CHAPTER 4 Optical Sources,Transducers,and Measurement Systems
Blackbody Radiation Heated surface emits electromagnetic radiation. The energy emitted increases, and the spectral distribution of the radiation shifts toward shorter wavelengths as the temperature of the surface increases. This is readily noted by a shift in the color of the surface from red to blue with increasing temperature. A blackbody absorbs all radiant energy incident upon it regardless of wavelength. None of the radiant energy is transmitted and none is reflected. Because a black body is in thermal equilibrium with its environment, it must emit as much radiation as it absorbs.
A blackbody, in fact, radiates more total power and more power per unit wavelength interval than any other thermal radiating source at a particular temperature. Although no material is an ideal blackbody, one can approximate a blackbody in the laboratory by an oven with a small exit hole in one wall. Radiation that enters the hole has little opportunity to escape by reflections If the oven is heated, it can serve as a source and emit radiation through the hole. In thermal equilibrium, absorption balances emission except for the small amount of energy that escapes through the hole.
Planck's Law With an ideal blackbody, absorption exactly balances emission. Since absorption occurs at the same rate as emission, the blackbody must contain an equilibrium density of radiation. We will call the spectral energy density of a blackbody radiator U (J cm -3 Hz -1 ) or U (J cm -3 nm -1 ), where the superscript Planck's final assumption was that each atomic oscillator could absorb or emit only discrete amounts of energy directly related to its oscillation frequency. Each energy value had to be an integer multiple of the basic quantum of energy h. Statistical arguments then led directly to Planck'sradiation law: U b u
Spectral radiance where c 1 = 2hc 2 = 1.190 x 10 16 W nm 4 cm -2 sr -1 And c 2 = hc/k = 1.438 x 10 7 nm K.
Spectral radiance of a blackbody as a function of wavelength for several temperatures. Note the wavelength shift to the blue and the increase in spectral radiance as the temperature increases.
Einstein Coefficients Einstein introduced three probability coefficients, A ji, B ji, and B ij, for transitions between two levels in an atomic system Radiative processes in a two-level atomic system. The transition probability for absorption is B ij ; that for spontaneous emission is A ji ; while that for stimulated emission is B ji
For the simple two-level system, an atom initially in level i can interact with a radiation field of frequency ij, absorbing the required energy and undergoing a transition to level j. Einstein coefficient for absorption It is the number of absorption transitions per second per unit energy density
Applications to Spectroscopy The blackbody equations are used to characterize and compare real radiation sources. They are generally used to describe broad spectral distributions from continuum sources or the continuum distribution from sources such as high-pressure arc lamps, Finally, Planck's law can be applied to describe the emission and self-absorption of atoms in flames under thermal equilibrium conditions
Real blackbody systems To describe real sources, gray bodies, The following was modified from Planck s law: Balckbody spectral radiance Spectral emissivity 0f the source Transmittanc factor For the source window This source shows imperfect absorption. Gery body does not absorb all radiation incident on it
Conventional Radiation Sources Characteristics of sources Sources used in spectroscopy can be distinguished by the types of spectra they produce. Quasi continuum sources: lasers emitting continuum over a relatively narrow spectral interval ~ 1A 0
Continuum Sources (a)nernst glower (b) tungsten filament lamp (d) conventional Xe arc lamp c) D 2 lamp EIMAC-type Xe arc lamp with parabolic reflector
spectral emittance of tungsten filament lamp vs. wavelength
Arc Lamps
spectral irradiance of D 2 lamp vs. wavelength measured at 50 cm from source.
Continuum Plus Line Sources These are other high-pressure arc lamps produce intense line spectra superimposed on an intense spectral continuum. The high pressure and the arc restriction considerably broaden the atomic lines. The most common of these lamps are mercury and xenon-mercury arc lamps. The high-pressure mercury arc contains many broad lines from atomic mercury superimposed on a continuum.
Mercury arc lamps
Spectral distribution of high intensity arc lamps: curve a, 75-W xenon arc lamp; curve b, 100 W mercury arc lamp. Strong self-ab sorption causes the lack of radiance at 254 nm
Line Sources The fine sources used in many atomic spectroscopic applications include low-pressure arc lamps, hollow cathode discharge tubes, electrodeless discharge lamps, and thermal gradient lamps.
Thermal Gradient Lamp The thermal gradient lamp (TGL) can be constructed for elements of high volatility (As, Cd, P, S, Se, Te, Zn). Intensities of TGLs are as high as those of EDLs and the lines produced can be narrower. The TGL has a much shorter warm-up time than an EDL of the same element. Because of these factors TGLs could replace EDLs in atomic absorption determinations of As, Se, and Te, and they could become popular atomic fluorescence sources for some elements.
Commercial thermal gradient lamp The lamp is made of glass with a silica viewing window (W) and contains a few torr of argon filler gas. The element to be excited is placed in bulb B, which is heated by furnace F. A discharge occurs between cathode filament C and anode wire A. The discharge (~ 0.5 A) through the atomic vapor provides intense resonance lines of the element.
Why a Laser source Laser sources
Principles of Lasers Lasers are highly useful sources in analytical in strumentation because of their high intensities, narrow bandwidths, the coherent nature of their outputs. It has been used in various applications such as high-resolution spectroscopy kinetic studies of processes with lifetimes in the range of 10-9 to 10-12 s, detection and determination of extremely small concentrations of species in the atmosphere, the induction of isotopically selective reactions. In addition, laser sources have become important in several routine analytical methods, including Raman spectroscopy, molecular absorption spectroscopy, emission spectroscopy, and as part of instruments for Fourier transform infrared spectroscopy.
The term laser is an acronym for light amplification by stimulated emission of radiation. As a consequence of their light-amplifying properties, lasers produce spatially narrow (a few hundredths of a micrometer), extremely intense beams of radiation. The process of stimulated emission produces a beam of highly monochromatic (bandwidths of 0.01 nm or less) and remarkably coherent ( Because of these unique properties, lasers have become important sources for use in the ultraviolet, visible, and infrared regions of the spectrum. A limitation of early lasers was that the radiation from a given source was restricted to a relatively few discrete wavelengths or lines. Currently, dye lasers are widely available; tuning of these sources provides a narrow band of radiation at any chosen wavelength within the range of the source
Components of lasers The heart of the device is a lasing medium. It may be a solid crystal such as ruby, a semiconductor such as gallium arsenide, a solution of an organic dye, or a gas such as argon or krypton. The lacing material is activated or pumped by radiation from an external source so that a few photons of proper energy will trigger the formation of a cascade of photons of the same energy. an electrical current or by an electrical discharge. Thus, gas lasers usually do not have the external radiation source shown in Figure, instead, the power supply is connected to a pair of electrodes contained in a cell filled with the gas.
A laser normally functions as an oscillator, or a resonator, in the sense that the radiation produced by the lacing action is caused to pass back and forth through the medium numerous times by means of a pair of mirrors as shown in the figure below. Additional photons are generated with each passage, thus leading to enormous amplification. The repeated passage also produces a beam that is highly parallel because nonparallel radiation escapes from the sides of the medium after being reflected a few times. One of the easiest ways to obtain a usable laser beam is to coat one of the mirrors with a sufficiently thin layer of reflecting material so that a fraction of the beam is transmitted rather than reflected.
Mechanism of laser action Laser action takes place through four processes: (a) pumping (b) spontaneous emission (fluorescence) (c) stimulated emission (d) absorption The figure shows the behavior of two of the many molecules that make up the lacing medium. Two of the several electronic energy levels of each are shown as having energies Ey, and Ex. Note that the higher electronic state for each molecule has several slightly different vibrational energy levels depicted as Ey, Ey, EY", and so forth.
Pumping Pumping is a process by which the active species of a laser is excited by means of an electrical discharge, passage of an electrical current, or exposure to an intense radiant source. During pumping, several of the higher electronic and vibrational energy levels of the active species will be populated. In the diagram (a) one molecule is shown as being promoted to an energy state EY"; the second is excited to the slightly higher vibrational level EY The lifetime of excited states is brief, and after 10-13 to 10-15 s relaxation to the lower excited vibrational level occurs with the production of an undetectable quantity of heat. Some excited electronic states of laser materials have lifetimes considerably longer(often 1 ms or more) than their excited vibrational counterparts; long-lived states are sometimes termed meta stable as a consequence.
a. Pumping: Excitation by electrical or radiant or chemical energy
Spontaneous Emission A species in an excited electronic state may lose all or part of its excess energy by spontaneous emission of radiation as shown in the fig. below It is also important to note that the instant at which emission occurs and the path of the resulting photon vary from excited molecule to excited molecule because spontaneous emission is a random process; thus, the fluorescent radiation produced by one of the molecules (b1) differs in direction and phase from that produced by the second molecule (b2) Consequently, spontaneous emission yields encoherent monochromatic radiation
Stimulated emission Stimulated emission, which is the basis of laser behavior, is shown in the figure below. The excited laser particles are struck by photons having precisely the same energies (E y - E x ) as the photons produced by spontaneous emission. Collisions of this type cause the excited species to relax immediately to the lower energy state and to simultaneously emit a photon of exactly the same energy as the photon that stimulated the process The emitted photon travels in exactly the same direction and is precisely in phase with the photon that caused the emission Thus, the stimulated emission is totally coherent with the incoming radiation
Absorption The absorption process, which competes with stimulated emission, is depicted below Here, two photons with energies exactly equal to (E y - E x ) are absorbed to produce the metastable excited state shown in diagram d(3); note that the state shown in diagram d(3) is identical to that attained in diagram a(3) by pumping.
Absorption Here, two photons with energies exactly equal to (E y - E x ) are absorbed to produce the metastable excited state shown in diagram d(3); Note that the state shown in diagram d(3) is identical to that attained in diagram a(3) by pumping. Thus, the absorption process competes with the stimulated emission process
Population inversion and light amplification In order to have light amplification in a laser, it is necessary that the number of photons produced by stimulated emission exceed the number lost by absorption. This condition will prevail only when the number of particles in the higher energy state exceeds the number in the lower In other words, a population inversion from the normal distribution of energy states must exist. Population inversions are brought about by pumping. The figure below contrasts the effect of incoming radiation on a noninverted population with that on an inverted one.
In each case the population is shown as being made up of nine molecules of the lasing medium. In the non inverted system, three molecules are in the excited state and six are in the lower energy level. Three of the incoming photons are absorbed by the medium, thus producing three additional excited molecules. The radiation also stimulates emission of two photons from excited molecules Thus, the beam is attenuated by one photon
When the population is inverted a net gain in photons is observed because stimulated emission goes onto a grater extent than does absorption
Three and four level laser systems In the three-levels system, the transition responsible for laser radiation is between an excited state E y and the ground state E 0 in a four-level system, on the other hand, radiation is generated by a transi ion from E y to a state E x that has a greater energy than the ground state. Furthermore, it is necessary that transitions between E x and the ground state be rapid. The advantage of the four-level system is that the population inversions necessary for laser action are more readily achieved.
To understand this fact, note that at room temperature a large majority of the laser particles will be in the ground-state energy level E o in both systems. Sufficient energy must thus be provided to convert more than 50% of the lasing species to the Ey level of a three-level system. n contrast, it is only necessary to pump sufficiently to make the number of particles in the Ey energy level exceed the number in E X of a four-level system. The lifetime of a particle in the Ex state is brief because the transition to E o is fast; thus, the number in the E x state will generally be negligible with respect to the number having energy E o and also with respect to the number in the E y state.
Some examples of useful lasers Solid state lasers The first successful laser, and one that still finds widespread use, was a three level device in which a ruby crystal was the active medium. Ruby is primarily Al 2 O 3 but contains approximately 0.05% chromium(iii) distributed among the aluminum(iii) lattice sites, which accounts for the red coloration. The chromium(iii) ions are the active lasing material. In early lasers, the ruby was machined into a rod about 4 cm in length and 0.5 cm in diameter. A flash tube (often a low-pressure xenon lamp) was coiled around the cylinder to produce intense flashes of light (A = 694.3 nm). Because the pumping was discontinuous, a pulsed beam was produced. Continuous wave ruby sources are now available.
The Nd:YAG laser is one of the most widely used solid state lasers. It consists of neodymium ion in a host crystal of yttrium aluminum garnet. This system offers the advantage of being a four-level laser, which makes it much easier to achieve population inversion than the ruby laser. The Nd:YAG laser has a very high radiant power output Nd:YAG laser
Gas Lasers These devices are of four types: 1. Neutral atom lasers such as He-Ne It is most widely used. because of its low initial and maintenance costs, its great reliability, and its low power consumption. The most important of its output lines is at 632.8 nm. It is generally operated in the continuous mode rather than a pulsed mode. He- Ne Laser
2. ion lasers in which the active species is Ar + or Kr +, The Ar + ion laser, which produces intense lines in the green (514.5 nm) and the blue (488.0 nm) regions, is an important example of an ion laser. This laser is a four-level device in which argon ions are formed by an electrical or radio-frequency discharge. The required input energy is high because the argon atoms must first be ionized and then excited from their ground state, with a principal quantum number of 3, to various 4p states. Laser activity occurs when the excited ions relax to the 4s state.
3. molecular lasers in which the lasing medium is CO 2 or N 2 The N 2 laser, which must be operated in the pulsed mode because pumping is carried out with a high potential spark source, provides intense radiation at 337.1 nm. This output has found extensive use for exciting fluorescence in a variety of molecules and for pumping dye lasers. The CO 2 gas laser is used to produce a band of infrared radiation in the region of 900 to 1100 cm -1. Nitrogen laser
Eximer lasers Eximer lasers contain a gaseous mixture of helium, fluorine, and one of the rare gases-argon, krypton, or xenon. The rare gas is electronically excited by an electrical current, then reacts with the fluorine to form excited ions such as ArF +, KrF +, or XeF +, which are called eximers because they are stable only in the excited state. Since the eximer ground state is unstable, rapid dissociation of the compounds occurs as they relax while giving off a photon. Thus, a population inversion exists as long as pumping is carried on Eximer lasers produce high energy pulses in ultraviolet (351 nm for XeF, 248 nm for KrF, and 193 nm for ArF).
Dye Lasers Dye lasers have become important radiation sources in analytical chemistry because they are continuously tunable over a range of 20 to 50 nm. The bandwidth of a tunable laser is typically a few hundredths of a nanometer or less. The active materials in dye lasers are solutions of organic compounds capable of fluorescing in the ultraviolet, visible, or infrared regions. Dye lasers are four-level systems. the lower energy level for laser action is not a single energy but a band of energies arising from the superposition of a large number of closely spaced vibrational and rotational energy states upon the base electronic energy state. Electrons in E y may then undergo transitions to any of these states, thus producing photons of slightly different energies.
Dye laser pumped by N 2 laser. The beam expanding telescope increases the number of grooves of the grating hit by the beam and decreases the power density so as to avoid damage to the grating. Tuning is accomplished by rotating the grating.
Tuning of dye lasers can be readily accomplished by replacing the nontransmitting mirror shown previously with a monochromator equipped with a reflection grating or a Littrow-type prism which will reflect only a narrow bandwidth of radiation into the laser medium; The peak wavelength can be varied by rotation of the grating or prism. Emission is then stimulated for only part of the fluorescent spectrum-namely, the wavelength reflected from the monochromator.
Laser Characteristics Laser radiation has significantly different properties from radiation emitted by conventional sources. Laser radiation is highly directional, spectrally pure, coherent, and of very high radiance. The directionality of lasers is a direct result of the resonant cavity. Only waves that propagate normal to the mirrors oscillate. As a consequence laser radiation can be accurately transmitted over large distances (e.g., to a spot on the moon and back). In the laboratory, the directionality of lasers simplifies the alignment of optical materials and enables samples in remote locations to be probed.
Spectral purity is a result of the resonant interaction of the medium with the pump source, In the visible region it is fairly routine to achieve linewidths on the order of 0.01 to 0.1 A. Laser radiation is also classified as coherent radiation. Radiation is said to be temporally coherent, if for a given point in space, there is always a constant phase difference between the amplitude of the wave at two successive instances in time. The final aspect of laser radiation that makes it different from conventional radiation is its high irradiance. Typically, lasers can achieve irradiances that are 4 to 10 orders of magnitude larger than those from conventional sources. This is a direct consequence of the power and directionality of the laser.
Chemical Lasers With chemical lasers no optical or electrical pumping is necessary. A chemical reaction produces molecules in excised states. If population inversion is achieved, lasing can occur on specific transitions. The hydrogen fluoride laser is the best known example; lasing here occurs on vibrational transitions. An electrical discharge is used to produce hydrogen atoms and fluorine atoms. The chemical reaction (H 2 + F ---> HF* + H) produces excited HF.
Semiconductor Lasers In the semiconductor or diode laser, population inversion occurs between the conduction band and the valence band of a pn-junction diode. Stimulated transitions of the electrons from the conduction band to the valence band are responsible for laser action, and stimulated emission results from electron-hole recombinations. Since the frequency emitted is directly related to the band gap energy, various semiconductor compositions can be used to give different wavelengths. Also, since the lasing occurs between bands, these lasers can be tuned over small intervals. Typical materials are GaAs, which laws at 0.84 [Lm, and lead salt diodes (PbSnTe), which lase in the mid- IR region Infrared lasers can achieve very narrow line widths (10-6 cm -1 ) but must be operated at low temperatures (10 to 20 K).
Optical transducers The purpose of the radiation transducer is to convert radiant power into an electrical signal or to another physical quantity (e.g., heat or resistance) that can readily be converted to an electrical signal. Regardless of the specific mechanism involved, the characteristics of the transducer (sensitivity, linearity, dynamic range, signal-to-noise ratio, etc.) play an important role in determining the accuracy and precision attainable in spectrochemical methods.
Optical transducers fall into three major categories: thermal detectors, photon detectors, multichannel detectors. Thermal detectors They sense the change in temperature that is produced by the absorption of incident radiation. The temperature change is converted into an electrical signal by methods that depend on the specific transducer. They are thus highly useful for direct radiometric measurements as well as for spectroscopy.
Photon detectors, respond to incident photon arrival rates rather than to photon energies. The spectral response of these transducers varies with wavelength. A major advantage of photon detectors is their rapid response time, generally submicro second, compared to thermal detectors, which have response times that are slower than milliseconds. Photon detectors can also detect lower radiant powers than thermal detectors in many cases.
Multichannel detectors are photographic emulsions, arrays of multiple semiconductor detectors, or arrays of thermal detectors. The elements in array detectors are arranged linearly or in a two-dimensional grid. The general characteristics of transducers are presented first in order to define many of the terms that are later used in describing specific detectors. Thermal detectors, photon detectors, and multichannel detectors are then discussed.
Transducer Characteristics Radiation transducers vary widely in their sensitivity, linearity, spectral response, response speed, electrical output domain, and noise figures. To evaluate a transducer for a particular application, the characteristics discussed below are commonly used.
Sensitivity and Responsitivity The responsivity R( ) is the ratio of the signal output X (voltage, current, charge) to the incident radiant power ( evaluated at a particular wavelength and incident power. The sensitivity Q( ) is the slope of a plot of electrical output X vs. incident radiant power :
For photon detectors and for many thermal detectors, Q( ) and R( ) are wavelength dependent and specified at a particular wavelength. A plot of Q( ) vs. or R( ) vs. is called the spectral response of the transducer.
Transfer function is the overall functional relationship between the output quantity X and the input quantity The sensitivity is the slope of the transfer function, whereas the responsivity is its magnitude at a given incident power [R( ) = X/ ]. If Q( ) is constant and independent of, the detector is said to exhibit linearity. If a transducer shows linearity and its transfer function passes through the origin, the sensitivity is equal to the responsivity. In many cases these terms are thus used interchangeably. Usually, transducers exhibit linearity over a limited range of incident radiant power. The total range, expressed in powers of 10, for which the transfer function in linear is called the linear dynamic range, while the dynamic range usually refers to the total range of incident radiant power over which the transducer is responsive.
The sensitivity of a transducer is often not only a function of wavelength and incident radiant power, but may depend on such variables as temperature, bias voltage, and component values. Stability of the transducer is expressed by the constancy of Q( ) or R( ) with time is Stability can be expressed as short term (hours) or long term (days or weeks). The stability depends highly on maintaining constant the variables upon which Q( ) or R( ) depend. Degradation is the long-term change in Q( ) at a constant Some transducers exhibit hysteresis in that the responsivity at a particular power changes if the incident power is increased and then brought back to its original value.
Response Speed Transducers vary widely in their ability to detect rapid changes in incident radiant power Quantitatively, the response time is evaluated in terms of the time constant = 1/(27 f c ), where f c is the frequency at which R( ) has fallen to 0.707 of its maximum value The rise time is the time for the output to rise from 10% to 90% of its final value when an instantaneous (step function) increase in radiant power is incident on the transducer.
Electrical Output Characteristics Transducers also differ in their electrical output domains and their output impedances. The same transducer can produce outputs in different data domains. The photomultiplier tube, discussed in detail below, can produce an output charge (an analog domain), an output current (analog), an output pulse rate (a time domain),or a specific number of output pulses (digital domain).
Dark Signal Characteristics Dark signal is the electrical output of a transducer in the absence of radiation Because of their different response mechanisms, radiation transducers vary widely in their dark signal characteristics. Photoemissive detectors, which depend on photoelectrons being emitted from photosensitive materials, exhibit dark signals due to thermal emission of electrons. Although dark signals can in principle be subtracted from the total signal in the presence of radiation, noise and unidirectional drifts in the dark signal can become major sources of error
Noise Characteristics The noise equivalent power (NEP or n) is the radiant power in watts of a sinusoidally modulated input incident on the detector that gives rise to a signal equal to the dark noise in a 1-Hz bandwidth. The dark noise d is thus given by d = R( ) n Since it is a function of wavelength, the NEP is usually specified at a particular wavelength. The modulation frequency, electrical bandwidth, and detector area should also be specified.
The detectivity D (W -1 ) is a measure of minimum detectability and is defined as D =1/ n As with the NEP, the wavelength, modulation frequency, band width, and detector area should be specified. The D star (D*), in cm Hz 1/2 W -1, is a normalization of detectivity to take into account the area and electrical bandwidth dependence. It is related to D by D* = DA 1/2 ( f) 1/2 where A is the detector area in cm 2 and f is the noise equivalent bandwidth in Hz.
Thermal Detectors
Pneumatic detector A pneumatic detector is based on a thin blackened membrane placed in a gas-filled, air tight chamber. As radiation strikes the detector, the gas is heated and expands against another membrane. The displacement of the membrane is detected in some types. In others the capacitance of the membrane serves as a measure of displacement. A commercial pneumatic cell is the Golay detector, which is widely used in infrared spectrometers. The Golay cell is sensitive (see D* in Table 4-5), but tends to be fragile.
Thermocouple detector A thin blackened strip or flake is connected thermally to the junction of two dissimilar metals. Radiation absorbed by the strip causes the junction to increase in temperature and a change in thermoelectric voltage is produced. The thermocouple detector has uniform spectral response in the region 1 to 40 m, reasonable sensitivity, excellent linearity. Thermocouples have high stability Since their output voltages are often on the order of microvolts, they require large amplification factors. Multiple-junction thermocouples, called thermopiles are also used.
Thermistor Bolometer The thermistor is made from an intrinsic semiconductor. As the temperature increases the number of valence-band electrons promoted to a conduction band increases, which increases the conductivity and decreases the resistance. A thin blackened tip allows the absorption of radiation, which heats the thermistor. The thermistor is normally placed in a bridge circuit with a reference thermistor that is not irradiated. The resistance can be measured by a null-comparison technique, or the out-of-balance voltage of the bridge can be monitored. Bolometers are rugged and exhibit moderate sensitivity and a wide linear range. The thermistor spectral response normally peaks in the near-ir region. Thermistor detectability is limited by thermal noise at frequencies above 20 Hz and by 1/f noise at lower frequencies.
Pyroelectric Detector A pyroelectric detector is typically made from triglycine sulfate (TGS). When placed in an electrical field, a surface charge results from alignment of electric dipoles. When a pulse of incident radiation heats the TGS, a change in surface charge results (pyroelectric effect), which is related to the incident radiant power. The output current is proportional to the rate of temperature change of the material dtldt The pyroelectric detector is fast (<1 ms response time) because only charge-reorientation limits the response speed for modulated inputs. For wavelengths below 2 m the TGS must be blackened, which can slow the response. The spectral response of a blackened detector is fairly flat over the region 1 to 36 m.
Photon Detectors Photon detectors include: (1) photovoltaic cells, in which the radiant energy generates a current at the interface of a semiconductor layer and a metal; (2) phototubes, in which radiation causes emission of electrons from a photo sensitive solid surface; (3) photomultiplier tubes, which contain a photoemissive surface as well as several additional surfaces that emit a cascade of electrons when struck by electrons from the photosensitive area; (4) photoconductivity detectors, in which absorption of radiation by a semiconductor produces electrons and holes, thus leading to enhanced conductivity; and (5) silicon photodiodes, in which photons increase the conductance across a reverse-biased pn junction.
Photon detectors Photon detectors can be broadly classified as photoemissive devices (photomultipliers and photo-tubes), pn-junction devices (photodiodes, phototran sistors), photoconductive cells, and photovoltaic cells.
Vacuum Phototubes vacuum phototube (PT), sometimes called a vacuum photodiode, consists of two electrodes sealed in an evacuated glass or silica envelope. The photosensitive cathode can be made from a number of photoemissive materials (e.g., Cs3Sb, alkali metal oxides, AgOCs). Thephoto cathode consists of a photosensitive material. It is biased negative with respect to the anode A Irradiation of the cathode causes photoelectrons to be emitted and attracted to the anode.
Only a certain fraction of the photons with greater than threshold energy yield photoelectrons with sufficient kinetic energy to escape the photocathode This fraction is called the quantum efficiency K( ) and is the ratio of the number of photoelectrons ejected to the number of incident photons. Typically, K( ) varies from0 to 0.5. In the absence of radiation, a small anodic dark current iad is obtained, where i ad = i cd and i cd is the cathodic dark current. At moderate bias voltages dark current can be due primarily to thermal emission of electrons at the photocathode. When voltages are low, ohmic leakage can dominate. At low temperatures luminescence and radioactivity can be the primary source of dark current. When high voltages are applied, ionization of residual gas and field emission can be additional sources.
Photomultiplier Tubes The photomultiplier tube (PMT), like the phototube, contains a photosensitive cathode and a collection anode. The cathode an anode are separated by several electrodes, called dynodes, that provide electron multiplication or gain. The cathode is biased negative by 400 to 2500 V with respect to the anode. A photoelectron ejected by the photocathode strikes the first dynode and releases two to five secondary electrons. Each secondary electron is accelerated by the field between the first and second dynode and strikes the next dynode with sufficient energy to release another two to five electrons.
Since each dynode down the chain is biased ~ 100 V more positive than the preceding dynode, this multiplication process continues until the anode is reached. The result is a large charge packet of a few nanoseconds' duration at the anode for each photoelectron collected by the first dynode. Dark current in PMTS arises from effects similar to those in phototubes. In addition, thermal emission from the dynodes can be a source of dark current in PMTs. Since the cathodic dark current is multiplied by the full gain of the tube, thermal emission from the photocathode (or early dynodes) is often a major component of the total dark current. Typically, dark currents in PMTs are in the range 10-7 to 10-11 A. For very low light levels, the dark current can be as large as the photocurrent, and noise in the dark current can limit precision.
Photodiode detectors (Silicon diode detectors) A silicon diode detector consists of a reverse-biased pn junction formed on a silicon chip. The reverse bias creates a depletion layer that reduces the conductance of the junction to nearly zero. If radiation is allowed to impinge on the chip, however, holes and electrons are formed in the depletion layer, and these provide a current that is proportional to radiant power. A silicon diode detector is more sensitive than a simple vacuum phototube but less sensitive than a photomultiplier tube Photodiodes have spectral ranges from about 190 to 1100 nm.
Photodiodes and Phototransistors In a photodiode, absorption of electromagnetic radiation by a pn-junction diode causes promotion of electrons from the valence band to the conduction band and thus the formation of electron-hole pairs in the depletion region, If the rate of light induced charge carrier production greatly exceeds that due to thermal processes, the limiting current under reverse bias is directly proportional to the incident radiant power. The photodiode acts as a current source when operated in the reverse-biased mode.
The spectral response of most photodiodes reaches a maximum in the near-ir region (0.85 to 1.0 m). They are very useful for UV-visible and near-ir detection. These devices often show excellent linearity over six to seven decades of incident radiant power. Responsivities are typically much lower than those of photomultiplier tubes. the simplicity, excellent linearity, and very small sizes of photodiodes make them attractive for applications where light levels are relatively high. They have practically replaced vacuum phototubes in all but a few applications. Photodiodes are extremely fast transducers. A special type, known as a pin junction, in which the p- and n-type semiconductor materials are separated by an insulating layer, has a subnanosecond response time.
Photoconductivity Detectors The most sensitive detectors for monitoring radiation in the near-infrared region of about 0.75 to 3 m are semiconductors whose resistances decrease when radiation within this range is absorbed. The useful range of photoconductors can be extended into the far-infrared region by cooling to suppress noise arising from thermally induced transitions among closely lying energy levels Crystalline semiconductors are formed from the sulfides, selenides, and tellurides of such metals as lead, cadmium, gallium, and indium. Absorption of radiation by these materials promotes some of their bound electrons into an energy state in which they are free to conduct electricity. The resulting change in conductivity can then be measured with an appropriate circuit
Photoconductive Cells The photoconductive cell is made of a semiconductor material such as CdS, PbS, PbSe, InAs, InSb, He-Cd-Te, or Pb-Sn-Te. It acts like a light-dependent resistor, which decreases in resistance when photons are absorbed. Incident photons release electron-hole pairs and increase conductivity. The PbS cell is still the most sensitive uncooled detector in the near-ir region of the spectrum 1.3 to 3 m. Cooling is necessary to avoid thermal excitation of electrons into the conduction band. The GeZn and GeCu detectors have D* values that exceed those of thermal detectors in the IR region although their spectral response is not flat.
Photovoltaic Cell The photovoltaic cell or barrier-layer cell converts radiant energy into electrical energy. In the open-circuit mode, no bias is required and a high impedance voltage measurement circuit is used to measure the potential difference. In this mode a limited linear dynamic range is achieved.
Photovoltaic Or Barrier-layer Cells The photovoltaic cell is used primarily to detect and measure radiation in the visible region. The typical cell has a maximum sensitivity at about 550 nm; the response falls off to perhaps 10% of the maximum at 350 and 750 nm Its range approximates that of the human eye. The photovoltaic cell consists of a flat copper or iron electrode upon which is deposited a layer of semi conducting material such as selenium The outer surface of the semiconductor is coated with a thin transparent metallic film of gold or silver, which serves as the second or collector electrode; the entire array is protected by a transparent envelope.
When radiation of sufficient energy reaches the semiconductor, covalent bonds are broken, with the result that conduction electrons and holes are formed. The electrons then migrate toward the metallic film and the holes toward the base upon which the semiconductor is deposited. The liberated electrons are free to migrate through the external circuit to interact with these holes. The result is an electrical current of a magnitude that is proportional to the number of photons striking the semiconductor surface. Ordinarily, the currents produced by a photovoltaic cell are large enough to be measured with a micrometer. The photocurrent is directly proportional to the power of the radiation striking the cell. Barrier-type cells find use in simple, portable instruments where ruggedness and low cost are important. For routine anal yses, these instruments often provide perfectly reliable analytical data.
Multichannel Detectors Multichannel photon detectors consist of an array of tiny photosensitive detectors that are arranged in such a pattern that all elements of a beam of radiation that has been dispersed by a grating can be measured simultaneously. Multichannel detectors when placed in the focal plane of a spectrograph can provide simultaneous detection of the dispersed radiation. The photographic emulsion was, of course, the original multichannel optical transducer. Photographic Detectors. Photographic films or plates are emulsions that contain silver halide crystals. Various emulsions can be obtained with responses from the UV to the near-ir region. For quantitative purposes the density of the exposed areas must be obtained. The major drawback of the photographic emulsion is the time required for development and densitometry.
Linear Photodiode Arrays In a linear photodiode array, the individual photosensitive elements are small silicon photodiodes, each of which consists of a reverse-biased pn junction The individual photodiodes are part of a large-scale integrated circuit formed on a single silicon chip. Light incident upon these elements creates charges in both the p and n regions. The positive charges are collected and stored in the p-type bars for subsequent integration (the charges formed in the n regions divide themselves equally between the two adjacent p regions). The number of sensor elements contained in a chip ranges from 64 to 4096, with 1024 being perhaps the most widely used. The integrated circuit making up a diode array also contains a storage capacitor and switch for each diode as well as a circuit for sequentially scanning the individual diode-capacitor circuits
Reversed-biased linera Photodiode Array detector Each element consists of a diffused p-type bar in an n-type silicon substrate to give a surface region that consists of a series of side-by-side elements having typical dimensions of 2.5 by 0.025 mm
Radiation impinging upon the depletion layer in either the p or the n region forms charges (electrons and holes) that create a current that partially discharges the capacitor in the circuit. The capacitor charge that is lost in this way is replaced during the next cycle. The resulting charging current is integrated by the preamplifier circuit, which produces a voltage that is proportional to the radiant intensity. After amplification, the analog signal from the preamplifier passes into ananalog-to-digital converter and to a microprocessor that controls the readout
In using a diode array detector, the slit width of the spectrometer is usually adjusted so that the image of the entrance slit just fills the surface area of one of the diodes making up the array. Thus the same information is obtained as would be by rotating the dispersing element so that a series of slit images is focused sequentially on the detector. With the array, information about the entire spectrum is accumulated essentially simultaneously.
Vidicons A vidicon tube contains an electron gun and a target, both of which are housed in a vacuum tube that is surrounded by a focusing coil that causes the electron beam to sweep systematically across the target A typical target has a diameter of 16 mm and is made up of over 15,000 silicon photodiodes per mm2. Each photodiode consists of a cylindrical section of p-type silicon surrounded by an insulating layer of silicon dioxide. All of the diodes are backed by a common layer of n-type silicon. The p-type layer of the target faces the electron beam, whereas the n-type layer faces the source of radiation.
When the surface of the target is swept by the electron beam, each of the p-type cylinders becomes charged to the potential of the beam and forms a miniature capacitor with the n-type layer directly behind it. When photons strike the n-type layer, positive holes and electrons are formed that partially discharge the capacitor. The next sweep of the beam then recharges the capacitors sequentially. The resulting charging currents are amplified and stored one by one in a computer as a function of time (or location of the beam along the focal plane of the spectrometer).
Vidicon Tube Detector
Target array of silicon diodes
Charge-transfer Detectors Neither photodiode arrays nor vidicons can match the performance of photomultiplier tubes in terms of sensitivity, dynamic range, and signal-to-noise ratio. A new class of multi channel imaging devices has been developed and applied in the fields of astronomy, astrophysics, and microscopy. these charge-transfer devices are useful in analytical spectroscopy as well, because they appear to offer not only the multichannel advantage but also performance characteristics that match those of the photomultiplier tube. Charge-transfer devices are made up of an array of tiny semiconductor capacitors that have been formed on a single silicon chip. Several types of these devices are currently marketed, which vary in design and in the way quantitative data are obtained
Signal Processors and Readouts The signal processor is ordinarily an electronic device that: amplifies the electrical signal from the detector; in addition, it may alter the signal from dc to ac (or the reverse), change the phase of the signal, and filter it to remove unwanted components. Furthermore, the signal processor may be called upon to perform such mathematical operations on the signal as differentiation, integration, or conversion to a logarithm. Several types of readout devices are found in modern instruments. Some of these include digital meters and the scales of potentiometers, recorders, and cathode ray tubes.
Photon Counting Frequently, the output from the photoelectric detectors is processed and displayed by analog techniques. That is, the average current, potential, or conductance associated with the detector is amplified and recorded or fed into a suitable meter. In some instances, however, it is possible and advantageous to employ direct digital techniques wherein electrical pulses produced by individual photons are counted. Here, radiant power is proportional to the number of pulses rather than to an average current or potential. Photon counting has been applied to ultraviolet and visible radiation.
Photon counting has a number of advantages over analog signal processing, including: improved signal-to noise ratio, sensitivity to low radiation levels, improved precision for a given measurement time, lowered sensitivity to voltage and temperature changes. The required equipment is more complex and expensive; the technique has not been widely applied for routine measurements in the ultraviolet and visible regions.
Types of Design of OpticalInstruments Spectroscopic instruments are designed to provide bands of radiation of known wavelength and, in most instances, to give information about the intensity or power of these bands. Three basic instrument designs for accomplishing these purposes are recognizable: temporal designs, spatial designs, multiplex designs. Within each of these major categories are two sub classes, nondispersive and dispersive.
Temporal Designs Temporal instruments operate with a single detector and are often termed single-channel devices as a consequence. In such systems, successive radiation bands are examined sequentially in time.
Nondispersive Instruments An example of a nondispersive temporal instrument is a photometer equipped with a series of narrow band filters of appropriate wavelength. Such an instrument can, for example, be used for the quantitative determination of each of the alkali metals by injection of a solution of the sample into a flame. By interchanging filters, the intensity of a line for each of the elements could be determined sequentially and used to calculate its concentration. Tunable lasers also permit the construction of a nondispersive temporal instrument for determining a portion of the absorption or emission spectrum of a compound. Here, a photomultiplier tube provides a series of light intensity data as the laser is tuned serially from one wavelength to the next.
Nondispersive temporal instruments generally offer the advantage of simplicity, low cost, high energy throughput (often leading to better signal-to-noise ratios), lower levels of stray radiation. On the other hand, they do not provide important spectral detail over a wide range of wavelengths, which is particularly important for qualitative and structural studies.
Dispersive Instruments The two monochromators depicted in Figure 6-11 can be operated as temporal dispersive instruments by lo cating a photoelectric detector at the exit slit. Spectra can be obtained by rotating the dispersing element (monochromator) manually or mechanically while monitoring the output of the detector. A sequential linear scan instrument often contains a motor-driven grating system that sweeps the spectral region of interest at a constant rate. The paper drive of a recorder is synchronized with the motion of the dispersing element, thus providing a wavelength scale based upon time. A sequential slew (slide) scan instrument may be similar to the one just described except that it is programmed to recognize and remain at significant spectral features such as peaks until a suitable signal-to-noise ratio is attained Regions in which the radiant power is not changing rapidly (that is, where the derivative of power with respect to time approaches zero) are scanned at high speed until the next peak is approached. Sequential slew scan instruments, while more complex and expensive than their linear scan counterparts, provide data more rapidly and efficiently.
Spatial Designs instruments are based upon multiple detectors or channels to obtain information about different parts or elements of the spectrum simultaneously. Nondispersive Systems As an example: a photometer for the simultaneous determination of sodium, potassium, and lithium has been described in which the radiation from a flame containing the sample is allowed to illuminate three slits arranged at different angles from the source A photomultiplier tube is placed at each slit, as well as interference filters that selectively transmit the peak radiation for one of the elements; simultaneous monitoring of the concentration of the three elements is thus possible.