Physics | Spectroscopy » P. F. Bernath - Infrared Emission Spectroscopy

Datasheet

Year, pagecount:2003, 48 page(s)

Language:English

Downloads:4

Uploaded:April 26, 2018

Size:950 KB

Institution:
-

Comments:

Attachment:-

Download in PDF:Please log in!



Comments

No comments yet. You can be the first!

Content extract

Source: http://www.doksinet 6 Infrared emission spectroscopy P. F Bernath Department of Chemistry, University of Waterloo, Waterloo, ON, Canada N2L 3G1 1 Introduction Infrared spectroscopy has been traditionally carried out mainly in absorption. The virtues of infrared emission spectroscopy have been largely overlooked, but there has been a recent surge of interest. The modern infrared Fourier transform spectrometer has made emission measurements much easier. This article is an attempt to provide a comprehensive view of the technique and builds on a shorter Chemical Society review.1 Previous reviews of infrared emission spectroscopy have concentrated on the basic principles and on applications in analytical chemistry.2^6 We shall cover some of this ground as well, but the main focus will be in applications in high resolution molecular spectroscopy. We shall not cover the infrared spectra of atoms although excellent infrared emission spectra have been recorded at Kitt Peak, Orsay

and Lund. This review has some overlap with reviews on the infrared spectra of transient molecules, including ions, free radicals and high temperature molecules,7^10 as well as the spectroscopic reviews of Barrow and Crozet.11 For the purposes of this review, ``infrared is arbitrarily de¢ned as 10^10 000 cmÿ1 (1^1000 mm), to include the far-infrared, mid-infrared and near-infrared region. The microwave and submillimeter regions are excluded, although laboratory microwave emission spectroscopy has become very popular with the proliferation of Flygare^Balle spectrometers.12 These instruments detect the coherent emission of microwave radiation. At slightly higher frequencies radio astronomers detect molecules in molecular clouds by microwave and millimeter wave emission spectroscopy. The ¢eld of astrochemistry13 is built almost entirely on centimeter wave and millimeter wave emission spectroscopy of species in various astronomical objects ranging from stellar envelopes to comets. We

shall focus our discussion mainly on emission from gases, although heated solids14 and liquids15 give useful spectra. The secret to recording useful spectra from condensed phases is to work with thin ¢lms or dispersed particles. For thick samples, multiple scattering of infrared photons results in a nearly featureless blackbody spectrum that depends only on the temperature of the emitter. DOI: 10.1039/b001200i Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 177 Source: http://www.doksinet This review will cover mainly high resolution spectroscopy. In this context, ``high resolution means that the rotational structure is at least partly resolved for a gas phase sample. The emission technique also works equally well for large molecules such as C60 and C70,16 and for species ranging from DNA bases17 to polycyclic aromatic hydrocarbons,18 in which the rotational structure is not resolved. For these molecules the solids are heated to about 200^300  C and emission from the vapour

provides excellent spectra even in the far-infrared region.18 The attraction of emission spectroscopy is the possibility of an improved signal-to-noise ratio compared to absorption spectroscopy. Ideally only photons emitted by the sample are detected (``zero background), free from the noise produced by the continuum lamp in an absorption experiment. This improvement in sensitivity is particularly useful for the spectroscopy of transient molecules because of their intrinsically low concentrations. This potential advantage of emission spectroscopy for the detection of ions and free radicals is well known is the visible and near-UV regions. For example, the violet emission from the CH radical (A 2 D!X 2P) is readily seen by eye in a £ame19 but the measurement of the A2 D X 2P absorption20 is much more dif¢cult. This emission advantage persists into the infrared region. The infrared region is unique because all molecules, with the exception of homonuclear diatomics, have at least one

allowed vibration^rotation transition. While it is true that there are infrared electronic transitions and a few light molecules have far-infrared rotational transitions, infrared spectroscopy is nearly synonymous with vibrational spectroscopy. Moreover, vibrational spectroscopy through the concept of group frequencies also provides chemical information in a way that rotational and electronic transitions do not. Even the weak electric quadrupole emission transitions of H2 have been seen by astronomers from molecular clouds experiencing shock waves.21 Infrared emission spectroscopy has the potential to be a sensitive, universal, molecule-speci¢c monitor of chemical composition. This review will be organized by sources, ranging from stars to microwave discharges, in which emission spectroscopy has been carried out. The coverage of the more recent high resolution spectroscopic work aims to be relatively complete. Older work and various applications in surface science, astronomy, chemical

dynamics or analytical chemistry are more illustrative than complete. But ¢rst a few basic principles and instrumental considerations will be discussed. 2 Basic principles Emission spectroscopy is based on a few basic equations and principles. The ¢rst is the interaction of monochromatic radiation with a sample (Fig. 1) (ignoring scattering and £uorescence). The beam can be re£ected, absorbed or transmitted, so a‡r‡tˆ1 1† in which a is the absorptance (absorption factor), r is the re£ectance (re£ection factor) and t is the transmittance (transmission factor) of a body.5,22 Note that 178 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet Fig. 1 Light striking an object will be re£ected, absorbed or transmitted (ignoring £uorescence and scattering). the difference between a re£ectance (``-ance) and a re£ectivity (``-ivity) is that the latter applies under some set of standard conditions (e.g, smooth surface, thick sample) while the former

applies to a speci¢c sample. Because we are generally discussing speci¢c objects, we will use the ``-ance terms Factors a, r and t are wavelength dependent numbers between 0 and 1 and there are three simple limits: a ˆ 1; r ˆ 1; t ˆ 1; rˆtˆ0 for a blackbody 2† aˆtˆ0 for a perfect mirror 3† aˆrˆ0 for a perfect window 4† Kirchhoffs law states that the emittance e of a sample is equal to the absorptance a (see above). The monochromatic emittance of a sample is de¢ned as: L ˆ eLBB 5† in which L is the radiance (or ``sterance, in units of watts per steradian per square meter of source per hertz of spectral bandwidth)22 of a sample, and LBB is the radiance of a blackbody. Thus LBB is related to the Planck function and e is a proportionality constant between 0 and 1 that converts it into the observed radiance, L. The expression for the radiance of a blackbody is: LBB ˆ 2hn3 c2 ehn=kT ÿ 1† 6† in units of W srÿ1 mÿ2 s. Because the emittance and absorptance

are equal, eˆa 7† all of the selection rules and optical properties of any material (gas, liquid or solid) that are generally de¢ned in terms of absorption transfer directly to emission. All of the above equations depend on both frequency (or wavenumber) and temAnnu. Rep Prog Chem, Sect C, 2000, 96, 177^224 179 Source: http://www.doksinet perature so Kirchhoffs law can be written as L n~ ; T † ˆ a n~ ; T †LBB n~ ; T † 8† in which ¬n is the wavenumber in customary non-SI units of cmÿ1. Kirchhoffs law has been tested experimentally using a hot CO2 gas sample.23 The absorptance was measured from the transmittance [e.g, eqn (1) with r ˆ 0], aˆ1ÿt 9† and the emittance measured by comparison with a blackbody source at the same temperature, eˆ L LBB 10† Direct comparison for different temperatures and concentrations of CO2 in N2 showed that a ˆ e. The direct application of Kirchhoffs law allows the determination of sample temperature24,25 as illustrated in Fig. 2

This ¢gure also illustrates the difference between gas phase emission and absorption measurements. In the upper panel, the absorption of a mixture of acetylene, butane and carbon dioxide at 555 K is displayed. In the middle panel, the measured radiance of the sample is plotted At the bottom, the ``normalized radiance, de¢ned as the radiance divided by absorptance, is plotted. By Kirchhoffs law this ratio is the blackbody radiance, ie, L L ˆ ˆ LBB T † a 1 ÿ t† 11† The radiance of a blackbody at a certain frequency is a function of only the temperature, which can be adjusted until the calculated Planck function matches the observed ``normalized radiance. The shaded regions mark regions that contain no molecular emission (i.e, no information) and are ignored The message of this ¢gure is clear: a thermal emission spectrum is nothing more than the ``inverted absorption spectrum modulated by the Planck function. The quantitative interpretation of sample emission is, in principle,

as simple as that of absorption. In practice, however, samples rarely have uniform temperatures and ``self-absorption is a problem. Fortunately sample emission increases strongly with temperature. For example, the total radiance (integrated over all frequencies) of a blackbody is proportional to the fourth power of temperature: Z1 Ltotal ˆ LBB dn ˆ sT 4 =p; 12† 0 in which s is the Stefan^Boltzmann constant. (The factor of p appears because the emission of the blackbody, eqn. (6), is per steradian and the normal version of the equation is simply the total emission from a hole.) This means that the highest temperature object in the ¢eld of view of the spectrometer dominates the appearance of the spectrum. The general equation for the observed radiance at a given frequency 180 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet Fig. 2 Spectra for a mixture of acetylene, carbon dioxide and butane in helium at 555 K: (a) absorption in %; (b) radiance; (c)

normalized radiance, with a blackbody ¢t for LBB at 565 K, ignoring the shaded regions.25 Reproduced from ref 25, with permission (Fig. 3) for the typical case of a gas sample in front of a wall26 is Z L 0† ˆ L z0 †t z0 † ÿ 0 t z0 † LBB Tz † dt 13† in which L(z0) is the radiance of an object at z0 away from the observer located at 0. Note that Kirchhoffs law gives de ˆ ÿdt. The transmittance t(z0) attenuates this Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 181 Source: http://www.doksinet Fig. 3 Radiance, L, seen by an observer at 0 looking through a gas sample to a back wall [radiance L(z0)] at z0. radiance from z0 and the integral accounts for the radiance of the intervening radiating elements. The transmittance is given by Beers law,  Zz  k z0 † dz0 t z† ˆ exp ÿ 0 14† in which k is the absorption coef¢cient. If the sample is uniform and the temperature is constant then eqn. (14) becomes, t z† ˆ eÿkz 15† and the general equation, eqn. (13),

reduces to: L 0† ˆ L z0 † eÿkz0 ‡ LBB 1 ÿ eÿkz0 † ˆ L z0 †t z0 † ‡ LBB ‰1 ÿ t z0 †Š 16† This equation has two simple limits: the optically thin case when t & 1 and L & L(z0), and the optically thick case when t & 0 and L ˆ LBB. In the optically thin case the radiance is that of the wall, while in the optically thick case only the front of the gas sample is seen. The discussion presented so far is on the macroscopic level. The connection to the microscopic world of atoms and molecules is through the Einstein equations; in particular, the Einstein A coef¢cient for emission from level 2 to 1 is: A21 ˆ 16p3 n3 jm21 j2 3he0 c3 17† in which m21 is the transition dipole moment, and e0 is the permittivity of free space. The Einstein A coef¢cient measures the rate of photon emission from the excited level of an atom or molecule, dN2 ˆ ÿA21 N2 dt 18† and has units of sÿ1. The above expression has been integrated over the lineshape of 182 Annu. Rep

Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet the transition and the more detailed expression27 is A21 †n ˆ 16p3 n3 jm21 j2 g n ÿ n0 † 3he0 c3 19† in which g(n ÿ n0) is the normalized lineshape function. The strong cubic frequency dependence of the emission rate means that emission work in the infrared and far-infrared is more dif¢cult than in the visible and ultraviolet. Moreover, transition dipole moments for vibration^rotation transitions tend to be smaller than typical values for allowed electronic transitions or typical values of permanent dipole moments for pure rotational transitions. 3 Methodology The basic requirement for emission spectroscopy is a source of radiation and a detector. For some experiments, a simple infrared ¢lter between the source and detector can provide useful results. For example, Chang and Klemperer28 excited the (HF)2 molecule in a jet expansion with a tuneable near-infrared laser and monitored the HF infrared emission.

This type of action spectroscopy obtained by scanning a laser and detecting the total emission is common at shorter wavelengths but also works in the infrared. The simple but effective technique of non-dispersive infrared absorption spectroscopy (NDIR) has an emission analog called £ame infrared emission (FIRE) spectrometry.29 FIRE is a simple ¢lter^detector combination that can be used as a detector in a gas chromatograph30 In this analytical application the hydrocarbon analyte from a gas chromatograph is burnt in a hydrogen^oxygen £ame and hot CO2 emission is detected.29,30 The basic requirement for any emission experiment is that the source and detector have different temperatures. The detector is also an emitter of radiation, and if the source and detector have the same temperature then there is no net £ux of radiation falling on the detector. Obviously, a room temperature detector such as triglycine sulfate (TGS) cannot be used to monitor a room temperature sample. But a

liquid-nitrogen-cooled InSb detector can certainly see a room temperature sample. Generally emission from a detector at 4 K (liquid He) or even 77 K (liquid N2) can be ignored. A slightly more sophisticated system for emission spectroscopy is based on the use of a circular or linear variable ¢lter between the source and detector. By rotating the circular ¢lter the peak transmission of the ¢lter can be changed. Such ¢lters can be purchased from, for example, Optical Coating Laboratory, Inc. (OCLI) of Santa Rosa, CA. The use of a circular variable ¢lter results in a simple, compact spectrometer but with a low resolving power, Rˆ n~ l ˆ Dl D~n Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 20† 183 Source: http://www.doksinet of typically less than 100. Circular variable ¢lters are still used in astronomy and for simple analytical applications in which their high optical throughput and relatively low cost are an advantage. Early high resolution infrared emission

measurements were made using classical grating spectrographs or spectrometers to disperse the emission. For example, the Ballik^Ramsay bands of the C2 molecule were discovered31 in 1963 using an infrared spectrometer constructed by Douglas and Sharma32 at the National Research Council of Canada. The C2 molecule was made by the evaporation of graphite from the walls of a carbon tube furnace (King furnace) operated at about 2900  C. The emission was detected using a dry-ice-cooled PbS detector The vibronic bands of the A 3Sÿg^X3Pu electronic transition were seen at 3800^7100 cmÿ1. The performance of any spectrometer can be improved in the thermal infrared region (n d 3000 cmÿ1) by cooling the entire instrument. Heroic early experiments were carried out by McDonald and co-workers33 to study the nascent products of reactions of F atoms with hydrocarbons such as ethylene. The chemiluminescence emitted by these free radical reactions at low pressures was very weak. A modern version of a

cooled spectrometer was developed originally by Pimentel and then used by Saykally and co-workers34 to study the weak emission from laser-excited polycyclic aromatic hydrocarbons (PAHs) in a supersonic free jet expansion. In this case the entire spectrometer was cooled to 4 K by liquid helium in order to take advantage of a new ultrasensitive blocked impurity band detector. This detector from Rockwell is based on Si:As and has an internal avalanche process that results in gain. In other words, it is a solid state photomultiplier that operates in the infrared. An even more sophisticated instrument can be made using modern infrared array detectors. The use of a large format array (typically InSb or HgCdTe) with a spectrograph is attractive because of the multiplex advantage. The most sensitive infrared instruments are the cryogenic echelle spectrographs that are in use or under construction at all major observatories. One example is Phoenix (Fig 4) at Kitt Peak National Observatory in

Tucson, Arizona.35 Phoenix is cooled to 50 K to eliminate the thermal emission from the spectrograph and uses an echelle grating in high order to obtain high resolution in a compact instrument. Order sorting is carried out with cooled infrared ¢lters Phoenix currently has a resolving power of about 70 000 or a resolution of 003 cmÿ1 at 2000 cmÿ1. The use of a 1024  1024 InSb array allows coverage of the 1800^10 000 cmÿ1 region. Phoenix is calculated to have a sensitivity advantage of nearly 100 over a conventional Fourier transform spectrometer. This high sensitivity originates from several factors including a very restricted spectral bandpass, cryogenic cooling and improved detector performance. The most important factor is that infrared array detectors are fundamentally different from conventional single element detectors. Because infrared arrays are integrating detectors, very low light levels can be handled. If the main noise source is read-out noise then the signal-to-noise

ratio grows linearly with time. This is in contrast to a conventional single element detector 184 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet Fig. 4 Phoenix cryogenic echelle spectrograph of Hinkle et al35 at Kitt Peak National Observatory, Tucson, AZ, USA. where the signal-to-noise ratio grows with the square root of time. The performance of Phoenix was tested by recording emission spectra of the NH free radical using a microwave discharge source.36 Most infrared emission measurements are made with Fourier transform spectrometers (FTSs). A typical experimental arrangement37 is illustrated in Fig 5. Light is admitted through the ``emission port, which is located near the internal sources used for absorption spectroscopy. Fourier transform spectroscopy has become ``conventional so no detailed descriptions are needed. Infrared emission spectroscopy requires simply that the internal glowbar source be replaced by the source of interest. The Fourier

transform interferometer is best viewed as a modulator that shifts infrared frequencies into audio frequencies at Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 185 Source: http://www.doksinet Fig. 5 Typical long wavelength infrared chemiluminescence emission experiment using a Fourier transform spectrometer equipped with a copper-doped Ge detector.37 Reproduced from ref. 37, with permission the detector. For example, the 0^10 000 cmÿ1 spectral region is mapped into the 0^10 000 Hz frequency range on the detector if the moving mirror changes the optical path difference at 1 cm sÿ1. Unusual infrared emission measurements are sometimes made by using a ``second modulation of the source. The most common of the double modulation experiments is time-resolved Fourier transform spectroscopy (TRFTS).38,39 A conceptually simple approach to TRFTS is the step-scan method using a periodic infrared source. If the ``moving mirror of an FTS is stopped then the infrared signal can be recorded as

a function of time. The moving mirror is then sent to the next sampling position and another infrared emission decay is recorded. By accumulating a set of emission decays, each recorded at the usual sampling position of the interferogram, a set of time-resolved spectra are obtained. 186 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet Fig. 6 The time-resolved Fourier transform emission spectrum of the reaction of O with HCl The highly excited OH molecules are quenched by collisions with HCl and this causes the OH vibration^rotation line intensities to decrease with time.38 Continuously scanning FTSs can also record time-resolved infrared spectra. In this case, the fringes of the internal He^Ne laser are used (with appropriate time delays) to trigger the infrared emission source and the sampling of the emission signal. In this way, an interferogram at a speci¢c time delay (or series of delays) from the excitation pulse is recorded. Time-resolved infrared

emission spectra are very useful in the study of reaction dynamics and collisional relaxation. In Fig 6 the time-resolved OH emission from the O ‡ HCl reaction is displayed.38 As time progresses the highly excited OH molecules produced by the chemical reaction are relaxed by collisions The development of schemes for various time-resolved FTS work continues with, for example, the recent development of the ``event-locked method by Weidner and Peale.40 Double modulation techniques are attractive in the emission spectroscopy of transient molecules because they discriminate against the more abundant precursor molecules. The selectivity of velocity,41,42 concentration43 and Zeeman modulation44,45 have all been demonstrated with FTSs. Unfortunately they generally do not also offer an increase in sensitivity and work most easily with step-scan instruments so they have not been widely adopted for high resolution work. An important consideration in emission work is the ``contrast between the

sample and background. This is particularly important in remote sensing of gases and in the far-infrared region. If the ¢eld-of-view of the spectrometer includes a gaseous sample against a background at a certain temperature, then the lines of the sample will appear in absorption or emission depending on the sample temperature. If the sample is warmer than the background then the lines will appear in emission, while if it is cooler then the lines will appear in absorption. If the sample and background have the same temperature then the sample lines will disappear, i.e, there is no contrast This spectral line reversal technique has been used to determine £ame temperature.19 A £ame is viewed against the continuum of a lamp and a metal atom such as Na is added to give a bright emission line. As the lamp ¢lament is increased in temperature from below the £ame temperature, the atomic emission line will disappear when the £ame and ¢lament temperatures are equal. Filament temperatures

are then easily determined with an optical pyrometer. Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 187 Source: http://www.doksinet The problem of contrast is particularly severe in the far-infrared region and this causes problems for both absorption and emission measurements. In the far-infrared sources operate usually in the Rayleigh^Jeans limit with hn << kT, so the blackbody radiance reduces to46 LBB ˆ 2n2 kT 2kT ˆ 2 : c2 l 21† Because the radiance is only linearly proportional to temperature, the contrast between a heated sample and the surroundings is much less than in the visible. Indeed, usually the spectrometer itself contributes a major portion of the ``signal in a far-infrared emission experiment. 4 Chemiluminescence To a chemist and spectroscopist, the light emitted from a chemical reaction has an undeniable fascination. Unfortunately chemiluminescence tends to be relatively weak and not very common. Chemiluminescence is closely related to the

spectroscopy of £ames and to the excitation of molecules by energy transfer from metastable species. These sources of infrared emission will be discussed separately below. A Nobel prize has been awarded to J. Polanyi (shared with Y T Lee and D Herschbach) for the study of reaction dynamics by infrared chemiluminescence.47 This work started in 1958 with the observation of HCl emission from the H ‡ Cl2 ! HCl ‡ Cl 22† reaction by Cashion and Polanyi.48 A commercial infrared spectrometer with an NaCl prism and a thermocouple detector was used. A similar chemiluminescent reaction (H2 ‡ Cl2 £ame) was later used by Clayton et al.49 at Penn State with a high resolution spectrometer. Infrared chemiluminescence in the style of Polanyi is still carried in a number of groups such as those of Setser and Leone. Butkovskaya and Setser50 studied the dynamics of the OH ‡ HBr ! H2 O ‡ Br 23† reaction, for example, while Klaassen et al.51 recorded the ¢rst high resolution spectrum of HOI

from the C2 H5 I ‡ O ! HOI ‡ C2 H4 24† chemical reaction. Much of the chemical dynamics work is now carried out by time-resolved Fourier transform spectroscopy. Recent examples include the work of the Sloan laboratory at 1 ms time resolution for reactions of H atoms with £uorochlorocarbons.52 With a step-scan FTS system even 10 ns time resolution is possible.53^55 188 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet Fig. 7 Chemiluminescence reactor used in the F ‡ O3 infrared emission experiment to make FO.37 The F atoms are created in a microwave discharge of F2 in He Reproduced from ref 37, with permission. Chemiluminescence has also proved to be very useful in studies devoted to new spectroscopy (rather than dynamics) of molecules. At long wavelengths, a remarkable high resolution spectrum of FO was recorded by Hammer et al37 from the energetic F ‡ O3 ! FO ‡ O2 25† reaction. The F atoms were made in a microwave discharge of F2 (Fig 7) FO

emission up to the 8!7 vibration^rotation band could be assigned near 1000 cmÿ1 (Fig 8). The analogous OH (or OD) reaction56^58 H ‡ O3 ! HO ‡ O2 26† gives an exceptionally ¢ne infrared emission spectrum with OH populated up to u ˆ 9. Interestingly this same reaction (26) is responsible for atmospheric nightglow (see below) and astronomers have detected emission from u ˆ 10.58 Infrared electronic emissions are also possible and the work of E. Fink and co-workers is particularly noteworthy. Most ``normal stable main group molecules do not have low-lying electronic states, with the exception of O2 and NO: O2 has a p2 con¢guration that leads to a X 3Sÿg ground state with a 1Dg and b 1S‡g states at 7882 and 13121 cmÿ1, respectively; NO has a regular X 2P ground state with two spin components, 2P1/2 and 2P3/2, split by 123 cmÿ1. Closed-shell main group molecules Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 189 Source: http://www.doksinet Fig. 8 Chemiluminescent

vibration^rotation emission from FO created in the F ‡ O3 reaction.37 The marks at the top indicate the vibrational band origins Reproduced from ref 37, with permission. generally have excited electronic states that give rise to UV spectra. By virtue of their unpaired electrons, free radicals often have low-lying electronic states. Most free radicals are very reactive molecules but there are a few ``stable free radicals such as O2 and NO. Fink and co-workers have recorded the infrared electronic spectra of a number of forbidden transitions of free radicals with the same p1 or p2 con¢gurations as NO or O2. The NH, PH, AsH, SbH and BiH family have p2 con¢gurations and Beutel et al. have measured the a 1D^X 3Sÿ infrared transitions of PH,59 AsH60 and SbH.61 These molecules were made by the reaction of H atoms with heated elemental solids. [BiH was also made in a similar fashion but emission was excited by energy transfer from O2(1D), see below.] Although the a 1D^X 3Sÿ transition is

forbidden by normal electric dipole selection rules, it is allowed by the magnetic dipole transition moment. Excellent high resolution spectra were obtained with a high purity Ge detector. The TeF, TeCl, TeBr and TeI molecules are isovalent with OH and have inverted X 2 P ground states from a p3 con¢guration. The spin^orbit coupling constants for TeF and TeCl, however, are about ÿ4000 cmÿ1, as compared to ÿ139 cmÿ1 for OH. Ziebarth et al.62 measured the magnetic dipole emission X 2P1/2 ! X 2P3/2 between the two spin components for TeF and TeCl at high resolution. The excited TeF and TeCl molecules were made by the reaction of a TeH/TeH2 mixture with F2 or Cl2. Low resolution infrared spectra for TeF, TeCl, TeBr and TeI are also known.63 Another interesting example from the Fink group is the infrared electronic transitions of BiP, BiAs and BiSb.64 These diatomics are isovalent with N2 and have X 1S‡ ground states. Unlike N2, however, the 3S‡ and 5S‡ states that also arise from

190 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet Fig. 9 Chemiluminescence from SrO from the Sr ‡ N2O reaction The bands are mainly due to the A0 1P ! X 1S‡ electronic transition and some isolated Sr atomic lines can also be seen. the lowest energy 4S ‡ 4S atomic asymptote are low-lying. The a 3S‡1 ! X 1S‡ electronic transitions of BiP, BiAs and BiSb were detected in the 7000^10 000 cmÿ1 region through chemiluminescence from the reaction of P, As and Sb atoms with Bix vapour. Very recently we have adapted the Broida oven £ow reactor to study the infrared chemiluminescence of the classic metal plus oxidizer reactions.65 In particular, the reactions Ca ‡ N2 O ! CaO ‡ N2 27† Sr ‡ N2 O ! SrO ‡ N2 28† yield excellent near-infrared electronic spectra. The A 1S‡ ! X 1S‡ and A0 1P ! X S‡ transitions of SrO are displayed in (Fig. 9) 1 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 191 Source: http://www.doksinet 5 Excitation by energy

transfer from metastables Excitation can be delivered to a molecule by a chemical reaction or by energy transfer from another excited molecule. Both of these processes occur in £ames so that the distinction that we have made between the three emission sources ^ chemiluminescence, excitation by energy transfer and £ames ^ is somewhat arti¢cial. Popular metastables include the rare gases, O2(1D), NF(1D) and active nitrogen. Of the rare gases, He is particularly popular because the metastable 23S state has 19.8 eV of available energy Active nitrogen is made by passing N2 gas through discharge and is a complex mixture of N and N2 in both ground and excited states O2(1D) and NF(1D) are unique in that both species can be made in relatively high concentrations by purely chemical means. This is attractive for the pumping of chemical lasers such as the COIL (chemical oxygen^iodine laser) system. 66 O2(1D) atoms are produced by the reaction of Cl2 gas with basic hydrogen peroxide, Cl2 g† ‡

H2 O2 l† ‡ 2KOH l† ! O2 1 D† g† ‡ 2KCl l† ‡ 2H2 O 29† Energy transfer to I2 causes dissociation and lasing on the I 2P1/2 ! 2P3/2 transition. NF(1D) can be formed by the thermolysis of the explosive gas £uorine azide67 at 1000 K, FN3 ! NF 1 D† ‡ N2 30† The chemical production of O2 and NF metastables is clearly not for the faint hearted! For laboratory purposes an electrical or microwave discharge is generally used to make metastable atoms and molecules. The COIL laser, in fact, was associated with a mysterious red and infrared emission that turned out to be CuCl2. Traces of chlorine reacted with heated copper tubing to produce CuCl2, which was excited by energy transfer from the O2(1D) molecule.68,69 Ultimately this spectroscopic mystery led to the high resolution analysis of the linear CuCl2 molecule70 Both visible and infrared electronic transitions of CuCl2 were analyzed using a variety of sources. The group of Fink has made extensive use of the O2(1D) metastable

to record spectra of an amazing number of main group molecules. The main types of main group free radicals that give rise to infrared electronic transitions have already been mentioned in the section on chemiluminescence. For diatomic molecules they generally have p1(2Pr), p2(3Sÿ, 1D, 1S‡) or p3(2Pi) con¢gurations as exempli¢ed by CH, NH and OH free radicals, respectively. In the CH family, the forbidden X 2P3/2 ! X 2P1/2 transitions71,72 were detected for PbF, PbCl, PbBr and PbI. These molecules were all made by the reaction of Pb vapour with halogens and then excited by O2(1D) metastables. The high quality of the spectra is illustrated with PbF (Fig 10). For the NH family, the a 1D ! X 3Sÿ transitions were analyzed for AsI,73 SbF,74 SbCl,74 SbBr,74 SbI,74 BiCl,75 BiBr75 and BiI.75 In addition the X2 3Sÿ1 ! X1 3Sÿ0‡ ¢ne structure transition of BiH76 (as well as for BiF,77 BiCl,78 BiBr78 and BiI78) was also measured. The X 2P1/2 ! X 2P3/2 transition79 of TeH and TeD was seen

near 4000 cmÿ1. 192 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet Fig. 10 Fourier transform emission of the X 2P3/2 ! X 2P1/2 transition of PbF72 Reproduced from ref. 72, with permission The O2 molecule has a similar energy level pattern to the NH family and has been studied extensively by magnetic dipole and collision-induced80 emission. The O2 a 1Dg^X 3Sÿg transition occurs near 8000 cmÿ1, so the 1D metastable carries about 1 eV of energy. The O2 b 1S‡g^a 1Dg electric quadrupole transition81 is seen near 5000 cmÿ1 (Noxon band82). The a 1D^X 3Sÿ or b 1S‡^ X 3Sÿ infrared electronic transitions were detected for the isovalent SO,83 S2,84,85 SeO,86,87 SeS,88 Se2,89 TeO,90 TeS,90 TeSe91,92 and Te293 molecules, all excited by O2(1D). The BiO molecule has a similar electronic structure to the isovalent NO. Extensive infrared electronic transitions of BiO have been studied94 including the X 2P3/2 ! X 2 P1/2 transition95,96 near 7000 cmÿ1. This

transition displays a remarkable hyper¢ne structure (Fig. 11) In the N2 family, the bands analogous to the Vergard^Kaplan system of N2 are shifted from the near-UV region into the near-infrared for the heavier members. The a 3S‡u^X 1S‡g transitions of Sb297 and Bi298 are found near 9000 and 5000 cmÿ1, respectively. Both ground and excited states correlate to ground state N(4S) atoms, and there is a large change in bond length and vibrational frequency. Analogous transitions have been seen for BiN, 99 BiP100 and BiAs100 All of the above emission spectra were induced by energy transfer from O2 metastables. The best known main group polyatomic that has an infrared electronic spectrum is ~ 2A00 electronic transition HO2. The bent hydroperoxyl free radical has the Aì 2A0 ^X ÿ1 near 7000 cm . Early high resolution emission measurements were made by Tuckett et al.101 using a SISAM spectrometer The de¢nitive analysis, however, is the recent work of Fink and Ramsay.102 Analogous

transitions of HS2,103 HSe2 and HTe279 have also been seen by Fink and co-workers. Another new free radical, BiOH, with an infrared electronic spectrum near 6000 cmÿ1 was discovered by Fink et al.104 BiOH forms when H2O is added to the reaction of Bi vapour and metastable O2. BiOH is similar to BiF (rather than Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 193 Source: http://www.doksinet Fig. 11 High resolution emission spectrum of the X2 2P3/2 ! X1 2P1/2 transition of BiO The lower panel displays the impressive Bi hyper¢ne structure in the rotational lines.94 Reproduced with permission, from ref. 94 the isovalent HNO molecule) in electronic structure. BiOH, however, is bent so the X 3 ÿ ~ 1 1 A0 , X ~ 2 1A00 and X ~ 3 1A0 electronic states. The observed S state of BiF splits into X ~2 infrared electronic transitions thus correspond to the ``¢ne structure transitions X ~ 1 and X ~3 ! X ~ 1. !X 194 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet

Fig. 12 Cossart source110 for the Penning excitation of N2 (in this case) with metastable rare gases. Reproduced from ref 110b, with permission Active nitrogen also gives interesting new spectra. For example, when Vilesov et al.105 added Xe to a nitrogen discharge, a new band appeared very near to the forbidden 2Po ! 2Do emission of the atom near 9600 cmÿ1 (the 2Po, 2Do and 4 o S states arise from the 2p3 con¢guration of N). They assigned the structure to a bound^bound transition of the Xe:N excimer molecule.105 Active nitrogen can also be used to excite the vibration^rotation emission of stable molecules106 such as CO2. He metastables are commonly used to create ions by Penning ionization. For example, the A 2P ! X 2S‡ near-infrared electronic transition of CS‡ was measured by Horani and Vervloet.107 One of the main problems with using He metastables in a £owing afterglow is that their concentration is low and thus the emission of the product ions or molecules tends to be weak.

A solution to this dif¢culty was found by Vervloet, who adopted the Engelking supersonic corona discharge 108 (see below) as the source of He metastables rather than the customary DC discharge. The Engelking source is a proli¢c generator of metastable He atoms and provided a much brighter source of CS‡ than usual in a £owing afterglow. This same technique was applied by Huber and Vervloet109 to record the emission spectrum of the b 3P ! a 3S‡ transition of NO‡ near 6000 cmÿ1. Another powerful source of rare gas metastables is the novel apparatus originally built by Cossart110 (Fig. 12) This source is often called the ``Cossart source, but this name is not unambiguous because Cossart has created a number of novel Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 195 Source: http://www.doksinet Fig. 13 Schematic of the tip of the Engelking corona-excited supersonic jet expansion source The vacuum pump, typically a Roots blower, is grounded to complete the circuit.108 Reproduced

from ref. 108b, with permission and useful sources of molecules. For example, the Cossart source was one of the sources used by Dabrowski et al. to record infrared emission experiments of ArH,111^114 ArD, KrH115^117 and KrD, XeH118 and XeD. These rare gas hydrides are examples of what Herzberg has called ``Rydberg molecules as distinct from Rydberg states in molecules.119 Rydberg molecules have no stable ground states and are based on putting an electron in a Rydberg orbital of a bound ion core. For example, ArH has no chemically bound ground state but ArH‡ is a deeply bound ion. By placing an electron in a Rydberg orbital built on the ArH‡ ion core, a series of bound excited states are obtained. The excimer molecule ArF (of laser fame) is another example of a Rydberg molecule. Rydberg^Rydberg transitions of ArH are readily detected in the infrared region. Indeed, ArH is so easy to make even as an impurity that we generally use Ne gas for infrared emission work with a hollow cathode

lamp. The rare gas hydride molecules were made in the Cossart source by reacting H2 (or D2) with a £ow of metastable rare gases.110^118 6 Engelking corona-excited supersonic jet source Droege and Engelking108 developed a simple but remarkably effective source for the emission spectroscopy of jet-cooled free radicals (Fig. 13) The source is particularly effective for the production of small non-metal free radicals. The corona excitation occurs on the high pressure side of the nozzle and the resulting plasma is rapidly expanded into vacuum. The high voltage, low current corona discharge is an ef¢cient source of electronically excited free radicals, which are rotationally cooled by jet expansion. The main dif¢culties with the source are that large molecules are not always cooled and that the source is prone to various oscillatory electrical modes.108 196 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet Vervloet has adapted this source for infrared

emission spectroscopy of electronically excited N2 and NO. Vervloet and co-workers120,121 studied the a 1 Pg^a0 1Sÿu, w 1Du^a 1Pg and C00 5Pu^A0 5S‡g (Herman) infrared bands of N2. The b 4Sÿ ! a 4P system of NO was analyzed by Huber and Vervloet.122 High-l Rydberg transitions (e.g, 5g^4f) of NO were also reported123,124 A completely unrelated method of recording emission spectra of cold molecules is to irradiate liquid and gaseous He with a proton beam. Tokaryk et al125 saw near-infrared emission from the d 3S‡u^c 3S‡g electronic transition of He2. He2 is another example of a Rydberg molecule. 7 Magnetically con¢ned plasma (Penning) sources The application of a magnetic ¢eld to a DC electrical discharge forces the ions and electrons into circular orbits. The plasma is thus con¢ned to a smaller volume and a brighter source results. The use of a magnetic ¢eld in this way was ¢rst advocated by Penning and Penning-type pressure gauges based on this principle are used still.

The magnetic ¢eld allows the Penning source to operate over a very wide pressure range. Cossart has used a number of Penning-type sources 126^128 for emission spectroscopy including the application of a Penning-type discharge to a jet expansion source.126 Penning-type sources are good sources of molecular ions. Bernard et al.129 used a Penning source to study the A 2P!X 2S‡ transition of ‡ N 2 in the near-infrared. Highly excited infrared electronic transitions of CO (E 1 P!B 1S‡ and C 1S‡!B 1S‡) were also investigated in this way.130,131 8 Flames The infrared spectroscopy of £ames has a long history, starting in 1890 with Julius, as discussed by Gaydon.19 Hydrocarbon £ames provide hot infrared emission spectra of H2O, CO2, CN, C2, CO and OH. For the purpose of this review the term ``£ame refers mainly to hydrocarbon plus oxygen (or air) £ames. Infrared spectra of many other types of £ames such as acetylene plus nitrous oxide132 or C2F4 plus oxygen133 have been recorded

mainly to study the combustion process (rather than spectroscopy). Worden et al134 recorded the infrared emission spectrum of a forest ¢re from an airplane at high resolution. They detected emission features due to several molecules, including hot H 2O. Most of the molecules detected (NH3, CO, CH3OH, etc.), however, appeared in absorption against the continuum emitted by the ¢re. Early use of £ames for high resolution infrared spectroscopy includes the work on the A 2P!X 2S‡ transition of CN by Bacis et al.,135 and the Phillips (A 1Pu^X 1 ‡ 136 S g) and Ballik^Ramsay (b 3Sÿg^a 3Pu)137 systems of C2. C2 was produced in an oxyacetylene torch and CN in a similar nitrous oxide^acetylene £ame. Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 197 Source: http://www.doksinet The detection of vibration^rotation spectra is also possible. Indeed, the infrared emission spectra of hydrocarbon £ames are dominated by the main combustion products CO2 and H2O. A low pressure (20 mbar)

methane plus oxygen £ame was used to study CO2 and CO emission in the 1800^5000 cmÿ1 region.138,139 Numerous highly excited energy levels could be assigned. An oxyacetylene torch was also used to record the emission spectrum of hot water140,141 and the OH142 free radical from about 3000 to 10 000 cmÿ1. The corresponding OD vibration^rotation bands were made in a D2 ‡ O2 £ame.143 The Meinel system of OH covers nearly the entire near-infrared region with bands in the Du ˆ 1, 2 and 3 vibrational sequences. The major advantage of £ames is that they are a bright source of hot molecules with temperatures ranging up to 3000 K. However, they have a number of drawbacks including large pressure-broadened linewidths (typically *0.1 cmÿ1) if they are operated at atmospheric pressure. Because CO2 and H2O so dominate the emission spectra, recording data for other species can be dif¢cult More recent spectroscopic work with £ames (e.g, for H2O) has been at reduced total pressures to reduce

the pressure-broadened linewidths.144 Finally, £ames have been used to measure the rotational emission spectrum of OH145 in the far-infrared between 50 and 375 cmÿ1. The rotational emission spectrum of hot H2O has also been measured at somewhat higher wavenumbers.144 9 Pyrotechnic and propellant combustion The visible spectra of ¢reworks are well known but infrared emission spectra of various solid propellants, £ares and other pyrotechnic materials have also been recorded. The goal of this work is an understanding of the combustion process and the measurement of the temperature. Infrared emission spectra of the combustion of boron-containing,146 silicon-containing147 and magnesium-containing148 pyrotechnics have been recorded by FTS emission spectroscopy at low resolution. Various combustion products ranging from HCl to BO could be identi¢ed depending on the composition of the £are. 10 Shock tubes Another interesting source for infrared emission is the shock tube. The shock

compression of an inert gas seeded with 1^10% or so of a precursor molecule results in temperatures of 1000^4000 K, depending on the experimental conditions. The heating is of short duration (*5 ms) so a rapid-scan infrared spectrometer is required to make measurements. The collision-induced quadrupole emission from H2 was measured near 4000 cmÿ1 using a shock tube.149,150 Emission spectra of shock-heated small hydrocarbons were recorded by Stephens and Bauer.151 A small HgCdTe infrared array detector has been used with a spectrograph to measure 198 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet Fig. 14 Schematic of the positive column discharge cell used to measure the vibration^rotation emission of CD.154 Reproduced with permission, from ref 154 the emission from shock-heated NO.152 All of these measurements were at low resolution, but a modern time-resolved FTS or a spectrograph like Phoenix35 would give much improved spectra. 11 Positive column

of a DC discharge In a typical direct current (DC) discharge, most of the cell is ¢lled by the positive column. The positive column is infrequently used for infrared emission spectroscopy because the molecular glow is much less bright than from the plasma inside a hollow cathode. The positive column is characterized by a nearly constant electric ¢eld and a modest voltage drop, leading to a relatively extended diffuse glow.153 By moving the electrodes to the side of the cell (Fig. 14), only the positive column emission is sent into the spectrometer.154 In addition, by injecting the carrier gas near the electrodes and the precursor molecule directly into the positive column, a steady discharge can be obtained. Although a hollow cathode discharge is brighter than the positive column, the cathode surface is not very tolerant of hydrocarbon ``impurities. The larger discharge volume and the more stable operation in the presence of CH4, etc., make the positive column an attractive emission

source for some applications.154 Tokaryk and Civis155,156 built a cooled hollow cathode similar to those used by Oka and co-workers for the laser spectroscopy of ions. They were able to record clean emission spectra of the b~ 3Pg ! a¬ 3Pu transition of 12C3 and 13C3. A methane^helium discharge was used to create a remarkably extensive series of bands in the Du2 ˆ 0 sequence. A similar source provided ArH‡ in emission, although in this case a hollow cathode gives a better signal.157 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 199 Source: http://www.doksinet Fig. 15 Vibration^rotation emission of NO excited in the positive column of a DC discharge161 The 12^9 band head is visible near 4830 cmÿ1 Reproduced from ref 161, with permission. In Orsay, France, a DC discharge through pure NO gave the spectra of several infrared electronic transitions of NO (M 2S‡^E 2S‡, D 2S‡^A 2S‡, E 2S‡^D 2 ‡ S and E 2S‡^A 2S‡).158,159 In addition, the vibration^rotation spectra of NO (Fig

15) could be recorded up to u ˆ 22 in the Du ˆ 3 sequence.160,161 The highly excited vibration^rotation bands of CO and CO2 were also recorded from the positive column of a DC discharge.162^167 In this case the CO2 was diluted with He and/or N2 to mimic the conditions found in a CO2 laser plasma. Time-resolved FTS spectra of CO emission have also been carried out to study energy transfer.168 12 Hollow cathode discharge There are a large number of different types of hollow cathode discharges. For example, Herzberg et al.169 built a hollow cathode that is cooled by liquid nitrogen in order to study the H2 and H3 molecules (Fig. 16) In this source it is possible to view either the hollow cathode or the positive column.169,170 This was very useful because H3 was found only in the cathode but H2 appeared in both the cathode and positive column regions. Bacis171 has another design for a liquid-nitrogen-cooled cathode but more typically water cooling is used. An uncooled hollow cathode

can run very hot and Davis et al. exploited this to record new infrared spectra of the Phillips (A 1Pu^X 1S‡g)172 and Ballik^Ramsay (b 3Sÿg^a 3Pu)173 systems of C2. In this case, the C2 was made largely by evaporation of the red hot cathode rather than by sputtering from the surface. Ferguson et al174 200 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet Fig. 16 Liquid-nitrogen-cooled hollow cathode that was used by Herzberg et al169 to study an infrared electronic transition of the Rydberg molecule, H3. Notice that the positive column (anode glow) could also be observed for comparison purposes. Ions and Rydberg molecules are more abundant in the cathode region. used a water-cooled steel cathode to record new spectra of the A 2Pu^X 2S‡g (Meinel) system of N2‡. The Meinel system of N2‡ is commonly seen in the spectra of aurora (see below). A composite wall (SiC/Cu) hollow cathode171 was used to record the ¢rst spectra of SiC175,176 (d 1S‡^b 1P, A

3Pÿ^X 3P), SiC is isovalent with C2 but was produced by sputtering rather than evaporation. Infrared electronic transitions of Rydberg molecules such as He2,177 XeH178 and H3169 can also be seen in hollow cathodes. Rydberg transitions (up to 6 h!5 g) of H2 and D2 are also prominent179^182 and the ionization potential of D2 was determined by extrapolation180 to be 124 745.353 cmÿ1 Transition-metal-containing molecules also commonly possess infrared electronic transitions. The presence of an open d shell leads to a large number of low-lying electronic states that are conveniently studied by infrared emission. Metal monoxides are made by adding a trace of O2 to the carrier gas (Ne or Ar, commonly) and the metal is sputtered from the cathode surface, e.g, for HfO,183 NiO,184 PtO,185 CoO,186 CuO,187 and AgO.188 For metal mononitrides (ScN,189 YN,190 HfN,191 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 201 Source: http://www.doksinet Fig. 17 The 1^1 band of the A 1S‡^X 1S‡

electronic transition of YN190 RuN192 and OsN193) or monohydrides (ScH,194 YH,195 LaH196 and PtH197) a trace of N2 or H2 is effective, but monohalides are better made in other ways (furnaces and microwave discharges) because the hollow cathode discharge is usually unstable. All of the metal mononitrides listed above are new molecules and a typical spectrum is illustrated for YN190 in Fig. 17 Vibration^rotation emission spectra are more dif¢cult to measure than electronic transitions because they are generally weaker and are at lower wavenumbers. Indeed, no emission spectrum of a transient molecule has been detected using a hollow cathode lamp below the InSb cut-off of 1800 cmÿ1. The protonated rare gases, NeH‡,198 ArH‡,199,200 KrH‡,200 and XeH‡,201 however, give excellent spectra from a discharge of a rare gas plus a trace of H2. HeH‡ is too weak and is best detected by laser spectroscopy. The vibration^rotation emission spectrum for CuH202 was inadvertently detected during the

course of experiments on NeH‡ and remains the only metal hydride detected in this way. Finally, a special high current hollow cathode discharge was made to detect the H3‡ and D3‡ ions in emission.203^206 This work used the clever trick of pressure labelling to distinguish H3‡ emission from that of H2. Experimentally it was found that H3‡ lines were prominent at 50 Torr of total pressure but nearly absent at 8 Torr. The interfering H2 lines, on the other hand, tended to decrease in intensity as the pressure increased. H3‡ emission can be seen in the spectra of gas giant planets as discussed below. 13 Radio-frequency and microwave discharges Radio-frequency (RF) and microwave discharges are similar (but not identical) sources of excited molecules. RF discharges generally operate at 27 MHz and microwave discharges at 2450 MHz to avoid interference with communications 202 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet Fig. 18 The 3 m long

radio-frequency plasma reactor described by Chollet et al208 with White-type mirrors to collect more emission. Note that a simple loop of wire around a quartz tube, as in the work of Vervloet or in the ICP source, also makes a satisfactory RF emission cell. Reproduced from ref 208, with permission users. The main advantage of these sources is that they can operate with electrodes external to the discharge tube. Although they can therefore work with very ``dirty precursor molecules such as hydrocarbons or metal halides, the build-up of deposits inside the tube does change the power coupling into the discharge cell. By switching the microwave source on and off, time-resolved spectra can be recorded.207 The groups of Guelachvili and Vervloet are particularly fond of RF discharges. A schematic of the reactor of Chollet et al.208 is provided in Fig 18 and some spectra from this source in Fig. 19 An RF discharge was used to record the vibronic bands ~ 2S‡ transition209 of C2H and the c

3P^b 3S‡ transition210 of of the Aì 2P^X CO. Vibration^rotation emission spectra of HNO(n1),211 NH2(2n2),212 HNSi(n1),213,214 HCN(n1),215 HNC(n1),215 NH,216 SiH217,218 and SH219 were all recorded with an RF plasma reactor (Fig. 18) The HNSi is a new transient molecule related to HCN, but with the H bonded to the N atom, and was recorded along with the A 2P^X 2S‡ spectrum of SiN220 from a SiH4 plus N2 mixture. A microwave discharge tube has also provided spectra of SH221 and NH36/ND222 as well as the BH,223 PH222,224 and CH225,226 molecules. Very recently the spectrum of SeH was recorded227 and this molecule was the last main group monohydride to be detected by vibration^rotation spectroscopy. The NH work includes a direct comparison of the performance of the cryogenic echelle spectrograph, Phoenix, with a Fourier transform spectrometer.36 There is another common RF discharge called the inductively coupled plasma (ICP) source that is used largely for analytical chemistry. The ICP is

generally operated at atmospheric pressure with a £ow of argon gas. The discharge operates at 27 MHz with 1^2 kW of power to achieve plasma temperatures of about Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 203 Source: http://www.doksinet Fig. 19 A time series of spectra recorded by Chollet et al208 with various gas mixtures Carbon-containing impurities in the cell are responsible for CO and HCN. Reproduced from ref. 208, with permission 5000^6000 K.228 The goal is to dissociate all molecules into atoms, which are then monitored by emission spectroscopy or mass spectrometry. The ICP plasma, unlike ``normal RF and microwave discharges that are operated at low pressure and *100 W of power, is a proli¢c source of ions. The ICP has, however, been occasionally used for molecular spectroscopy The infrared emission spectra of OH229 and OD230 have been recorded with an ICP. The microwave discharge is a particularly useful source for the infrared emission of non-metal diatomics and a

few triatomics. The precursor molecules are usually introduced as gases. Molecules studied in this way include BN,231 BF,232 C2,233^236 C3,237 CN,238 CP,239,240 SiF,241,242 NO,243^245 N2,246^250 P2251,252 and O2.253 Of particular note are the Rydberg^Rydberg transitions of NO and SiF, and the discovery of two new infrared electronic transitions236 of C2 (B 1Dg^A 1Pu, B0 1S‡g^A 1Pu) and a triplet^triplet transition237 of C3. The electronic emission spectra of metal oxides, sul¢des and halides are readily produced. The secret is to ¢nd a relatively volatile precursor molecule (eg, organometallics or halides) to aid in the introduction of the metal into the plasma. For example, FeO can be made by reacting a mixture of ferrocene, argon and oxygen in the discharge region.254 Early work in this area was carried out, for example, by Merer and co-workers. Molecules studied in this way are TiO,255 VO,256 CrO,257 NbO,258 AlO,259 ZrS,260 HfS,261 TaS,262 TiCl,263 ZrCl264,265 and YI.266 An

interesting example is the discovery of the b0 3P^a 3D system of ZrS by Jonsson and Lindgren.260 The ZrS emission was detected in the 7400^9700 cmÿ1 region from 204 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet the microwave discharge of a mixture of ZrCl4 and sulfur powders without a carrier gas. These new bands of ZrS matched absorption spectra of an unknown molecule in the S-type star, R And. 14 Laser-excited infrared emission With the exception of some vibration^rotation emission work on acetylene267 and the HF dimer,28 almost all work has been on laser-excited electronic transitions. The technique was pioneered by Verges and Amiot, who, along with the Lyon group in France, continue to dominate the ¢eld. The typical experimental apparatus used by Amiot268 for Rb2 is displayed in Fig. 20 Metal dimers are made in a heat pipe oven and a laser beam is focussed through a hole in a mirror. The mirror is used to send the laser-induced £uorescence

into a high resolution Fourier transform spectrometer. To access highly excited states two lasers can be used Verges et al269 have written a recent review of the technique with examples. Fig. 20 The apparatus used by Amiot268 to record the infrared laser-induced £uorescence of K2. The l meter is a wavemeter and FP is a Fabry^Perot etalon that are used to monitor the laser wavelength. Reproduced from ref 268, with permission Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 205 Source: http://www.doksinet Fig. 21 The laser-induced £uorescence of the F 1S‡g ! A 1S‡u and E 1S‡g ! A 1S‡u transitions of 6Li2.272 Fluorescence to the last few bound vibrational levels of the A state is seen on the left. Reproduced from ref 272, with permission The heat pipe oven has allowed numerous electronic states of all of the alkali dimers, Li2,270^274 Na2,275^277 K2,278^281 Rb2,282^287 and Cs2,288^292 (except the radioactive Fr2), to be studied. The ground states have been followed nearly to

dissociation and highly reliable dissociation energies have been extracted Fig 21 shows a typical long vibrational progression in the A 1S‡u state of 6Li2. In this experiment272 the F 1S‡g and E 1S‡g states are populated by optical^optical double resonance spectroscopy. The £uorescence could be observed to the last few bound levels of the A 1S‡g state272 and the dissociation energy De was found to be 8517.03 cmÿ1 for the ground state of 6Li2 The technique has also been applied to the heteronuclear dimers, LiNa,293 NaK294,295 and RbCs.296,297 The problem with using a heat pipe oven for these molecules is the difference in vapour pressures of the alkali metals. A heat pipe oven can thus make LiNa but LiCs is very dif¢cult The original work on FTS detection of laser-induced £uorescence was carried out on I2. The B 3P0u‡^X 1S‡g transition was excited with an argon ion laser and a long ground state progression in the visible and near-infrared was seen.298 Similar experiments were

carried out for Te2,299 CsH300 and Bi2301 A classic experiment was carried out by Vervloet302 who made NH2 in an RF discharge and excited various levels with a dye laser. The bending levels of NH2 could be assigned up to u2 ˆ 10 and clearly showed the re-ordering of the K levels at the barrier to linearity. The heat pipe oven was also used in a series of experiments on the alkaline earth monohalides, CaF,303,304 BaF,305,306 BaCl307^310 and BaI.311 The monohalides were made by heating a mixture of the metal and the metal dihalide to about 1000  C, 206 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet e.g, Ba ‡ BaF2 ! 2BaF 31† For CaF, BaF and BaCl, the B0 2D states were located by laser excitation of the C 2P X 2S‡ transition and Fourier transform emission spectroscopy of the C 2P ! B0 2 D infrared transitions. For BaF and BaCl extensive measurements were made on all of the low-lying electronic states. Laser-induced £uorescence can also be measured

with the time-resolved FTS technique. Typically a pulsed ultraviolet laser dissociates a molecule such as HCCH and the infrared emission from a product such as C2H is detected,312 hn C2 H2 ! C2 H ‡ H 32† These experiments are carried out mainly to investigate photodissociation dynamics although they could be adapted for high resolution spectroscopy. 15 Furnaces Like the hollow cathode and arc discharges, the furnace is a traditional source of molecular spectra. When Fourier transform spectrometers became available it was natural to extend emission spectroscopy of electronic transitions into the infrared. (The arc discharge is dif¢cult to use in the infrared because source £uctuations are a problem for FTSs.) The early work on the Ballik^Ramsay system of C2 in a King furnace has already been cited31 and similar spectra of Si2 (d 1S‡g^b 1 Pu) have been measured.313 Thermal emission spectra of the new molecule BaLi314,315 were measured through two near-infrared electronic

transitions (2) 2 ‡ S ^X 2S‡ and (2) 2P^X 2S‡. A similar heat pipe oven source was used for Bi2316 and the forbidden B0 2D^X 2S‡ transition of BaH.317 Other molecules studied in this way are MnH,318 MnCl,319 CrF,320 CrCl,321 CoF,322 ScF,323,324 ScCl,325,326 ScI,327 LaF,328 LaCl,329 LaS,330 TiS,331 CuS332 and AlS.333 Most notable here is the analysis of the complex B 6P^X 6S‡ transition of CrF by Wallin et al.320 CrF was made by mixing Cr powder with CrF3 and heating the mixture to 1500 K in a ceramic furnace. The most interesting development, however, is the adaptation of thermal emission for rotational and vibration^rotation spectroscopy. The method works for stable molecules such as HF,334,335 HCl,335,336 HBr,337 DCN338 and H2O.339^347 For HF and HCl, pure rotational emission was also measured.334,335 The H2O emission spectra allowed the identi¢cation of hot H2O absorption in sunspots (``water on the Sun)339,340 (Fig. 22) Water emission spectra have now been assigned from 400

to 6000 cmÿ1 for pure rotational as well as vibration^rotation emission. The number of known energy levels of water was more than doubled in this work.339^347 The very complex spectra of hot water led to the use of a direct variational prediction of energy levels from a state-of-the-art ab initio surface in order to assign quantum numbers.340 A King furnace was used for the vibration^rotation emission spectra of CN,348 AlH,349 SH,350 CS,351,352 CuH,353 AgH,353 and AuH.353 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 207 Source: http://www.doksinet Fig. 22 The absorption spectrum of a sunspot (*3000 K) is on top and the laboratory emission spectrum of hot water (*1800 K) is on the bottom.347 All of the strong lines have been assigned and the asterisks mark lines seen in the laboratory but not in the sunspot. Early work (1989) in vibration^rotation emission spectroscopy below the 1800 cmÿ1 cut-off of the InSb detector was carried out on GeS by Uehara et al.,354 in addition to

the work on FO.37 This work was followed by the SiS emission measurements by Frum et al.355 in 1990 This spectacular emission measurement (Fig. 23) was recorded inadvertently and convinced the author of this review of the power and utility of the technique. Similar work on SiO356 and GeO357 followed In the case of SiO,356 the laboratory emission measurements at 1400 C were combined with SiO absorption measurements in a sunspot at 3200 K. The laboratory SiO data were found by accident because of the reaction of molten gallium with a mullite (2Al2O3SiO2) ceramic tube. Vibration^rotation emission spectra of the metal hydrides LiH,358 CaH,359 SrH,360 BaH,361 AlH,362 GaH,363 InH364 and BiH365 were recorded using ceramic (alumina or mullite) furnaces. All of the above molecules were formed by the reaction of molten metals with H2 gas (or D2, for deuterides) at high temperature. Interestingly, the spontaneous reactions of Ca and Sr with H2 did not occur,360,361 presumably because of a large

energy barrier to reaction. An electrical discharge was then used to promote the reaction (Fig. 24) in the furnace The pure rotational emission spectrum of LiH and LiD (Fig. 25) was also recorded in the far-infrared region.358 Clearly infrared emission spectroscopy has the sensitivity to detect transient molecules even at very long wavelengths Metal halides also give excellent 208 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet Fig. 23 Vibration^rotation thermal emission spectrum of SiS355 observed at 13 mm obtained from a mixture of Si and SiS2 powders at 1300 K. Fig. 24 Schematic of the furnace used to record the vibration^rotation emission spectrum of CaH359 by discharging a mixture of Ca vapour and H2. spectra, as illustrated by LiF,366 NaF,367 KF,368 MgF,369 CaF,370 SrF,371 BaF,372 BF,373 AlF,373,374 GaF,375 InF,376,377 TlF,378 NaCl,379 KCl,379,380 AlCl,381 NaBr,382 AlBr,383 LiI,384 CsI385 and BeF2.386 The CsI molecule gave a spectrum near 100

cmÿ1 but only the vibration^rotation band heads could be measured.385 The BeF2 antisymmetric stretching mode and associated hot bands near 1500 cm ÿ1 allowed an equilibrium Be^F bond length of 1.373 Ð to be determined386 16 Atmospheric science and remote sensing Emission spectroscopy is relatively common in atmospheric science but the term has a peculiar meaning in this area. For example, the infrared radiance of the Earth can be detected from an aircraft or a satellite viewing in the nadir direction. This is often Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 209 Source: http://www.doksinet Fig. 25 Pure rotational lines of a mixture of 6LiH, 6LiD, 7LiH and 7LiD obtained by detection of far-infrared thermal emission with a Fourier transform spectrometer.358 Reproduced from ref. 358, with permission called an ``emission spectrum although the atmospheric molecules of interest appear in absorption against the blackbody radiance of the Earth. This review, however, will adopt

the usual de¢nition of emission spectroscopy. The most spectacular infrared emission spectra of our atmosphere are recorded in the far-infrared region. Typically, a high resolution Fourier transform spectrometer in a balloon records stratospheric spectra by looking out at the limb of the Earth. An example of such a spectrum is presented as Fig. 26 The groups of Carli387^389 in Florence and Traub390 at the Smithsonian Astrophysics Observatory record far-infrared spectra of this type. Speci¢c molecules such as OH can be monitored with a simpler system based on Fabry^Perot etalons, but nothing beats a Fourier transform spectrometer for wide spectral coverage.389 Mid-infrared emission experiments of the limb of the Earth will be made with two satellite instruments, MIPAS391 (Michelson Interferometer for Passive Atmospheric Sounding) and TES392 (Tropospheric Emission Sounder). The dif¢culty with these measurements is the weak radiance emitted by the atmospheric molecules at 200^300 K.

The geometry is as drawn in Fig. 3, but the radiance of the end wall is very small because the view is to deep space in the limb geometry. The goal in these experiments is to understand the chemistry, particularly ozone chemistry, in our upper atmosphere. Emission measurements of our atmosphere can also be made from the ground with a Fourier transform spectrometer. A typical spectrum (Fig 27) recorded by Evans and Puckrin shows both emission and absorption lines.393 Highly excited infrared emission from our atmosphere can be recorded from airglow (nightglow and dayglow) as well as from aurora. Prominent infrared airglow emission394,395 is from the a 1Dg^X 3Sÿg system of O2 and the Meinel system of OH. 210 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet Fig. 26 Far-infrared emission spectrum of the stratosphere recorded by viewing the limb of the Earth from a balloon.26 Notice the excellent signal-to-noise ratio for the OH lines obtained with a Fourier

transform spectrometer. Reproduced from ref 26, with permission Fig. 27 Low resolution infrared emission spectrum of the zenith sky in winter393 Most of the lines appear in absorption but some are in emission. Reproduced from ref 393, with the permission of the American Meteorological Society Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 211 Source: http://www.doksinet The OH emission is prominent at night from the mesopause region at 85^90 km and this emission can be used to monitor atmospheric gravity waves.395 Infrared auroral emissions include the Meinel system396 of N2‡. The remote sensing of forest ¢res by infrared emission spectroscopy has already been discussed in the section on £ames.134 Other types of infrared emission remote sensing are also possible. For example, the emission of stack gas has been monitored at low resolution with a portable Fourier transform spectrometer.397 A number of recent emission spectra of volcanic plumes have also been recorded.398,399

Love et al.399 noted that the ratio of SiF4/SO2 seems to decrease just before a volcanic eruption. There are also military applications for the remote sensing of rocket plumes and other engine exhaust in the infrared. 17 Matrices Free radicals and other transient molecules can be trapped in rare gas matrices. Upon excitation, generally with a laser, the molecule will sometimes emit radiation in the infrared. Two reviews by Bondybey et al400,401 have covered both the spectroscopy and photophysics of recent matrix work. Electronic emission of the Phillips and Ballik^Ramsay systems of C2402 as well as the BN,403 BC404 and CuCl2405 molecules were seen in the Bondybey group. Vibrational emission from WO406 and ND407 was measured, and ND displayed nearly unperturbed rotational structure. 18 Liquids Infrared emission spectroscopy of liquids has been carried out for thin ¢lms of molten salts.15 For example, Bates and Boyd studied nitrate and chlorate melts408 Strong bands show some

distortions but the use of a thick sample as a reference alleviates this problem.15 Infrared emission of molecules in solution is also possible and Ogilby and co-workers, for example, have studied O2(1Dg) emission in various solvents.409 19 Surfaces and solids The presence of thin ¢lm, even monolayer, surfaces can be readily monitored by emission spectroscopy. As already mentioned, the best spectra are obtained with thin samples of a few micrometers in thickness. Some band distortions are often noticeable due to self-absorption410 etc., but the use of a thick sample as a reference eliminates the problem.15 Sullivan et al411 have reviewed the use of emission spectroscopy for surface analysis. A wide variety of ¢lms have been studied, including CO, metal oxides and organic molecules on a variety of substrates including metals, non-metals such as Si, metal oxides and zeolites. A typical application would be the study of a chemical reaction412 (e.g, NO ‡ NH3 on a catalyst surface,

such as 212 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet V2O5). Emission spectroscopy has been used to monitor thin ¢lms formed on silicon wafers as part of the fabrication of semiconductor devices.413 The emission of solid particles in combustion has proved to be a useful diagnostic.24,25 The spectra of bulk solids are most easily obtained by grinding the sample to a powder and sprinkling it on a heated metal substrate.414,415 Photoluminescence of direct band gap semiconductors like GaAs is an important tool in solid state physics.416 Impurities in solids, eg, ZnO:Cu2‡ (some of these materials can lase), can also be studied through their emission.417 Time-resolved FTS spectroscopy has also been applied to study the emission of solids, particularly for solid state laser media.418 An interesting variation on the technique is the use of a laser to heat a thin surface layer of a solid in order to record emission.419 20 Astronomy Most astronomical

infrared spectra appear in absorption against a bright continuum due to a star or dust. Some sources, however, such as comets, provide spectacular infrared emission spectra. Solar radiation evaporates the ``icy snowball as it nears the Sun. The hot evaporated molecules and their photolysis products appear in emission against the dark sky Infrared electronic spectra of C2 and CN are readily seen420 and more recently vibration^rotation lines of many molecules were found in the bright comets Hale-Bopp and Hyakutake.421 In a recent review Crovisier421 has summarized the species detected through vibration^rotation transitions so far and they include H2O, CO, CO2, CH4, C2H2, C2H, CH3OH, HCN and OCS. To this list can be added CH422 Fig. 28 The spectrum of Jupiters north tropical zone showing stratospheric emission from the n9 band of ethane on the left.427 Reprinted, with permission, from ref 427, # 1984 by Annual Reviews www.AnnualReviewsorg Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224

213 Source: http://www.doksinet Fig. 29 The emission of the n2 fundamental band of H3‡ in the aurora of Uranus and Jupiter428 Reproduced from ref. 428, with permission Most of these new molecules were detected with cryogenic echelle spectrographs such as Phoenix422 and a few observations were made from space with the Infrared Space Observatory, ISO.423 A particular surprise was the unexpectedly high abundance of ethane, C2H6, in Hyakutake424 In our solar system, atomic and molecular features in the Sun and sunspots appear in absorption although a few high-l Rydberg transitions of Mg and other atoms appear in emission near 1000 cmÿ1. The lines have very large Zeeman effects and can be used to study solar magnetic ¢elds.425 Spectra of most of the planets, satellites and asteroids are also absorption spectra, because the features are viewed against the planetary blackbody emission. Typical spectra of rocks and gaseous 214 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source:

http://www.doksinet CO2 were recorded by the Thermal Emission Spectrometer (TES) on Mars.426 As in atmospheric science, the word ``emission is not always used in the same sense as in this review. Some of the giant planets, however, display some interesting emission spectra427 For example, the n9 band of C2H6 appears in emission in the stratosphere of Jupiter427 (Fig 28) The most beautiful spectra, however, are those of H3‡ in the aurora of Jupiter428,429 (Fig. 29) Saturn430 and Uranus431 also display H3‡ emission. On Jupiter (and to a lesser extent on Saturn and Uranus)431 the H3‡ is formed by the interaction of the solar wind with the planetary magnetic ¢eld and emission is, therefore, localized near the magnetic poles. H3‡ has become an important tool for the study of Jupiters atmosphere.428 The three most important molecules in astronomy are H2, CO and H2O. The forbidden H2 vibration^rotation and pure rotational lines can be detected in shocked regions.21,432 H2, OH, H2O and

CO emission can be observed in supernova remnants.432,433 H2 emission is even detectable in starburst and Seyfert galaxies434 Star-forming regions in the Orion nebula also show H2,21 H2O435 and CO436 in infrared emission. Circumstellar envelopes can also show infrared emission as seen for CO in the carbon star IRC‡10216437 and H2O in the Mira variable (red giant), R. Cas438 Planetary nebulae also show very strong emission spectra of mainly atoms and a set of broad features that have been assigned to gaseous PAHs. 18,439^441 The idea is that PAHs absorb ultraviolet radiation and after internal conversion emit in the infrared.439,440 This PAH assignment, however, is not universally accepted An alternative assignment is a carbonaceous solid441 and even a giant molecule/small particle is quite possible. References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 P. F Bernath, Chem Soc Rev, 1996, 25, 111 P. V Huong, in Advances in Infrared and Raman Spectroscopy, Heyden, London, 1978, vol 4, p

85 F. J DeBlase and S Compton, Appl Spectrosc, 1991, 45, 611 J. Mink and G Keresztury, Appl Spectrosc, 1993, 47, 1446 G. Keresztury, in Encyclopaedia of Analytical Chemistry: Instrumentation and Applications, ed R A Meyers, J. Wiley & Sons, New York, 2000 J. B Bates, in Fourier Transform Infrared Spectroscopy, ed J R Ferraro and L J Basile, Academic Press, New York, 1978. P. Bernath, Annu Rev Phys Chem, 1990, 41, 91 E. Hirota, Annu Rep Prog Chem, Sect C, 1994, 91, 3; E Hirota, this volume E. Hirota, Chem Rev, 1992, 92, 141 M. Jacox, J Phys Chem Ref Data, Monograph No 3, 1994 R. F Barrow and P Crozet, Annu Rep Prog Chem, Sect C, 1992, 89, 353; R F Barrow and P Crozet and, Annu. Rep Prog Chem, Sect C, 1997, 93, 147 A. C Legon, Annu Rev Phys Chem, 1983, 34, 275 E. Herbst, Annu Rev Phys Chem, 1995, 46, 27 For example,P. F Bernath, S A Sinque¢eld, L L Baxter, G Sclippa, C M Rohl¢ng and M Bar¢eld, Vibr. Spectrosc, 1998, 16, 95 J. Hvistendahl, E Rytter and H A Oye, Appl Spectrosc,

1983, 37, 182 L. Nemes, R S Ram, P F Bernath, F A Tinker, M C Zumwalt, L D Lamb and D R Huffman, Chem. Phys Lett, 1994, 218, 295 P. Colarusso, K-Q Zhang, B Guo and P F Bernath, Chem Phys Lett, 1996, 269, 39 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 215 Source: http://www.doksinet 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 216 K.-Q Zhang, B Guo, P Colarusso and P F Bernath, Science, 1996, 274, 582 A. G Gaydon, The Spectroscopy of Flames, Chapman and Hall, London, 2nd edn, 1974 L. Lynds and B A Woody, Appl Opt, 1988, 27, 1225 N. Z Scoville, D N B Hall, S G Kleinman and S T Ridgway, Astrophys J, 1982, 253, 136 W. L Wolfe, in The Infrared and Electro-Optical System Handbook, ed G J Zissis, SPIE, Bellingham, WA, 1993. J. Bak and S Clausen, J Quant Spectrosc Radiat Transfer, 1999, 61, 687 J. R Markham, K Kinsella, R M Carangelo, C R Brouillette, M D Carangelo, P E Best and P R

Solomon, Rev. Sci Instrum, 1993, 64, 2515 P. R Solomon and P E Best, in Combustion Measurements, ed N Chigier, Hemisphere, New York, 1991, p. 385 B. Carli and M Carlotti, in Spectroscopy of the Earths Atmosphere and Interstellar Medium, ed K N Rao and A. Weber, Academic Press, San Diego, 1992 P. F Bernath, Spectra of Atoms and Molecules, Oxford University Press, New York, 1995, ch 1 H.-C Chang and W Klemperer, JChem Phys, 1994, 100, 1 M. A Busch and K W Busch, Appl Spectrosc, 1993, 47, 912 Y. Zhang, K W Busch and M A Busch, Appl Spectrosc, 1992, 46, 930 E. A Ballik and D A Ramsay, Astrophys J, 1963, 137, 61 A. E Douglas and D Sharma, J Chem Phys, 1953, 21, 448 J. G Moehlmann, J T Gleaves, J W Hudgens and J D McDonald, J Chem Phys, 1974, 60, 4790 D. J Cook, S Schlemmer, N Balucani, D R Wagner, J A Harrison, B Steiner and R J Saykally, J Phys. Chem A, 1998, 102, 1465 K. H Hinkle, R Cuberly, N Gaughan, J Heynssens, R R Joyce, S Ridgway, P Schmitt and J E Simmons, Proc. SPIE, 1998, 3354,

810; see also website: www.noaoedu/kpno/phoenix/phoenixhtm1 R. S Ram, P F Bernath and K H Hinkle, J Chem Phys, 1999, 110, 5557 P. D Hammer, A Sinha, J B Burkholder and C J Howard, J Mol Spectrosc, 1988, 129, 99 J. J Sloan and E J Kruus, in Time-Resolved Spectroscopy, ed R J H Clark and R E Hester, Wiley, New York, 1989. R. A Palmer, G D Smith and P Chen, Vibr Spectrosc, 1999, 19, 131 H. Weidner and R E Peale, Appl Spectrosc, 1997, 51, 1106 P. A Martin and G Guelachvili, Phys Rev Lett, 1990, 65, 2535 N. Picquë and G Guelachvili, Vibr Spectrosc, 1999, 19, 295; N Picquë and G Guelachvili, Appl Opt., 1999 38, 1224 A. Benidar, G Guelachvili and P A Martin, Chem Phys Lett, 1991, 177, 563 G. Guelachvili, J Opt Soc Am B, 1986, 3, 1718 M. Elhanine, R Farrenq and G Guelachvili, Appl Opt, 1989, 28, 4024 M. A Janssen, in Atmospheric Remote Sensing by Microwave Radiometry, ed M A Janssen, Wiley, New York, 1993. J. C Polanyi, Science, 1987, 236, 680 J. K Cashion and J C Polanyi, J Chem Phys,

1958, 28, 455 C. M Clayton, D W Merdes, J Pliva, T K McCubbin and R H Tipping, J Mol Spectrosc, 1983, 98, 168. N. I Butkovskaya and D W Setser, J Chem Phys, 1997, 106, 5028 J. J Klaassen, J Lindner and S R Leone, J Chem Phys, 1996, 104, 7403 C. A Carere, W S Neil and J J Sloan, Appl Opt, 1996, 35, 2857 D. E Heard, R A Brownsword, D G Weston and G Hancock, Appl Spectrosc, 1993, 47, 1438 T. J Johnson, A Simon, J M Weil and G W Harris, Appl Spectrosc, 1993, 47, 1376 C. D Pibel, E Sivota, J Brenner and H-L Dai, J Chem Phys, 1998, 108, 1297 D. D Nelson, A Schiffman, D J Nesbitt, J J Orlando and J B Burkholder, J Chem Phys, 1990, 93, 7003. M. C Abrams, S P Davis, M L P Rao and R Engleman, J Mol Spectrosc, 1994, 165, 57; M C Abrams, S. P Davis, M L P Rao, R Engleman and J W Brault, Astrophys J Suppl, 1994, 93, 351. D. E Osterbrock, J P Fulbright, P C Crosley and T A Barlow, Publ Astron Soc Paci¢c, 1998, 110, 1499. M. Beutel, K D Setzer, O Shestakov and E H Fink, Chem Phys Lett, 1996, 249,

183 M. Beutel, K D Setzer, O Shestakov and E H Fink, J Mol Spectrosc, 1996, 178, 165 M. Beutel, K D Setzer, O Shestakov and E H Fink, J Mol Spectrosc, 1996, 179, 79 K. Ziebarth, K D Setzer and E H Fink, J Mol Spectrosc, 1995, 173, 488 E. H Fink, K D Setzer, D A Ramsay and M Vervloet, Chem Phys Lett, 1991, 177, 265 R. Breidohr, O Shestakov and E H Fink, J Mol Spectrosc, 1994, 168, 126 H. Li, C Focsa, R Skelton and P F Bernath, unpublished W. E McDermott, N R Pchelkin, D J Bernard and R R Bousek, Appl Phys Lett, 1978, 32, 469 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet 67 D. J Bernard, B K Winker, T A Seder and R H Cohen, J Phys Chem, 1989, 93, 4790 68 R. Huang, R Zhang and R N Zare, Chem Phys Lett, 1990, 170, 437 69 H. P Yang, Y Qin, T J Cui, Q N Yuan, X B Xie, Q Zhuang and C H Zhang, Chem Phys Lett, 1992, 191, 130. 70 A. J Ross, P Crozet, R Bacis, S Churassy, B Erba, S H Ashworth, N M Lakin, M R Wickham, I R. Beattie and J M Brown, J Mol Spectrosc,

1996, 177, 134 71 K. Ziebarth, R Breidohr, O Shestakov and E H Fink, Chem Phys Lett, 1992, 190, 271 72 K. Ziebarth, K D Setzer, O Shestakov and E H Fink, J Mol Spectrosc, 1998, 191, 108 73 M. Beutel, K D Setzer and E H Fink, J Mol Spectrosc, 1999, 194, 250 74 M. Beutel, K D Setzer and E H Fink, J Mol Spectrosc, 1999, 195, 147 75 M. Beutel, K D Setzer and E H Fink, J Mol Spectrosc, 1996, 175, 48 76 E. H Fink, K D Setzer, D A Ramsay, M Vervloet and J M Brown, J Mol Spectrosc, 1990, 142, 108. 77 E. H Fink, K D Setzer, D A Ramsay, J P Towle and J M Brown, J Mol Spectrosc, 1996, 178, 143. 78 E. H Fink, K D Setzer, D A Ramsay and M Vervloet, Chem Phys Lett, 1991, 179, 95 79 E. H Fink, K D Setzer, D A Ramsay and M Vervloet, J Mol Spectrosc, 1989, 138, 19 80 J. Wildt, E H Fink, P Briggs, R P Wayne and A F Vilesov, Chem Phys, 1992, 159, 127 81 E. H Fink, H Kruse, D A Ramsay and M Vervloet, Can J Phys, 1986, 64, 242 82 J. F Noxon, Can J Phys, 1961, 39, 1110 83 K. D Setzer, E H Fink and D A

Ramsay, J Mol Spectrosc, 1999, 198, 163 84 I. Barnes, K H Becker and E H Fink, Chem Phys Lett, 1979, 67, 314 85 E. H Fink, H Kruse and D A Ramsay, J Mol Spectrosc, 1986, 119, 377 86 E. H Fink, K D Setzer, U Kottsieper, D A Ramsay and M Vervloet, J Mol Spectrosc, 1988, 131, 127. 87 E. H Fink, K D Setzer, D A Ramsay and M Vervloet, J Mol Spectrosc, 1987, 125, 66 88 E. H Fink, H Kruse, D A Ramsay and D-C Wang, Mol Phys, 1987, 60, 277 89 E. H Fink, K D Setzer, D A Ramsay and Q-S Zhu, Can J Phys, 1994, 72, 919 90 R. Winter, I Barnes, E H Fink, J Wildt and F Zabel, J Mol Struct, 1982, 80, 75 91 R. Winter, E H Fink, J Wildt and F Zabel, Chem Phys Lett, 1983, 94, 335 92 E. H Fink, K D Setzer, D A Ramsay and M Vervloet and G Z Xu, J Mol Spectrosc, 1989, 136, 218. 93 R. Winter, I Barnes, E H Fink, J Wildt and F Zabel, Chem Phys Lett, 1982, 86, 118 94 O. Shestakov, R Breidohr, H Demes, K D Setzer and E H Fink, J Mol Spectrosc, 1998, 190, 28 95 C. Linton, R Bacis, F Martin, S Rosenwaks and J

Verges, J Chem Phys, 1992, 96, 3422 96 E. H Fink, K D Setzer, D A Ramsay and M Vervloet, Chem Phys Lett, 1991, 179, 103 97 R. Breidohr, O Shestakov and E H Fink, Chem Phys Lett, 1993, 218, 13 98 R. Breidohr, K D Setzer, O Shestakov, E H Fink and W Zyrnicki, J Mol Spectrosc, 1994, 166, 251. 99 R. Breidohr, K D Setzer, O Shestakov, E H Fink and W Zyrnicki, J Mol Spectrosc, 1994, 166, 471. 100 R. Breidohr, O Shestakov, K D Setzer and E H Fink, J Mol Spectrosc, 1995, 172, 369 101 R. P Tuckett, P A Freedman and W J Jones, Mol Phys, 1979, 37, 379; R P Tuckett, P A Freedman and W. J Jones, Mol Phys 1979, 37, 403 102 E. H Fink and D A Ramsay, J Mol Spectrosc, 1997, 185, 304 103 K. J Holstein, E H Fink, J Wildt and F Zabel, Chem Phys Lett, 1985, 113, 1 104 E. H Fink, O Shestakov and K D Setzer, J Mol Spectrosc, 1997, 183, 163 105 A. F Vilesov, J Wildt and E H Fink, Chem Phys, 1991, 153, 531 106 D. Bailly, S A Tashkun, V I Perevalov, J L Teffo and Ph Arcas, J Mol Spectrosc, 1998, 190, 1 107 M.

Horani and M Vervloet, Astron Astrophys, 1992, 256, 683 108 (a) A. T Droege and P C Engelking, Chem Phys Lett, 1983, 96, 316; (b) P C Engelking, Rev Sci Instrum., 1986, 57, 2274 109 K. P Huber and M Vervloet, J Mol Spectrosc, 1991, 146, 188 110 (a) D. Cossart, J Mol Spectrosc, 1994, 167, 11; (b) D Cossart, C Cossart-Magos, B Gandara and J M. Robbe, J Mol Spectrosc, 1985, 109, 166 111 I. Dabrowski, D W Tokaryk and J K G Watson, J Mol Spectrosc, 1998, 189, 95 112 I. Dabrowski, D W Tokaryk, R H Lipson and J K G Watson, J Mol Spectrosc, 1998, 189, 110 113 I. Dabrowski, G Di Lonardo, G Herzberg, J W C Johns, D A Sadovskii and M Vervloet, J Chem Phys., 1992, 97, 7093 114 I. Dabrowski, D W Tokaryk, M Vervloet and J K G Watson, J Chem Phys, 1996, 104, 8245 115 I. Dabrowski, G Herzberg, B P Hurley, R H Lipson, M Vervloet and D C Wang, Mol Phys, 1988, 63, 269. 116 I. Dabrowski and D A Sadovskii, Mol Phys, 1994, 81, 291 117 I. Dabrowski and D A Sadovskii, J Chem Phys, 1997, 107, 8874 Annu. Rep

Prog Chem, Sect C, 2000, 96, 177^224 217 Source: http://www.doksinet 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 218 I. Dabrowski, G Herzberg and R H Lipson, Mol Phys, 1988, 63, 289 G. Herzberg, Annu Rev Phys Chem, 1987, 38, 27 F. Roux, F Michaud and M Vervloet, J Mol Spectrosc, 1994, 164, 510 K. P Huber and M Vervloet, J Mol Spectrosc, 1992, 153, 17 K. P Huber and M Vervloet, J Mol Spectrosc, 1988, 129, 1 M. Vervloet, A L Roche and Ch Jungen, Phys Rev A, 1988, 38, 5489 K. P Huber, M Vervloet, Ch Jungen and A L Roche, Mol Phys, 1987, 61, 501 D. W Tokaryk, G R Wagner, R L Brooks and J L Hunt, J Chem Phys, 1995, 103, 10439 D. Cossart and C Cossart-Magos, Chem Phys Lett, 1996, 250, 128 D. Cossart, M Bonneau and J M Robbe, J Mol Spectrosc, 1987, 125, 413 D. Cossart and M Elhanine, Chem Phys Lett, 1998,

285, 83; D Cossart, J Chim Phys, 1979, 76, 1045. A. Bernard, M Larzilliere, C Effantin and A J Ross, Astrophys J, 1993, 413, 829 C. Amiot, J-Y Roncin and J Verges, J Phys B, 1986, 19, L19 J.-Y Roncin, A Ross and E Boursey, J Mol Spectrosc, 1993, 162, 353 S. T Darian and M Vanpee, Combust Flame, 1987, 70, 65 C. H Douglass, H D Ladouceur, V A Shamanian and J R McDonald, Combust Flame, 1995, 100, 529. H. Worden, R Beer and C P Rinsland, J Geophys Res, 1997, 102, 1287 R. Bacis, D Cerny, J dIncan, G Guelachvili and F Roux, Astrophys J, 1977, 214, 946 J. Chauville, J-P Maillard and A W Mantz, J Mol Spectrosc, 1977, 68, 399 C. Amiot, J Chauville and J-P Maillard, J Mol Spectrosc, 1979, 75, 19 D. Bailly, C Camy-Peyret and R Lanquetin, J Mol Spectrosc, 1997, 182, 10 D. Bailly, S A Tashkun, V I Perevalov, J L Teffo and Ph Arcas, J Mol Spectrosc, 1999, 197, 114 J.-M Flaud, C Camy-Peyret and J-P Maillard, Mol Phys, 1976, 32, 499 C. Camy-Peyret, J-M Flaud, J-P Maillard and G Guelachvili, Mol Phys,

1977, 33, 1641 J.-P Maillard, J Chauville and A W Mantz, J Mol Spectrosc, 1976, 63, 120 C. Amiot, J-P Maillard and J Chauville, J Mol Spectrosc, 1981, 87, 196 R. Lanquetin, L H Coudert and C Camy-Peyret, J Mol Spectrosc, 1999, 195, 54 J. L Hardwick and G C Whipple, J Mol Spectrosc, 1991, 147, 267 K. J Smit, R J Hancox, D J Hatt, S P Murphy and L V De Yong, Appl Spectrosc, 1997, 51, 1400 K. J Smit, L V De Yong and R Gray, Chem Phys Lett, 1996, 254, 197 J. Wang, Z Chen, Y Luo, C Zhu, X Jin and T Wang, Spectrosc Lett, 1992, 25, 1355 R. Krech, G Caledonia, S Schertzer, K Ritter, T Wilkerson, L Cotnoir, R Taylor and G Birnbaum, Phys. Rev Lett, 1982, 49, 1913 G. E Caledonia, R H Krech, T D Wilkerson, R L Taylor and G Birnbaum, Phys Rev A, 1991, 43, 6010. K. M Stephens and S H Bauer, Spectrochim Acta, 1994, 50A, 741 W. T Rawlins, R R Foutter and T E Parker, J Quant Spectrosc Radiat Transfer, 1993, 49, 423 A. von Engel, Electric Plasmas: Their Nature and Uses, Taylor and Francis, London, 1983

I. Morino, K Matsumura and K Kawaguchi, J Mol Spectrosc, 1995, 174, 123 D. W Tokaryk and S Civis, J Chem Phys, 1995, 103, 3928 S. Civis and D W Tokaryk, J Mol Spectrosc, 1995, 172, 543 P. A Martin and G Guelachvili, Chem Phys Lett, 1991, 180, 344 C. Amiot and J Verges, Phys Scr, 1982, 26, 422 C. Amiot, Chem Phys Lett, 1981, 83, 40 C. Amiot and J Verges, J Mol Spectrosc, 1980, 81, 424 C. Amiot, J Mol Spectrosc, 1982, 94, 150 D. Bailly, J Mol Spectrosc, 1998, 192, 257 D. Bailly, C Rossetti and G Guelachvili, Chem Phys, 1985, 100, 101 D. Bailly and N Legay, J Mol Spectrosc, 1993, 157, 1 A. Campargue, D Bailly, J-L Teffo, S A Tashkun and V I Perevalov, J Mol Spectrosc, 1999, 193, 204. R. Farrenq, C Rossetti, G Guelachvili and W Urban, Chem Phys, 1985, 92, 389 D. Bailly and C Rossetti, J Mol Spectrosc, 1984, 105, 215; D Bailly and C Rossetti, J Mol Spectrosc., 1984, 105,229; D Bailly and C Rossetti, J Mol Spectrosc, 1984, 105, 331 T. Nakanaga, F Ito and H Takeo, Chem Phys Lett, 1993, 206,

73 G. Herzberg, H Lew, J J Sloan and J K G Watson, Can J Phys, 1981, 59, 428 S. P Reddy and C V V Prasad, J Phys E: Sci Instrum, 1989, 22, 308 R. Bacis, J Phys E: Sci Instrum, 1976, 9, 1081 S. P Davis, M C Abrams, J G Phillips and M L P Rao, J Opt Soc Am B, 1988, 5, 2280 S. P Davis, M C Abrams, Sandalphon, J W Brault and M L P Rao, J Opt Soc Am B, 1988, 5, 1838. D. W Ferguson, K N Rao, P A Martin and G Guelachvili, J Mol Spectrosc, 1992, 153, 599 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet 175 P. F Bernath, S A Rogers, L C OBrien, C R Brazier and A D McLean, Phys Rev Lett, 1988, 60, 197. 176 C. R Brazier, L C OBrien and P F Bernath, J Chem Phys, 1989, 91, 7384 177 S. A Rogers, C R Brazier, P F Bernath and J W Brault, Mol Phys, 1988, 63, 901 178 M. Douay, S A Rogers and P F Bernath, Mol Phys, 1988, 64, 425 179 I. Dabrowski and G Herzberg, Acta Phys Hung, 1984, 55, 219 180 Ch. Jungen, I Dabrowski, G Herzberg and M Vervloet, J Mol Spectrosc, 1992, 153,

11 181 Ch. Jungen, I Dabrowski, G Herzberg and D Kendall, J Chem Phys, 1989, 91, 3926 182 Ch. Jungen, I Dabrowski, G Herzberg and M Vervloet, J Chem Phys, 1990, 93, 2289 183 R. S Ram and P F Bernath, J Mol Spectrosc, 1995, 169, 268 184 R. S Ram and P F Bernath, J Mol Spectrosc, 1992, 155, 315 185 C. I Frum, R Engleman and P F Bernath, J Mol Spectrosc, 1991, 150, 566 186 R. S Ram, C N Jarman and P F Bernath, J Mol Spectrosc, 1993, 160, 574 187 L. C OBrien, R L Kubicek, S J Wall, D E Koch, R J Friend and C R Brazier, J Mol Spectrosc, 1996, 180, 365. 188 L. C OBrien, S J Wall and M K Sieber and J Mol Spectrosc, 1997, 183, 57 189 R. S Ram and P F Bernath, J Chem Phys, 1992, 96, 6344 190 R. S Ram and P F Bernath, J Mol Spectrosc, 1994, 165, 97 191 R. S Ram and P F Bernath, J Mol Spectrosc, 1997, 184, 401 192 R. S Ram, J Liëvin and P F Bernath, J Chem Phys, 1998, 109, 6329 193 R. S Ram, J Liëvin and P F Bernath, J Chem Phys, 1999, 111, 3449 194 R. S Ram and P F Bernath, J Chem Phys,

1996, 105, 2668; R S Ram and P F Bernath, J Mol Spectrosc., 1997, 183, 263 195 R. S Ram and P F Bernath, J Chem Phys, 1994, 101, 9283; R S Ram and P F Bernath, J Mol Spectrosc., 1995, 171, 169 196 R. S Ram and P F Bernath, J Chem Phys, 1996, 104, 6444 197 M. C McCarthy, R W Field, R Engleman and P F Bernath, J Mol Spectrosc, 1993, 158, 208 198 R. S Ram, P F Bernath and J W Brault, J Mol Spectrosc, 1985, 113, 451 199 J. W Brault and S P Davis, Phys Scr, 1982, 25, 268 200 J. W C Johns, J Mol Spectrosc, 1984, 106, 124 201 S. A Rogers, C R Brazier and P F Bernath, J Chem Phys, 1987, 87, 159 202 R. S Ram, P F Bernath and J W Brault, J Mol Spectrosc, 1985, 113, 269 203 W. A Majewski, M D Marshall, A R W McKellar, J W C Johns and J K G Watson, J Mol Spectrosc., 1987, 122, 341 204 W. A Majewski, P A Feldman, J K G Watson, S Miller and J Tennyson, Astrophys J Lett, 1989, 347, L51. 205 W. A Majewski, A R W McKellar, D Sadovskii and J K G Watson, Can J Phys, 1994, 72, 1016 206 T. Amano, M-C Chan,

S Civis, A R W McKellar, W A Majewski, D Sadovskii and J K G Watson and Can. J Phys, 1994, 72, 1007 207 G. Durry and G Guelachvili, J Mol Spectrosc, 1994, 168, 82 208 P. Chollet, G Guelachvili, M Morillon-Chapey, P Gressier and J P M Schmitt, J Opt Soc Am B, 1986, 3, 687. 209 M. Vervloet and M Herman, Chem Phys Lett, 1988, 144, 48 210 I. Dabrowski, M Vervloet and D-C Wang, Can J Phys, 1987, 65, 1171 211 J. C Petersen and M Vervloet, Chem Phys Lett, 1987, 141, 499 212 A. R W McKellar, M Vervloet, J B Burkholder and C J Howard, J Mol Spectrosc, 1990, 142, 319. 213 M. Elhanine, R Farrenq and G Guelachvili, J Chem Phys, 1991, 94, 2529 214 M. Elhanine, B Hanoune and G Guelachvili, J Chem Phys, 1993, 99, 4970 215 N. Picquë and G Guelachvili, Spectrochim Acta, 2000, 56A, 681 216 D. Boudjaadar, J Brion, P Chollet, G Guelachvili and M Vervloet, J Mol Spectrosc, 1986, 119, 352 217 J. C Knights, J P M Schmitt, J Perrin and G Guelachvili, J Chem Phys, 1982, 76, 3414 218 M. Betrencourt, D

Boudjaadar, P Chollet, G Guelachvili and M Morillon-Chapey, J Chem Phys, 1986, 84, 4121. 219 A. Benidar, R Farrenq, G Guelachvili and C Chackerian, J Mol Spectrosc, 1991, 147, 383 220 M. Elhanine, B Hanoune, G Guelachvili and C Amiot, J Phys II (France), 1992, 2, 931 221 R. S Ram, P F Bernath, R Engleman and J W Brault, J Mol Spectrosc, 1995, 172, 34 222 R. S Ram and P F Bernath, J Mol Spectrosc, 1996, 176, 329 223 F. S Pianalto, L C OBrien, P C Keller and P F Bernath, J Mol Spectrosc, 1988, 129, 348 224 R. S Ram and P F Bernath, J Mol Spectrosc, 1987, 122, 275 225 P. F Bernath, J Chem Phys, 1987, 86, 4838 226 P. F Bernath, C R Brazier, T Olsen, R Hailey, W T M L Fernando, C Woods and J L Hardwick, J. Mol Spectrosc, 1991, 147, 16 227 R. S Ram and P F Bernath, unpublished Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 219 Source: http://www.doksinet 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260

261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 220 L. M Faires, B A Palmer, R Engleman and T M Niemczyk, Spectrochim Acta, 1984, 39B, 819 M. C Abrams, S P Davis, M L P Rao and R Engleman, Astrophys J, 1990, 363, 326 M. C Abrams, S P Davis, M L P Rao and R Engleman, Pramana, 1992, 39, 163 R. S Ram and P F Bernath, J Mol Spectrosc, 1996, 180, 414 H. Brehdohl, I Dubois, F Mëlen and M Vervloet, J Mol Spectrosc, 1988, 129, 145 C. Amiot and J Verges, Astrophys J, 1982, 263, 993 K. Islami and C Amiot, J Mol Spectrosc, 1986, 118, 132 M. Douay, R Nietmann and P F Bernath, J Mol Spectrosc, 1988, 131, 250 M. Douay, R Nietmann and P F Bernath, J Mol Spectrosc, 1988, 131, 261 H. Sasada, T Amano, C Jarman and P F Bernath, J Chem Phys, 1991, 94, 2401 C. Amiot and J Verges, Chem Phys Lett, 1983, 95, 189 R. S Ram and P F Bernath, J Mol Spectrosc, 1987, 122, 282 R. S Ram, S Tam and P F Bernath, J Mol Spectrosc, 1992, 152, 89 H. Bredohl, I Dubois, F Mëlen

and M Vervloet, J Mol Spectrosc, 1988, 129, 232 M. Vervloet, H Bredohl, I Dubois and F Mëlen, J Mol Spectrosc, 1988, 131, 53 C. Amiot and J Verges, Chem Phys Lett, 1979, 66, 570 C. Amiot and J Verges, Phys Scr, 1982, 25, 302 A. Bernard, C Effantin, J dIncan, C Amiot and J Verges, Mol Phys, 1991, 73, 221 F. Roux, C Effantin, J dIncan and J Verges, J Mol Spectrosc, 1982, 91, 238 F. Roux and F Michaud, J Mol Spectrosc, 1991, 149, 441 F. Roux, D Cerny and J Verges, J Mol Spectrosc, 1982, 94, 302 F. Roux, F Michaud and J Verges, J Mol Spectrosc, 1983, 97, 253 A. Faye, Q Kou, R Farrenq and G Guelachvili, J Mol Spectrosc, 1999, 197, 147 C. Effantin, R Bacis, C Amiot and J Verges, J Mol Spectrosc, 1978, 69, 79 C. Amiot, C Effantin, J dIncan and J Verges, J Mol Spectrosc, 1978, 72, 189 C. Amiot and J Verges, Can J Phys, 1981, 59, 1391 A. W Taylor, A S-C Cheung and A J Merer, J Mol Spectrosc, 1985, 113, 487 D. C Galehouse, J W Brault and S P Davis, Astrophys J Suppl, 1980, 42, 241 A. J Merer,

G Huang, A S-C Cheung and A W Taylor, J Mol Spectrosc, 1987, 125, 465 A. S-C Cheung, W Zyrnicki and A J Merer, J Mol Spectrosc, 1984, 104, 315 O. Launila, B Schimmelpfenning, H Fagerli, O Gropen, A G Taklif and U Wahlgren, J Mol Spectrosc., 1997, 186, 131 O. Launila and J Jonsson, J Mol Spectrosc, 1994, 168, 1 J. Jonsson and B Lindgren, J Mol Spectrosc, 1995, 169, 30 O. Launila, J Jonsson, G Edvinsson and A G Taklif, J Mol Spectrosc, 1996, 177, 221 S. Wallin, G Edvinsson and A G Taklif, J Mol Spectrosc, 1998, 192, 368; S Wallin, G Edvinsson and A. G Taklif, J Mol Spectrosc, 1997, 184, 466 R. S Ram and P F Bernath, J Mol Spectrosc, 1997, 186, 113 J. G Phillips, S P Davis and D C Galehouse, Astrophys J Suppl, 1980, 43, 417 R. S Ram and P F Bernath, J Mol Spectrosc, 1997, 186, 335 A. Bernard, F Roux and J Verges, J Mol Spectrosc, 1980, 80, 374 M. Saarinen, D Permogorov and L Halonen, J Chem Phys, 1999, 110, 1424 C. Amiot, J Chem Phys, 1990, 93, 8591 J. Verges, C Amiot, R Bacis and A J

Ross, Spectrochim Acta, 1995, 51A, 1191 J. Verges, R Bacis, B Barakat, P Carrot, S Churassy and P Crozet, Chem Phys Lett, 1983, 98, 203 I. Russier, F Martin, C Linton, P Crozet, A J Ross, R Bacis and S Churassy, J Mol Spectrosc, 1994, 168, 39. C. Linton, F Martin, I Russier, A J Ross, P Crozet, S Churassy and R Bacis, J Mol Spectrosc, 1996, 175, 340. I. Russier, A Yiannopoulou, P Crozet, A J Ross, F Martin and C Linton, J Mol Spectrosc, 1997, 184, 129. A. J Ross, P Crozet, C Linton, F Martin, I Russier and A Yiannopoulou, J Mol Spectrosc, 1998, 191, 28. J. Verges, C Effantin, J dIncan, A Topouzkhanian and R F Barrow, Chem Phys Lett, 1983, 94, 1 J. Verges, C Effantin, J dIncan, D L Cooper and R F Barrow, Phys Rev Lett, 1984, 53, 46 R. F Barrow, C Amiot, J Verges, J dIncan, C Effantin and A Bernard, Chem Phys Lett, 1991, 183, 94. C. Amiot, J Mol Spectrosc, 1991, 146, 370 C. Amiot, J Verges and C E Fellows, J Chem Phys, 1995, 103, 3350 I. Russier, M Aubert-Frëcon, A J Ross, F Martin, A

Yiannopoulou and P Crozet, J Chem Phys, 1998, 109, 2717. I. Russier-Antoine, A J Ross, M Aubert-Frëcon, F Martin, P Crozet and S Magnier, J Phys B, 1999, 32, 4039. C. Amiot, P Crozet and J Verges, Chem Phys Lett, 1985, 121, 390 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 C. Amiot and J Verges, Mol Phys, 1987, 61, 51 C. Amiot, Chem Phys Lett, 1995, 241, 133 C. Amiot and J Verges, Chem Phys Lett, 1997, 274, 91 C. Amiot, Mol Phys, 1986, 58, 667 C. Amiot, O Dulieu and J Verges, Phys Rev Lett, 1999, 83, 2316 C. Amiot, C Crepin and J Verges, Chem Phys Lett, 1984, 106, 162; C Amiot, C Crepin and J Verges, Chem. Phys Lett, 1983, 98, 608 C. Amiot and J Verges, Chem Phys Lett, 1985, 116, 273 C. Amiot, J Chem Phys, 1988, 89, 3993 J. Verges and C Amiot, J Mol Spectrosc, 1987,

126, 393 C. Amiot, W DemtrÎder and C R Vidal, J Chem Phys, 1988, 88, 5265 C. E Fellows, J Verges and C Amiot, Mol Phys, 1988, 63, 1115; C E Fellows, J Verges and C Amiot, J. Chem Phys, 1990, 93, 6281 A. J Ross, R M Clements and R F Barrow, J Mol Spectrosc, 1988, 127, 546 R. F Barrow, R M Clements, G Delacretaz, J dIncan, A J Ross, J Verges and L WÎste, J Phys B, 1987, 20, 3047. T. Gustavsson, C Amiot and J Verges, Mol Phys, 1988, 64, 279; T Gustavsson, C Amiot and J Verges, Chem. Phys Lett, 1988, 143, 101 C. E Fellows, R F Gutterres, A P C Campos, J Verges and C Amiot, J Mol Spectrosc, 1999, 197, 19. D. Cerny, R Bacis and J Verges, J Mol Spectrosc, 1986, 116, 458 J. Verges, C Effantin, O Babaky, J dIncan, S J Prosser and R F Barrow, Phys Scr, 1982, 25, 338 C. Crepin, J Verges and C Amiot, Chem Phys Lett, 1984, 112, 10 R. F Barrow, F Taher, J dIncan, C Effantin, A J Ross, A Topouzkhanian, G Wannous and J Verges, Mol. Phys, 1996, 87, 725 M. Vervloet, Mol Phys, 1988, 63, 433 J. Verges,

C Effantin, A Bernard, A Topouzkhanian, A R Allouche, J dIncan and R F Barrow, J Phys. B, 1993, 26, 279 C. M Gittens, N A Harris, R W Field, J Verges, C Effantin, A Bernard, J dIncan, W E Ernst, P BÏndgen and B. Engels, J Mol Spectrosc, 1993, 161, 303 R. F Barrow, A Bernard, C Effantin, J dIncan, G Fabre, A El Hachimi, R Stringat and J Verge s, Chem. Phys Lett, 1988, 147, 535 C. Effantin, A Bernard, J dIncan, G Wannous, J Verges and R F Barrow, Mol Phys, 1990, 70, 735. C. Amiot and J Verges, Chem Phys Lett, 1991, 185, 310 M. Ha¢d, C Amiot and J Verges, Chem Phys Lett, 1993, 210, 45 C. Amiot, M Ha¢d and J Verges, J Phys B, 1993, 26, L407 C. Amiot, M Ha¢d and J Verges, J Mol Spectrosc, 1996, 180, 121 R. F Gutterres, J Verges and C Amiot, J Mol Spectrosc, 1999, 196, 29 T. R Fletcher and S R Leone, J Chem Phys, 1989, 90, 871 S. P Davis and J W Brault, J Opt Soc Am B, 1987, 4, 20 R. Stringat, G Fabre, A Boulezhar, J dIncan, C Effantin, J Verges and A Bernard, J Mol Spectrosc., 1994,

168, 514 J. dIncan, C Effantin, A Bernard, G Fabre, R Stringat, A Boulezhar and J Verge s, J Chem Phys, 1994, 100, 945. R. F Barrow, J Chevaleyre, C Effantin, M A Lebeault-Dorget, A J Ross, G Wannous and J Verges, Chem. Phys Lett, 1993, 214, 293 R. F Barrow, B J Howard, A Bernard and C Effantin, Mol Phys, 1991, 72, 971 W. J Balfour, O Launila and L Klynning, Mol Phys, 1990, 69, 443 O. Launila, Mol Phys, 1992, 76, 319 S. Wallin, R Koivisto and O Launila, 1996, 105, 388 M. Bencheikh, R Koivisto, O Launila and J P Flamand, J Chem Phys, 1997, 106, 6231 R. S Ram, P F Bernath and S P Davis, J Chem Phys, 1996, 104, 6949 E. A Shenyavskaya, J Verges, A Topouzkhanian, M-A Lebeault-Dorget, J dIncan, C Effantin and A. Bernard, J Mol Spectrosc, 1994, 164, 129 E. A Shenyavskaya, M-A Lebeault-Dorget, C Effantin, J dIncan, A Bernard and J Verges, J Mol Spectrosc., 1995, 171, 309 F. Taher, C Effantin, A Bernard, E A Shenyavskaya and J Verge s, J Mol Spectrosc, 1996, 199, 223. F. Taher, A Bernard, C

Effantin, J dIncan, E A Shenyavskaya and J Verges, J Mol Spectrosc, 1996, 179, 229. F. Taher, C Effantin, A Bernard, J dIncan and J Verges, J Mol Spectrosc, 1998, 189, 220 J. Verges, C Effantin, J dIncan, A Bernard and E A Shenyavskaya, J Mol Spectrosc, 1999, 198, 196. Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 221 Source: http://www.doksinet 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 222 J. Xin and L Klynning, Phys Scr, 1994, 49, 209 R. J Winkel, S P Davis and M C Abrams, Appl Opt, 1996, 35, 2874 J. Jonsson and O Launila, Mol Phys, 1993, 79, 95 L. C OBrien, M Dulick and S P Davis, J Mol Spectrosc, 1999, 195, 328 O. Launila and J Jonsson, J Mol Spectrosc, 1994, 168, 483 R. S Ram, Z Morbi, B Guo, K-Q Zhang, P F Bernath, J Vander Auwera, J W C Johns and S P Davis, Astrophys. J Suppl, 1996, 103, 247 R. B LeBlanc, J

B White and P F Bernath, J Mol Spectrosc, 1994, 164, 574 T. Parekunnel, T Hirao, R J Le Roy and P F Bernath, J Mol Spectrosc, 1999, 195, 185 V. Braun and P F Bernath, J Mol Spectrosc, 1994, 167, 282 W. Quapp, M Hirsch, G C Mellau, S Klee, M Winnewisser and A Maki, J Mol Spectrosc, 1999, 195, 284. L. Wallace, P Bernath, W Livingston, K Hinkle, J Busler, B Guo and K Zhang, Science, 1995, 268, 1155. O. L Polyansky, N F Zobov, S Viti, J Tennyson, P F Bernath and L Wallace, Science, 1997, 277, 346. O. L Polyansky, J R Busler, B Guo, K Zhang and P F Bernath, J Mol Spectrosc, 1996, 176, 305 O. L Polyansky, J Tennyson and P F Bernath, J Mol Spectrosc, 1997, 186, 213 O. L Polyansky, N F Zobov, J Tennyson, J A Lotoski and P F Bernath, J Mol Spectrosc, 1997, 184, 35. N. F Zobov, O L Polyansky, J Tennyson, J A Lotoski, P Colarusso, K-Q Zhang and P F Bernath, J. Mol Spectrosc, 1999, 193, 118 O. L Polyansky, N F Zobov, S Viti, J Tennyson, P F Bernath and L Wallace, J Mol Spectrosc, 1997, 186, 422.

O. L Polyansky, N F Zobov, S Viti, J Tennyson, P F Bernath and L Wallace, Astrophys J Lett, 1997, 489, L205. N. F Zobov, O L Polyansky, J Tennyson, S V Shirin, R Nassar, T Hirao, T Imajo, P F Bernath and L. Wallace, Astrophys J Suppl, 2000, 530, 994 S. P Davis, M C Abrams, M L P Rao and J W Brault, J Opt Soc Am B, 1991, 8, 198 J. L Deutch, W S Neil and D A Ramsay, J Mol Spectrosc, 1987, 125, 115 R. J Winkel and S P Davis, Can J Phys, 1984, 62, 1420 R. Winkel, S P Davis, R Pecyner and J W Brault, Can J Phys, 1984, 62, 1414 R. S Ram, P F Bernath and S P Davis, J Mol Spectrosc, 1995, 173, 146 J. Y Seto, Z Morbi, F Charron, S K Lee, P F Bernath and R J Le Roy, J Chem Phys, 1999, 110, 11756. H. Uehara, K Horiai, K Sueoka and K Nakagawa, Chem Phys Lett, 1989, 160, 149 C. I Frum, R Engleman and P F Bernath, J Chem Phys, 1990, 93, 5457 J. M Campbell, D Klapstein, M Dulick, P F Bernath and L Wallace, Astrophys J Suppl, 1995, 101, 237. E. G Lee, J Y Seto, T Hirao, P F Bernath and R J Le Roy, J

Mol Spectrosc, 1999, 194, 197 M. Dulick, K-Q Zhang, B Guo and P F Bernath, J Mol Spectrosc, 1998, 188, 14 C. I Frum and H M Pickett, J Mol Spectrosc, 1993, 159, 329 C. I Frum, J J Oh, E A Cohen and H M Pickett, J Mol Spectrosc, 1994, 163, 339 K. A Walker, H G Hedderich and P F Bernath, Mol Phys, 1993, 78, 577 J. B White, M Dulick and P F Bernath, J Chem Phys, 1993, 99, 8371 J. M Campbell, M Dulick, D Klapstein, J B White and P F Bernath, J Chem Phys, 1993, 99, 8379 J. B White, M Dulick and P F Bernath, J Mol Spectrosc, 1995, 169, 410 H. G Hedderich and P F Bernath, J Mol Spectrosc, 1993, 158, 170 H. G Hedderich, C I Frum, R Engleman and P F Bernath, Can J Phys, 1991, 69, 1659 A. Muntianu, B Guo and P F Bernath, J Mol Spectrosc, 1996, 176, 274 M.-C Liu, A Muntianu, K-Q Zhang, P Colarusso and P F Bernath, J Mol Spectrosc, 1996, 180, 188. B. E Barber, K-Q Zhang, B Guo and P F Bernath, J Mol Spectrosc, 1995, 169, 583 F. Charron, B Guo, K-Q Zhang and Z Morbi and P F Bernath, J Mol

Spectrosc, 1995, 171, 160 P. Colarusso, B Guo, K-Q Zhang and P F Bernath, J Mol Spectrosc, 1996, 175, 158 B. Guo, K-Q Zhang and P F Bernath, J Mol Spectrosc, 1995, 170, 59 K.-Q Zhang, B Guo, V Braun, M Dulick and P F Bernath, J Mol Spectrosc, 1995, 170, 82 H. G Hedderich and P F Bernath, J Mol Spectrosc, 1992, 153, 73 H. Uehara, K Horiai, K Nakagawa and H Suguro, Chem Phys Lett, 1991, 178, 553 T. Karkanis, M Dulick, Z Morbi, J B White and P F Bernath, Can J Phys, 1994, 72, 1213 H. Uehara, K Horiai, T Mitani and H Suguro, Chem Phys Lett, 1989, 162, 137 H. Uehara, K Horiai, A Kerim and N Aota, Chem Phys Lett, 1992, 189, 217 R. S Ram, M Dulick, B Guo, K-Q Zhang and P F Bernath, J Mol Spectrosc, 1997, 183, 360 H. Uehara, K Horiai, T Konno and K Miura, Chem Phys Lett, 1990, 169, 599 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 Source: http://www.doksinet 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413

414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 H. G Hedderich, M Dulick and P F Bernath, J Chem Phys, 1993, 99, 8363 H. Uehara, K Horiai, A Kerim, Y Ozaki and T Konno, Chem Phys Lett, 1993, 213, 101 H. Uehara, K Horiai, Y Osaki and T Konno, Chem Phys Lett, 1993, 214, 527 B. Guo, M Dulick, S Yost and P F Bernath, Mol Phys, 1997, 91, 459 V. Braun, B Guo, K-Q Zhang, M Dulick, P F Bernath and G A McRae, Chem Phys Lett, 1994, 228, 633. C. I Frum, R Engleman and P F Bernath, J Chem Phys, 1991, 95, 1435 B. Carli, F Mencaraglia and A Bonetti, Int J Infrared Millimeter Waves, 1980, 1, 263; B Carli, F Mencaraglia and A. Bonetti, Int J Infrared Millimeter Waves, 1982, 3, 385 B. Carli, F Mencaraglia, A Bonetti, B M Dinelli and F Forni, Int J Infrared Millimeter Waves, 1983, 4, 475. B. Carli, M Carlotti, B M Dinelli, F Mencaraglia and J H Park, J Geophys Res, 1989, 94, 11049 D. G Johnson, K W Jucks, W A Traub and K V Chance, J Geophys Res, 1995, 100, 3091 H. Fischer and H

Oelhaf, Appl Opt, 1996, 35, 2787 R. Beer, Jet Propulsion Laboratory, see website: tesjplnasagov W. F J Evans and E Puckrin, J Climate, 1995, 8, 3091 W. F J Evans, D M Hunten, E J Llewellyn and A Vallance Jones, J Geophys Res, 1968, 73, 2885 R. P Lowe and D N Turnbull, Geophys Res Lett, 1995, 22, 2813 A. Vallance Jones, in Auroral Physics, ed C-I Meng, M J Rycroft and L A Frank, Cambridge University Press, Cambridge, 1991. R. C Carlson, A F Hayden and W V Telfair, Appl Opt, 1988, 27, 4952 C. Oppenheimer, P Francis, M Burton, A J H Maciejewski and L Boardman, Appl Phys B, 1998, 67, 505. S. P Love, F Goff, D Counce, C Siebe and H Delgado, Nature, 1998, 396, 563 V. E Bondybey, A M Smith and J Agreiter, Chem Rev, 1996, 96, 2113 V. E Bondybey, M RÌsÌnen and A Lammers, Annu Rep Prog Chem, Sect C, 1999, 95, 331 G. M Lask, J Agreiter, R Schlachta and V E Bondybey, Chem Phys Lett, 1993, 205, 31 M. Lorenz, J Agreiter, A M Smith and V E Bondybey, J Chem Phys, 1996, 104, 3143 A. M Smith, M

Lorenz, J Agreiter and V E Bondybey, Mol Phys, 1996, 88, 247 M. Lorentz, N Caspary, W Foeller, J Agreiter, A M Smith and V E Bondybey, Mol Phys, 1997, 91, 483. M. Lorentz, J Agreiter, N Caspary and V E Bondybey, Chem Phys Lett, 1998, 291, 291 N. Caspary, B E Wurfel, A M Smith and V E Bondybey, Chem Phys Lett, 1997, 220, 241 J. B Bates and G E Boyd, Appl Spectrosc, 1973, 27, 204 R. D Scurlock and P R Ogilby, J Phys Chem, 1987, 91, 4599 P. R Grif¢ths, Appl Spectrosc, 1972, 26, 73 D. H Sullivan, W C Conner and M P Harold, Appl Spectrosc, 1992, 46, 811 D. H Sullivan, M P Harold and W C Conner, J Catal, 1998, 178, 108 J. E Franke, T M Niemczyk and D M Haaland, Spectrochim Acta, 1994, 50A, 1687 A. M Vassallo, P A Cole-Clarke, L S K Pang and A J Palmisano, Appl Spectrosc, 1992, 46, 73 A. M Vassallo and K S Finnie, Appl Spectrosc, 1992, 46, 1477 A. Bignazzi, E Grilli, M Radice, M Guzzi and E Castiglioni, Rev Sci Instrum, 1996, 67, 666 B. M Kimpel and H-J Schulz, Phys Rev B, 1991, 43, 9938 H.

Weidner and R E Peale, in OSA TOPS on Advanced Solid-State Lasers, ed S A Payne and C Pollock, Optical Society of America, Washington, D. C, 1996, vol 1 L. T Lin, D D Archibald and D E Honigs, Appl Spectrosc, 1988, 42, 477 J. R Johnson, U Fink and H P Larson, Astrophys J, 1983, 270, 769 J. Crovisier, Faraday Discuss, 1998, 108, 437 D. Reuter, D Jennings, K Hinkle, S Ridgway, L Wallace, P Bernath and J Waller, IAU Circular, 6681, 1997. J. Crovisier, K Leech, D Bockelëe-Morvan, T Y Brooke, M S Hanner, B Altieri, H U Keller and E. Lellouch, Science, 1997, 275, 1904 M. J Mumma, M A DiSanti, N Dello Russo, M Fomenkova, K Magee-Sauer, C D Kaminski and D. A Xie, Science, 1996, 272, 1310 D. Deming, R J Boyle, D E Jennings and G Wiedemann, Astrophys J, 1988, 333, 978 P. R Christensen, D L Anderson, S C Chase, R T Clancy, R N Clark, B J Conrath, H H Kieffer, R. O Kuzmin, M C Malin, J C Pearl, T L Roush and M D Smith, Science, 1998, 279, 1692 S. T Ridgway and J W Brault, Annu Rev Astron

Astrophys, 1984, 22, 291 S. Miller, H A Lam and J Tennyson, Can J Phys, 1994, 72, 760 T. Oka and T R Geballe, Astrophys J Lett, 1990, 351, L53 T. R Geballe, M-F Jagod and T Oka, Astrophys J Lett, 1993, 408, L109 H. A Lam, S Miller, R D Joseph, T R Geballe, L A Trafton, J Tennyson and C H Ballester, Astrophys. J Lett, 1997, 474, L73 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224 223 Source: http://www.doksinet 432 M. A Richter, J R Graham, G S Wright, D M Kelly and J H Lacy, Astrophys J Lett, 1995, 449, L83. 433 W. T Reach and J Rho, Astrophys J Lett, 1998, 507, L93 434 H. Mouri, Y Tamiguchi, K Kawara and M Nishida, Astrophys J Lett, 1989, 346, L73 435 M. Harwit, D A Neufeld, G J Melnick and M J Kaufman, Astrophys J Lett, 1998, 497, L105 436 T. Geballe and R Garden, Astrophys J, 1987, 317, L107 437 R. Sahai and P G Wannier, Astrophys J, 1985, 299, 424 438 Truong-Bach, R. J Sylvester, M J Barlow, Nguyen-Q-Rieu, T Lim, X W Liu, J P Baluteau, S Deguchi, K. Justtanont and A G G M

Tielens, Astron Astrophys, 1999, 345, 925 439 J. L Puget and A Lëger, Annu Rev Astron Astrophys, 1989, 27, 161 440 L. J Allamandola, A G G M Tielens and J R Barker, Astrophys J Suppl, 1989, 71, 733 441 C. Joblin, Faraday Discuss, 1998, 109, 349 224 Annu. Rep Prog Chem, Sect C, 2000, 96, 177^224