In recent years, the synchronously mode-locked cw dye laser has emerged as one of the most promising sources of laser pulses of picosecond and subpicosecond duration. The great interest in synchronously mode-locked cw dye lasers is due to their numerous attractive features, including their reliability and stability, broad spectral tunability, easily controllable pulse durations, and their amenability to synchronization and frequency modulation. We briefly review here the pulse formation mechanism and general performance characteristics of such lasers, as well as several recent advances in their source capabilities, and some recent applications.

A synchronously-pumped, ring dye laser has produced pulses as short as 0.65 picoseconds with peak powers over 4 kilowatts and with improved tuneability and average powers when compared with linear dye lasers.

In synchronously pumped or passively mode-locked cw dye lasers, traveling-wave operation in the gain-jet is found to lead to better locking than standing wave. On the other hand, standing-wave in the saturable absorber-jet gives better locking in passively mode-locked lasers. Key considerations leading to these conclusions are reviewed. 0.4 psec for synchronously pumped tunable dye lasers and 60 fsec for passively locked nontunable dye lasers are obtained experimentally. *

This paper reviews the recent progress in producing picosecond optical pulses from semi-conductor laser diodes. The discussion concentrates on the mode-locking of a semiconductor laser diode in an external resonator. Transform-limited optical pulses ranging from several picoseconds to subpicosecond durations have been observed with active and passive mode-lock-ing. Even though continuing research on the influence of impurities and defects on the mode-locking process is still needed, this technique has good promise for being utilized in fiber-optic communication systems. Alternative methods of direct electrical and optical excitation to produce ultrashort laser pulses are also described. They can generate pulses of similar widths to those obtained by mode-locking. The pulses generated will find applications in laser ranging and detector response measurement.

Tunable CW laser action of platelet semiconductors is reported in both mode-locked and unmode-locked configurations. The gain media are platelets of CdS, CdSe, CdSSe and InGaAsP, cooled to 85K and longitudinally pumped by argon-ion and krypton-ion lasers. Anti-reflection coating of the crystal face and external bandwidth restriction have been used to generate pulses as short as 4 ps in CdS. The pulses observed are chirped, with non-transform limited time-bandwidth products of about 1.7. The energy conversion efficiency is 20% into the TEM_oo mode, with output powers of over 10 mW from CdS. Pulses as short as 7 ps tunable over a 26 nm range have been obtained in InGaAsP.

Pulses 7 picoseconds or less in duration have been generated from a bulk GaAs crystal by a synchronous mode-locking technique. The GaAs crystal was optically pumped by two-photon absorption of the emission from a mode-locked Nd:glass laser. Two-photon absorption as the means of excitation increases the volume of the gain medium by increasing the pene-tration depth of the pump intensity, enabling generation of intra-cavity pulses with peak power in the megawatt range. Tuning of the wavelength of the GaAs emission is achieved by varying the temperature. A tuning range covering 840 nm to 885 nm has been observed over a temperature range from 97°K to 260°K. The intensity of the GaAs emission has also been observed to decrease as the temperature of the crystal is increased.

The characteristics of state-of-the art cw mode-locked Nd:YAG systems at 1064 nm are described, including their properties when simultaneously Q-switched and mode-locked. The wavelength region around 1320 nm is discussed in detail and simultaneous mode-locking of both the 1319 nm and 1338 nm lines is reported for the first time. Several examples of the laser's success as a pump source for tunable picosecond and subpicosecond lasers are given.

Design considerations are discussed for a simple, easy to use and relatively efficient high gain dye laser amplifier chain for CW mode-locked dye lasers. The amplifier boosts the output of a synchronously mode-locked dye laser to obtain ≈005 mj, ≤ 1 psec pulses over a ≈ 400 Å bandwidth. These pulses are suitable for efficient Raman Shifting, frequency mixing and continuum generation to vastly extend the spectral range of the system. Our amplifier is pumped by a frequency doubled Nd:YAG oscillator only, which longitudinally pumps three identical brewster cells with the same flowing dye solution in each. Contrary to popular belief, high small signal gains (≥ 10⁵) are easily attained in a single stage with longitudinal pumping, with better beam homogeneity and easier alignment than transverse pumping. Gain saturation measurements are presented which agree well with calculations. Factors which relax the pump timing sensitivity are examined. The importance of gain saturation for both efficient amplification and for amplitude stability is also discussed. The need for isolated amplifier stages is stressed and optimal amplifier cell areas for a given stage are calculated.

This paper reports a potpourri of recent results on mode, locking (primarily of Nd:YAG lasers and phase conjugate resonators) and on mode-locked pulse applications (to photo-acoustic spectroscopy and to laser annealing) from the author's group in the Edward L. Ginzton Laboratory at Stanford University.

A convenient method of optically exciting and monitoring coherent acoustic waves in transparent or light-absorbing liquids and solids is described. The acoustic frequency is easily and continuously tunable from ≈ 3 MHz to at least 30 GHz with our experimental apparatus and in principle over a considerably wider range. In anisotropic materials any propagation direction can be selected. The optically generated acoustic waves can be optically amplified, cancelled, or phase shifted.

We report the solution kinetics for the photo-excited, intramolecular electron transfer of malachite green leucocyanide (MGCN) to form the malachite green dye cation (MG⁺) and a cyanide anion (CN^-). By analyzing picosecond resolved fluorescence emission and transient absorbance we have identified the ionizing state as the MGCN excited singlet. The solvent dielectric constant dramatically affected the kinetics of ion formation and the ultimate yield of ionization. The MG⁺ rise kinetics are not simple first order in the first 300 ps. The kinetics show a very fast rise that is almost independent of solvent dielectric constant (E) which is then followed by a slower rise that is very dependent on dielectric constant. More experiments to refine the solvent dependence are in progress. The apparatus included a synch pumped dye laser for fluorescence lifetime analysis and an amplified, synch pumped dye laser for transient absorption analysis.

Fluorescence decay measurements using a mode-locked argon laser and single photon time-correlated detection, help us understand energy migration in photosynthesis. For intact spinach chloroplasts we find multiple exponential decays. There is some hint of a very fast component (<100 ps), but our data reveal little about it. We would expect this to be due to "photosystem I units." In addition we find two components at longer times, both of which lengthen as conditions are changed from "open" reaction centers (0.1 and 0.7 ns) to "closed" reac-tion centers (1 and 2 ns). This may reflect two kinds of organization for "photosystem II units."

A nonlinear optical microscope based on CARS using picosecond dye lasers is described. Applications to biological studies are demonstrated.

We treat picosecond laser induced damage as a non-equilibrium phase transition and propose a new damage mechanism. This model which includes energy transfer by resonant surface plasmons on small electron density droplets is shown to be consistent with existing experimental data. New experimental data has been taken which demonstrates the nucleation and growth aspects of the laser damage process.

Phenomenon of picosecond photoconductivity has gained in importance in the recent past due to ultrafast optoelectronic switching. Conductive as well as dielectric properties of the photoinduced plasma can be used to control the wave propagation depending on the wave-length of the electromagnetic waves. Possible applications of this phenomenon are enumer-ated and our recent work on multikilovolt picosecond optoelectronic switching in CdS_0.5 Se_0.5 is described in detail.

The generation of high power picosecond risetime electrical pulses that are in picosecond synchronism with optical pulses using laser-induced photoconductivity in high-resistivity semiconductors) has resulted in a number of applications. These include jitter-free streak camera operation, picosecond active pulse shaping, active prepulse suppression, and more recently the generation of picosecond microwave bursts.

Recent advances in the field of subpicosecond light pulse aeneration and amplification have opened the way to a variety of new optical devices and applications. A new type of Kerr shutter responding with a speed limited by our laser pulses (300 femtoseconds) and an opening dynamics of 10⁴ has been developed. This new component is expected to lead to more accurate results for time resolved luminescence but also for medical spectrotomography. The technique of photoconductive switching to drive in the averaqing mode a streak camera with subpicosecond pulses has been tested. Our subpicosecond laser has also been used to trigger photoconductive switches driving a dipole antenna producing a burst of microwave of duration of 3 psec.

The present work uses picosecond optical pulses from a synchronously pumped modelocked dye laser to switch pulse-forming transmission line networks. Transfer efficiency of over 30 percent and pulse width of 50 ps has been obtained using a switching junction fabricated on an Fe-doped semi-insulating InP substrate. Propagation of these short pulses are subject to dispersion along the microstrip lines. Numerical solution using the Fast Fourier. Transform is performed in the time domain to visualise the deformation of the pulses as they propagate along the microstrip line.

Two devices based on picosecond photoconductivity are reported. One is a germanium on sapphire device with a response time (FWHM) faster than 50 ps and a sensitivity in excess of 0.1 ma/mW, and is suitable for use when ultra-high speed gating of electrical signal is required. The other is an optically triggered GaAs transferred-electron device that generates oscillation bursts of controllable duration and frequency over the ranges of 1.5 to 4.0 nsec and 6.7 to 10.5 GHz, respectively.

The electron-optical streak camera is one of the few instruments capable of directly measuring a picosecond optical signal both linearly and in real time. The goal of this research has been the development of a synchronously operating streak camera system suit-able for use with mode-locked dye laser sources. The major accomplishments of this research are threefold: 1) A complete single-event streak camera system was constructed. The laser source was a mode-locked dye laser and amplifier system producing picosecond pulses with gigawatt powers. The camera was based on a Photochron II streak tube and microchannel plate intensifier. An optical Multichannel Analyzer was used to generate a real-time digital and graphic representation of the picosecond optical event. The system provides a powerful tool for future research in picosecond high excitation and nonlinear processes. 2) A synchronously operating streak camera system was constructed for use with an actively mode-locked R6G dye laser producing tunable picosecond pulses at a repetition rate of 82 MHz. The streak camera was operated at the laser repetition rate by employing a sinusoidal high-voltage RF deflection signal. The high data-accumulation rate was shown to yield a substantial increase in sensitivity and dynamic range over conventional single-event streak cameras. Time resolution was ≈25 ps and appeared to be limited by laser jitter rather than by the streak camera technical time resolution. 3) The viability of the synchronous streak camera for picosecond spectroscopy of solids was demonstrated by performing the first direct time-resolved measurement of bound exciton luminescence in CuCl. Single crystals at 8°K were excited by the frequency-doubled dye laser. The lifetime of the bound exciton of I line was typically ≈130 ps with a formation time of ≈ 10 ps. Excellent signal-to-noise ratios were obtained with excitation energies even as low as 10^-13 J per pulse.

The use of subpicosecond passively mode locked dye lasers, combined with nonlinear detection is shown to open the field of "range gating" to microscopic dimensions. We present a one dimensional application of this technique to the evaluation of fiber connectors.

Gigabit coherent optical pulse generations in injection synchronized sinusoidally modulated laser diodes as well as in multiple bit FM mode-locked miniature solid state LiNdP₄O₁₂ (LNP) lasers are investigated. Bandwidth-limited 40-50 psec optical pulses at a repetition rate up to 2-3 GHz were obtained. Experimental results on grating of picosecond optical pulses with a wideband LiNbO₃ directional coupler modulator are shown.

Transient recombination kinetics of photogenerated elementary excitations in semiconductors can be studied with subpicosecond time resolution. This note describes the different time resolved luminescence techniques which are applicable to semiconductors, such as the optical Kerr gate, the up-conversion gate, the streak camera, the three-pulse excitation method and the population mixing method.

The dynamics of electron localization in amorphous media are of considerable interest to fundamental studies of electron transport as well as to applications in picosecond optoelectronics. Since efficient photoionization at 10^-12 s usually requires at least Mwatt power densities, we describe two complementary routes to obtaining ultrafast, high power laser pulses. We first report some novel applications of the optical Kerr effect in CS₂ for picosecond pulse generation in ruby and pulse gating on the femtosecond time scale, and then describe an amplified, subpicosecond, synchronously mode-locked tunable dye laser system. Recent results on photoionization and electron trapping in liquids and clusters are then discussed.

We describe direct observations on a picosecond time scale of NO₂ fragment formation from 264 nm photolysis of low-pressure nitromethane. Using laser induced fluorescence probing, we find that the population of ground-state NO₂ products rises and saturates within <5 ps, with a measured yield of near unity. The experiments were conducted with a newly developed mode-locked Nd:phosphate glass oscillator/amplifier system which can generate energetic short pulses of high beam quality at repetition rates of 1/5 Hz.

Predissociation and recombination kinetics of Iodine in both liquid (7°C) and fluid (19°C) Xenon have been studied by picosecond absorption spectroscopy. The results indicate lifetimes of ≈14 ps and ≈40 ps for predissociation and geminate recombination, respectively. Transient spectra of I₂/Xe and I₂/CC1₄ were also obtained, indicating the existence of a transient Iodine-solvent complex in CC1₄ that was not observed in Xenon. The kinetics of the prediss ciation of diazene compound and the spectroscopy of the resulting biradical have been studied by picosecond fluorescence spectroscopy. The data indicates diazene 2 has a lifetime of 38 ps and the singlet biradical 1S has a lifetime of approximately 280 ps. These results and the fluorescence spectrum are discussed in terms of known chemistry and theoretical calculations.

We report results of recent measurements of excited state decay kinetics in molecules which are commonly used as commercial polymer photostabilizers. Discussion centers on molecules in the 2-(2'-hydroxypheny1)-benzotriazole and 2-hydroxybenzophenone classes. Determination of room temperature fluorescence decay, transient absorption and ground state absorption recovery kinetics are presented. In addition, low temperature spectra and variable temperature fluorescence decay kinetics are also presented. Solvent effects on the spectra and kinetics give considerable insight into the radiationless processes that occur in these systems.

A rate theory and its application to liquid state chemical reactions on the picosecond timescale are outlined. A new feature of picosecond reaction dynamics in solution is the strong participation of solvent molecules. They mainly get in the way of an otherwise rapid combination of reactants. The displacement or reorganization of solvent molecules during a reaction may require hundreds or even thousands of picoseconds, particularly when the reactants contain electrical charges. The rate equations presented here include these solvent displacement rates. Solution of the rate equations for light induced reactions are obtained for continuously applied light and for light in the form of picosecond pulses. The former solutions provide quantum yields, while the latter give dynamical information. Numerical input to these solutions show that experiments should be quite sensitive to solvent displacement rates. Apparently overlooked by photochemists is the great sensitivity of quantum yield data to these rate parameters. Comparison of the theoretical results with a few experimental data will be made.

There is now a four order of magnitude time range (≈100 fs to 1 ns) over which molecular dynamic calculations can overlap short light pulse experimental measurements to probe the microscopic nature of chemical processes. In this paper we combine theoretical molecular dynamics calculations of picosecond transient spectra with experimental measurements of such spectra to probe the molecular dynamics of a chemical reaction in solution. The example illustrated is the photodissociation of Ι₂ followed by solvent caging, radical recombination to form a new highly vibra-tionally excited Ι₂ molecule, and the subsequent decay of this vibrational energy to the solvent. We suggest, as Nes-bitt and Hynes do also, that the lifetimes observed in Ι₂ picosecond absorption experiments may not be due to gem-inate recombination times, but perhaps to the time necessary for the recombined Ι₂ molecules to vibrationally decay to levels with high Franck-Condon factors for absorption.