate and the body. The NOMFET is a Nanoparticle Organic Memory Field-Effect Transistor. The OFET is an Organic Field-Effect Transistor using an organic semiconductor in its channel. The GNRFET is a Field-Effect Transistor that uses a graphene nanoribbon for its channel. The VeSFET (Vertical-Slit Field-Effect Transistor) is a square-shaped junction-less FET with a narrow slit connecting the source and drain at opposite corners. Two gates occupy the other corners, and control the current through the slit. The TFET (Tunnel Field-Effect Transistor) is based on band to band tunneling.

156 Advantages of FET The main advantage of the FET is its:
high input resistance, on the order of 100M ohms or more. Thus, it is a voltage-controlled device, and shows a high degree of isolation between input and output. It is a unipolar device, depending only upon majority current flow. It is less noisy and is thus found in FM tuners and in low-noise amplifiers for VHF and satellite receivers. It is relatively immune to radiation. It exhibits no offset voltage at zero drain current and hence makes an excellent signal chopper. It typically has better thermal stability than a bipolar junction transistor (BJT).[3]

157 Advantages of FET over "BJT"
Very high input impedance ( In the order of a few Mega-Ohm ) FET is unipolar High miniaturisation is possible Very high switching speed FET is less noisy Low temperature sensitivity FET is voltage controlled device The characteristics will not change with time

158 Disadvantages of FET It has a relatively low gain-bandwidth product compared to a BJT. The MOSFET has a drawback of being very susceptible to overload voltages, thus requiring special handling during installation.

159 Uses IGBTs see application in switching internal combustion engine ignition coils, where fast switching and voltage blocking capabilities are important. The most commonly used FET is the MOSFET. The CMOS (complementary metal oxide semiconductor) process technology is the basis for modern digital integrated circuits. This process technology uses an arrangement where the (usually "enhancement-mode") p-channel MOSFET and n-channel MOSFET are connected in series such that when one is on, the other is off. The fragile insulating layer of the MOSFET between the gate and channel makes it vulnerable to electrostatic damage during handling. This is not usually a problem after the device has been installed in a properly designed circuit. In FETs electrons can flow in either direction through the channel when operated in the linear mode, and the naming convention of drain terminal and source terminal is somewhat arbitrary, as the devices are typically (but not always) built symmetrically from source to drain. This makes FETs suitable for switching analog signals between paths (multiplexing). With this concept, one can construct a solid-state mixing board, for example. A common use of the FET is as an amplifier. For example, due to its large input resistance and low output resistance, it is effective as a buffer in common-drain (source follower) configuration.

160 High electron mobility transistor
Cross section of a GaAs/AlGaAs/InGaAs pHEMT

161 Band structure in GaAs/AlGaAs heterojunction based HEMT

162 High electron mobility transistor (HEMT), also known as heterostructure FET (HFET) or modulation-doped FET (MODFET), is a field effect transistor incorporating a junction between two materials with different band gaps (i.e., a heterojunction) as the channel instead of a doped region, as is generally the case for MOSFET. A commonly used material combination is GaAs with AlGaAs, though there is wide variation, dependent on the application of the device. Devices incorporating more indium generally show better high-frequency performance, while in recent years, gallium nitride HEMTs have attracted attention due to their high-power performance. HEMT transistors are able to operate at higher frequencies than ordinary transistors, up to millimeter wave frequencies, and are used in high frequency products such as cell phones, satellite television receivers, and radar equipment.

163 Explanation To allow conduction, semiconductors are doped with impurities which donate mobile electrons (or holes). However, these electrons are slowed down through collisions with the impurities (dopants) used to generate them in the first place. HEMTs avoid this through the use of high mobility electrons generated using the heterojunction of a highly-doped wide-bandgap n-type donor-supply layer (AlGaAs in our example) and a non-doped narrow-bandgap channel layer with no dopant impurities (GaAs in this case).

164 The electrons generated in the thin n-type AlGaAs layer drop completely into the GaAs layer to form a depleted AlGaAs layer, because the heterojunction created by different band-gap materials forms a quantum well (a steep canyon) in the conduction band on the GaAs side where the electrons can move quickly without colliding with any impurities because the GaAs layer is undoped, and from which they cannot escape. The effect of this is to create a very thin layer of highly mobile conducting electrons with very high concentration, giving the channel very low resistivity (or to put it another way, "high electron mobility"). This layer is called a two-dimensional electron gas. As with all the other types of FETs, a voltage applied to the gate alters the conductivity of this layer.

165 Quantum mechanism Since GaAs has higher electron affinity, free electrons in the AlGaAs layer are transferred to the undoped GaAs layer where they form a two dimensional high mobility electron gas within 100 ångström of the interface. The n-type AlGaAs layer of the HEMT is depleted completely through two depletion mechanisms: Trapping of free electrons by surface states causes the surface depletion. Transfer of electrons into the undoped GaAs layer brings about the interface depletion. The Fermi energy level of the gate metal is matched to the pinning point, which is 1.2 eV below the conduction band. With the reduced AlGaAs layer thickness, the electrons supplied by donors in the AlGaAs layer are insufficient to pin the layer. As a result, band bending is moving upward and the two-dimensional electrons gas does not appear. When a positive voltage greater than the threshold voltage is applied to the gate, electrons accumulate at the interface and form a two-dimensional electron gas.

166 Versions of HEMTs pHEMT
Ideally, the two different materials used for a heterojunction would have the same lattice constant (spacing between the atoms). In practice, e.g. AlGaAs on GaAs, the lattice constants are typically slightly different, resulting in crystal defects. As an analogy, imagine pushing together two plastic combs with a slightly different spacing. At regular intervals, you'll see two teeth clump together. In semiconductors, these discontinuities form deep-level traps, and greatly reduce device performance. A HEMT where this rule is violated is called a pHEMT or pseudomorphic HEMT. This is achieved by using an extremely thin layer of one of the materials – so thin that the crystal lattice simply stretches to fit the other material. This technique allows the construction of transistors with larger bandgap differences than otherwise possible, giving them better performance.

167 mHEMT Another way to use materials of different lattice constants is to place a buffer layer between them. This is done in the mHEMT or metamorphic HEMT, an advancement of the pHEMT. The buffer layer is made of AlInAs, with the indium concentration graded so that it can match the lattice constant of both the GaAs substrate and the GaInAs channel. This brings the advantage that practically any Indium concentration in the channel can be realized, so the devices can be optimized for different applications (low indium concentration provides low noise; high indium concentration gives high gain).

168 Induced HEMT In contrast to a modulation-doped HEMT, an induced high electron mobility transistor provides the flexibility to tune different electron densities with a top gate, since the charge carriers are "induced" to the 2DEG plane rather than created by dopants. The absence of a doped layer enhances the electron mobility significantly when compared to their modulation-doped counterparts. This level of cleanliness provides opportunities to perform research into the field of Quantum Billiard for quantum chaos studies, or applications in ultra stable and ultra sensitive electronic devices.

169 Applications Applications are similar to those of MESFETs – microwave and millimeter wave communications, imaging, radar, and radio astronomy – any application where high gain and low noise at high frequencies are required. HEMTs have shown current gain to frequencies greater than 600 GHz and power gain to frequencies greater than 1 THz. (Heterojunction bipolar transistors were demonstrated at current gain frequencies over 600 GHz in April 2005.) Numerous companies worldwide develop and manufacture HEMT-based devices. These can be discrete transistors but are more usually in the form of a 'monolithic microwave integrated circuit' (MMIC). HEMTs are found in many types of equipment ranging from cellphones and DBS receivers to electronic warfare systems such as radar and for radio astronomy.

170 ترانزیستور اثر میدان ترانزیستور اثر میدان، دسته‌ای از ترانزیستورها هستند که مبنای کار کنترل جریان در آن‌ها توسط یک میدان الکتریکی صورت می‌گیرد. با توجه به اینکه در این ترانزیستورها تنها یک نوع حامل بار (الکترون آزاد یا حفره) در ایجاد جریان الکتریکی دخالت دارند، می‌توان آن‌ها را جزو ترانزیستورهای تک‌قطبی محسوب کرد که در مقابل ترانزیستورهای دوقطبی (که حامل‌های اکثریت و اقلیت همزمان در آن‌ها نقش دارند) قرار می‌گیرند.[۱] ترانزیستورهای اثر میدان دارای سه پایهٔ سورس، درین و گیت هستند. این دسته از ترانزیستورها خود به دو گروه ماسفت و جی‌فت تقسیم می‌شوند. در این نوع ترانزیستورها، برخلاف ترانزیستورهای دو قطبی پیوندی که کنترل جریان امیتر و کلکتور با جریان ورودی به بیس صورت می‌گیرد، کنترل جریان سورس و درین با اعمال ولتاژ به گیت صورت می‌گیرد.

171 کاربردها کاربرد اصلی این ترانزیستورهای در مدارهای مجمتع به‌ویژه تراشه‌های دیجیتال است. در بیشتر این تراشه‌ها هزاران ماسفت استفاده شده است که نه تنها به عنوان عنصر فعال بلکه به عنوان مقاوت و خازن نیز به کار می‌روند. هرچند مدارهای ساخته‌شده با ماسفت نسبت به مدارهای ساخته‌شده با بی‌جی‌تی‌ها پیچیده‌تر هستند و سرعت کمتری دارند، اما هزینهٔ کمتری نیز دارند و فضای کمتری اشغال می‌کنند و بنابراین در فناوری‌های یکپارچه‌سازی کلان‌مقیاس (VLSI) کاربرد گسترده‌ای دارند و در فرآیندهای کنترل صنعتی، ابزارهای الکترونیک خودکار، الکترونیک نوری، مدارهای سوئیچینگ تلفنی و… نقش مهمی ایفا می‌کنند. جی‌فت‌ها همچنین به علت مقاومت ورودی زیاد و اغتشاش کم در الکترونیک خطی (غیر دیجیتال) اهمیت ویژه‌ای دارند. از ترانزیستورهای اثر میدان در مدارهای کاربردی‌ای نظیر تقویت‌کننده‌ها، کلیدها، منابع جریان، بار فعال و… استفاده می‌شود.[۲] فت‌ها در ساخت فرستنده باند اف‌ام رادیو نیز کاربرد فراوانی دارند.

172 پایانه‌ها یکی از پایه‌های (پایانه‌های) ترانزیستور FET است. فت دارای سه پایه با نام‌های درِین D - سورس S و گیت G است که پایه گیت، جریان عبوری از درین به سورس را کنترل می‌نماید. فت‌ها دارای دو نوع N کانال و P کانال هستند. در فت نوع N کانال زمانی که گیت نسبت به سورس مثبت باشد جریان از درین به سورس عبور می‌کند. FETها معمولاً بسیار حساس بوده و حتی با الکتریسیته ساکن بدن نیز تحریک می‌گردند. به همین دلیل نسبت به نویز بسیار حساس هستند.

173 اهمیت و ویژگی‌های ترانزیستورهای اثر میدان
در مقایسه با ترانزیستورهای پیوندی، ترانزیستورهای اثر میدان را می‌توان یک وسیلهٔ حساس به ولتاژ دانست که امپدانس ورودی آن بسیاربسیار زیاد (در حدود ۱۰۱۴ اهم) و امپدانس خروجی‌اش نیز به‌نسبت زیاد است. ماسفت‌ها در مقایسه با بی‌جی‌تی‌ها بسیار کوچکند به طوری که تنها ۱۰ تا ۲۰ درصد از فضای اشغال‌شده در توسط آن‌ها را اشغال می‌کنند و بنابراین در یک تراشه تعدادی زیادی ماسفت را می‌توان جای داد و به همین خاطر است که ماسفت‌ها کاربرد گسترده‌ای در یکپارچه‌سازی بزرگ‌مقیاس دارند. تفاوت دیگر ماسفت‌ها با ترانزیستورهای پیوندی در این است که ماسفت‌ها در بازه‌ای از گسترهٔ عملکردشان مانند یک مقاومت کنترل‌شده با ولتاژ عمل می‌کنند و نسبت به آی‌سی‌های مقاومتی مشابه جای کمتری در تراشه‌ها می‌گیرند. ویژگی سوم ماسفت‌ها مقاومت بسیار زیاد ورودی‌شان است؛ این ویژگی بدین معنا خواهد بود که ثابت زمانی مدار ورودی به قدری بالا است که اجازه می‌دهد بار الکتریکی ذخیره‌شده در خازن کوچک ورودی برای مدت طولانی باقی بماند و بتوان از آن به عنوان یک ابزار ذخیره‌سازی در مدارهای دیجیتال استفاده کرد. ویژگی چهارم ماسفت‌ها قابلیتشان در مصارف توان‌بالا است که می‌توانند جریان‌ها زیاد را در چند نانوثانیه کلیدزنی کنند که این مدت نسبت به آنچه در بی‌جی‌تی‌ها امکان‌پذیر است بسیار سریع‌تر است. ویژگی چهارم امکان استفاده از ماسفت‌ها را در مدارهای فرکانس‌بالایِ توان‌بالا ممکن می‌سازد.[۳] مزیت دیگر ماسفت‌ها در مدارهای دیجیتال این است که توان مصرفی آن‌ها در حالت خاموش تقریباً صفر است.[

174 نماد مداری یک جی‌فت با کانال پی
جریان الکتریکی از سورس به درین در یک جی‌فت با کانال پی هنگام اعمال ولتاژ به گیت محدود می‌شود. نماد مداری یک جی‌فت با کانال ان نماد مداری یک جی‌فت با کانال پی نمودار I-V خصوصیات و طرح‌های خروجی از JFETنوع n.

175 ترانزیستور پیوندی اثر میدان یا جِی‌فِت (به انگلیسی: junction gate field-effect transistor یا JUGFET یا JFET)‏ به گونه‌ای از ترانزیستورهای اثر میدان گفته می‌شود که از یک کانال عبور و یک گیت تشکیل شده‌اند؛ دو پایهٔ درین و سورس با اتصال اهمی به دو طرف کانال متصل هستند و پایهٔ گیت اتصال یکسوساز دارد.[۱] بسته به اینکه کانال از جنس نیمه‌هادی نوع ان باشد یا پی، به جی‌فت حاصل ان‌جی‌فت یا پی‌جی‌فت گفته می‌شود. در جی‌فت جریان الکتریکی عبوری از کانالِ بین سرهای سورس و درین، با اعمال ولتاژ به سر گیت کنترل می‌شود.[۲]

176 در جی‌فت‌ها جریان بین سرهای سورس و درین با کنترل مقاومت ناحیهٔ کانال مهار می‌شود. برای تغییر مقاومت ناحیهٔ کانال، عرض ناحیهٔ تهی (بدون حامل الکتریکی) را با اعمال بایاس معکوس بین پیوند گیت و کانال تغییر می‌دهند.[۳] جی‌فت ساده‌ترین نوع ترانزیستور اثر میدان است. می‌توان آن را به عنوان یک سوئیچ الکترونیکی تحت کنترل و یا به عنوان یک مقاومت کنترل شده به وسیله ولتاژ استفاده کرد. بار الکتریکی از طریق یک کانال نیمه هادی بین ترمینال های"source" و "drain"جریان می‌یابد. با اعمال کردن یک ولتاژ بایاس معکوس به ترمینال "gate"، این کانال "pinched"(فشرده) می‌شود. به طوری که جریان الکتریکی کاهش یافته یا به طور کامل قطع می‌شود.

177 ساختار هنگامی که یک ولتاژ به گیت اعمال شده است، جریان الکتریکی از سورس به درین در یک JFET کانال p محدود می‌شود. JFETیک کانال طولانی از مواد نیمه هادی است، که طوری الاییده می‌شود که حاملین اکثریت شامل بار مثبت یا حفره ها (در نوع p)باشد و یا حاملین اکثریت، بار منفی یا الکترون (نوع n) باشد.. تماس اهمی در پایان هرسرمنبع (Source) و تخلیه (Drain)را تشکیل می‌دهند. یک پیوندpn در یک یا در دو طرف ویا در اطراف کانال تشکیل می‌شود، به گونه‌ای که ناخالصی آن مخالف ناخالصی کانال باشد.

178 عملکرد عملکرد JFET شبیه شلنگ باغ است. جریان آبی که از شلنگ عبور می‌کند را می‌توان با فشردن آن کنترل کرد تا سطح مقطع آنرا کاهش داد. جریان بار الکتریکی در یک JFET با محدود کردن کانال جریان انتقالی کنترل می‌شود. جریان همچنین به میدان الکتریکی بین سورس و درین بستگی دارد (شبیه به تفاوت در فشار بر هر انتهای شلنگ). ساخت کانال هدایت با استفاده از اثر میدان انجام می‌شود: برای بایاس معکوس کردن گیت-سورس پیوند pnیک ولتاژ بین گیت و سورس اعمال می‌شود. بدین وسیله ناحیه تهی این اتصال را وسیعتر می‌کند و به کانال هدایت تجاوز می‌کند وسطح مقطع را محدود می‌کند. ناحیه تهی بدلیل خالی بودن از حامل‌های متحرک است که به این نام خوانده می‌شود. و بنابراین برای کاربردهای عملی از لحاظ الکتریکی نارسانا می‌باشد. زمانی که ناحیه تهی، عرض کانال رسانش را گسترش می‌دهد و"pitch off" بدست می‌آید و رسانش بین درین و سورس متوقف می‌شود. "pinch off" در یک ولتاژ بایاس معکوس خاص اتصال گیت و سورس اتفاق می‌افتد. و ولتاژ Vp حتی در ابزارهای مشابه به مقدار قابل ملاحظه‌ای تغییر می‌کند،. به عنوان مثال VGS(off) برای قطعه Temic J201 از -۰٫۸ تا-۴ ولت متغییر است.(مقادیر معمول آن بین -۰٫۳ تا -۱۰ ولت می‌باشد.(برای خاموش کردن دستگاه کانال N-نیاز به یک منبع ولتاژ منفی بین گیت و سورس (VGS) در مقابل، برای خاموش کردن دستگاه کانال p-نیاز به VGS مثبت است. در عملیات معمول، میدان الکتریکی توسعه یافته توسط گیت، جریان سورس درین را تا حدودی متوقف می‌کند. برخی از دستگاه‌های JFET، نسبت به سورس و درین متقارن هستند.

179 مقایسه با ترانزیستورهای دیگر
در دمای اتاق جریان گیت JFET (نشت معکوس از پیوند گیت به کانال) مشابه جریان در MOSFET (که بین گیت و کانال عایق اکسید وجود دارد)، اما بسیار کمتر از جریان بیس در یک ترانزیستور اتصال دوقطبی است. رسانایی متقابل JFET بیشتر از MOSFET است. بنابراین در بیشتر نویزهای پایین، مقاومت درونی ظاهری بالا وتقویت کننده‌های عملیاتی استفاده می‌شود. تاریخچه JFET JFET توسط Julius Lilienfed درسال ۱۹۲۵ پیش بینی شد و در اواسط سال ۱۹۳۰ تئوری عملکرد آن به اندازه کافی به خوبی شناخته شد تا بتواند حق امتیاز آن را تصدیق کند. با این حال، برای چندین سال ایجاد کریستال‌های آلائیده شده با حرارت کافی برای نشان دادن این اثر امکان پذیر نبود. در سال ۱۹۴۷، پژوهشگرانJohn Bardeen، Walter Houser Brattainو William Shockley زمانی که ترانزیستور نقطه تماسی را کشف کرده بودند، برای ساخت اولین JFET تلاش می‌کردند. اولین نسل JFETهای عملی برخلاف انتظار آنها سال‌ها بعد از ترانزیستورهای پیوندی ساخته شد.JFET تا حدودی می‌تواند به عنوان ترکیبی از یک MOSFET و BJT رفتار کند. اگرچه ترانزیستور دوقطبی با گیت عایق شده شباهت بیشتری به بیشتر ویژگی‌های هیبرید دارد.

180 مدل ریاضی جریان در N-JFET به دلیل یک ولتاژ کوچک VDS شده توسط فرمول زیر بدست می‌آید: که جریان اشباع درین-سورس =IDSS ضخامت کانال = 2a عرض w = طول L= شارژ الکترونی Q = 1.6*10^-9= الکترون تحرک μn = نئودیمیم = N نوع غلظت دوپینگ در ناحیه اشباع: در منطقه خطی: یا جایی که Vp ولتاژ مصرف شده است، حداقل ولتاژ از گیت –سورس برای به طور کامل خاموش کردن، عبور می‌کند. هنگامی که در مقایسه با - کوچک است، دستگاه مانند یک مقاومت کنترل شده باولتاژ عمل می‌کند.

181 ماسفت شامل پایه های گیت(G)،درین(D)،سورس(S) و بدنه(B)

182 ماسفت یا ترانزیستور اثرمیدانی نیمه‌رسانا-اکسید-فلز (به انگلیسی: metal–oxide–semiconductor field-effect transistor ٫ MOSFET )‏ معروف‌ترین ترانزیستور اثرمیدان در مدارهای آنالوگ و دیجیتال است.این گونه از ترانزیستور اثر میدان نخستین بار در سال ۱۹۲۵ میلادی معرفی شد. در آن هنگام، ساخت و به کارگیری این ترانزیستورها، به سبب نبود علم و ابزار و امکان، با دشواری همراه بود و از همین روی، برای پنج دهه فراموش شدند و از میدانِ پیش‌رفت‌های الکترونیک بر کنار ماندند. در آغازِ دههٔ ۱۹۷۰م، بارِ دیگر نگاه‌ها به MOSFETها افتاد و برای ساختنِ مدارهای مجتمع به کار گرفته شدند. در ترانزیستور اثرِ میدان ( FET ) چنان که از نام اش پیدا است، پایهٔ کنترلی، جریانی مصرف نمی‌کند و تنها با اعمال ولتاژ و ایجاد میدان درون نیمه رسانا، جریان عبوری از FET کنترل می‌شود. از همین روی ورودی این مدار هیچ اثر بارگذاری بر روی طبقات تقویت قبلی نمی‌گذارد و امپدانس بسیار بالایی دارد.عمده تفاوت ماسفت با ترانزیستور JFET در این است که گیت ترانزیستورهای ماسفت توسط لایه‌ای از اکسید سیلیسیم (SiO2) از کانال مجزاشده است. به این دلیل به ماسفتها فِت با گیت مجزا (به انگلیسی: IGFET ، Insulated Gate FET)‏ نیز گفته میشود.[۱] مدارهای مجتمع بر پایهٔ فناوری ترانزیستورهای اثرِ میدانِ MOS، را می‌توان بسیار ریزتر و ساده‌تر از مدارهای مجتمع بر پایهٔ ترانزیستورهای دوقطبی ساخت؛ بی آن که (حتی در مدارها و تابع‌های پیچیده و مقیاس‌های بزرگ ) نیازی به مقاومت، دیود، یا دیگر قطعه‌های الکترونیکی داشته باشند.[۲] همین ویژگی، تولیدِ انبوهِ آن‌ها را آسان می‌کند، چندان که هم اکنون بیش‌تر از ۸۵ درصدِ مدارهای مجتمع، بر پایهٔ فناوریِ MOS طراحی و ساخته می‌شوند. ترانزیستورهای MOS، بسته به کانالی که در آن‌ها شکل می‌گیرد، NMOS یا PMOS نامیده می‌شوند. در آغازِ کار، PMOS ترانزیستورِ پرکاربردتر در فناوری MOS بود. اما از آن جا که ساختنِ NMOS آسان‌تر است و مساحتِ کم‌تری هم می‌گیرد، از PMOS پیشی گرفت. بر خلافِ ترانزیستورهای دوقطبی، در ترانزیستورهای MOSFET، جریان، نتیجهٔ شارشِ تنها یک حامل ( الکترون یا حفره) در میانِ پیوندها است و از این رو، این ترانزیستورها را تک‌قطبی هم می‌نامند.

183 ساختار و کارکرد ماسفت افزایشی
فت دارای سه پایه با نام‌های درین D، سورس S و گیت G است که پایه گیت، جریان عبوری از درین به سورس را کنترل می‌کند. فت‌ها دارای دو نوع N کانال و P کانال هستند. در فت نوع N کانال زمانی که گیت نسبت به سورس مثبت باشد جریان از درین به سورس عبور می‌کند. FET‌ها معمولاً بسیار حساس بوده و حتی با الکتریسیته ساکن بدن نیز تحریک می‌گردند. به همین دلیل نسبت به نویز بسیار حساس هستند. نوع دیگر ترانزیستورهای اثر میدانی MOSFET‌ها هستند (ترانزیستور اثرمیدانی نیمه‌رسانای اکسید فلز) یکی از اساسی‌ترین مزیت‌های ماسفت‌ها نویز کمتر آن‌ها در مدار است. فت‌ها در ساخت فرستنده باند اف ام رادیو نیز کاربرد فراوانی دارند. برای تست کردن فت کانال N با مالتی متر، نخست پایه گیت را پیدا می‌کنیم. یعنی پایه‌ای که نسبت به دو پایه دیگر در یک جهت مقداری رسانایی دارد و در جهت دیگر مقاومت آن بی نهایت است. معمولاً مقاومت بین پایه درین و گیت از مقاومت پایه درین و سورس بیشتر است که از این طریق می‌توان پایهٔ درین را از سورس تشخیص داد.

184 ماسفت کاهشی ساختار این گونهٔ ترانزیستورِ MOS، همانند ساختار ترانزیستورهای افزایشی است، تنها با این تفاوت که هنگامِ ساخت آن، کانال را، به وسیلهٔ یک نوار از جنس سیلیسیم، میانِ سورس و درین تعبیه می‌کنند. از این رو، اگر اختلاف پتانسیل میان آن دو اعمال شود، جریانی از سورس به درین خواهیم داشت؛ هرچند که ولتاژ اعمال شده به گیت صفر باشد.

185 مدارهای CMOS MOSFET در مکمل های دیجیتال، نیمه هادی اکسید فلزی (CMOS) منطق مورد استفاده قرار می گیرد،که به عنوان بلوک های ساختمانی با استفاده از p-و ماسفت کانال N-است. بیش از حد یک نگرانی عمده ای را در مدارهای مجتمع است، از ترانزیستورهای بیشتری را در تراشه های کوچکتر بسته بندی شده است. CMOS منطق مصرف برق را کاهش می دهد، زیرا هیچ جریان (ایده آل)، و در نتیجه هیچ قدرت مصرف می شود، به جز زمانی که ورودی به گیت های منطقی در حال تغییر است. CMOS انجام این کاهش در حال حاضر با تکمیل هر nMOSFET با یک pMOSFET و اتصال هر دو گیتس و هر دو درن با هم است. یک ولتاژ بالا بر روی دروازه باعث خواهد شد nMOSFET برای انجام و pMOSFET به انجام و ولتاژ پایین بر ​​روی دروازه باعث معکوس. در طول زمان سوئیچینگ ولتاژ از یک کشور به کشور دیگر می رود، هر دو ماسفت به طور خلاصه انجام خواهد شد. این ترتیب تا حد زیادی مصرف برق و تولید گرما را کاهش می دهد. برنامه های کاربردی CMOS دیجیتال و آنالوگ در زیر توضیح داده شده است.

186 دیجیتال رشد فن آوری های دیجیتال مانند ریزپردازنده فراهم کرده است انگیزه برای پیشبرد تکنولوژی MOSFET سریع تر از هر نوع دیگری از سیلیکون مبتنی بر ترانزیستور است. یک مزیت بزرگ از ماسفت برای تعویض دیجیتال این است که لایه ی اکسید بین دروازه و کانال مانع جریان از جریان از طریق دروازه DC، بیشتر به کاهش مصرف برق و دادن امپدانس ورودی بسیار بزرگ است. اکسید عایق بین دروازه و کانال به طور موثر جدا در یک منطق مرحله از مراحل اولیه و بعد از آن، که اجازه می دهد تا یک خروجی MOSFET تنها به درایو تعداد قابل توجهی از ورودی MOSFET MOSFET. دو قطبی ترانزیستور مبتنی بر منطق (مانند TTL) چنین ظرفیت بالا fanout ندارد. این انزوا نیز باعث می شود آن را آسان تر برای طراحان به برخی از اثرات حد بارگذاری بین مراحل منطق به طور مستقل را نادیده گرفت. این حد با فرکانس عامل تعریف می شود: به عنوان افزایش فرکانس، امپدانس ورودی ماسفت ها کاهش می یابد.

187 آنالوگ مزایای MOSFET در مدارهای دیجیتال را به برتری در تمام مدارهای آنالوگ تبدیل شدنی نیست. دو نوع مدار بر اساس ویژگی های مختلف ترانزیستور رفتار می کنند است. مدارات دیجیتال سوئیچ، صرف بیشتر وقت خود را خارج از منطقه تعویض، در حالی که مدارهای آنالوگ بر رفتار MOSFET دقیقا در منطقه سوئیچینگ از عملیات برگزار می شود بستگی دارد. اتصال ترانزیستور دو قطبی (BJT) به طور سنتی ترانزیستور طراح آنالوگ از انتخاب، به دلیل آن transconductance بالا و پایین امپدانس خروجی (تخلیه ولتاژ استقلال) در منطقه تعویض. با این وجود، ماسفت ها به طور گسترده ای در بسیاری از انواع مدارات آنالوگ به دلیل مزایای خاصی استفاده می شود. عملکرد و ویژگیهای بسیاری از مدارهای آنالوگ را می توان با تغییر اندازه (طول و عرض) از ماسفت استفاده می شود طراحی شده است. در مقایسه، در اکثر ترانزیستورهای دوقطبی اندازه دستگاه قابل توجهی بر عملکرد. ویژگی های ایده آل ماسفت در مورد gate فعلی (صفر) و تخلیه منبع ولتاژ افست (صفر) نیز آنها را عناصر نزدیک به ایده آل سوئیچ، و همچنین ایجاد تغییر خازن مدارهای آنالوگ عملی. در ناحیه خطی خود، ماسفت ها را می توان به عنوان مقاومتهای دقیق، که می تواند مقاومت در برابر کنترل بسیار بیشتر از BJT ها استفاده می شود. در مدارهای قدرت بالا، ماسفت گاهی اوقات استفاده از فراری حرارتی رنج می برند نه به عنوان BJT ها داشته باشد. همچنین آنها را می توان به خازن و مدارهای چرخنده است که اجازه می دهد عملیات، آمپر ساخته شده از آنها به عنوان سلف ظاهر می شود شکل گرفته است، در نتیجه اجازه می دهد تمام آنالوگ دستگاه های طبیعی، به جز برای دیودها (که می تواند به صورت کوچکتر از یک MOSFET به هر حال)، به به طور کامل از ماسفت ساخته شده است. این اجازه می دهد تا برای کامل مدارات آنالوگ را بر روی یک تراشه سیلیکونی در یک فضای بسیار کوچکتر ساخته شده است.

188 بعضی از IC ها ترکیب مدارات MOSFET در یک مخلوط سیگنال مدار مجتمع آنالوگ و دیجیتال، ساخت فضای مورد نیاز هیئت مدیره و حتی کوچکتر است. این یک نیاز برای منزوی ساختن مدارهای آنالوگ مدارهای دیجیتال در سطح تراشه، منجر به استفاده از حلقه انزوا و سیلیکون بر روی عایق (SOI). مزیت اصلی BJT ها در مقابل ماسفت در فرایند طراحی آنالوگ، توانایی BJT ها که مسئولیت رسیدگی به یک جریان بزرگتر در یک فضای کوچکتر است. فرآیندهای ساخت وجود دارد که ترکیب BJT و MOSFET را به یک دستگاه واحد. دستگاه ها ترانزیستور مختلط به نام بی FET ها بیین (دوقطبی-FET ها بیین) اگر تنها یکی از آنها حاوی BJT-FET و میکنند Bicmos (دوقطبی-CMOS) اگر آنها حاوی مکمل BJT-FET ها بیین. دستگاه های چنین مزایای استفاده از هر دو گیتس عایق و بالاتر چگالی جریان است

189 پیشرفت ماسفتها در طول دهه های گذشته، MOSFET به طور مستمر در اندازه کوچک و طول کانال MOSFET معمولی یک بار میکرومتر، اما مدارهای مجتمع مدرن ترکیب MOSFET های با طول کانال از دهها نانومتر است. کار در تئوری پوسته پوسته شدن رابرت Dennard محوری در شناخت است که این کاهش مداوم ممکن بود بود. اینتل شروع به تولید یک فرایند شامل 32 نانومتر اندازه از ویژگی های (با کانال که حتی کوتاه تر) در اواخر سال 2009 است.صنعت نیمه هادی را حفظ می کند "نقشه راه"، که مجموعه سرعت برای توسعه MOSFET. از لحاظ تاریخی، مشکلات با کاهش اندازه MOSFET شده اند با دستگاه نیمه هادی فرآیند ساخت، نیاز به استفاده از ولتاژهای بسیار پایین، و با فقیرتر عملکرد الکتریکی طراحی مدار ایجاب می و نوآوری (MOSFET های کوچک نمایشگاه جریان نشتی بالاتر و پایین خروجی مقاومت، در زیر مورد بحث).

Download ppt "Cross-sectional view of single wafer capacitive microphone"

Similar presentations

Ads by Google