Instrumentation. 1) Sources 2) Wavelength Selectors 3) Detectors Recommended Reading: Applied Spectroscopy

Transcription

1 Applied Spectroscopy Instrumentation 1) Sources 2) Wavelength Selectors 3) Detectors Recommended Reading: Spectrophysics, Thorne Chapters 11 and 12 Principles of Instrument Analysis, Skoog Holler and Nieman Chapter 7

2 Basic Components Five basic elements in all spectroscopic instruments. 1) A stable source of radiation 2) A sample 3) A wavelength selector (spectrometer) 4)A radiation detector 5) A signal processor Two main categories of spectrometer: 1) Dispersive spectrometers, use prisms and gratings to spread the wavelengths out spatially 2) Interferometric spectrometers, eg Michelson and Fabry-Perot.

3 Radiation Sources

4 Radiation Sources Radiation (light) sources for spectroscopy must satisfy the following two requirements 1) sufficinet power2) stability Two types of Source 1. Continuum Sources: give a broad featureless continuous distribution of radiation. UV region Ar, Xe, He discharge lamps Visible Tungsten Filament Lamp Infra Red Blackbody radiation from heated inert bodies, Nernst, Globar 2. Line Sources: produce relatively narrow bands at specific wavelengths generating structured emission spectrum UV / Visible Hg and Na vapour lamps Uv/Visible/IR Lasers 1+2) Line plus continuum sources contain lines superimposed on continuum background - medium pressure arc lamps, D 2 lamp Sources may be continuous or pulsed in time

5 Continuum Sources Continuum sources are preferred for spectroscopy because of their relatively flat radiance versus wavelength curves Nernst Glower Tungsten Filament D 2 Lamp Arc Lamp Arc Lamp with Parabolic Reflector

6 Blackbody Sources A hot material, such as an electrically-heated filament in a light bulb, emits a continuum spectrum of light. The spectrum is approximated by Planck's radiation law for blackbody radiators: where h is Planck's constant, ν is 3 2hν 1 frequency, c is the speed of light, k P = 2 c exp( hν kt) 1 is the Boltzmann constant, and T is temperature in K

7 Blackbody Sources Infra Red Globar : 1-40 µm Silicon Carbide (SiC) rod (50mm long 50 mm diameter) electrically heated to about 1400K Nernst glower (ZrO 2, YO 2 ): 400 nm - 20 µm Cylinder of rare earth oxide electrically heated to about 2000K K in air - λ max lies in IR - relatively fragile -low spectral radiance ( ~10-4 W cm -2 nm -1 sr -1 )

8 Blackbody Sources Heated filaments (W incandescent lamp, QTH) K in evacuated envelope - greater radiance (U=σ T 4 ) P λ ~ 10-2 W cm - 2 nm-1 sr -1 - greater UV-Vis output - λ max still in IR - QTH heated up to 3600 K WI 3600K ( ) W( g) ( ) + I WI ( g) W s W g 2 2 Hot W ( g) W( s) + I2 2

9 Discharge Lamps Discharge lamps, such as neon signs, pass an electric current through a rare gas or metal vapor to produce light. The electrons collide with gas atoms, exciting them to higher energy levels which then decay to lower levels by emitting light. Low-pressure lamps have sharp line emission characteristic of the atoms in the lamp, and High-pressure lamps have broadened lines superimposed on a continuum. Common discharge lamps and their wavelength ranges are: hydrogen or deuterium : nm mercury : nm, and weaker lines in the near-uv and visible Ne, Ar, Kr, Xe discharge lamps : many sharp lines throughout the near-uv to near-ir xenon arc : nm The sharp lines of the mercury and rare gas discharge lamps are useful for wavelength calibration of optical instrumentation. Mercury and xenon arc lamps are used to excite fluorescence.

10 Hydrogen Discharge Lamp Discharge Lamps

11 Hollow Cathode Lamps (HCL) Hollow-cathode lamps are a type of discharge lamp that produce narrow emission from atomic species. They get their name from the cup-shaped cathode, which is made from the element of interest. The electric discharge ionizes rare gas atoms, which are accelerated into the cathode and sputter metal atoms into the gas phase. Collisions with gas atoms or electrons excite the metal atoms to higher energy levels, which decay to lower levels by emitting light. Hollow-cathode lamps have become the most common light source for atomic absorption (AA) spectroscopy. They are also sometimes used as an excitation source for atomic-fluorescence spectroscopy (AFS).

12 Lasers A laser is a coherent and highly directional radiation source. LASER stands for Light Amplification by Stimulated Emission of Radiation. A laser consists of at least three components: 1. a gain medium that can amplify light that passes through it 2. an energy pump source to create a population inversion in the gain medium. Requires at least a three state medium 3. two mirrors that form a resonator cavity Pump Pump so that N 1 > N 0 (population inversion) Pumping Methods: Optical - flashlamp, laser Electrical - capacitive electrical discharge Chemical - reaction leaving product in excited state 2 Fast Decay 0 1 Lasing

13 Gas Lasers Gas lasers are typically excited by an electrical discharge. Some gas lasers and their dominant lasing wavelength(s): nitrogen : 337 nm (pulsed) He-Ne : nm (cw) Ar ion : 488, 541 nm (cw) CO 2 : 10.6 µm (cw or pulsed) excimer: ArF* nm, XeCl* nm (pulsed)

14 Dye Lasers The gain medium in a dye laser is an organic dye molecule that is dissolved in a solvent. The dye and solvent are circulated through a cell or a jet, and the dye molecules are excited by flashlamps or other lasers. Pulsed dye lasers use a cell and cw dye lasers typically use a jet. The organic dye molecules have broad fluorescence bands and dye lasers are typically tunable over 30 to 80 nm. Dyes exist to cover the nearuv to near-infrared spectral region: nm.

15 Semiconductor Lasers Semiconductor lasers are light-emitting diodes within a resonator cavity that is formed either on the surfaces of the diode or externally. An electric current passing through the diode produces light emission when electrons and holes recombine at the p-n junction. These lasers are used in optical-fiber communications, CD players, and in high-resolution molecular spectroscopy in the near-infrared. Diode laser arrays can replace flashlamps to efficiently pump solid-state lasers. Diode lasers are tunable over a narrow range and different semi-conductor materials are used to make lasers at 680, 800, 1300, and 1500 nm.

16 Solid State Lasers The gain medium in a solid-state laser is an impurity center in a crystal or glass. Solid-state lasers made from semiconductors are described below. The first laser was a ruby crystal (Cr 3+ in Al 2 O 3 ) that lased at 694 nm when pumped by a flashlamp. The most commonly used solid-state laser is one with Nd 3+ in a Y 3 Al 5 O 8 (YAG) or YLiF 4 (YLF) crystal or in a glass. These Nd 3+ lasers operate either pulsed or cw and lase at approximately 1064 nm. The high energies of pulsed Nd 3+ :YAG lasers allow efficient frequency doubling (532 nm), tripling (355 nm), or quadrupling (266 nm), and the 532 nm and 355 nm beams are commonly used to pump tunable dye lasers.

17 Wavelength Selectors

18 I 0 I 0 /2 Resolving Power A more fundamental concept than dispersion. Full Width at Half Maximum FWHM Even an ideal spectrometer, illuminated by an ideal source of monochromatic light still has a finite width set by diffraction limits of the optics instrument function. If λ and λ +δλ are the wavelengths of two monochromatic lines that can be just separated by a spectrometer, then the λ R = resolving power of the δλ spectrometer is defined as = Rayleigh Criterion Note that R is dimensionless. Also λ ν E ν R = = = = ~ δλ δν δe δν ~

19 Wavelength Selectors Ideally the output from a wavelength selector should be radiation of a single wavelength or frequency (monochromatic). No real wavelength selector approaches this ideal. Instead a band of wavelengths is obtained. The effective bandwidth is an inverse measure of the performance of a wavelength selector. Narrow bandwidth better performance Two types of wavelength selectors usually encountered: 1) Filters Interference Filters Interference Wedges Absorption Filters 2) monochromators Prism type Grating type

20 Characterisation of Filters Interference and Absorption filters are characterized by three parameters: 1) The centre wavelength of the transmitted radiation λ o 2) Percentage of λ o transmitted by the filter = % transmittance 3) Effective Bandwidth i.e. The Full Width at Half Maximum (FWHM) of the transmitted line shape. Range from UV visible IR Bandwidths (Δλ / λ) ~ 1.5% with ~ 90% transmittance but can also get (Δλ / λ) ~ 0.15% with ~ 10% transmittance

21 Interference Filters White Radiation Glass Plate Metal film Dielectric Layer, eg MgF, CaF λ λ λ θ θ Narrow band of radiation t nλ = 2t / cos(θ) for θ small then cos(θ) 1 nλ 2t refractive index of medium = η 2tη λ = n λ = ηλ

22 Interference Wedges continuous distribution of wavelengths wavelength is now a function of position along the wedge λ(t) = 2tη n

23 Absorption Filters Transmit a narrow band of radiation. Used in Visible Region only. usually made of 1) Colored Glass or 2) Dye suspended in gelatin % transmittance Cutoff Filters wavelength Performance of absorption filters is inferior to interference filters Filters only give a fixed band of wavelengths but usually need to SCAN the spectrum, e.g. to measure absorption (or reflection) as a function of wavelength.

24 Monochromators Monochromators - instruments designed for spectral scanning. Separate EM radiation into individual wavelength components. Monochromators for the UV, vis, IR similar in construction but different materials used. All monochromators have the following common elements 1) Entrance Slits 2) Collimating lens or Mirror 3) Dispersing element (Prism or Grating) 4) Focusing element 5) Exit slit

25 Monochromator Components f 1 f Disperser 2 Prism or Grating λ 1 λ 2 Entranc e Slit S Collimating Lens L1 Focusing Lens L2 Exit Slit or Photographi c Plate

26 Dispersion Dispersive spectrometers separate different wavelengths by spreading them out spatially. Dispersion is a measure of this spreading How do we quantify the spatial separation of wavelengths on the exit focal plane? Dispersive Element (Prism, grating ) θ θ +dθ If rays of wavelength λ and λ + dλ emerge from the disperser at angles θ and θ + dθ, then the angular dispersion D a, of the spectrometer is defined as dθ = Units: rad.nm -1. D a dλ f λ λ+dλ The linear dispersion D, is a measure of the linear separation of the two wavelengths in the focal plane f of the lens or concave grating, D dy = dλ y 1 y+dy Units: mm.nm -1.

27 Reciprocal Dispersion A relationship between angular and linear dispersion can be obtained from the fact that for small angles dy = f dθ, where f is the focal length of the instrument. Then D = f D a. It is more usual to use reciprocal dispersion D -1 = dλ / dy d 1d D 1 λ λ = = Units: nm.mm -1. dy f dθ D -1 is typically around nm.mm -1 in UV/Visible. Explicit quantities for these expressions depend on the type of dispersive instrument used, e.g. prism or grating and will be derived below.

28 Prism Instruments Prism can serve several different purposes in a spectrome - change the direction of a beam - change the polarization of a beam - split a beam into two - disperse the beam A variety of shapes and materials are available to perfor functions. S L 1 Pris m L 2 λ 1 λ 2

29 i 1 r 1 Deviation and Dispersion θ = d1 + d2 d1 = i1 r1 and d2 = i2 r2 α = r1 + r2 θ = i1 + i2 α If prism is at or close to position of minimum deviation then i1 = i2 and θ = 2 i1 and α = 2 r Combine these relationships with Snell s Law of refraction, Sin(i) = n.sin(r) for refractive index n gives θ + α () α sin i = sin = n.sin 2 2 α b α θ d 1 d 2 dθ dθ dn = and from Snell s law dλ dn dλ r 2 i 2 r1 α = r2 the change in deviation θ with wavelength determines the angular dispersion of the prism dθ dn = 2 sin( α 2) { 2 2 ( )} 1 n sin α 2 1 2

30 therefore dθ dλ = dθ dn dn dλ = 2 sin( α 2) { ( )} 1 2 λ 1 n 2 sin 2 α 2 dn d for a 60 prism sin(α/2) = 1/2 and then dθ dλ dθ dn = = dn dλ 1 dn d { ( ) 2 } 1 2 λ 1 n 2 Resolving Power dn dλ dn R = b dλ for values of n in the range 1.4 to 1.6 the first term on the RHS is approximately n to with in 4%, then dθ dn D A = = n dλ dλ angular dispersion where b = length of the base of the prism

31 because d θ dn = n dλ dλ there will be no dispersion if n(λ) is constant - dispersion in prism occurs because of the change in refractive index of the prism material as a function of wavelength - if prism material exhibits normal dispersion, higher frequency (shorter wavelength) light experiences a higher refractive index than lower frequency (longer wavelength) light dn dλ dn dλ (glass@357 nm) =1.94 x 10-4 nm -1 (glass@825 nm) =1.78 x 10-1 nm -1 Prisms not often used as dispersion elements because of non-constant D A with wavelength - produces non-constant bandwidth - means range of λ's projected onto exit slit varies with λ

32 Littrow Mounting P R S L Reflecting back surface on prism Wadsworth Mounting Prism 90 Mirror This is a constant deviation mounting. The Mirror rotates with the prism such that the deviation of the beam is always 90 degrees.

33 Diffraction Gratings Two types (1) Transmission Grating, (2) Reflection Grating α β α β α β d d for light of wavelength λ the condition for constructive interference is mλ = d ( sinα + sinβ ) Grating Equation β is positive if it is on the same side of the normal as α, otherwise it is negative m is an integer, the diffraction order Zero order (m = 0) means straight through transmission or specular reflection

34 Definition of Blaze Angle γ the angle γ between the groove facet and the horizontal is called the blaze angle of the grating

35 Important points about Diffraction Gratings - diffraction angle depends on d - longer λ's diffracted more than shorter ones (β 600 nm > β 500 nm ) - When m = 0 (zero order), sinα = -sin β or α = - β. In this case, all λ's are diffracted at the same angle If blaze was parallel to the grating plane (γ = 0 ), the zero order beam would also appear in the specular direction (most of the reflected light not dispersed) (see diagram on next page) If blaze angle 0, specular and zero-order angles do not correspond and majority of the light is dispersed In the special case when incident beam is along the surface normal, α = 0 and first-order beam is in specular direction - in this case, β is twice the blaze angle, γ. The wavelength at this angle is called the blaze wavelength. (see diagram on next page)

36 In the special case when incident beam is along the surface normal, α = 0 and first-order beam is in specular direction mλ λ blaze blaze = = = ( α + sinβ) d sin dsinβ dsin2γ

37 Dispersion and Resolving Power of a Grating Angular dispersion can be found by differentiation the grating equation ( α + sinβ) ( sinα sinβ) dβ m d sin + D A = = = = dλ dcosβ dλcosβ λcosβ Near the grating normal, cos(β) 1, the dispersion has an almost constant value of m/d giving an almost linear wavelength scale gratings for the visible and UV typically have between 600 and 1200 lines/mm. Resolving Power R = order number of grooves on the grating R = mn = W m d number of grooves = width of grating / distance between grooves

38 Example a 600 lines/mm grating used in first order has an angular dispersion of rad.nm -1 which gives a reciprocal dispersion of 1.6 nm.mm -1 with a 1m focal length spectrometer. a 1200 lines/mm grating used in second order in the same spectrometer has an angular dispersion of rad.nm -1 which gives a reciprocal dispersion of 0.4 nm.mm -1 with a 1m focal length spectrometer. The same 1200 grating in a 3m spectrometer gives a reciprocal dispersion of 0.13 nm.mm -1. For fixed values of α and β, nλ is constant. Example: nλ = nm n: λ (nm):

39 Monochromators Comprised of - dispersive element - image transfer system (mirrors, lenses and adjustable slits) an image of the entrance slit is transferred to the exit slit after dispersion. One of the most common arrangements is the Czerny- Turner monochromator:

40 Other Grating Mounts G Littrow Mount D R S M Ebert Mount S 1 G M S 2

41 Wavelength Selection Wavelength selection is accomplished by rotating the grating Since angle between the entrance slit, grating and exit slit is fixed (2φ ), grating formula can be expressed in terms of the grating rotation angle θ (between grating normal and optical axis) Since α = θ - φ and β = θ + φ, mλ = d [sin(θ - φ ) + sin(θ + φ )] = 2d.sinθ.cosφ (the trigonometric identity 1/2(sin(A+B)+sin(A-B)) = sina cosb) Grating formula now in experimental variables: θ (the grating rotation angle) and φ (half-angle between the entrance, grating and exit and slit).

42 Dispersive Characteristics in the Focal Plane for monochromator operation we are much more interested in dispersion at focal plane (exit slit), defined by the linear dispersion, D l = dx/dλ For a Czerny-Turner arrangement, the linear dispersion is: D l = f D A where f is the focal length of the focusing (exit) optic -1 1 λcosβ RD = D = f DA = f sinα + sinβ inverse linear dispersion ( ) ( )

43 Dispersion: Grating vs. Prism Grating : dλ/dy is constan t Prism dλ/dy varies with wavelength.

44 Spectral Bandpass and Slit Function The spectral bandpass (nm) is the half-width of the range of wavelengths passing through the exit slit. The geometric spectral bandpass 1 Sg = D W where D -1 is the inverse linear dispersion W is slit width In a monochromator, an image of entrance slit is focused at the exit slit: - when input is polychromatic, a monochromated version of the image appears at the exit slit - when input is monochromatic image, rotating the grating angle θ will sweep monochromatic image across the exit slit

45 Slit Function

46 Slit Function The total intensity t(λ) measured at the exit slit as image is translated is called the slit function - for equal entrance and exit slits, shape is triangular - for unequal entrance and exit slits, shape is trapezoidal with a base of s and half-width of S g Mathematically, the slit function is where t t ( λ) 1 λ λ = sg 0 when ( λ) = 0 everywhere else λ 0 s g λ λ 0 + s λ is the incident (monochromatic) wavelength at entrance slit λ 0 is the wavelength setting of the monochromator (the wavelength directed to the center of the exit slit) g

47 Resolution Resolution quantifies how well separated two features are at the exit slit Resolution is related to - linear dispersion (D l ) (or angular dispersion (D A ), and physical dimensions of the monochromator, through the focal length f) and - the slit width W If the width of a single peak base is S (= 2s g ), then two features will just be completely separated when the wavelength difference between them is S 1 Δλ s = S = 2sg = 2D W = 2R D W Alternatively, we may adjust slit width to obtain resolution of two features separated by Δλ s Δλ W = s 2R D

48 Effect of Slit Width on Spectra 2.0 nm Bandwidth Δλ s 0.5 nm Bandwidth Absorbance Absorbance

49 Radiation Detectors

50 Detectors /Radiation Transducers Transducer: Devices to convert radiant energy (electromagnetic radiation) into an electrical signal. Ideal properties 1. High Sensitivity 2. High signal to noise ratio (S/N) 3. Constant response over a wide wavelength range 4. Fast response time 5. Zero output in the absence of radiation 6. Electrical signal, S, should be directly proportional to incident radiant power P S = kp

51 Detectors /Radiation Transducers Two general types of radiation transducer 1) Photon transducers Used in visible and UV spectroscopy - respond to incident photon rate - highly variable spectral response (determined by photosensitive material) - respond quickly (microseconds or faster) - single or multichannel (1-D or 2-D) 2) Thermal transducers Used for IR spectroscopy - respond to incident energy rate - relatively flat spectral response curves (determined by window and coating) - generally slow (milliseconds or slower) - usually single channel

52 Responsivity R(λ) ) and Sensitivity Q(λ): X () ( λ) dx R λ = () ( λ) Q λ = Φ() λ dφ() λ where X(λ) is output signal (voltage, current, charge) Φ(λ) is incident flux (W) Plot of R(l) or Q(l) versus l is called the spectral response

53 Photo Detector Characteristics Photon detectors are based on - photoconductive materials (MCT transducer) - photovoltaic cells (Si, Se photocell) - photoemissive materials (PMT's, phototubes) - semiconductor pn junctions (photodiodes)

54 Transducers Two common types: 1. Photoemissive: - Based on photoelectric effect: photon electron electrons released only if hn > E min ; number of electrons number of photons 2. Photoconductive -photons striking device cause an increase in electrical conductivity -e.g., photodiodes, semiconductors Two classes of detector to consider: 1. Single-Channel - monitor intensity of a single resolution element at a time. 2. Multi-Channel - monitors intensities of many resolution elements at a time.

55 Photoconductive cell: Photoconductive Cell - semiconductor material (CdS, PbS, PbSe, InSb, InAs, HgCdTe, or PbSnTe) behaves like resistor - in series with constant voltage source and load resistor - voltage across load resistor used to measure the resistance of the semiconductor - incident radiation causes band-gap excitation and lowers the resistance of the semiconductor - most sensitive in near IR (PbS) - sometimes cooling is necessary to reduce thermal band-gap excitation

56 Photovoltaic Cell : Photovoltaic Cell - thin layer of crystalline semiconductor (Se, Si, Cu 2 O, HgCdTe)sandwiched between two different metal electrodes. - no bias but irradiation causes formation of electron hole-pair formation. - electron migrates one way, holes migrate in opposite direction - if resistance of external circuitry is small, microamps produced - high sensitivity in near IR to UV ( V W -1 ) - eg Fe-Se-Ag nm R(λ) peaking near 550 nm.

57 two electrodes enclosed in glass or silica envelope - bias ( V) is applied between two electrodes - cathode is a photoemissive material (Cs 3 Sb, NaO, AgOCs) -emits photoelectrons - current collected by anode Phototube - photoemission only if hν > surface workfunction (1-5eV) - High sensitivity ( A W -1 ) - Dark currents (typically A) caused by - thermionic emission - field ionization (high bias) - ohmic resistance

58 Photomultipler Tubes (PMTs( PMTs) - similar to phototubes - photoemissive cathode and anode - multiple secondary electron emissive dynodes (MgO, GaP) - each dynode is biased ~100 V more positive than previous to accelerate electrons from dynode to dynode - gain per dynode, g, is typically total gain m = g n is charge pulse at anode is few ns wide

59 PMT Spectral Response Curves

60 - R(λ) is a function of photocathode material - very high sensitivity ( A W -1 ) - Alternatively, the rate of charge pulses can be counted, a technique called photon counting. - Dark currents in PMT's result from similar processes to phototube - thermionic emission associated with the photocathode can be significant (multiplied by dynodes) A - cooling the PMT (0 to -60 C) helps.

61 Photodiode Detectors contains a reverse-bias semiconductor pn junction - p-type semiconductor has excess holes (eg B-doped Si) - n-type semiconductor has excess electrons (eg P- doped Si) - under reverse bias, depletion layer formed (resistivity of depletion layer is very high) - under irradiation, electron-hole pairs created that move under bias (holes p-type, electrons n- type) - momentary current is produced - ns or sub-ns - spectral response of a typical photodiode depends on band-gap of semiconductors used (typically near IR into near UV) - R(λ) less than PMT (no internal gain) but Q(λ) constant over 6-7 orders of magnitude - poor sensitivity ( A W -1 ) - can be made very small, ideal for use in multichannel devices

62 Multi-Channel Detectors Monitors intensities of many resolution elements simultaneously -similar to FT-interferometry (multiplexed measurement), but in the frequency domain. Examples: - photographic plates - photodiode arrays (PDA) -Charge Integrating Devices (CID) and -Charge Coupled Detectors (CCD) Most limited to UV/Vis

63 Photo-Diode Arrays - based on pn-photodiodes constructed by semiconductor chip techniques - regions of p-type Si deposited onto n-type Si crystal - distance between elements is typically 25 or 50 um (up to 4096 elements per array) - usually operated in the depletion layer is formed around each p-type islands - upon irradiation (integration time) bias is turned off and electrons and holes are created in depletion regions - holes migrate to the p-type islands and accumulate (max ~10 6 )

64 Photo-Diode Arrays - during read-out period, each p-type region is interrogated - thermal excitation of electron-hole pairs creates difficulties with long integration times - array is often cooled - dynamic range 2-4 orders of magnitude - at detection limit thermal excitation dominates - sensitivity can be increased by coupling diode array with microchannel plate (MCP)

65 Charge Transfer Devices - photodiode arrays are inferior to PMT's in respect to sensitivity, dynamic range and signal-to-noise ratio - charge transfer devices approach the performance of PMT's - each pixel is metal oxide semiconductor - negative bias is applied to each electrode, a potential well collects photogenerated holes - more than 10 6 holes bleed onto adjacent pixels - charge accumulated during the integration time can be integrated in two ways: - charge-injection device (CID) - charge-coupled device (CCD)

66 Charge Injection Device (CID) (two electrode per pixel): (1) during integration, one electrode (B) more negative than theother (A) - all photogenerated holes are accumulated under B (2) voltage applied to A is removed and the surface charge measured on A. (3) potential on electrode B is switched to a positive potential, causing the holes to migrate to electrode A (4) charge under A is remeasured and the signal is the difference between the two measurements (5) positive voltage applied to electrode A to repel accumulated holes and return system to initial state

67 Charge Coupled Device (CCD) -three electrodes per pixel substrate is p-type Si so electrons (not holes) are accumulated - each pixel contains three electrodes - following the integration period, a three-phase voltage transfers electrons in a step wise manner along a row - readout process is destructive

68 Thermal Transducers Cannot use photon transducers in IR region Thermal Transducers Operation : Infrared radiation is absorbed by a blackbody and the resultant temperature rise is measured. Radiant power levels are small Watts Need to detect heat changes as small as 10-3 K need blackbody with low heat capacity and small size. In general thermal detectors are very noisy must chop signal and use Phase Sensitive Detection (PSD) Methods.

69 Thermal Transducers

70 Thermal Transducers: Thermocouple - Thermocouple: based on thermoelectric potential when two dissimilar metal wires e.g. Bi/Sb are in contact. - junction attached to blackened disc of known area but small heat capacity ( mm). - output is nv-mv range (limited sensitivity) - Q(λ) constant over modest temperature range ( W) - moderate responsivity R(λ) 5-25 V W -1 - junctions with different sensitivities are available - response time limited by capacitance of wires to ms - multiple junction thermocouples called thermopiles Thermopile: Can detect changes as small as 10-6 Sensitivity ~ 10μV / μw K

71 Thermal Transducers: Bolometer Thermistor bolometer: Measure change in resistance as a function of temperature. Either a resistance thermometer made from metals (Pt or Ni) or semiconductor material (thermistor). - blackened metal or semiconductor with narrow bandgap ( mev) - radiation excites electron-hole pairs which decrease resistance - decrease in resistance is compared with unirradiated bolometer - difference is amplified - Q(λ) constant W - high responsivity R(λ) 1000 V W -1 - response time 1-10 ms, Slow response no good for FTIR spectroscopy - Example: Ge bolometer operated at 1.5 K ideal for μm range.

72 Thermal Transducers: Pyroelectric Pyroelectric detector: Pyroelectric Infrared Detectors (PIR) convert the changes in incoming infrared light to electric signals. Pyroelectric materials are characterized by having spontaneous electric polarization, which is altered by temperature changes as infrared light illuminates the elements. - based on a piezoelectric material - eg Triglycine Sulphate, DTGS - non-centrosymmetric crystal has permanent dipole moment across unit cell - acts like a capacitor - when irradiated crystal expands slightly, capacitance decreases, current flows - high responsivity R(λ) up to 10 4 V W -1 - Q(λ) constant W - fast response time <10 ms, good for FTIR spectroscopy.

73 COBE The COBE dewar was a 660 liter liquid helium cryostat. It provided a stable 1.4 Kelvin environment for the two cold instruments, the Far Infrared Absolute Spectrophotometer (FIRAS) and the Diffuse Infrared Background Experiment (DIRBE). The first phase of the COBE science mission came to an end on Friday, September 21, 1990, after 306 days of cryogenic operations as the last of the superfluid helium contained within the dewar was consumed.

74 COBE Map of the sky The colors represent temperature variations with red indicating regions that are a hundredth of a percent warmer and blue indicating regions that are a hundredth of a percent cooler than the average temperature of 2.7 degrees above absolute zero.