Intracavity Laser Spectroscopy (ICLS) is one of the most sensitive techniques of absorption spectroscopy. ICLS has a wide field of applications: molecular spectroscopy, spectroscopy of atmospheric gases and pollutants, investigation of chemical reaction's kinetics and nonlinear optical phenomena, such as stimulated Raman scattering, two- photon absorption, etc. The sensitivity of ICLS is limited by nonlinear processes in laser active media. The investigations of these processes permit achieving sensitivity of ICLS up to 10^-11 cm^-1.

The method of broad-band Intracavity Laser (ICL) Spectroscopy suggested in 1970 [1] consists of quenching the laser emission at the absorption-line frequencies of the species placed in a broad-band laser cavity, the generation band width of such a laser being much greater than the half-width of spectral line studied. In this case the laser emission spectrum has sharp dips at the frequencies of absorption lines, which can be recorded with an ordinary spectroscopic instrumentation. In the intracavity laser spectroscopy a laser itself is a nonlinear detector of a weak absorption and therefore the parameters of an intracavity spectrometer are first of all determined by the mechanism of the lasing process. The most important are the resonator mechanism caused by multipass travelling of radiation during the process ofgeneration and the threshold character oflasing process. The models describing the formation of a dip in the laser generation spectrum [2-8] can be divided into two groups. The first one asumes high sensitivity of the ICL-spectrscopy to be due to the threshold mechanism for which it is supposed the laser system is in a stationary stable state and all the parameters determining the sensitivity of the intracavity absorption are constant in time [7-8]. These models predict a strong influence of the spatial inhomogeneity of the inversion decay and a sharp increase of the ICL-spectrometer sensitivity near the generation threshold. The second group of the models is based on the assumption that even cw multimode lasers never reach a stable state of generation due to its breaks at a mode caused by different perturbation mechanisms, the life time of a mode being the basic quantity which determines the ICL-spectrometer sensitivity (so called resonator mechnism). Time sweep of the laser emission spectra confirm the fact that longitudinal modes of a cw laser have a fmite life time [9]. The analysis ofFourier spectra ofsuch a laser has revealed that at high spectral power density single modes oscillate at about 100 kHz frequency which corresponds to a 10 ts lifetime of a mode [10]. The decrease of the spectral power density results in the increase of a mode's lifetime, that in turn causes the increase of the ICLspectrometer sensitivity near the generation threshold. Experiments carried out under careftilly controlled conditions showed that the depth of a dip in the intracavity spectrum increases exponentially with the increase of mode's life time.9 This convicingly shows the validity of the resonator mechanism. The resonator mechnism allows one to explain the possibility of recording absorption coefficients as low as Kmin 1010...l01 'cm' during a 0.1 s time interval. As has been shown in Ref. 1 1 detection limit of ICLspectrometer based on broadband ranning wave dye laser is determined by quantum nature of radiation. However, it is only possible if the frequency variations of the laser gain coefficient in the vicinity of an absorption line is less than Kmin. The spectral stability of the gain coefficient is, in turn, determined by the broadening mecahnism in the active medium, i.e. by the ratio ofthe width ofhomogeneous amplification contour (or frequency filter width in resonator) to an absorption line halfwidth. The simplest way of excluding the frequency variation of gain coefficient is decreasing of the numbers of resonator optical elements. In that case the resolnator scheme becomes very simple and consists of 2 mirrors and an active element. Simultaneously, such simplicity decreases the spectral range of the spectrometer, so generation band covers only 20-30% ofgain profile width. The increase ofthe spectral tunable region using dispersive resonator leads to decreasing of the spectrometer sensitivity, so a scheme of the resonator of the ICL-spectrometer is determined by concrete task of investigation. Begining from 1970 the method of ICL-spectroscopy has been extensively developing and now there are about 500 papers in this field including several reviews13'7, devoted to the analysis of sensitivity of ICL-spectroscopy to the determination of quantitative information and to use of different schemes of the spectrometers. There was comprehensive analysis of the ICL-spectrometers for the aim of obtaining quantitative information performed at the Institute of Atmospheric Optics beginning 1972 when first ruby laser ICL-spectrometer had been designed here.12 It includes: 1) design of a complex of ICL-spectrometers based on Nd-glass, ruby, dye, and color 22 center lasers; 2) determination of absorption line parameters; 3) development of efective methods for elimination of spurious selection; 4) use of dispersive resonator in ICLspewometers. Obtained results were summarized in the monograph "Intracavity laser spectroscopy. Method and application" which was published in Russian in 1985. ' This monograph is the only monograph on Intracavity laser spectroscopy in the world but scientists abroad the Russia practically have no information on it. In this paper the main materials of the monograph are presented and the last results in ICLspectroscopy obtained at the Institute of Atmospheric Optics and at others scientific centers during last 10 years are su.mmurised.

Broadband lasers simultaneously generating a large number of longitudinal cavity modes are extremely sensitive to frequency-selective absorption by the media in the laser cavity. The spectroscopic technique based on this phenomenon is known as intracavity laser spectroscopy. The technique is capable of detecting absorption coefficients in the range of 10^-10 cm^-1. The sensitivity limit has been shown to result from a nonlinear interaction. A number of theoretical models describing the nonlinear interaction has been proposed, however, none of these models can adequately describe all experimental data. The present work overviews the theoretical models in a logical connection with experimental evidence obtained under various conditions.

Recent years there appeared many experimental and theoretical studies of highly excited vibrational states of small polyatomic molecules (Refs. (1 -4) and references therein). These studies have been stimulated by current interest in intramolecular kinetics, photoselective chemistry, and atmospheric optics. The investigation of highly excited vibrationrotational states has become important for searching a quantum analog of classic chaotic motion, classic-quantum correspondence, nonlinear resonance studies and for various application in the laser chemistry, environment, atmospheric optics, and astrophysics. Rotational-vibrational lines caused by transitions to excited vibrational states (energy is larger than 7000 cm') are very weak, their strengths, as a rule, are 5 -7 order of magnitude less than line strengths of fundamental bands. Therefore, high sensitive spectrometers must be used to record these weak bands. Intracavity laser spectrometers due to their high sensitivity to absorption become one of the powerful tools for studying highly excited molecular states. The method of broad-band Intracavity Laser (ICL) Spectroscopy consists of quenching the laser emission at the absorption-line frequencies of the species placed in a broad-band laser cavity. In this case the laser emission spectrum has sharp dips at the frequencies of absorption lines, which can be recorded with an ordinary spectroscopic instrumentation. In the intracavity laser spectroscopy a laser itself is a nonlinear detector of weak absorption. The laser imitates a multipass absorption cell, but with much greater effective length Leff Of the absorbing layer than in a simple multipass cell where it is limited by radiation losses on mirrors. In this case Leff S proportional to the duration t of continuous generation in the region of an absorption line under study. ICL -spectroscopy was widely used to various applications some of them are described in the monograph5 and in other papers of the present issue. One of the most effective applications of ICL - spectroscopy is study of weak molecular spectra caused by transitions to highly excited states. This paper summarized investigation of highly excited molecular states using the intracavity laser spectroscopy.

The high sensitivity of intracavity spectroscopy has been used for recording very weak electronic molecular spectra in the visible regime. Spectra of mononitrides of transition metals have been chosen as objects of investigation. Numerous new bands in spectral studied have been detected. The great number of molecular constants additionally to that previously reported in literature have been obtained for various electronic states of molecules studied. Spectra of the isotopic molecules have been used to make the unambiguous assignment of the bands recorded.

The task to control plasmas parameters is very important for many applications and research setups. Plasmas are characterized by a set of parameters as an electron density, an electron temperature, and gas, ion, "vibrational", "rotational" temperatures and so on. A most commonly used method for diagnostics of plasmas is a spectroscopy based on dependence of atom spectra on the plasma parameters. These methods [1] have a very significant advantage since they do not perturb the plasma state. The another important advantage is rather a large amount of information containing in the plasmas' spectrum. We are going to introduce the ICLS as a new plasma diagnostics technique because of its high sensitivity that could allow to perform measurements then other spectroscopic methods fail. We skip describing of the ICLS principles because the other part of this book doing it perfectly. Now we mention briefly the plan of our part. First of all, using of ICLS is required to know the errors of measurements of plasma parameters, we estimate some essential types of errors in Chapter 2. Investigation of non-equilibrium molecular plasmas are in Chapter 3. It is a plasma electric field that determines most of all the physical processes in plasmas. On the other hand, the problems of propagation of strong electromagnetic waves in plasma attract great interest in connection with the study of heating in confined plasmas [2]. The problems of wake-field excitation [3] are under presentday consideration. So, the problem of determination of the electric field inside plasma is one of the utmost importance. The atom levels affected by the electrical field are shifted due to the Stark effect, the value of this shift, being dependent of the field intensity and the constants of atom interaction with the electric field. If an a.c. field is present along with the level shifts, there are satellites of lines due to amplitude and frequency modulation of absorption. Thus, to determine the intensity of the electric field, it is required to measure the shifts and the intensity of satellites of lines. So, it is essential to have a method with high sensitivity, because the satellite intensities in low fields are very small, and with high spectral resolution, because the shifts of levels are small in low fields too. Plasma satellites arising due to Langmuir plasma waves are located near the forbidden lines at the distance of plasma frequency w = (4iie2Ne/me)11'2. For the case of fluctuations of electron density to registration of Langmuir satellites requires a method is needed which can register the whole spectrum of the atom by one shot . These very features are characteristic for the method of intracavity laser spectroscopy. Chapter 4 is concerned about the plasma field diagnostics by the ICLS. The interferometry methods of plasma diagnostics in the optical, infrared and millimeter bands of the electro-magnetic waves attract the researches' steady attention, as they assume simple interpretation of the obtained results, and are free from different modeling considerations, do not require any a priori data about the investigated plasma parameters. The application of these methods, however, in the most convenient in all respects optical range (high resolution, simplicity and reliability of the registration, etc.) is restricted by quite low sensitivity. This drawback is caused by the small refraction of the electromagnetic waves in the optical band and by the possibility of ising only double-slit interferometer, framing interference pictures with dark and light bands of the same width and that can not but hampers the registration of their small shifts. 167 168 The nature of selective losses is not essential for the ICLS so, it is easy to extend ICLS method into interferometry of plasmas, we deal with it in chapter 5. Here is what the present issue is about. We, with some small exceptions will refer to the works, done in our group. Though quite a number of researchers use the methods of intracavity laser spectroscopy, practically only our group is working at the implementing of this methods in plasma diagnostics.

Because a light wave, together with frequency and amplitude, is also characterized by phase and polarization state, in intracavity laser spectroscopy (ICLS) it is appropriate to attempt to realize not only the effect of the accumulation of the values of the absorption of the light wave, as this was done in Ref. 1, but also the changes of phase and state of polarization of laser radiation upon placing a phase or optically active object, respectively, in the laser cavity. In addition, one assumes that numerous passages of the generated radiation through a spectral apparatus (for example, an interference device) placed in the laser cavity might improve its spectral characteristics radically. However, along the way of appreciably broadening the possibilities of ICLS it is clear one might meet serious difficulties. Among them are: 1) The absence in existing lasers of polarization coherence of the generated radiation. 2) The possibility of generating nonlinear processes in the object placed in one of the arms of an intracavity interferometer. 3) With the anticipated increase in the accuracy of spectral interferometric measurements in ICLS, there might appear those effects which existed previously (in classical spectral interferometry) but were hidden under the broad contour of interference fringes (IF). For example, the appearance of the distinction of the real shape of the absorption line from Lorenlzian in the refraction curve, which might be unsymmetric and not be described by the generally customary Selmeier equation. 4) It was not evident earlier that, when solving applied problems (because of the increased accuracy of measurements of the positions of interference fringes in the spectrum), one is able to transfer this level of accuracy to the whole complex of accompanying spectroscopic and interferometric measurements (the methods of treating of the experimental data, errors of determining ofthe associated quantities which enter the computational equations). 5)Itis not clear how one can define the linear state of polarization if radiation in the cavity at the beginning of the generation pulse, which subsequently might undergo optical rotation inside the laser in correspondence with the number ofpassages ofthe generated radiation through the optically active object. This enumeration might be extended. In addition to the indicated difficulties, others, which can not always be predicted beforehand, might be expected. As will be clear from the following, this was confirmed completely during the performance of subsequent investigations. It is necessary to note that solutions to the problem of increasing the sensitivity of optical measurements in ICLS were proposed earlier. These are works2 which used certain types of laser processing a resonant response to the modulation of the quality factor of their cavities at the frequency of the relaxation oscillations for a significant increase in the sensitivity of polarization and interference measurements. This idea might be used, for example, for investigating the optical activity induced in an object and of phase objects whose thickness might be modulated in time. In Ref.3 they increased (to a factor of 100) the sensitivity of polarization measurements as the result of placing of the object being investigated inside an anisotropic cavity of an optical quantum generator of monochromatic radiation. Physical problems arising from the realization of ideas that might lead to the establishment of phase, polarization, and interference ICLS with linear (in correspondence with the number of passages of the laser radiation through the cavity) increase in the sensitivity and precision of optical measurements go far beyond the realm of problems which arose earlier during the establishment of absorption ICLS. However, a lot of work is obviously needed to realize possibilities that might reveal themselves to researchers after the development of these new directions in ICLS. The review of work performed under the direction of the authors Optics Department of St.Petersburg University for the last ten years on the development of the physical basis of phase, polarization, and interference ICLS is contained in this exposition. For a discussion of the material, we start from the scheme of the investigations on intracavity laser spectroscopy shown in Fig. 1.

Intracavity Laser Absorption Spectroscopy (ICLAS) combined with flash photolysis has been used over 20 past years as a powerful tool for time and quantum state resolved spectroscopy of vibrational relaxation, elementary gas phase chemical reaction, heterogeneous decay, rotational temperature and absorption cross section measurements, and in some cases for spectroscopic identification of free radicals as well. Detailed description of experimental setups and methods of ICLAS spectra recording and processing enable to critically evaluate data obtained by ICLAS technique for NH₂, PH₂, HCO, HSO, HNO, HO₂, CH₃O₂ radicals in visible and near IR region.