Indium gallium arsenide

From Wikipedia, the free encyclopedia

https://en.wikipedia.org/wiki/Indium_gallium_arsenide

 

Indium gallium arsenide (InGaAs) (alternatively gallium indium arsenide, GaInAs) is a ternary alloy (chemical compound) of indium arsenide (InAs) and gallium arsenide (GaAs). Indium and gallium are (group III) elements of the periodic table while arsenic is a (group V) element. Alloys made of these chemical groups are referred to as "III-V" compounds. InGaAs has properties intermediate between those of GaAs and InAs. InGaAs is a room-temperature semiconductor with applications in electronics and photonics.

The principal importance of GaInAs is its application as a high-speed, high sensitivity photodetector of choice for optical fiber telecommunications.

Nomenclature

Indium gallium arsenide (InGaAs) and gallium-indium arsenide (GaInAs) are used interchangeably. According to IUPAC standards the preferred nomenclature for the alloy is Ga_xIn_1-xAs where the group-III elements appear in order of increasing atomic number, as in the related alloy system Al_xGa_1-xAs. By far, the most important alloy composition from technological and commercial standpoints is Ga_0.47In_0.53As, which can be deposited in single crystal form on indium phosphide (InP). 

Materials Synthesis

GaInAs is not a naturally-occurring material. Single crystal material is required for electronic and photonic device applications. Pearsall and co-workers were the first to describe single-crystal epitaxial growth of In_0.53Ga_0.47As on (111)-oriented  and on (100)-oriented  InP substrates. Single crystal material in thin-film form can be grown by epitaxy from the liquid-phase (LPE), vapour-phase (VPE), by molecular beam epitaxy (MBE), and by metalorganic chemical vapour deposition (MO-CVD). Today, most commercial devices are produced by MO-CVD or by MBE. 

The optical and mechanical properties of InGaAs can be varied by changing the ratio of InAs and GaAs, In
_1-xGa
_xAs. Most InGaAs devices are grown on indium phosphide (InP) substrates. In order to match the lattice constant of InP and avoid mechanical strain, In
_0.53Ga
_0.47As is used. This composition has an optical absorption edge at 0.75 eV, corresponding to a cut-off wavelength of λ=1.68 μm at 295 K.

By increasing the mole fraction of InAs further compared to GaAs, it is possible to extend the cut-off wavelength up to about λ=2.6 µm. In that case special measures have to be taken to avoid mechanical strain from differences in lattice constants.

GaAs is lattice-mismatched to germanium (Ge) by 0.08%. With the addition of 1.5% InAs to the alloy, In_0.015Ga_0.985As becomes latticed-matched to the Ge substrate, reducing stress in subsequent deposition of GaAs. 

Electronic and optical properties

Fig.1 Energy gap versus gallium composition for GaInAs

 

InGaAs has a lattice parameter that increases linearly with the concentration of InAs in the alloy. The liquid-solid phase diagram shows that during solidification from a solution containing GaAs and InAs, GaAs is taken up at a much higher rate than InAs, depleting the solution of GaAs. During growth from solution, the composition of first material to solidify is rich in GaAs while the last material to solidify is richer in InAs. This feature has been exploited to produce ingots of InGaAs with graded composition along the length of the ingot. However, the strain introduced by the changing lattice constant causes the ingot to be polycrystalline and limits the characterization to a few parameters, such as bandgap and lattice constant with uncertainty due to the continuous compositional grading in these samples.

Fig.2 Lattice parameter of GaInAs vs GaAs alloy content

 

Fig.3 Photoluminescence of n-type and p-type GaInAs

Properties of single crystal GaInAs

Single crystal GaInAs

Single crystal epitaxial films of GaInAs can be deposited on a single crystal substrate of III-V semiconductor having a lattice parameter close to that of the specific gallium indium arsenide alloy to be synthesized. Three substrates can be used: GaAs, InAs and InP. A good match between the lattice constants of the film and substrate is required to maintain single crystal properties and this limitation permits small variations in composition on the order of a few per cent. Therefore, the properties of epitaxial films of GaInAs alloys grown on GaAs are very similar to GaAs and those grown on InAs are very similar to InAs, because lattice mismatch strain does not generally permit significant deviation of the composition from the pure binary substrate. 

Ga
_0.47In
_0.53As is the alloy whose lattice parameter matches that of InP at 295 K. GaInAs lattice-matched to InP is a semiconductor with properties quite different from GaAs, InAs or InP. It has an energy band gap of 0.75 eV, an electron effective mass of 0.041 and an electron mobility close to 10,000 cm²·V⁻¹·s⁻¹ at room temperature, all of which are more favorable for many electronic and photonic device applications when compared to GaAs, InP or even Si. Measurements of the band gap and electron mobility of single-crystal GaInAs were first published by Takeda and co-workers.

Property	Value at 295 K
Lattice Parameter	5.869 Å
Band Gap	0.75 eV
Electron effective mass	0.041
Light-hole effective mass	0.051
Electron mobility	10,000 cm2·V−1·s−1
Hole mobility	250 cm2·V−1·s−1

FCC lattice parameter

Like most materials, the lattice parameter of GaInAs is a function of temperature. The measured coefficient of thermal expansion  is 5.66×10⁻⁶ K⁻¹. This is significantly larger than the coefficient for InP which is 4.56×10⁻⁶ K⁻¹. A film that is exactly lattice-matched to InP at room temperature is typically grown at 650 °C with a lattice mismatch of +6.5×10⁻⁴. Such a film has a mole fraction of GaAs = 0.47. To obtain lattice matching at the growth temperature, it is necessary to increase the GaAs mole fraction to 0.48. 

Bandgap energy

The bandgap energy of GaInAs can be determined from the peak in the photoluminescence spectrum, provided that the total impurity and defect concentration is less than 5×10¹⁶ cm⁻³. The bandgap energy depends on temperature and increases as the temperature decreases, as can be seen in Fig. 3 for both n-type and p-type samples. The bandgap energy at room temperature is 0.75 eV and lies between that of Ge and Si. By coincidence the bandgap of GaInAs is perfectly placed for photodetector and laser applications for the long-wavelength transmission window, (the C-band and L-band) for fiber-optic communications.

Effective mass

The electron effective mass of GaInAs m^*/m° = 0.041 is the smallest for any semiconductor material with an energy bandgap greater than 0.5 eV. The effective mass is determined from the curvature of the energy-momentum relationship: stronger curvature translates into lower effective mass and a larger radius of delocalization. In practical terms, a low effective mass leads directly to high carrier mobility, favoring higher speed of transport and current carrying capacity. A lower carrier effective mass also favors increased tunneling current, a direct result of delocalization. 

The valence band has two types of charge carriers: light holes: m^*/m° = 0.051 and heavy holes: m^*/m° = 0.2. The electrical and optical properties of the valence band are dominated by the heavy holes, because the density of these states is much greater than that for light holes. This is also reflected in the mobility of holes at 295 K, which is a factor of 40 lower than that for electrons. 

Fig.4 Electron and hole mobilities of GaInAs vs impurity concentration at 295 K.

Mobility of electrons and holes

Electron mobility and hole mobility are key parameters for design and performance of electronic devices. Takeda and co-workers were the first to measure electron mobility in epitaxial films of InGaAs on InP substrates. Measured carrier mobilities for electrons and holes are shown in Figure 4.

The mobility of carriers in Ga
_0.47In
_0.53As is unusual in two regards:

• The very high value of electron mobility
• The unusually large ratio of electron to hole mobility.

The room temperature electron mobility for reasonably pure samples of Ga
_0.47In
_0.53As approaches 10×10³ cm²·V⁻¹·s⁻¹, which is the largest of any technologically important semiconductor, although significantly less than that for graphene. 

The mobility is proportional to the carrier conductivity. As mobility increases, so does the current-carrying capacity of transistors. A higher mobility shortens the response time of photodetectors. A larger mobility reduces series resistance, and this improves device efficiency and reduces noise and power consumption. 

The minority carrier diffusion constant is directly proportional to carrier mobility. The room temperature diffusion constant for electrons at 250 cm²·s⁻¹ is significantly larger than that of Si, GaAs, Ge or InP, and determines the ultra-fast response of Ga
_0.47In
_0.53As photodetectors.

The ratio of electron to hole mobility is the largest of currently-used semiconductors. 

Applications

Fig.5 upper: Ge photodiode lower: GaInAs photodiode in the wavelength range 1 µm to 2 µm.

 

Photodetectors

The principal application of GaInAs is as an infrared detector. The spectral response of a GaInAs photodiode is shown in Figure 5. GaInAs photodiodes are the preferred choice in the wavelength range of 1.1 µm < λ < 1.7 µm. For example, compared to photodiodes made from Ge, GaInAs photodiodes have faster time response, higher quantum efficiency and lower dark current for the same sensor area. GaInAs photodiodes were invented in 1977 by Pearsall.

Avalanche photodiodes offer the advantage of additional gain at the expense of response time. These devices are especially useful for detection of single photons in applications such as quantum key distribution where response time is not critical. Avalanche photodetectors require a special structure to reduce reverse leakage current due to tunnelling. The first practical avalanche photodiodes were designed and demonstrated in 1979.

In 1980, Pearsall developed a photodiode design that exploits the uniquely short diffusion time of high mobility of electrons in GaInAs, leading to an ultrafast response time. This structure was further developed and subsequently named the UTC, or uni-travelling carrier photodiode. In 1989, Wey and co-workers designed and demonstrated a p-i-n GaInAs/InP photodiodes with a response time shorter than 5 picoseconds for a detector surface measuring 5 µm x 5 µm.

Other important innovations include the integrated photodiode – FET receiver and the engineering of GaInAs focal-plane arrays.

Lasers

Semiconductor lasers are an important application for GaInAs, following photodetectors. GaInAs can be used as a laser medium. Devices have been constructed that operate at wavelengths of 905 nm, 980 nm, 1060 nm, and 1300 nm. InGaAs quantum dots on GaAs have also been studied as lasers. GaInAs/InAlAs quantum-well lasers can be tuned to operate at the λ = 1500 nm low-loss, low-dispersion window for optical fiber telecommunications  In 1994, GaInAs/AlInAs quantum wells were used by Jérôme Faist and co-workers who invented and demonstrated a new kind of semiconductor laser based on photon emission by an electron making an optical transition between subbands in the quantum well. They showed that the photon emission regions can be cascaded in series, creating the quantum cascade laser (QCL). The energy of photon emission is a fraction of the bandgap energy. For example, GaInAs/AlInAs QCL operates at room temperature in the wavelength range 3 µm < λ < 8 µm. The wavelength can be changed by modifying the width of the GaInAs quantum well. These lasers are widely used for chemical sensing and pollution control. 

Photovoltaics and transistors

GaInAs is used in triple-junction photovoltaics and also for thermophotovoltaicpower generation.

In
_0.015Ga
_0.985As can be used as an intermediate band-gap junction in multi-junction photovoltaic cells with a perfect lattice match to Ge. The perfect lattice match to Ge reduces defect density, improving cell efficiency.

HEMT devices using InGaAs channels are one of the fastest types of transistor.

 

In 2012 MIT researchers announced the smallest transistor ever built from a material other than silicon. The Metal oxide semiconductor field-effect transistor (MOSFET) is 22 nanometers long. This is a promising accomplishment, but more work is needed to show that the reduced size results in improved electronic performance relative to that of silicon or GaAs-based transistors.

In 2014, Researchers at Penn State University developed a novel device prototype designed to test nanowires made of compound semiconductors such as InGaAs. The goal of this device was to see if a compound material would retain its superior mobility at nanoscale dimensions in a FinFET device configuration. The results of this test sparked more research, by the same research team, into transistors made of InGaAs which showed that in terms of on current at lower supply voltage, InGaAs performed very well compared to existing silicon devices.

In Feb 2015 Intel indicated it may use InGaAs for its 7 nanometer CMOS process in 2017.

Safety and toxicity

The synthesis of GaInAs, like that of GaAs, most often involves the use of arsine (AsH
₃), an extremely toxic gas. Synthesis of InP likewise most often involves phosphine (PH
₃). Inhalation of these gases neutralizes oxygen absorption by the bloodstream and can be fatal within a few minutes if toxic dose levels are exceeded. Safe handling involves using a sensitive toxic gas detection system and self-contained breathing apparatus.

Once GaInAs is deposited as a thin film on a substrate, it is basically inert and is resistant to abrasion, sublimation or dissolution by common solvents such as water, alcohols or acetones. In device form the volume of the GaInAs is usually less than 1000 μm³, and can be neglected compared to the volume of the supporting substrate, InP or GaAs.

The National Institutes of Health studied these materials and found:

• No evidence of carcinogenic activity of gallium arsenide in male F344/N rats exposed to 0.01, 0.1, or 1.0 mg/m³
• Carcinogenic activity in female F344/N rats
• No evidence of carcinogenic activity in male or female B6C3F1 mice exposed to 0.1, 0.5, or 1.0 mg/m³.

The World Health Organization's International Agency for Research on Cancer's review of the NIH toxicology study concluded:

• There is inadequate evidence in humans for the carcinogenicity of gallium arsenide.
• There is limited evidence in experimental animals for the carcinogenicity of gallium arsenide.
• The gallium moiety may be responsible for lung cancers observed in female rats

REACH (Registration, Evaluation, Authorisation and Restriction of Chemicals) is a European initiative to classify and regulate materials that are used, or produced (even as waste) in manufacturing. REACH considers three toxic classes: carcinogenic, reproductive, and mutagenic capacities.

The REACH classification procedure consists of two basic phases. In phase one the hazards intrinsic to the material are determined, without any consideration of how the material might be used or encountered in the work place or by a consumer. In phase two the risk of harmful exposure is considered along with procedures that can mitigate exposure. Both GaAs and InP are in phase 1 evaluation. The principal exposure risk occurs during substrate preparation where grinding and polishing generate micron-size particles of GaAs and InP. Similar concerns apply to wafer dicing to make individual devices. This particle dust can be absorbed by breathing or ingestion. The increased ratio of surface area to volume for such particles increases their chemical reactivity.

Toxicology studies are based on rat and mice experiments. No comparable studies test the effects of ingesting GaAs or InP dust in a liquid slurry.

The REACH procedure, acting under the precautionary principle, interprets "inadequate evidence for carcenogenicity" as "possible carcinogen". As a result, the European Chemicals Agency classified InP in 2010 as a carcinogen and reproductive toxin:

• Classification & labelling in accordance with Directive 67/548/EEC
• Classification: Carc. Cat. 2; R45
• Repr. Cat. 3; R62

and ECHA classified GaAs in 2010 as a carcinogen and reproductive toxin:

• Classification & labelling in accordance with Directive 67/548/EEC:
• Classification3: Carc. Cat. 1; R45
• Repro. Cat. 2; R60

Fiber-optic cable

From Wikipedia, the free encyclopedia

https://en.wikipedia.org/wiki/Fiber-optic_cable

 

A TOSLINK optical fiber cable with a clear jacket. These cables are used mainly for digital audio connections between devices.

A fiber-optic cable, also known as an optical-fiber cable, is an assembly similar to an electrical cable, but containing one or more optical fibers that are used to carry light. The optical fiber elements are typically individually coated with plastic layers and contained in a protective tube suitable for the environment where the cable will be deployed. Different types of cable are used for different applications, for example, long distance telecommunication, or providing a high-speed data connection between different parts of a building.

Design

A multi-fiber cable

 

Optical fiber consists of a core and a cladding layer, selected for total internal reflection due to the difference in the refractive index between the two. In practical fibers, the cladding is usually coated with a layer of acrylate polymer or polyimide. This coating protects the fiber from damage but does not contribute to its optical waveguide properties. Individual coated fibers (or fibers formed into ribbons or bundles) then have a tough resin buffer layer or core tube(s) extruded around them to form the cable core. Several layers of protective sheathing, depending on the application, are added to form the cable. Rigid fiber assemblies sometimes put light-absorbing ("dark") glass between the fibers, to prevent light that leaks out of one fiber from entering another. This reduces cross-talk between the fibers, or reduces flare in fiber bundle imaging applications.

Left: LC/PC connectors
Right: SC/PC connectors
All four connectors have white caps covering the ferrules.

 

For indoor applications, the jacketed fiber is generally enclosed, with a bundle of flexible fibrous polymer strength members like aramid (e.g. Twaron or Kevlar), in a lightweight plastic cover to form a simple cable. Each end of the cable may be terminated with a specialized optical fiber connector to allow it to be easily connected and disconnected from transmitting and receiving equipment.

Fiber-optic cable in a Telstra pit

 

Investigating a fault in a fiber cable junction box. The individual fiber cable strands within the junction box are visible.

 

An optical fiber breakout cable

 

For use in more strenuous environments, a much more robust cable construction is required. In loose-tube construction the fiber is laid helically into semi-rigid tubes, allowing the cable to stretch without stretching the fiber itself. This protects the fiber from tension during laying and due to temperature changes. Loose-tube fiber may be "dry block" or gel-filled. Dry block offers less protection to the fibers than gel-filled, but costs considerably less. Instead of a loose tube, the fiber may be embedded in a heavy polymer jacket, commonly called "tight buffer" construction. Tight buffer cables are offered for a variety of applications, but the two most common are "Breakout" and "Distribution". Breakout cables normally contain a ripcord, two non-conductive dielectric strengthening members (normally a glass rod epoxy), an aramid yarn, and 3 mm buffer tubing with an additional layer of Kevlar surrounding each fiber. The ripcord is a parallel cord of strong yarn that is situated under the jacket(s) of the cable for jacket removal.^[3] Distribution cables have an overall Kevlar wrapping, a ripcord, and a 900 micrometer buffer coating surrounding each fiber. These fiber units are commonly bundled with additional steel strength members, again with a helical twist to allow for stretching.

A critical concern in outdoor cabling is to protect the fiber from contamination by water. This is accomplished by use of solid barriers such as copper tubes, and water-repellent jelly or water-absorbing powder surrounding the fiber.

Finally, the cable may be armored to protect it from environmental hazards, such as construction work or gnawing animals. Undersea cables are more heavily armored in their near-shore portions to protect them from boat anchors, fishing gear, and even sharks, which may be attracted to the electrical power that is carried to power amplifiers or repeaters in the cable.

Modern cables come in a wide variety of sheathings and armor, designed for applications such as direct burial in trenches, dual use as power lines, installation in conduit, lashing to aerial telephone poles, submarine installation, and insertion in paved streets. 

Capacity and market

In September 2012, NTT Japan demonstrated a single fiber cable that was able to transfer 1 petabit per second (10¹⁵bits/s) over a distance of 50 kilometers.

Modern fiber cables can contain up to a thousand fibers in a single cable, with potential bandwidth in the terabytes per second. In some cases, only a small fraction of the fibers in a cable may be actually "lit". Companies can lease or sell the unused fiber to other providers who are looking for service in or through an area. Companies may "overbuild" their networks for the specific purpose of having a large network of dark fiber for sale, reducing the overall need for trenching and municipal permitting. They may also deliberately under-invest to prevent their rivals from profiting from their investment. 

The highest strand-count singlemode fiber cable commonly manufactured is the 864-count, consisting of 36 ribbons each containing 24 strands of fiber.

Reliability and quality

Optical fibers are very strong, but the strength is drastically reduced by unavoidable microscopic surface flaws inherent in the manufacturing process. The initial fiber strength, as well as its change with time, must be considered relative to the stress imposed on the fiber during handling, cabling, and installation for a given set of environmental conditions. There are three basic scenarios that can lead to strength degradation and failure by inducing flaw growth: dynamic fatigue, static fatigues, and zero-stress aging.

Telcordia GR-20, Generic Requirements for Optical Fiber and Optical Fiber Cable, contains reliability and quality criteria to protect optical fiber in all operating conditions. The criteria concentrate on conditions in an outside plant (OSP) environment. For the indoor plant, similar criteria are in Telcordia GR-409, Generic Requirements for Indoor Fiber Optic Cable.

Cable types

• OFC: Optical fiber, conductive
• OFN: Optical fiber, nonconductive
• OFCG: Optical fiber, conductive, general use
• OFNG: Optical fiber, nonconductive, general use
• OFCP: Optical fiber, conductive, plenum
• OFNP: Optical fiber, nonconductive, plenum
• OFCR: Optical fiber, conductive, riser
• OFNR: Optical fiber, nonconductive, riser
• OPGW: Optical fiber composite overhead ground wire
• ADSS: All-Dielectric Self-Supporting
• OSP: Fiber optic cable, outside plant
• MDU: Fiber optics cable, multiple dwelling unit

Jacket material

The jacket material is application-specific. The material determines the mechanical robustness, chemical and UV radiation resistance, and so on. Some common jacket materials are LSZH, polyvinyl chloride, polyethylene, polyurethane, polybutylene terephthalate, and polyamide.

Fiber material

There are two main types of material used for optical fibers: glass and plastic. They offer widely different characteristics and find uses in very different applications. Generally, plastic fiber is used for very short-range and consumer applications, whereas glass fiber is used for short/medium-range (multi-mode) and long-range (single-mode) telecommunications.

Color coding

Patch cords

The buffer or jacket on patchcords is often color-coded to indicate the type of fiber used. The strain relief "boot" that protects the fiber from bending at a connector is color-coded to indicate the type of connection. Connectors with a plastic shell (such as SC connectors) typically use a color-coded shell. Standard color codings for jackets (or buffers) and boots (or connector shells) are shown below:

Cord jacket (or buffer) color

Color		Meaning
	Orange	multi-mode optical fiber
	Aqua	OM3 or OM4 10 G laser-optimized 50/125 µm multi-mode optical fiber
	Erika violet	OM4 multi-mode optical fiber (some vendors)
	Lime green	OM5 10 G wideband 50/125 µm multi-mode optical fiber
	Grey	outdated color code for multi-mode optical fiber
	Yellow	single-mode optical fiber
	Blue	Sometimes used to designate polarization-maintaining optical fiber

Remark: It is also possible that a small part of a connector is additionally color-coded, e.g. the lever of an E-2000 connector or a frame of an adapter. This additional colour coding indicates the correct port for a patchcord, if many patchcords are installed at one point.

Multi-fiber cables

Individual fibers in a multi-fiber cable are often distinguished from one another by color-coded jackets or buffers on each fiber. The identification scheme used by Corning Cable Systems is based on EIA/TIA-598, "Optical Fiber Cable Color Coding." EIA/TIA-598 defines identification schemes for fibers, buffered fibers, fiber units, and groups of fiber units within outside plant and premises optical fiber cables. This standard allows for fiber units to be identified by means of a printed legend. This method can be used for identification of fiber ribbons and fiber subunits. The legend will contain a corresponding printed numerical position number or color for use in identification.

Propagation speed and delay

Optical cables transfer data at the speed of light in glass. This is the speed of light in vacuum divided by the refractive index of the glass used, typically around 180,000 to 200,000 km/s, resulting in 5.0 to 5.5 microseconds of latency per km. Thus the round-trip delay time for 1000 km is around 11 milliseconds.

Losses

Signal loss in optic fiber is measured in decibels (dB). A loss of 3 dB across a link means the light at the far end is only half the intensity of the light that was sent into the fiber. A 6 dB loss means only one quarter of the light made it through the fiber. Once too much light has been lost, the signal is too weak to recover and the link becomes unreliable and eventually ceases to function entirely. The exact point at which this happens depends on the transmitter power and the sensitivity of the receiver. 

Typical modern multimode graded-index fibers have 3 dB per kilometre of attenuation (signal loss) at a wavelength of 850 nm, and 1 dB/km at 1300 nm. Singlemode loses 0.35 dB/km at 1310 nm and 0.25 dB/km at 1550 nm. Very high quality singlemode fiber intended for long distance applications is specified at a loss of 0.19 dB/km at 1550 nm. Plastic optical fiber (POF) loses much more: 1 dB/m at 650 nm. POF is large core (about 1 mm) fiber suitable only for short, low speed networks such as TOSLINK optical audio or for use within cars.

Each connection between cables adds about 0.6 dB of average loss, and each joint (splice) adds about 0.1 dB.

Invisible infrared light (750 nm and larger) is used in commercial glass fiber communications because it has lower attenuation in such materials than visible light. However, the glass fibers will transmit visible light somewhat, which is convenient for simple testing of the fibers without requiring expensive equipment. Splices can be inspected visually, and adjusted for minimal light leakage at the joint, which maximizes light transmission between the ends of the fibers being joined.

The charts Understanding wavelengths in fiber optics and Optical power loss (attenuation) in fiber illustrate the relationship of visible light to the infrared frequencies used, and show the absorption water bands between 850, 1300 and 1550 nm. 

Safety

The infrared light used in telecommunications cannot be seen, so there is a potential laser safety hazard to technicians. The eye's natural defense against sudden exposure to bright light is the blink reflex, which is not triggered by infrared sources. In some cases the power levels are high enough to damage eyes, particularly when lenses or microscopes are used to inspect fibers that are emitting invisible infrared light. Inspection microscopes with optical safety filters are available to guard against this. More recently indirect viewing aids are used, which can comprise a camera mounted within a handheld device, which has an opening for the connectorized fiber and a USB output for connection to a display device such as a laptop. This makes the activity of looking for damage or dirt on the connector face much safer.

Small glass fragments can also be a problem if they get under someone's skin, so care is needed to ensure that fragments produced when cleaving fiber are properly collected and disposed of appropriately. 

Hybrid cables

There are hybrid optical and electrical cables that are used in wireless outdoor Fiber To The Antenna (FTTA) applications. In these cables, the optical fibers carry information, and the electrical conductors are used to transmit power. These cables can be placed in several environments to serve antennas mounted on poles, towers, and other structures.

According to Telcordia GR-3173, Generic Requirements for Hybrid Optical and Electrical Cables for Use in Wireless Outdoor Fiber To The Antenna (FTTA) Applications, these hybrid cables have optical fibers, twisted pair/quad elements, coaxial cables or current-carrying electrical conductors under a common outer jacket. The power conductors used in these hybrid cables are for directly powering an antenna or for powering tower-mounted electronics exclusively serving an antenna. They have a nominal voltage normally less than 60 VDC or 108/120 VAC. Other voltages may be present depending on the application and the relevant National Electrical Code (NEC).

These types of hybrid cables may also be useful in other environments such as Distributed Antenna System (DAS) plants where they will serve antennas in indoor, outdoor, and roof-top locations. Considerations such as fire resistance, Nationally Recognized Testing Laboratory (NRTL) Listings, placement in vertical shafts, and other performance-related issues need to be fully addressed for these environments.

Since the voltage levels and power levels used within these hybrid cables vary, electrical safety codes consider the hybrid cable to be a power cable, which needs to comply with rules on clearance, separation, etc. 

Innerducts

Innerducts are installed in existing underground conduit systems to provide clean, continuous, low-friction paths for placing optical cables that have relatively low pulling tension limits. They provide a means for subdividing conventional conduit that was originally designed for single, large-diameter metallic conductor cables into multiple channels for smaller optical cables. 

Types

Innerducts are typically small-diameter, semi-flexible subducts. According to Telcordia GR-356, there are three basic types of innerduct: smoothwall, corrugated, and ribbed. These various designs are based on the profile of the inside and outside diameters of the innerduct. The need for a specific characteristic or combination of characteristics, such as pulling strength, flexibility, or the lowest coefficient of friction, dictates the type of innerduct required.

Beyond the basic profiles or contours (smoothwall, corrugated, or ribbed), innerduct is also available in an increasing variety of multiduct designs. Multiduct may be either a composite unit consisting of up to four or six individual innerducts that are held together by some mechanical means, or a single extruded product having multiple channels through which to pull several cables. In either case, the multiduct is coilable, and can be pulled into existing conduit in a manner similar to that of conventional innerduct. 

Placement

Innerducts are primarily installed in underground conduit systems that provide connecting paths between manhole locations. In addition to placement in conduit, innerduct can be directly buried, or aerially installed by lashing the innerduct to a steel suspension strand.

As stated in GR-356, cable is typically placed into innerduct in one of three ways. It may be

1. Pre-installed by the innerduct manufacturer during the extrusion process,
2. Pulled into the innerduct using a mechanically assisted pull line, or
3. Blown into the innerduct using a high air volume cable blowing apparatus.

PIN diode

From Wikipedia, the free encyclopedia

https://en.wikipedia.org/wiki/PIN_diode

 

Layers of a PIN diode

 

A PIN diode is a diode with a wide, undoped intrinsic semiconductor region between a p-type semiconductor and an n-type semiconductor region. The p-type and n-type regions are typically heavily doped because they are used for ohmic contacts.

The wide intrinsic region is in contrast to an ordinary p–n diode. The wide intrinsic region makes the PIN diode an inferior rectifier (one typical function of a diode), but it makes it suitable for attenuators, fast switches, photodetectors, and high voltage power electronics applications. 

Operation

A PIN diode operates under what is known as high-level injection. In other words, the intrinsic "i" region is flooded with charge carriers from the "p" and "n" regions. Its function can be likened to filling up a water bucket with a hole on the side. Once the water reaches the hole's level it will begin to pour out. Similarly, the diode will conduct current once the flooded electrons and holes reach an equilibrium point, where the number of electrons is equal to the number of holes in the intrinsic region. When the diode is forward biased, the injected carrier concentration is typically several orders of magnitude higher than the intrinsic carrier concentration. Due to this high level injection, which in turn is due to the depletion process, the electric field extends deeply (almost the entire length) into the region. This electric field helps in speeding up of the transport of charge carriers from the P to the N region, which results in faster operation of the diode, making it a suitable device for high frequency operations. 

Characteristics

The PIN diode obeys the standard diode equation for low frequency signals. At higher frequencies, the diode looks like an almost perfect (very linear, even for large signals) resistor. The P-I-N diode has a relatively large stored charge adrift in a thick intrinsic region. At a low enough frequency, the stored charge can be fully swept and the diode turns off. At higher frequencies, there is not enough time to sweep the charge from the drift region, so the diode never turns off. The time required to sweep the stored charge from a diode junction is its reverse recovery time, and it is relatively long in a PIN diode. For a given semiconductor material, on state impedance, and minimum usable RF frequency, the reverse recovery time is fixed. This property can be exploited; one variety of P-I-N diode, the step recovery diode, exploits the abrupt impedance change at the end of the reverse recovery to create a narrow impulse waveform useful for frequency multiplication with high multiples.

The high-frequency resistance is inversely proportional to the DC bias current through the diode. A PIN diode, suitably biased, therefore acts as a variable resistor. This high-frequency resistance may vary over a wide range (from 0.1 Ω to 10 kΩ in some cases; the useful range is smaller, though). 

The wide intrinsic region also means the diode will have a low capacitance when reverse-biased.

In a PIN diode the depletion region exists almost completely within the intrinsic region. This depletion region is much larger than in a PN diode and almost constant-size, independent of the reverse bias applied to the diode. This increases the volume where electron-hole pairs can be generated by an incident photon. Some photodetector devices, such as PIN photodiodes and phototransistors (in which the base-collector junction is a PIN diode), use a PIN junction in their construction.

The diode design has some design trade-offs. Increasing the area of the intrinsic region increases its stored charge reducing its RF on-state resistance while also increasing reverse bias capacitance and increasing the drive current required to remove the charge during a fixed switching time, with no effect on the minimum time required to sweep the charge from the I region. Increasing the thickness of the intrinsic region increases the total stored charge, decreases the minimum RF frequency, decreases the reverse bias capacitance, but doesn't decrease the forward bias RF resistance and increases the minimum time required to sweep the drift charge and transition from low to high RF resistance. Diodes are sold commercially in a variety of geometries for specific RF bands and uses. 

Applications

PIN diodes are useful as RF switches, attenuators, photodetectors, and phase shifters.

RF and microwave switches

A PIN Diode RF Microwave Switch.

 

Under zero- or reverse-bias (the "off" state), a PIN diode has a low capacitance. The low capacitance will not pass much of an RF signal. Under a forward bias of 1 mA (the "on" state), a typical PIN diode will have an RF resistance of about 1 ohm, making it a good RF conductor. Consequently, the PIN diode makes a good RF switch.

Although RF relays can be used as switches, they switch relatively slowly (on the order of 10s of milliseconds). A PIN diode switch can switch much more quickly (e.g., 1 microsecond), although at lower RF frequencies it isn't reasonable to expect switching times in the same order of magnitude as the RF period.

For example, the capacitance of an "off"-state discrete PIN diode might be 1 pF. At 320 MHz, the capacitive reactance of 1 pF is 497 ohms:

${\displaystyle Z_{diode}={\frac {1}{2\pi fC}}={\frac {1}{2\cdot \pi \cdot 320\times 10^{6}\cdot 1\times 10^{-12}}}={497}\ \Omega }$

As a series element in a 50 ohm system, the off-state attenuation in dB is:

${\displaystyle A=20\log _{10}\left({\frac {Z_{load}}{Z_{source}+Z_{diode}+Z_{load}}}\right)=20\log _{10}\left({\frac {50}{50+497+50}}\right)={21.5}\ dB}$

This attenuation may not be adequate. In applications where higher isolation is needed, both shunt and series elements may be used, with the shunt diodes biased in complementary fashion to the series elements. Adding shunt elements effectively reduces the source and load impedances, reducing the impedance ratio and increasing the off-state attenuation. However, in addition to the added complexity, the on-state attenuation is increased due to the series resistance of the on-state blocking element and the capacitance of the off-state shunt elements.

PIN diode switches are used not only for signal selection, but also component selection. For example, some low phase noise oscillators use them to range-switch inductors.

RF and microwave variable attenuators

An RF Microwave PIN diode Attenuator.

 

By changing the bias current through a PIN diode, it is possible to quickly change the RF resistance.

At high frequencies, the PIN diode appears as a resistor whose resistance is an inverse function of its forward current. Consequently, PIN diode can be used in some variable attenuator designs as amplitude modulators or output leveling circuits.

PIN diodes might be used, for example, as the bridge and shunt resistors in a bridged-T attenuator. Another common approach is to use PIN diodes as terminations connected to the 0 degree and -90 degree ports of a quadrature hybrid. The signal to be attenuated is applied to the input port, and the attenuated result is taken from the isolation port. The advantages of this approach over the bridged-T and pi approaches are (1) complementary PIN diode bias drives are not needed—the same bias is applied to both diodes—and (2) the loss in the attenuator equals the return loss of the terminations, which can be varied over a very wide range. 

Limiters

PIN diodes are sometimes designed for use as input protection devices for high frequency test probes and other circuits. If the input signal is small, the PIN diode has negligible impact, presenting only a small parasitic capacitance. Unlike a rectifier diode it does not present a nonlinear resistance at RF frequencies which would give rise to harmonics and intermodulation products. If the signal is large, then when the PIN diode starts to rectify the signal, the forward current charges the drift region and the device RF impedance is a resistance inversely proportional to the signal amplitude. That signal amplitude varying resistance can be used to terminate some predetermined portion the signal in a resistive network dissipating the energy or to create an impedance mismatch that reflects the incident signal back toward the source. The latter may be combined with an isolator, a device containing a circulator which uses a permanent magnetic field to break reciprocity and a resistive load to separate and terminate the backward traveling wave. It should be emphasized that when used as a shunt limiter the PIN diode is a low impedance over the entire RF cycle, unlike paired rectifier diodes that would swing from a high resistance to a low resistance during each RF cycle clamping the waveform and not reflecting it as completely. The ionization recovery time of gas molecules that permits the creation of the higher power spark gap input protection device ultimately relies on similar physics in a gas. 

Photodetector and photovoltaic cell

The PIN photodiode was invented by Jun-ichi Nishizawa and his colleagues in 1950.

PIN photodiodes are used in fibre optic network cards and switches. As a photodetector, the PIN diode is reverse-biased. Under reverse bias, the diode ordinarily does not conduct (save a small dark current or I_s leakage). When a photon of sufficient energy enters the depletion region of the diode, it creates an electron-hole pair. The reverse bias field sweeps the carriers out of the region, creating current. Some detectors can use avalanche multiplication.

The same mechanism applies to the PIN structure, or p-i-n junction, of a solar cell. In this case, the advantage of using a PIN structure over conventional semiconductor p–n junction is better long-wavelength response of the former. In case of long wavelength irradiation, photons penetrate deep into the cell. But only those electron-hole pairs generated in and near the depletion region contribute to current generation. The depletion region of a PIN structure extends across the intrinsic region, deep into the device. This wider depletion width enables electron-hole pair generation deep within the device, which increases the quantum efficiency of the cell.

Commercially available PIN photodiodes have quantum efficiencies above 80-90% in the telecom wavelength range (~1500 nm), and are typically made of germanium or InGaAs. They feature fast response times (higher than their p-n counterparts), running into several tens of gigahertz, making them ideal for high speed optical telecommunication applications. Similarly, silicon p-i-n photodiodes have even higher quantum efficiencies, but can only detect wavelengths below the bandgap of silicon, i.e. ~1100 nm. 

Typically, amorphous silicon thin-film cells use PIN structures. On the other hand, CdTe cells use NIP structure, a variation of the PIN structure. In a NIP structure, an intrinsic CdTe layer is sandwiched by n-doped CdS and p-doped ZnTe; the photons are incident on the n-doped layer, unlike in a PIN diode. 

A PIN photodiode can also detect X-ray and gamma ray photons. 

Example PIN photodiodes

SFH203 and BPW43 are cheap general purpose PIN diodes in 5 mm clear plastic cases with bandwidths over 100 MHz. RONJA telecommunication systems are an example application.