recent advances in ultrafast-laser-based spectroscopy and ... · topical review recent advances in...

Topical Review

Recent advances in ultrafast-laser-basedspectroscopy and imaging for reactingplasmas and flames

Anil K Patnaik1,2, Igor Adamovich3, James R Gord4 and Sukesh Roy1,5

1 Spectral Energies, LLC, 4065 Executive Dr, Dayton, OH 45430, United States of America2Department of Physics, Wright State University, Dayton, OH 45435, United States of America3Department of Mechanical and Aerospace Engineering, Ohio State University, Columbus, OH 43210,United States of America4Air Force Research Laboratory, Aerospace Systems Directorate, WPAFB, OH 45433, United States ofAmerica

E-mail: [email protected]

Received 19 July 2016, revised 10 July 2017Accepted for publication 10 August 2017Published 25 September 2017

AbstractReacting flows and plasmas are prevalent in a wide array of systems involving defense,commercial, space, energy, medical, and consumer products. Understanding the complex physicaland chemical processes involving reacting flows and plasmas requires measurements of keyparameters, such as temperature, pressure, electric field, velocity, and number densities of chemicalspecies. Time-resolved measurements of key chemical species and temperature are required todetermine kinetics related to the chemical reactions and transient phenomena. Laser-based,noninvasive linear and nonlinear spectroscopic approaches have proved to be very valuable inproviding key insights into the physico-chemical processes governing reacting flows and plasmasas well as validating numerical models. The advent of kilohertz rate amplified femtosecond lasershas expanded the multidimensional imaging of key atomic species such as H, O, and N in asignificant way, providing unprecedented insight into preferential diffusion and production of thesespecies under chemical reactions or electric-field driven processes. These lasers not only provide2D imaging of chemical species but have the ability to perform measurements free of variousinterferences. Moreover, these lasers allow 1D and 2D temperature-field measurements, whichwere quite unimaginable only a few years ago. The rapid growth of the ultrafast-laser-basedspectroscopic measurements has been fueled by the need to achieve the following whenmeasurements are performed in reacting flows and plasmas. They are: (1) interference-freemeasurements (collision broadening, photolytic dissociation, Stark broadening, etc), (2) time-resolved single-shot measurements at a rate of 1–10 kHz, (3) spatially-resolved measurements, (4)higher dimensionality (line, planar, or volumetric), and (5) simultaneous detection of multiplespecies. The overarching goal of this article is to review the current state-of-the-art ultrafast-laser-based spectroscopic techniques and their remarkable development in the past two decades inmeeting one or all of the above five goals for the spectroscopic measurement of temperature,number density of the atomic and molecular species, and electric field.

Keywords: ultrafast, plasma, multiphoton imaging, CARS spectroscopy, TALIF

Plasma Sources Science and Technology

Plasma Sources Sci. Technol. 26 (2017) 103001 (41pp) https://doi.org/10.1088/1361-6595/aa8578

5 Author to whom any correspondence should be addressed.

0963-0252/17/103001+41$33.00 © 2017 IOP Publishing Ltd1

mailto:[email protected]

https://doi.org/10.1088/1361-6595/aa8578

http://crossmark.crossref.org/dialog/?doi=10.1088/1361-6595/aa8578&domain=pdf&date_stamp=2017-09-25

http://crossmark.crossref.org/dialog/?doi=10.1088/1361-6595/aa8578&domain=pdf&date_stamp=2017-09-25

1. Introduction

Research and development in reacting plasmas and flames isdiverse and interdisciplinary [1–3]. Over the past decades, thecontinuously evolving field of plasmas has taken a number ofleaps forward in the development of many new plasmatechnologies associated with defense, commercial, space,energy, medical, and consumer products. The parameter spaceof these plasmas ranges from low pressure (millibars) and lowtemperature [4] to extremely high pressure (up to gigabars)and high temperature (up to millions of K) [5]. The complexreacting plasma and flow environments involve a number ofparameters that depend on their applications, which rangefrom plasma-assisted combustion [6] to materials processing[7] to biomedical applications [8] to plasma reactors in fusion[5]. To characterize and understand the underlying physicsand chemistry of these plasmas, it is vital to measure directlykey performance parameters such as temperature, pressure,velocity, number densities of reacting species, and electricfield. Laser-based, noninvasive, linear and nonlinear spec-troscopic methods have proved to be valuable in providinginsight into key physico-chemical processes that governreacting plasmas. The limited spatial and temporal resolutionof the conventional diagnostics technologies has preventedcomplete understanding of the plasma interactions [2],thereby limiting innovation and complicating product mod-ification of plasma-based systems [9]. Plasma sources usedfor various applications are discussed below.

1.1. Brief discussion of plasma sources for various applications

Over the past two decades, extensive experimental and kin-etic-modeling studies have been conducted in nonequlibriumplasmas over a wide range of experimental conditions [2, 10].These efforts were prompted by a variety of potential appli-cations such as plasma-flow control [11–13], plasma-assistedcombustion [6, 14, 15], plasma-induced liquid-phase chem-istry [16, 17], and plasma medicine [8, 18, 19]. The range ofnonequlibrium plasma sources used for these applications isremarkably wide and includes numerous configurations ofvolumetric and surface dielectric-barrier discharges sustainedby AC waveforms and short (nanosecond (ns)-duration)pulses, atmospheric-pressure plasma jets sustained by DC,AC, RF, and microwave waveforms, and arrays of micro-plasmas generated by microhollow-cathode DC discharges[20], to name a few. A detailed overview of these plasmasources can be found elsewhere (e.g. see [2, 6, 8, 11–19] andreferences therein) and is outside the scope of the presentwork. The focus of the present review is mainly on highly-transient plasmas such as those generated in streamers (seefigures 1 and 2) [21, 22], high-peak-voltage ns-pulse-durationdischarges (both volumetric and surface, see figures 3 and 4)[23, 24], and atmospheric-pressure plasma jets and plasmabullets (see figure 5) [25]. These discharges are known toexhibit highly complex behavior on short timescales. Recentadvances in both experiments and kinetic modeling in thisrapidly developing field can be found in a special issue onfast-pulsed discharges (see [26] and references therein).

One of the difficulties in understanding the kinetics ofthese plasmas is the lack of noninvasive, spatio-temporallyresolved measurements for their characterization. To beginwith, rapidly varying distribution of the electric field in theseplasmas controls the electron-energy distribution function(EEDF) and energy partition among the internal energymodes of atoms and molecules during electron-impact pro-cesses. At these highly-transient conditions and in the pre-sence of strong spatial gradients (such as in streamer-dominated plasmas), the EEDF may also be nonlocal, whichsignificantly complicates the plasma kinetics. These dominanteffects control the number densities of a variety of excitedspecies and radicals, the kinetics of molecular-energy transfer,and the energy-thermalization rate. Quantitative insight intothese processes requires spatio-temporally resolved mea-surements of density and temperature of electron, electricfield, gas temperature, and species number densities in theplasma using nonintrusive, in situ experimental techniques.

The next subsection introduces the current state-of-the-artdiagnostics and the diagnostic needs for reacting plasmasand flows.

Figure 1. Images from different types of pin-to-plane discharges inatmospheric-pressure air for voltage-rise rates between 6 and26 kV cm−1 and peak voltages between 40 and 67 kV. Camera gateis 15 ns. Pin electrode is on the top. Discharge gap is 20 mm. (a)Glow-like regime. (b) Multi-channel regime. (c) Diffusion regime.Reproduced from [21]. © IOP Publishing Ltd. All rights reserved.

Figure 2. Left: Broadband plasma-emission images from a pulseddielectric-barrier discharge (DBD) during voltage-pulse rise and fall(t10%–90%=45 ns, ICCD camera gate 150 ns) in 0.1% O2–N2

mixture at 1 atm. Right: streak camera images of broadband plasmaemission of second-positive system of N2 (0–0 transition at 337 nm)under these conditions. Reproduced from [22]. © IOP PublishingLtd. All rights reserved.

2

Plasma Sources Sci. Technol. 26 (2017) 103001 Topical Review

1.2. Diagnostic needs for reacting plasmas and flames

For most of the plasmas discussed above, only a very smallfraction of the neutral species at the target volume are ionized(see figure 6 in [2]), except for extreme conditions encounteredin nuclear-fusion reactors [5]. Hence, in most cases the dominantspecies in the plasmas are neutral atoms and molecules,including vibrationally and electronically excited species andhighly reactive radicals. Depending on specific energy loading,mixture composition, temperature, pressure, flow conditions, theelectron-impact process, as well as energy transfer amongexcited species results in a wide range of plasma-chemicalreactions. A common interest is shared in the research anddevelopment of many plasma applications to understand thespatial distribution and temporal evolution of key intermediateplasma species, including electrons, ions, metastable excitedspecies, and highly reactive atoms and radicals [10]. Many of theexcited species generated by electron impact and molecularcollisions within the plasma decay spontaneously to emit

radiation. Hence, the passive detection of the emitted light byoptical emission spectroscopy is the simplest form of plasmadiagnostics. However, this technique does not provide quanti-tative data unless the excitation and de-excitation mechanisms ofthe plasma are known exactly, such as those in well-calibratedequilibrium plasmas and flows [27]. However, most of therecent applications discussed in the earlier section are highly-transient and nonequilibrium in nature with rapidly changingreacting intermediates, chemical products, and their chemicalkinetics. It is particularly important to quantify concentrations ofthe intermediates such as atomic neutrals, e.g. hydrogen, oxy-gen, nitrogen, and carbon. These intermediates are highly reac-tive and can be generated in significant concentrations to affectthe macroscopic plasma conditions. Thus, absolute concentra-tion measurements of important intermediate plasma species innonequilibrium plasmas provide valuable insight that aids theunderstanding of the physical and chemical nature of theseplasma systems and facilitates the development of model-basedpredictive capabilities.

Figure 3. Applied voltage–pressure diagram that separates regions ofquasi-uniform diffusive surface discharge (under curve) andfilamentary surface discharge (above curve) in synthetic air fornegative pulse polarity. Reproduced from [23]. © IOP PublishingLtd. All rights reserved.

Figure 4. Plasma-emission images from volumetric ns-pulsedischarge in nitrogen, sustained in plenum of nonequilibrium plasmain small-scale, Mach-5-blowdown supersonic wind tunnel (350 Torrpressure). Pulse peak voltage is 25 kV; pulse-repetition rateν=100 kHz; pulse #1000. Electrodes shown schematically bydashed lines. Reproduced with permission from [24].

Figure 5. High-speed broadband emission images of nonequilibriumplasma bullet generated in effluent of 20% O2–He atmospheric-pressure plasma jet. Voltage-pulse-repetition rate is 10 kHz.Pulsewidth is 500 ns. Peak voltage is 9 kV. Flow rate is 0.4 L min−1.Reproduced from [25]. © IOP Publishing Ltd. All rights reserved.

3


In the review article entitled ’The 2012 Plasma Roadmap’,Czarnetzki explicitly outlined the limitations in the currentplasma-characterization capabilities and described the need fordiagnostics and model development for different types ofplasmas [2]. For quantifying the intermediates in reactingplasma and flow species, ideal diagnostic techniques should benonintrusive, in situ, and species-selective, with a largedynamic range to permit the detection of short-lived reactivespecies with low densities, while avoiding saturation effects athigher densities. Conventional continuous-wave diode-laser-and ns-pulsed-laser-based spectroscopic techniques exhibitfundamental limitations, despite their ability to provide non-intrusive, in situ, and species-selective measurement platforms.For example, line-of-sight absorption-spectroscopy techniquesoften lack sufficient spatial resolution. Laser-induced fluores-cence (LIF) is generally based on single-photon absorption andoffers high spatial resolution. However, because of relativelystrong absorption cross-sections, large concentrations canprove to be optically thick, resulting in significant probe-beamattenuation or stimulated-emission (SE) effects [28]. In addi-tion, many key intermediates such as atomic hydrogen, oxy-gen, and nitrogen have a large energy spacing between theground and excited electronic states. Such large energy gapsrequire high single-photon energies with wavelengths in thevacuum-ultraviolet (VUV) region, which are relatively difficultto generate and pose significant problems for propagationthrough air. Furthermore, the laser-based, linear optical tech-niques such as laser-induced emission and absorption in plas-mas and flows require extensive dependence on models forcomplex-flow characterization; still, such techniques yield onlylimited spatial and temporal resolution [10, 29]. Most state-of-the-art plasma diagnostics are limited to emission or line-of-sight absorption spectroscopy, which yields mainly qualitativeinformation or time-averaged and spatially-averaged quantita-tive information that is suitable for equilibrium plasmas only[30]. Employing calibration techniques such as Rayleighscattering with inert gases or with LIF of a mixture with knownconcentration of the target molecule and extensive modeling ofthe collisional processes, spatially-resolved absolute con-centration measurements of radical species such as OH and NOemploying LIF- and planar-LIF (PLIF) have been demon-strated in different plasma conditions [31–35]. However,ns-laser-based LIF is still limited to lower plasma densities,because of strong single-photon resonant absorptions [27].

In nonequilibrium reacting plasmas, the rapidly evolvingphysical and chemical processes that involve many reactingintermediate species, such as O, N, and H, strongly influencethe outcome of the plasma process [10]. To address the above-mentioned limitations with linear optical methods, multiphotonexcitation has been developed and employed. Multiphotonapproaches offer two significant advantages: (1) red-shiftedwavelengths (away from the VUV wavelengths) allow beampropagation with minimal absorption in air, and (2) smallerabsorption cross-sections permit high-resolution reacting-spe-cies measurements at higher concentrations. Two-photon-absorption LIF (TALIF) was first demonstrated for atomichydrogen and deuterium by Bokor et al [36] and has since beendramatically expanded to the detection of many ground-state

atomic species. Traditionally, ns-laser systems have beenemployed to probe TALIF transitions utilizing numerousexcitation schemes. However, the relatively high fluence oftypical ns-laser pulses that are used to overcome the relativelysmall multiphoton-absorption cross-sections can result in sig-nificant interference from photodissociation and photoioniza-tion within the medium, resulting in substantial photolyticinterference of the TALIF signal. For example, photo-dissociation of O2 could create an O atom, which wouldinterfere with the chemically produced O atom in the TALIF-based concentration measurement [27]. Similar interference isalso prevalent in reacting flames where methyl (CH3) and water(H2O) can photodissociate to produce an H or O atom and,hence, create significant interference during the probing pro-cesses [37–40]. A partial suppression of this photolytic inter-ference was reported for H- and O-atom TALIF schemes usingps-duration pump pulses [41], emphasizing the potential ofultrafast-pulse excitation in reducing or removing photolyticinterferences during the measurement of atomic or molecularspecies via TALIF. The advantage of moving towards a shorterpulse is that higher irradiance can be realized with orders-of-magnitude lower fluence to achieve a similar level of signal-to-noise ratio (SNR) and, therefore, interference-free single-shotimaging capabilities can be extended to higher dimensions.Section 3 contains a detailed discussion of how ultrafast-laser-based interference-free TALIF detection has significantlyexpanded measurement capabilities.

Furthermore, a variety of promising nonlinear opticaltechniques, such as four-wave mixing (FWM) and coherentanti-Stokes Raman scattering (CARS), have been developedin the past four decades and extensively used for thermo-metric studies in nonequilibrium plasmas [42] and flowdiagnostics [28]. Such FWM techniques generally providehigher spectral and spatial selectivity than the linear spec-troscopic techniques. However, most of those efforts werecarried out through the use of ns-duration laser pulses [43].Typically, ns-laser-based CARS causes an array of signalinterferences and other limitations such as: (a) contributionfrom nonresonant background (NRB) that limits the accuracyand degrades the sensitivity of the technique, (b) dependenceof the signal on the collisional environment via collisionaldephasing, where measurements are strongly dependent onthe accuracy of the collisional models, (c) inability to respondto the dynamical changes of temperature and number den-sities of reacting species in highly-transient reacting plasmasand flows, because of the low repetition rate of the laser, and(d) insufficient intensity to perform measurements along a lineor in a plane. During the past decade, it has been demon-strated that ultrafast-laser-based CARS has the potential toovercome the shortcomings generally encountered withns-CARS [3], which is discussed in detail in section 4.

Remarkable advancements in kilohertz-rate amplifiedfemtosecond (fs) lasers and ultrafast nonlinear optical techni-ques have significantly expanded the diagnostics capability forreacting plasmas and flames [3]. Employing fs-laser-basedTALIF, photolytic-interference-free multidimensional imagingof the key reacting species H, O, N, OH, and CO has beendemonstrated, providing unprecedented insight into the

4


production and preferential diffusion of these species in elec-trically- or chemically-driven processes [44, 45]; fs-TALIF isdescribed in detail in section 3. In addition, high-power ultra-fast lasers have enabled NRB-free, single-shot, multi-dimensional temperature measurements at higher than kHzrepetition rates, which was quite unimaginable only a few yearsago (see section 4). The intense growth in ultrafast-laser-basedspectroscopy during the past decade has been driven by thefollowing goals: (1) interference-free measurement, (2) time-resolved single-shot measurement for dynamic environments,(3) spatially-resolved measurement, (4) increased dimension-ality, and (5) measurement of multiple species simultaneously.Throughout this article, the paths followed to reach each of theabove goals of ultrafast-diagnostics development for reactingplasmas and flows will be discussed.

1.3. Background physics of the effect of laser duration on laser-plasma interactions

Laser-based diagnostics of plasma relies on the interaction ofthe laser with ions, electrons, molecules, and atomic species atthe target volume of the plasma. The specific approach dependson the species under consideration, environmental conditions(temperature, pressure, neighboring species), and laser para-meters (laser energy, pulse duration, and bandwidth). While theimpact of the laser parameters and other environmental con-ditions will be discussed throughout this review, in the fol-lowing we present a brief overview of various laser parametersof interest related to a range of pulse durations.

The laser parameter that is often used to quantify LIF andPLIF excitation is laser fluence (F), which is defined as time-integrated photon flux or laser energy (ε) delivered per unitarea (A); i.e. F=ε/A and typically expressed in terms of Jcm–2. Typically, LIF/PLIF signal scales with the fluence in thelinear regime (i.e. below saturation). However, with ps and fsexcitations, the fluence-based description fails to explain thelaser-plasma interaction, because of the absence of the pulseduration in the underlying formulation. In the short-pulseregime, the underlying physics is governed by the proximity ofthe laser timescales with respect to the collisional timescales.

Furthermore, laser pulse with electric field E t

( ) coher-ently interact with the plasma species with dipole moments(di

) that determine the Rabi frequency t d E ti i W = ⋅

( ) ( )/

that defines the rate at which the coherent population transfertakes place within the plasma species at a given instant t; here,

h 2 p= / and h is Planck’s constant. Laser intensity is squareof the electric field, which is defined as follows:

I t c E t2 . 1.102e=

( ) ∣ ( )∣ ( )

Here, c is the speed of light in free space, and 0e is thevacuum permittivity. Nonlinear interactions such as two-photon and multiphoton processes are governed by cycle-averaged laser irradiance or intensity (Iav)=p/A, rather thanthe laser fluence. The laser parameter that captures the pulseduration is average laser power pav

1 2e p t=( ) ( )/ / (assuminga Gaussian pulse shape); here τ is the duration of the pulse. Inthe ultrafast regime, the electrical energy is packed tightlywithin the short durations and hence the peak electric fields

are extremely high and have the ability to saturate themolecular transitions with the population oscillating at Rabifrequency Ωi.

For a typical Nd:YAG-based ns laser with duration of8 ns, laser energy of 2 mJ, beam diameter of 50 μm, laserintensity is 7.2×109W cm−2; whereas, for a Ti:Sapphire-based fs-laser with energy of 6 μJ per pulse, pulse duration of100 fs, and beam diameter of 50 μm, laser intensity is1.7×1012W cm−2. Clearly, with orders-of-magnitude lessenergy, shorter pulses can be highly efficient for excitation ofthe plasma species producing stronger diagnostics signal.

In terms of the spectral properties, fs lasers are associatedwith broad bandwidth (∼150 cm−1 for a ∼100 fs-durationpulse) to ultra-broad bandwidth (>1000 cm−1 for a ∼10 fs-duration pulse), whereas bandwidth of a 10 ps pulse is∼2 cm−1. In typical plasma species, e.g. in molecular N2, theground rovibrational states are separated by ∼8 cm−1. Hence,fs-laser pulses couple many molecular levels simultaneouslyallowing single-shot spectral measurement of measured spe-cies, whereas ns and ps lasers generally couple only onetransition at a time, because of their narrow bandwidth.Hence, ns or ps-laser frequencies are scanned to obtain thecomplete spectra of the target species, or broadband modelesslasers are used to obtain single-shot spectra. Furthermore, fs-laser-based diagnostics measurements are inherently collisionindependent if the detection time is short, since typical col-lisional timescales at atmospheric pressure are a few tens of psto a few hundred ps. From the diagnostics perspective, fslasers have a myriad of other advantages that are discussed indetail in sections 3.1 and 4.

The overarching goal of this review is to discuss currentstate-of-the-art ultrafast-laser-based spectroscopic techniques andtheir remarkable development over the past two decades inmeeting at least one or all of the above five goals for the mea-surement of electric field, temperature, number density of theatomic, molecular, and plasma species. New avenues thatresulted from a detailed understanding of plasmas and that aidedthe validation of plasma models will also be highlighted. Futureopportunities of ultrafast diagnostics for characterizing andunderstanding extreme plasma conditions will also be discussed.

2. Current state-of-the art laser-based approachesfor plasma characterization and technicalchallenges

As outlined earlier, most of the current state-of-the-art optical-diagnostic techniques for plasma characterization rely on lightscattering from the reacting plasmas excited mainly by laserpulses with ns duration. Since the development of highly-transient plasmas are sustained by ns-duration voltage pulses,the use of ultrashort-pulse (picosecond (ps) and fs-laserdiagnostics has considerable advantages, as discussed earlier.Ultrashort laser pulses can induce a small optogalvanic effectthat influences the electron density in the plasma via theinverse Bremsstrahlung process. However, the ultrashortpulses prevent the generation of cascaded ionization andlaser-induced breakdown since the pulsewidth of the laser is

5


shorter than the collisional timescales at atmospheric pressureor higher [46]. Thus, the nonlinear diagnostics (CARS, FWM,and second-harmonic generation (SHG)) employing ultrashortpulses are typically independent of laser-induced breakdown.In the following, we briefly discuss the current state-of-the-artspectroscopic methods used for the measurement of the keyplasma parameters such as electric field, temperature, con-centration of plasma species, and flow velocity.

2.1. Spectroscopic methods for electric-field measurements

The rapid development of ultrashort-pulse laser diagnosticshas made it possible to address several critical challenges inthe characterization of highly-transient plasmas. One suchchallenge concerns the measurement of the spatio-temporaldistribution of the electric field in high-peak-voltage pulseddischarges with ns and sub-ns voltage-rise times. Thesemeasurements are essential for understanding the kinetics ofionization and charge transport in these plasmas, since rapidlyvarying electric field and strong spatial gradients (e.g. in fast-ionization wave discharges and streamer discharges) mayhave a significant effect on the EEDF. In this case, the pre-diction of both the EEDF and the electron-impact-rate coef-ficients requires the use of advanced theoretical and modelingapproaches such as Monte Carlo–particle-in-cell [47] than thewidely used two-term expansion solution of the Bolzmannequation for plasma electrons [48]. Experimental data underthese conditions are critical for the validation of these mod-eling approaches.

2.1.1. Emission spectroscopy. The use of optical-emissionspectroscopy for reduced electric-field interference (e.g. basedon the comparison of nitrogen second-positive and first-negative emission-band intensities [49]) is affected by twomajor uncertainties that are rather difficult to quantify. First,the emission-spectroscopy signal is averaged over the line ofsight. In many high-pressure air plasmas such as streamersand surface dielectric-barrier discharges, optical-emissionspectra are collected across a region with very steep spatialgradients, which necessitates the incorporation of the line-of-sight integration procedure into the data analysis andmodeling predictions. This also makes it a challenge todetermine whether the two emission bands originate from thesame spatial location, which may have a significant effect onthe electric-field value inferred from the emission-intensityratio. Second, the procedure of electric-field inference istypically based on the steady-state solution of the two-termexpansion of the Boltzmann equation for the symmetric partof the EEDF [48]. The applicability and accuracy of thisapproach under the conditions of strong temporal and spatialgradients that are typical for ns-pulse streamers and surfacesare highly uncertain. The higher-order expansion of theBoltzmann-equation solution and Monte Carlo simulationsare necessary to estimate the effects of temporal and spatialelectron-energy relaxation on the EEDF [50].

2.1.2. Pockels probe. Another electric-field measurementtechnique based on the use of an electro-optic (Pockels effect)

probe has been employed recently. Utilizing the fact that therefractive index is proportional to the applied electric field[51], Pockels probe enabled simultaneous measurements oftwo electric-field-vector components in a pulsed dielectric-barrier capillary discharge in helium [52]. In this approach,however, the electric field is typically measured outside theplasma, because the insertion of the probe within the plasmaperturbs the electric-field distribution [53].

2.1.3. FWM. A nonintrusive laser-diagnostic method ofmaking electric-field measurements in high-pressuredischarges is the FWM technique [53], which is similar toCARS (see section 4 for more details). Briefly, in this method,the probe beam used in CARS is replaced by an externallyapplied electric field that acts as a ‘zero-frequency’ probewave, generating a coherent signal beam with the wavelengthcorresponding to the energy difference between the ground andthe first excited vibrational state of the molecules [42]. Similarto CARS, the intensity of the IR beam generated from FWMscales quadratically with the electric field and pressure. Theduration of the IR signal, which limits the temporal resolutionof this method, depends on the coherence-dephasing time ofthe molecules excited by the pump and Stokes beams, typicallyof the order of a few hundred ps. Thus, this approach isespecially effective for electric-field measurements in high-pressure discharges, such as volumetric and near-surfacedielectric-barrier discharges that are sustained by high-peak-voltage ns-duration pulses. Most FWM measurements inplasmas have been conducted on H2 or N2 as a probespecies using ns-laser systems [54–59]. The use of ns-laserlimits the time resolution of this approach and FWM signalintensities are also typically poor, limited by the intensity of thelaser. Significant enhancement of the signal compared to the nsapproach is achieved when the FWM signal is produced usingshorter duration (ps) lasers. Higher signal allows the electric-field measurements of individual electric-field components[60–63] with improved SNR and improved sub-ns temporalresolution [64]. Furthermore, time resolution in the repetitivelypulsed-discharge experiments is limited by the voltage-pulse‘jitter’ (typically of the order of ∼1 ns). Higher temporalresolution is achieved by saving the relevant waveforms(voltage, current, pump laser, and CARS signal) for each lasershot and using short-duration ‘time bins’ (of the order ofcoherence decay time, ∼200 ps) during data post-processing[60]. Although measurements in atmospheric-pressure airplasmas using nitrogen as a probe species are considerablymore challenging, because of nitrogen’s lower Raman cross-section compared to other plasma species in air [58, 62], theyhave significant potential for characterizing atmospheric-pressure air plasmas (e.g. see figure 6), with applications inplasma aerodynamics, biology, and medicine as well as high-pressure fuel–air plasmas for plasma-assisted combustion.

2.1.4. SHG. Recently, another nonintrusive laser-diagnosticmethod for electric-field measurements was demonstratedusing SHG by a fs-laser beam passing through an electrostaticelectric field in atmospheric air (see figure 7) [65]. This

6


method is based on the asymmetry in polarizability that isinduced by an applied electric field that allows theobservation of SHG, which is otherwise forbidden inhomogeneous media. The use of an fs-laser with arelatively low pulse energy (∼500 μJ, ∼50 fs pulseduration) considerably increases the SHG signal intensityand prevents laser-induced breakdown. One of the principaladvantages of this method is that, unlike FWM, it is species-independent and can be used in many different gas mixtures.Also, it is possible that a detectable SHG signal induced bythe electric field can be generated using a ps laser with pulseenergy of a few tens of mJ and pulse duration of a few tensof ps.

The electric-field measurements discussed above havebeen conducted in well-characterized, repetitively pulsed, ns-duration discharges (or repetitively pulsed electrostaticelectric fields), with FWM signal accumulated over multiplepulses (of the order of a few hundred). In dielectric-barrierdischarges and atmospheric-pressure plasma jets, sustained bymore conventional AC waveforms rather than ns-durationpulses, an additional difficulty arises, since these dischargesare often generated as an ensemble of micro-discharges that isdistributed randomly over space and time. In this case, themeasured electric field would represent a value averaged overthe laser-beam path unless plasma self-organization occurs[61], forming a reproducible surface-charge distribution andelectric-field pattern [66]. Electric-field measurements in‘single-shot’ discharges with relatively poor shot-to-shotreproducibility (such as isolated streamers or sparks), mayrequire discharge guidance by an auxiliary ultrashort laserpulse [67]. Finally, the spatial resolution of the FWM andSHG methods that employ collinear phase-matching geome-tries may be fairly low, up to several centimeters along thedirection of the laser beam unless a different phase-matchingscheme is used [62].

2.2. Measurement of electron temperature and electron density

The second challenge encountered in the characterization ofhighly-transient plasmas is making nonintrusive, time-resolved,spatially-resolved measurements of electron density and elec-tron temperature and, possibly, EEDF. These measurementsare critical for quantitative insight into the electron kinetics anddischarge-energy partition both in ns-pulse discharges and inmore commonly used dielectric-barrier discharges as well asatmospheric-pressure plasma jets driven by AC, RF, andmicrowave waveforms. Again, the use of emission spectrosc-opy for the electron-density inference from Stark spectral-linebroadening is hindered by the uncertainty in spatial resolutionand by the relatively low temporal resolution. Thomson scat-tering, which is used for electron-density and electron-temp-erature measurements in nonequlibrium plasmas, is limited bylow SNR that requires signal accumulation over a large numberof laser shots in steady-state or repetitively pulsed experiments(∼105 laser shots at electron density of ∼1011 cm−3 [68],∼103–104 laser shots at electron densities of 1014–1015 cm−3

[69–71]). In addition, the SNR and spectral resolution shouldbe sufficiently high to permit the deconvolution of overlappingThomson and Raman scattering in high-pressure gas mixtures[69] (e.g. see figure 8). Since the Thomson-scattering signal isproportional to the laser fluence (i.e. the total number of pho-tons passed through the probe volume), the use of ultrashort-pulse lasers would have a rather limited advantage over themore widely used ns lasers. On the other hand, narrow-line-width burst-mode lasers (BMLs) that are operated at high pulseenergies and high-repetition rate would offer a significantadvantage over lasers operated continuously at pulse-repetition

Figure 6. Time-resolved electric field measured in a negative polarityns-pulse discharge between a razor blade high-voltage electrode anda quartz plate in ambient air, ∼100 μm below the electrode.Reproduced from [62]. © IOP Publishing Ltd. All rights reserved.

Figure 7. Normalized fs localized electric-field measurement signalfor different species, measured at atmospheric pressure in a uniformelectrostatic field, yielding a quadratic calibration curve. Reprintedwith permission from [65], Copyright (2017) by the AmericanPhysical Society.

7


rates ranging from a few tens of hertz [68] to a few kilohertz[69–71]. As an example, a recently developed custom-designedlaser system generates ∼100 ps duration, 0.15 cm−1 pulseswith pulse energies of up to ∼500 mJ/pulse, pulse-repetitionrate of ∼10 kHz–1MHz, and ∼100–10 000 pulses/burst (totalburst energy of ∼100 J) [72]. The use of these lasers forThomson scattering would considerably improve the accuracyof the data collected, which would not be affected by thevariation of plasma conditions on a longer timescale. In addi-tion, these lasers are very well suited for nonlinear frequencyconversion, because of their inherently high-peak intensity.Their frequency-conversion efficiency (exceeding 50%) ismuch higher than that of commercial-grade systems and isessential for FWM and CARS measurements, where two orthree separate frequency-conversion processes may be utilizedfor the generation of the necessary laser output.

2.3. Measurement of vibrational-level populations

The third challenge in the characterization of molecular plas-mas that are sustained at high-specific-energy loading (e.g. inair or in fuel–air mixtures) is spatio-temporally resolvedtemperature and vibrational-population measurements. Theresults provide quantitative insight into the kinetics of ‘rapidheating’ in short-pulse-duration discharges, which is causedprimarily by the collisional quenching of the excited electronicstates of nitrogen molecules [73–77]. This process is essentialin high-speed-plasma-flow control [78, 79] that is driven bylocalized volumetric heating of the flow in the pulsed discharge[80, 81]. It also has a significant impact on plasma-assistedcombustion—in particular, the competition of plasma-gener-ated radical reactions and thermal ignition [82]. The use ofemission spectroscopy for the translational–rotational temper-ature inference (from partially rotationally resolved structure ofN2 second-positive emission bands [76, 78, 80, 82]) is hinderedby the uncertainty in the spatial resolution and the relativelylow time resolution. The use of ultrashort-pulse (ps) CARS

spectroscopy, on the other hand, considerably improves boththe spatial and the temporal resolutions, along with improvedSNR; while improvement in the temporal resolution resultsfrom the usage of ultrashort pulses, the spatial-resolutionimprovement is afforded by proper selection of the phase-matching condition to discriminate the coherent CARS signalbeam from the plasma emission and stray light (see figure 9)[83–85]. Comparison of the results of these measurements withkinetic-modeling predictions (see figure 9) identified severaldominant kinetic processes that control rapid heating(quenching of electronically excited N2 molecules) and slowheating (vibrational relaxation of N2 molecules by O atoms) inthe afterglow of ns-pulse discharges in air [77, 85].

The timescale of the rapid heating process in atmospheric-pressure discharges in air is several ns [82]; thus, resolving itaccurately would necessitate the use of ultrashort-pulse-duration(ps and fs) lasers. In high-pressure pulsed plasmas, the SNR ofthe CARS signal can be sufficiently high to enable single-laser-shot measurements for which the time resolution is limited onlyby the duration of the laser pulse (e.g. see [3] and referencestherein). Recent works have demonstrated the capabilities of fs-CARS thermometry in high-pressure environments, includingsingle-shot measurements in atmospheric-pressure flames at1 kHz [86], single-shot measurements of 1D temperature dis-tributions in flames [87], and time-resolved temperature mea-surements at pressures up to 50 bar [46]. CARS-basedmeasurements will be discussed in detail in sections 4 and 5.

2.4. Measurement of number density of atomic and radicalspecies

The fourth challenge in the development of diagnostics ofchemically reacting plasmas that are sustained in air and infuel–air mixtures involves time-resolved measurements of theabsolute number densities of atomic and radical species andtheir spatial distributions. The reacting plasmas include bothns-pulse discharges and more commonly used discharges

Figure 8. Left: overlapping Raman and Thomson-scattering spectra in the center of the argon atmospheric-pressure microwave plasma jetexhausting into ambient air (plasma parameters inferred from the spectra are ne=1.8·1020 m−3, Te=1.7 eV, Trot=386 K, air molefraction 9.7%). Right: estimated parameter space in which Thomson and Raman scattering measurements can be applied. Reproduced from[69]. © IOP Publishing Ltd. All rights reserved.

8


driven by AC, RF, and microwave waveforms—e.g. atmo-spheric-pressure plasma jets involving reacting species O, H,and N atoms as well as radicals such as NO, OH, NO2, andHO2 and long-lived metastable species such as O2(a

1Δ). It iswell known that the yield of atomic and radical speciesgenerated in the plasma is significantly enhanced at high-peakintensities with reduced electric fields, which are typical forns-pulse discharges. These measurements are critical forplasma-assisted combustion applications and for plasmaapplications in biology and medicine. Specifically, the gen-eration of O and H atoms as well as OH radicals in none-quilibrium plasmas is essential for the initiation of reactionchains with fuel species, plasma oxidation, and subsequentignition. HO2 radicals, on the other hand, are much lessreactive, and their formation during other radical recombi-nation in high-pressure, low-temperature plasmas is detri-mental to plasma-assisted combustion. Furthermore, emissionfrom laser-generated plasmas, known as laser-inducedbreakdown spectroscopy (LIBS), has been used for measuringratios of the concentrations of reaction species in a reactingflame under high-pressure (up to 11 bar) conditions [88]. TheLIBS-based diagnostics is pursued for ps- and fs-laser excitedplasmas in reacting flames [89].

Excited species and radicals are also known to play asignificant role in biochemical processes [19]. However, lackof credible in situ characterization of plasma sources—inparticular, range and number densities of species generated—has been one of the most significant deficiencies in manyprevious studies of low-temperature plasmas for applicationssuch as plasma sterilization, wound healing, and cancertherapies. Lack of characterization limits the fundamentalunderstanding of the mechanism of plasma interaction withviruses, bacteria, cells, and tissue. This uncertainty resulted inan ongoing discussion among researchers in the field con-cerning what may be considered a ‘plasma dose.’ Most pre-vious studies in the field relied on a purely empiricalapproach, i.e. the generation of unspecified amounts of

‘reactive oxygen and/or nitrogen species’ by the plasma neara cell culture or an aqueous solution being perfectly studied todetect and isolate the effect of generated species. Althoughthis approach may have been justified at an early stage, fur-ther development of plasma–biomedical applications requiresaccurate quantitative data on the chemical composition ofplasma. Specifically, in plasma–liquid interaction environ-ments, the measurement of fluxes of dominant species gen-erated in the plasma towards the liquid surface is critical.

The accuracy of ‘conventional’ LIF and TALIFemploying ns-pulse lasers, which are widely used for radicaland atomic-species measurements, is known to be limited to asignificant extent by the uncertainty of the collisional-quenching rate of the excited states populated by the laserpulse. The quenching rate may vary significantly with thetemperature and chemical composition of the plasma, whichis controlled both by plasma-chemical reactions [31, 90, 91]and by mixing and evaporation processes that occur inatmospheric-pressure plasma jets [92] and plasmas generatedover liquid surfaces [93]. In addition, laser-induced photolyticeffects (photodissociation of molecules) may introduce asignificant additional uncertainty into the results. The use ofps and fs lasers with low pulse energies and pulse durationscomparable to or shorter than the collisional timescales makesit possible to obtain photolytic interference-free TALIFmeasurements of atomic-species distributions, as well asdirect measurements of the fluorescence quenching rates,which will be discussed in detail in section 3 [44, 94, 95].

Quenching-independent measurements of OH employingfs fully resonant, electronically enhanced CARS (FREE-CARS) have also been demonstrated recently [96], which willbe discussed at length in section 6. Comparison of theseresults with those of kinetic modeling would help the devel-opment of predictive-modeling tools and reduce the need forthe empirical approach that has been prevalent in previouswork in this field.

Figure 9. Left: Ps broadband N2 vibrational CARS spectra in the afterglow of a ns-pulse discharge in nitrogen at 100 Torr, with N2 (v=0–8)detected (Source: [83]). Right: comparison of N2 (v=0–8) vibrational-level populations with kinetic-modeling predictions in a ns-pulsedischarge afterglow in air at 100 Torr. Reproduced from [77]. © IOP Publishing Ltd. All rights reserved.

9


3. Recent development in fs-laser-based imaging ofkey reacting species

As outlined in section 1, the spatial distribution and temporalevolution of key intermediate plasma species including ions,neutral metastables, and reactive species are vital to theunderstanding of plasmas and reacting flows [10]. In part-icular, quantification of atomic neutrals such as hydrogen,oxygen, nitrogen, and carbon are crucial, because of theirhigh concentrations and reactivities. Many low-temperatureand highly nonequilibrium plasma sources, operating inatmospheric-pressure gas and liquid environments [2], are inthe initial stages of development. This section contains adiscussion of the recent developments in 2D imaging of keyspecies, as mentioned earlier.

3.1. Comparison of ns, ps, and fs-laser-based imaging—theirsimilarities and shortcomings

Noninvasive and in situ laser-based spectroscopic techniquessuch as LIF are widely used for species-dependent measure-ments and can be extended to 2D imaging via PLIF in manysituations [28]. The LIF-based diagnostics signal is linearlydependent on the species concentration, compared to thequadratic dependency in nonlinear techniques such as inCARS, FWM, etc [97]. Hence, LIF-based techniques areparticularly suitable for minor-species imaging [29]. Never-theless, as discussed in section 1, LIF-based measurements ofkey atomic species such as H, O, and N as well as molecularspecies such as CO are particularly challenging, because thefrequencies necessary for single-photon electronic excitationfall into the VUV region, where the medium is optically densein most practical devices and poses a significant problem topropagation through air [27]. Therefore, to avoid strongabsorption of the VUV excitation pulse, the excited states ofthese key species must be accessed through multiphotonexcitation [98], which requires visible or ultraviolet (UV)pulses with high pulse peak intensities to overcome therelatively weak multiphoton-absorption cross-sections, as wasfirst shown in H and D atoms [36]. The TALIF technique wasdramatically expanded to measurements of other species inreacting flows [99, 100] and plasmas [101, 102]. TALIF hasbeen employed for diagnostics in a variety of nonequilibriumplasmas [103, 104].

Unfortunately, since the two-photon excitation rate scalesas the square of the irradiance, all of the ns-laser-based TALIFmeasurements require high irradiance of visible or UV photons[27]. The high-irradiance excitation causes unwanted photo-dissociation of other species in the medium, generating thesame species that are being probed resulting in significantinterference [37–40]. Settersten et al carefully probed theextent of the photolytic interference caused by ns-laser-basedexcitation for TALIF measurements of the oxygen atom in ahydrogen-fueled laminar diffusion flame using a weak ps probe[105]. The results are presented in figure 10; a 3-GW cm–2 ns-photolysis laser tuned at 225.58 nm with 3.5 ns duration gen-erates the TALIF signal by exciting the p P p P3 23 3¬¬transition of the O atom; and the LIF signal is detected from the

p P s S3 33 3 transition at 845 nm. The same transition wasprobed by the 55 ps-duration dye laser tuned to J=2 line ofthe p P p P3 23 3¬¬ two-photon transition. The figure clearlyshows significant contribution of photolytic O-atom emissionto the TALIF signal, which is not present when the TALIFsignal is generated by the weak ps-probe excitation only. Theauthors also verified that the major contribution of the ns-laser-based photolytic interference in the TALIF signal was causedby photochemical production of the O atom from CO2, asopposed to photolysis of O2. A partial suppression of thisphotolytic interference had been reported for H- and O-atom-detection TALIF schemes using ps-duration pump pulses [41],which emphasizes the potential of ultrafast-pulse excitation inreducing or removing photolytic interferences in the detectionof species of interest via TALIF. The advantage of movingtowards a shorter pulse arises from the fact that higher irra-diance can be realized with orders-of-magnitude lower fluenceto obtain a similar level of SNR. Also, it is worth noting herethat the threshold of laser intensity for the saturation of dipoleexcitation is inversely dependent on the square of the pulseduration of the excitation pulse; hence the shorter pulse exci-tation is preferable to avoid signal saturation [106, 107].

Fs-duration pulses, which exhibit high-peak intensities butlow average energies, provide three particular advantages: (1)high-peak intensity allowing efficient two-photon excitation; (2)low average energies virtually eliminating photolytic productionof additional H or O atoms; and (3) commercially available fslasers with kilohertz repetition rate offering single-shot diag-nostics with TALIF for transient or turbulent environments.

Stauffer et al first demonstrated the fs-TALIF of theradical OH that is produced as a reaction intermediate in alaminar premixed flame [108]. Fs-TALIF-based atomic ima-ging of hydrogen was demonstrated, and a detailed study ofthe role of photolytic interference is presented in [109].

Figure 10. LIF signal as a function of distance from the surface ofthe flame. Presented are TALIF data due only to ns-photolysis laser(open squares), ps-probe laser (open circles), and both photolysisand probe laser (closed circle). Thin solid curves are spline fits of thedata. Thick solid line is numerically simulated profile of O-atomdensity. Dashed line depicting photolytic O atom is extracted bysubtracting photolysis-only and probe-only signal from the com-bined photolysis-probe signal. [105] (2003) (© Springer-Verlag2003). With permission of Springer.

10


Considering TALIF of the H atom, the advantages of transi-tioning to fs-duration-based excitation can be understood asfollows: the two-photon excitation of the n n1 3= =transition of the H atom with 205 nm photons is used topopulate the n=3 level, and the subsequent H

αfluorescence

at 656 nm from the n n3 2= = decay is detected. Shownin figure 11 are schematics of several competing processesthat can complicate the quantitative interpretation of mea-sured TALIF signals, including three-photon photoionization,SE along the signal-propagation direction, collisionalquenching, and photolytic production of additional H atomsthat contribute to the observed LIF signal. Typically, photo-dissociation of numerous H-containing flame radicals andmolecules can produce substantial quantities of additional Hin the medium. In these conditions, fs-duration pulses becomefavorable for TALIF, because the signal is proportional to thesquare of the laser intensity, whereas single-photon-inducedphotodissociation processes scale linearly. Therefore, high-peak powers but low average energy associated with fs pulsesvirtually eliminate photolytic interference. In addition, thebroad bandwidths of nearly Fourier-transform–limited fspulses contribute collectively to enhance multiphoton exci-tation through the combination of a large number of photonpairs with energies summing to the resonant energy of then n1 3= = transition (see figure 11(b)), since allcontributing photon frequencies have the same spectral phase[86]. Several studies that have been documented on fs-TALIFimaging of different atoms and reacting intermediates will bediscussed later in this section.

Schmidt et al recently investigated the merits of applyingthe fs-TALIF technique to the measurement of atomic oxygenin an atmospheric-pressure plasma jet (APPJ) source with ahelium–oxygen feed gas for direct comparison with conven-tional ns-TALIF [44]. For this comparison, high-pressuremicroplasma sources with small characteristic length scaleswere considered, because of the presence of strong spatialgradients for atomic-species concentration within an order of100 μm or less. For fs-TALIF, atomic oxygen and xenon were

excited by an optical parametric amplifier (OPA) pumped bya Ti:sapphire regenerative amplifier generating ∼25 μJ/pulseat 225 nm for the two-photon atomic transitions. In ns-TALIF, the required UV beam was produced by mixing theoutput of the dye laser with the frequency-tripled output of theNd:YAG laser in a beta-barium-borate crystal. The detectionsystem consisted of two main parts: (1) a camera that wasused to image the TALIF signal in the discharge volume ofthe APPJ and (2) a gated photomultiplier tube (PMT) that wasused to make point-based collisional-quenching measure-ments of the excited electronic state. A comparison of ns- andfs-TALIF will now be presented.

3.1.1. Accuracy of quenching measurement. The timescalesassociated with collisional quenching can vary from a few psto hundreds of ps, depending on the collisional environment.Thus, accurate calibrations for quantitatively measuring theconcentrations of atomic species are limited only to very low-pressure regimes in ns-laser-based TALIF, because of thedifficulty in deconvolving the time-resolved fluorescencedecay signal from the excitation-pulse envelope (shown infigure 12) [44]. However, in fs-TALIF, the collision is frozenduring the excitation enabling direct measurement of thequenching rate following the excitation. As shown infigure 12(a), the effective quenching rate must be extractedfrom the ns-TALIF signal by modeling the known naturaldecay rates and the excitation rate. However, fs-TALIF signalcan provide a direct measure of the quenching rate from thetime-resolved decay of the fluorescence signal.

3.1.2. Significant reduction in photodissociation. As mentionedearlier, the photodissociation or photolysis in ns-TALIFproduces significant background interference. In figure 13, thecomparison of ns-TALIF and fs-TALIF for atomic oxygen ispresented where radial profiles are obtained for 4%-O2/Hedischarge under the same operating conditions as in figure 12.Clearly, when the ns-pulse energy is increased, the additionaloff-axis signal increasingly perturbs the normalized O-atomradial profile. With 4% O2, 11 kV applied voltage, and a pulse-repetition frequency of 15 kHz, the He–O2 discharge primarilyproduces ozone that enshrouds the plasma jet up to ∼1 cm fromthe discharge axis. Since O3 has more than a five-orders-of-magnitude higher absorption cross-section than O2, a higherconcentration of O atom is generated within the duration of thens pulse, reducing the accuracy of the concentrationmeasurement. However, in fs-TALIF, the laser energy hasalmost no effect on the signal mainly because of the lower laserfluence. The laser fluence in fs-TALIF is orders-of-magnitudelower than the ns system. Since photodissociation linearlyscales with fluence, fs-TALIF has a significant advantage ofproviding higher accuracy in concentration measurement. It isto be noted that the signal level with 2.3μJ of laser energy issufficient to perform single-shot measurements for the fsexcitation case, whereas even 70 μJ of energy requires

Figure 11. (a) TALIF H-atom detection. (b) Excitation usingmultiple pairs of photons from fs pulse. Reproduced with permissionfrom [109].

11


significant averaging to achieve decent SNR for the nsexcitation scheme.

3.1.3. Superiority in 2D-imaging capability. The advantagesof high-repetition rate, low average energy, and highirradiance in the fs-laser-based TALIF have also been testedfor the planar-imaging capability. Superior 2D images withlarger image area averaging over significantly shorter durationare achieved with fs-TALIF compared to ns-TALIF. Theimages were reconstructed from a vertical scanning of the fullimage, as shown in figure 14. The time required forreconstruction of the 2D images is significantly less in fs-TALIF (∼20 min) than ns-TALIF (>3 h) dictated by highersignal level and kilohertz repetition rate of the laser system.Details of the 2D imaging will be discussed in detail for

specific atomic and molecular species in the followingsubsection.

Furthermore, in recent years efforts have been made tomove the development of BML towards generating pulseswith shorter durations that have the ability to produce intensepulses up to megahertz repetition rates. The short-pulse BMLwill pave the way for high-speed (>100 kHz) diagnostics ofturbulent reacting and nonreacting flows [72, 110]. Suchcapabilities with high-energy and high-bandwidth laserswould be extremely useful for investigating transient andturbulent flows. These capabilities could be further extendedto 3D measurements and to 4D with temporal evolutionrecorded [111, 112].

Thus, the benefits of fs-TALIF over ns-TALIF can besummarized as: (1) higher two-photon excitation efficiency,(2) reduction in interference from photodissociation, (3) direct

Figure 12. (a) Atomic-oxygen ns-TALIF signal collected from 2%-O2/He-mixture APPJ discharge with fast-rise time PMT and 12 ns-FWHM laser-pulse duration, illustrating limitations of ns-duration laser pulses to resolve sub-ns excited-state effective lifetime accurately.(b) Fs-TALIF for similar conditions in (a); the exponential decays demonstrate the strength of fs-duration excitation as collected with high-speed PMT. Reproduced from [44]. © IOP Publishing Ltd. All rights reserved.

Figure 13. Comparison of induced photodissociation levels for different incident laser-pulse energies indicated in the legend for (a) ns-TALIFand (b) fs-TALIF. Independent of incident energy, fs-TALIF shows no discernible levels of induced photodissociation. Reproduced from[44]. © IOP Publishing Ltd. All rights reserved.

12


measurement of the quenching rates of the excited states thatare limited only by the detector bandwidth and not by thepulsewidth of the laser, (4) potential for 2D imaging of atomicspecies, (5) higher dynamic range, because of a lower single-shot detection limit, and (6) superior measurement precisionafforded by the all-solid-state fs-laser systems. Pulse-to-pulseenergy fluctuation for fs lasers is typically 2% compared to∼10% fluctuation for ns lasers.

3.2. Atomic imaging in low-pressure and atmospheric-pressureplasmas

As discussed previously, the measurement of intermediateatomic species such as H, O and N atoms, as well as mole-cular species such as OH, NO and CO in chemically reactingflows such as plasmas and flames provides key insight intothe physical and chemical nature of the system [113]. Forexample, the generation of O and H atoms as well as OHradicals in nonequilibrium plasmas is essential for the initia-tion of reaction chains involving fuel species and plasmaoxidation, and subsequent ignition. The single-shot imagingcapabilities of atomic and molecular species described abovewill significantly aid in enhancing the understanding ofhighly-transient reacting flows and their kinetics. In thissubsection, we will review the development of the TALIFtechnique for a few atomic species and molecular radicals ofinterest for plasmas and reacting flows and discuss the recentdevelopments in spatially and temporally resolved fs-TALIF-based concentration measurements and imaging in detail.

3.2.1. H-atom imaging. Hydrogen atom is a key plasmaspecies that is easily generated, e.g. in a low-pressure plasma

discharge by electron-impact dissociation

e e eH H 2H 1s . 3.12 2*+ + + ( ) ( )

The H atom generated in the plasma plays a very importantrole in subsequent reaction chain [113]. Hence, quantitativemeasurement of the H atom is crucial to the understanding ofthe chemical kinetics in reactive plasmas. Ever since the firstobservation of two-photon n n1 3= = excitation of theH atom by 205 nm laser excitation and detection offluorescence signal at 656.3 nm was reported by Bokor et al[36], H-atom TALIF has been used as the most effectivedetection technique and applied to many different plasmasand flames [10]. Using ns-duration laser excitation for TALIFemission, the absolute H-atom concentration is measured in apure hydrogen RF discharge between a pair of planarelectrodes in a parallel-plate configuration [114]. From thetemporal evolution of the TALIF signal, the authors observedthat the H atoms are produced near both electrodes at theonset of the discharge, but over a period of 2.5 ms, theH-atom concentration becomes flat between the electrodes.The laser-induced dissociation of molecules during theH-atom TALIF measurement is discussed in [115]. Atomichydrogen is measured using TALIF in a laboratory model ofNASA’s hydrogen arcjet thruster, and the hydrogendissociation fraction was reported to be 31%±14% [116].TALIF measurement of the H atom in an expanding Ar–Hcascaded arc plasma showed that the signal is strongly selfquenched [117]. With TALIF measurement in an expandingthermal Ar–H plasma, Boogarts et al proposed to use rare-gasKr calibration as opposed to a flow-tube reactor method, andthe detection limit was reported to be 1015 m−3 [104]. Many

Figure 14. Comparison of 2D image of the O atom employing (a) ns-TALIF and (b) fs-TALIF for the same discharge in 2%-O2/He-mixtureAPPJ. Minimum detection limit of fs-TALIF is factor-of-three lower than ns-TALIF. Reproduced from [44]. © IOP Publishing Ltd. All rightsreserved.

13


other ns-TALIF measurements of the H atom in plasmas arereported, e.g. in highly diluted SiH4–H2 RF discharges thatare used in thin-film deposition [118], in a large-scalemicrowave plasma reactor [119], in a pulsed microwavedischarge [120], in a diamond-depositing DC arcjet [121], in ahydrogen gas flow into a wall-stabilized cascaded arc [122],in helicon plasmas [123], in ns-pulse discharges at a liquid–vapor interface [93], in an Ar–O2–H2 mixture excited by ns-pulse discharges [124], etc. Ns-TALIF suffers from rapidquenching (particularly in higher pressures) and photolyticinterferences, as discussed earlier. Recently, Schmidt et aldemonstrated that 2D imaging of absolute concentration ofthe H atom can be achieved using fs-laser-based TALIF [95].Those authors demonstrated that the high-peak intensity ofthe fs-laser pulses with low average energy and increasedspectral bandwidth compared to that of the ns lasers provideshigh-fidelity TALIF signal. The 2D TALIF image is obtainedin a low-temperature ns-pulse discharge plasma from a‘canonical’ pin-to-pin geometry at 100 Torr pressure with a1% hydrogen-in-helium mixture, as shown in figure 15. Thepresented composite 2D image is generated from 100 shot-averaged images that were collected 25 μs after the onset ofthe applied discharge voltage. The concentration of the atomicH peaks within ∼1.5 mm of the cathode, as shown in the axialdistribution of the generated H atom (shown in the left side offigure 15). Note that the absolute concentration of the H atomis obtained by calibrating the signal with Kr, where theabsolute number densities ranged from 2×1012 cm−3 to6×1015 cm−3. It is also reported that the spatial extent of theatomic H is extended in the He mixture compared to the Armixture. Selecting the TALIF signal from a small400×400 μm region in the middle of the plasma volume,

the temporal decay of the atomic-hydrogen number density isobtained for two separate mixtures, 1% H2/He and 1%H2/Ar. It is shown that the decay rate of atomic hydrogen inthe H2/He mixture is slower than that in the H2/Ar mixture.

In reacting flames, H-atom TALIF imaging has beeninvestigated using ps- and fs-laser excitation. From the fs-TALIF line profiles of the H atom in a CH4/O2/N2 Bunsenflame, it is shown that the line-profile shape is independent ofthe pulse energy, unlike the previous ps- and ns-TALIFexperiments—confirming the absence of photolytic interfer-ence in fs-TALIF signal [109]. Quantitative measurement ofthe H atom in more complex, laminar, premixed tubularflames has been demonstrated, offering data to validate thepseudo-1D model for tubular premixed flames [125].Furthermore, by developing an efficient fourth-harmonic,generator-based, high-power fs UV source for the two-photonexcitation (2×205 nm) of the H atom, 2D TALIF has beendemonstrated in a laminar Bunsen flame [126].

3.2.2. O-atom imaging. Like the hydrogen atom, the oxygenatom is also a key reacting atomic species in the chemicalkinetics of most plasmas. The O atom is responsible for theproduction of reaction intermediates, such as OH and NO,that advances the reaction chain in plasmas. Hence,quantitative 2D imaging of the O atom gives crucialinformation on the plasma kinetics. The O atom can begenerated in atmospheric plasma by electron impact

ee e

O O OO O O. 3.2

2

2

+ ++ + +

-

( )

Mcllrath et al were the first to show the possibility of usingthe two-photon transition p P p P2 33 3 of the O atom for

Figure 15. Composite planar image of the H atom in a 1% hydrogen-in-helium mixture at 100 Torr total pressure. The image is generatedfrom fluorescence collected 25 μs after voltage onset. Broadband image obtained under the same condition is shown on the right. Reproducedfrom [95]. © IOP Publishing Ltd. All rights reserved.

14


TALIF-based concentration measurements [127]. Followingthat demonstration, numerous ns-TALIF-based O-atommeasurements have been reported in a variety of plasmas;e.g. a plasma-etching environment [101], a microwavedischarge [128], an APPJ [129], a corona discharge [130],methane–air and ethylene–air nonequilibrium plasmas [131],a streamer discharge afterglow [132], an RF-discharge cell[133], a diffuse-filament electric discharge and afterglow[134], ns-pulse discharges in combustion mixtures [135], acapillary ns discharge [136]. However, ns-TALIF of the Oatom is plagued by significant photolytic interference[129, 130]. The primary contribution (90%) to photolyticinterference in O-atom signal is caused by [10]

hO O D O a . 3.3g31

21u+ + D( ) ( ) ( )

Since a large amount of ozone is produced in atmosphericplasmas, the dissociation produces significant interference inns-TALIF signal of the O atom [129].

Recently, fs-laser-based TALIF measurements have beenextended to obtain the spatially and temporally resolvedO-atom distribution in a ns, repetitively pulsed, externallygrounded APPJ in a capillary-dielectric-barrier-discharge(CDBD) configuration flowing helium with a variable-oxygenadmixture [94]. Calibrating the signal and imaging systemwith Xe that has similar two-photon transition frequencies tothe O atom, the absolute O-atom density could be obtained.With their setup, 2D planar imaging of the O atom has beendemonstrated, with number densities varying by three orders-of-magnitude and a detection limit of 2×1012 cm−3; seefigure 16. It is reported that when the concentration of the O2

in the feed gas was changed, a transition was observed in thedischarge morphology from annular to filamentary. Further-more, as shown in figure 17, a planar fs-TALIF image of theO atom was also obtained for the same setup as in figure 15(discussed in the previous subsection) with a different mixtureof 1%-O2/He at 100 Torr in the ‘canonical’ pin-to-pin

geometry [95]. The O-atom concentration is surprisinglyfound to be highest near the anode, which is invertedcompared to the distribution of the H atom (figure 15). Thisinverted O-atom density distribution is attributed to dissocia-tion of metastable excited electronic states of O2, both byelectron impact and quenching of excited helium atoms. Itappears that these low-energy excited states could beproduced more efficiently near the anode.

A detailed comparison and analysis of ns-TALIF versus fs-TALIF of the O atom has been presented for the APPJ dischargeconfiguration in a 4%-O2/He mixture [44], which is discussed indetail in sections 3.1.1–3.1.3. The advantages of fs-TALIFspecific to O-atom imaging inferred from this study are: (1) fs-TALIF signal is found to be 15 times stronger than ns-TALIFsignal when the signals are normalized with respect to theincident laser intensity, (2) fs-TALIF allows measurements overa wider range of pressures than ns-TALIF, (3) experimentaluncertainties in O-atom densities are notably smaller using fs-TALIF (±17%) compared to ns-TALIF (∼±33%), and (4) 2DO-atom images can be acquired in minutes with fs-TALIFcompared to multiple hours with ns-TALIF.

In reacting flames, Settersten and co-workers comparedns- and ps-TALIF of the O atoms to demonstrate thephotolytic-interference-reduction advantage for short-pulselaser excitations in a flame [41, 105]. Note that the ns- and ps-laser-based TALIF measurements of the O atom in flames arealso not completely free from photolytic interference. Inreacting flames, the O atom is primarily generated fromphotodissociation of the vibrationally excited CO2. Photo-lytic-interference-free 2D imaging of the O atom is demon-strated with fs-TALIF at a kilohertz rate for a range ofCH4–air conditions in a Bunsen flame [137].

3.2.3. N-atom imaging. Nitrogen-containing plasmas areprevalent in many applications. Electron impact cangenerate nitrogen atoms N(4S) easily by direct dissociation

Figure 16. Image of the CDBD for 2% oxygen-in-helium core-flow condition (left) and corresponding 2D TALIF image of O-atomdistribution (right). Reproduced from [94]. © IOP Publishing Ltd. All rights reserved.

15


of ground-state N2 molecules,

e eN X N S N S . 3.424 4+ + +( ) ( ) ( ) ( )

At low temperatures, charge exchange reaction between N+

and N2(X) and dissociative recombination of the molecularion are also very favorable, e.g.

e

N N X N S N

N N S N S . 3.5

24

2

24 4

+ +

+ +

+ +

+

( ) ( )( ) ( ) ( )

Nitrogen atoms are also crucial in reactive kinetics, becausethey contribute to the production of intermediates such asNOx.

Bischel et al have demonstrated that the two-photon2P S 3P S3 04 4 0 transition of the N atom can be excitedusing 211 nm and the TALIF signal is generally acquired at869 nm [99]. N2 TALIF is employed to study flowcharacteristics in free-nitrogen plasma that was created bycascaded arc [138]. The pressure dependence of N-atomrecombination probability is studied using ns-TALIF-basedmeasurement [139]. N-atom TALIF in an atmospheric-pressure pulse corona discharge is measured to investigatethe production mechanism of atomic nitrogen [140]. Since theN atom is responsible for the generation of NO, a combinedstudy of NO LIF and N-atom TALIF is carried out in diffuse-filament electric discharge, and it is concluded that NOformation in the ns-pulse discharge is dominated by reactionsof excited states of nitrogen [134]. The study is furtherextended to determine the effect of nitrogen on ns discharges

in air, H2–air and C2H4–air mixtures by measuring N-atomTALIF [141].

All of these N-atom TALIF measurements are limited tons excitation. With fs-TALIF it is expected that more accurateand higher-dimensional imaging can be realized, which willbe crucial for understanding the plasma kinetics.

3.2.4. Other radicals and molecule imaging (OH, NO, CO).Radicals and molecules such OH and CO and the generation ofregulated pollutants such as NO are of significance to thediagnostics of reacting flows and plasmas as well. Because oftheir importance, these molecules have been extensivelystudied by employing both ns-laser–based LIF and PLIF in avariety of plasmas [33, 34, 93, 142–144] and ps-time-resolvedLIF [134, 145]. Settersten, Dreizler, and Farrow used ps-TALIF to measure temperature-dependent and collision-partner-dependent quenching cross-sections of the CO in aheated cell [134, 146]. The development of sub-ps-laser-basedLIF, however, holds promise for time-gated detection ofmolecular species on timescales much shorter than typicalcollisional lifetime, even at higher pressures. These approachesmay, therefore, enable time-resolved, single-shot, and high-repetition-rate LIF or PLIF measurements (in addition to thephotolytic-interference-free advantage, as discussed earlier).

Broadband TALIF measurement was first demonstratedin OH to investigate its viability of fs-TALIF for minor-species detection [108]. Stauffer et al modeled two-photonabsorption in the OH A X2 2å ¬ P electronic transitioninvolved in the TALIF measurement and showed that the

Figure 17. Axial distribution of O-atom number density (left) and the fs-TALIF image of the O atom in 1%-O2/He mixture at 100 Torrpressure (right). Signal is collected 25 μs after the discharge within single ns duration of the discharge pulse. Reproduced from [95]. © IOPPublishing Ltd. All rights reserved.

16


temperature-dependent initial population is needed foraccurate derivation of the relative concentrations from theobserved TALIF signal [147]. Recently, planar (2D) photo-lytic-interference-free fs-TALIF imaging of CO using theB X1 1å ¬ å+ + electronic transition was demonstrated in anear-adiabatic methane–air flame stabilized over a Henckenburner [45]. Figure 18(a) shows the normalized spectrally-resolved TALIF signal of CO at three different heights of theflame, as recorded with an intensified charge-coupled device(CCD). Visualization of the transient dynamics of the jetflame was captured in the single-shot planar TALIF imagespresented in figure 18(b). From the photolytic interferencesobserved, it was concluded that the laser should operatebelow an intensity of 1010W cm−2 to avoid interference. Thisstudy demonstrated the ability to obtain photolytic-interfer-ence–free 2D TALIF images of reaction intermediates at1 kHz or higher repetition rate.

4. Recent development in fs-CARS for temperatureand species concentrations: advantages ofultrafast CARS

Amongst the various nonlinear spectroscopic techniquesdiscussed earlier, CARS spectroscopy has emerged as themost useful technique for measuring a wide range of theplasma and flow parameters (e.g. temperature, number densityof species, and electric field) with high spatial and temporalresolution. The advantages of CARS over other laser-diag-nostic methods that were discussed earlier are: (1) highspectral selectivity that results from the CARS signal beinggenerated only from the species that are in Raman resonancewith the pump and Stokes beams (i.e. E Eg 1 2 w = -( )/ ), (2)coherent laser-like signal dictated by the phase-matchingcriterion, removing any interference from the interrogatinglaser pulses, (3) low absorption of interacting laser beams andgenerated signal that results from their being individually off-resonant from the electronic transitions and, hence, notabsorbed by the medium, (4) capability to probe high-density

environments or major species in the target volume that isenabled by the fact that the CARS signal is quadraticallydependent on the number density of the target molecule, and(5) high spatial resolution resulting from the Raman interac-tion mediated by multiple laser beams, generating CARSsignal at the interaction volume.

In a typical CARS configuration, like all FWM techni-ques, photons from three laser beams interact via nonlinearinteraction with the target species to generate the fourth anti-Stokes laser-like signal beam, while the first two laser beams,pump (E1) and Stokes (E2) interact via two-photon resonancewith the Raman (vibrational or rotational) transition to gen-erate the coherence ωR in the probe medium. The third pulse,a probe pulse E3, interrogates the coherence, generating thelaser-like coherent anti-Stokes signal E4, as shown infigure 19(a). It has been recognized from the late 1970s that

Figure 18. (a) Spectrally-resolved CO TALIF signal from three positions in jet flame for different excitation energies. (b) Single-shot COTALIF signal recorded at 1 kHz in unsteady jet flame for different energies. Reproduced with permission from [45].

Figure 19. (a) Ultrafast vibrational CARS configuration to generateanti-Stokes signal E4; (b) three pulses in BOXCARS configuration togenerate a new beam at the target volume, which is laser-like CARSsignal, in the direction determined by the phase-matching conditionshown in (c). Adapted from [3], Copyright (2010), with permissionfrom Elsevier.

17


the CARS technique is capable of generating a highly species-specific and directional signal, because of the uniquearrangement of the lasers in a so-called BOXCARS geometry,where the pump, Stokes, and probe beams coming from threedifferent directions with their wave vectors k k k, and ,1 2 3

respectively, intersect at the interaction region to generate thesignal beam precisely at a fourth direction k4

following the

phase-matching condition

k k k k . 4.11 2 3 4- + =

( )

The CARS signal is calculated from the Maxwell equationusing standard approximations where the slowly varyingenvelope of the instantaneous CARS signal can be calculatedas [3]

I t Lc

k L P t L, , . 4.2CARS0

42 2

03 2

e=( ) ∣ ( )∣ ( )( )

Here, P0(3) is the third-order molecular polarization that is

induced by the excitation laser pulses, and L is the inter-rogation length at the target where the effective nonlinearinteraction takes place. The temporal evolution of molecularpolarization P(3) is obtained by expanding the source term inthe Maxwell wave equation, which can be written as

P t E t t E t t

E t t t

,

d . 4.3

33

31

2

òt c t

t

= ¢ + - ¢

+ - ¢ ¢-¥

¥( ) ( ) ( ) ( )

( ) ( )

( ) ( )

Here, τ is the delay of the probe pulse with respect to thearrival of the time-overlapped pump and Stokes. Calculationof the third-order susceptibility χ(3) involves solving thecoupled differential equations using the Schrödinger waveequation in a perturbative approach [148] or the density-matrix equations [149, 150]. Furthermore, the ultrafast time-resolved CARS signal is typically measured as an integrationof the above polarization for a fixed probe delay τ to obtain[151]

I P t t_ , d . 4.4CARS int3 2ò t¢ ¢

-¥

¥ ∣ ( )∣ ( )( )

Except for very specific limiting cases, the above integrationis calculated numerically.

The significance of fs-CARS spectroscopy has beendiscussed and CARS has been employed for studying gas-phase molecules over the past two decades [151]. Thepotential advantages of fs-CARS in plasma diagnostics andimaging were first outlined in the review article by Zheltikov[152]. Since that time, fs-laser technology has matured so thatcurrent systems can provide energy in excess of 1 mJ orhigher at a rate of 10 kHz. The high energy, stability, andtunability via the OPA of amplified fs-laser pulses haveopened many new avenues for providing insight into the flowphysics of plasmas and reacting flows with unprecedentedtemporal resolution. A typical fs-CARS setup is presented infigure 20.

A few of the distinct advantages of ultrafast CARS willnow be discussed in further detail.

4.1. Multiphoton excitation for stronger coherence

Raman excitation being a two-photon process, fs-laser-basedpump and Stokes provide unique advantage over other long-pulse laser-based CARS configurations, because of the asso-ciated frequency bandwidth of these pulses. For example,rovibrational lines of molecular nitrogen are separated by∼2320±20 cm−1; in a traditional ns-CARS configuration,such Raman transitions are typically accessed by using532 nm laser as the pump and a tunable broadband dye laseras the Stokes pulse. Thus, the Raman-resonance condition

E ER 1 2 w = -( )/ is satisfied by only one pair of photonenergies from each pair of pump and Stokes pulses (seefigure 21(a)). However, with fs pulses having a spectral widthof the order of 10 nm (∼200–300 cm−1), the Raman conditionis satisfied by many pairs of photons having different fre-quencies within the broadband pulses, as depicted infigure 21(b). Thus, a strong Raman coherence [149] is gen-erated with modest excitation-pulse energies through utiliza-tion of the full bandwidth [153, 154] and, hence, a strongersignal [3, 149, 155]. The broad bandwidth has also enabledcoherent detection of multiple species simultaneously [156].This multiphoton excitation advantage of fs-laser-basedpumping is not limited to CARS but is applicable to anydiagnostic method involving multiphoton laser excitations, asdiscussed in section 3.

4.2. NRB-free detection

One of the major issues with the early works in frequency-domain CARS spectroscopy was the contribution from NRB,which generally appears at the CARS signal frequency. NRBsignal is generated when the pump, Stokes, and probe pulseoverlapped in time at the target volume. This added greatercomplexity, particularly for high-pressure conditions wherethe broadening conditions are not well known. The utilizationof pulse sequencing with shorter pulses was proposed toremove the NRB in condensed-phase systems [157]. In 2005,Roy et al showed that with the ability to delay the probe fromthe pump and Stokes pulses when their durations are 100 psor shorter, the NRB-free CARS signal can be extracted (seefigure 22) [158]. Time-resolved nonresonant signals wereobtained by flowing pure O2, CO2, and Ar; and the resonant-CARS signal was acquired for the Q-branch transition in thev=0→v=1 band of N2 as a function of the time delaybetween the probe and time-overlapped pump-Stokes pulses.It was observed that the magnitude of the nonresonant signaldecreased by three orders-of-magnitude relative to resonant-CARS signal when the probe was delayed by ∼110 ps. Notethat the pump and probe-pulse durations were ∼135 ps andthat the Stokes duration was ∼106 ps. The NRB suppressionis easily extended to fs-CARS diagnostics [159].

4.3. Saturation-free fs-CARS spectroscopy

The intensity of the generated CARS signal grows linearlywith the intensities of the pump, Stokes, and probe pulses inthe weak field limit. It is well known, however, that in ns-CARS, the Raman coherence can saturate and experience a

18


Stark shift, modifying the spectral lineshape of the CARSsignal at peak intensities above 1014Wm−2 [160]. However,it has been shown theoretically that in vibrational and rota-tional fs-CARS studies of gas-phase nitrogen, the threshold oflaser intensities for saturation are significantly higher com-pared to those of other fs nonlinear processes (such as fila-ments, etc); hence, fs-CARS signal will rarely be saturated[150]. This saturation-free regime for higher intensities hasbeen exploited to increase the CARS sensitivity up to single-molecule conditions in plasmonics [161] and to reach super-spatial and -spectral resolutions in vibrational CARS-micro-scopy imaging [162].

Also, it has been shown that in an electronic-resonance-enhanced CARS (ERE-CARS) configuration, where the

probe pulse resonantly couples the ground and electronicallyexcited electronic states of the molecule, a strong ns-durationprobe can saturate the signal [163]. However, in a detailedtheoretical study of ERE-CARS for different pulse regimeswith durations ranging from 10 ns–10 fs, it was shown thatthe signal saturation threshold associated with probe pulsebegins to increase for pulse durations shorter than the mole-cular decay and dephasing times; the threshold intensity wasshown to increase quadratically with the inverse of the pulseduration [164]. It has also been shown that the saturationthreshold of the pump and Stokes intensities for Ramancoherence using ultrafast pulses has a similar dependence[106]. The saturation-free regimes in standard fs-CARS orhigh-saturation-threshold regimes in ultrafast ERE-CARS are

Figure 20. Typical fs-CARS setup where photons from three beams (pump, Stokes and probe) interact at the target region to generate CARSsignal. The laser and optical components with the typical laser parameters marked in the figure are used in [86]. Reproduced with permissionfrom [86].

Figure 21. Raman excitation of N2 rovibrational line (a) ns-CARS, (b) fs-CARS configurations. Because of multiple-photon pairs ofexcitation with fs-laser, Raman coherence ρab is almost two orders-of-magnitude higher than ns-CARS. Reproduced with permissionfrom [153].

19


helpful in achieving high SNR. ERE-CARS and all-resonantfs-CARS will be discussed in further detail in section 6.

4.4. Collision-independent CARS signal

Collisional quenching of the population and dephasing of theRaman polarization are the major drawbacks in spectroscopywith long-pulse (ns or longer) excitations. Since collisionaltimescales range from tens of ps to ns, in a long-pulse (ns orlonger) CARS configuration, several collision cycles canoccur, causing collisional energy transfer of population anddephasing of coherence during the interrogation period thatwill require accurate collisional modeling to extract thedesired diagnostic information. However, in fs-CARS, thesignal is generated in timescales where the collisions arealmost frozen. This collision-independent signal enables time-resolved studies of vibrations and rotations in gas-phasemolecules, which were first conducted by Hayden andChandler in benzene and 1,3,5-hexatriene gases [151].Motzkus and co-workers have shown from the beatingstructure of the experimental transients of H2 that the rota-tional dynamics could be accurately modeled. Such featurescould be used to extract detailed information on theJ-dependent collisional broadening, line-shifting coefficients,and other molecular parameters, even at high temperature[165] and pressure [159, 166].

Wrzesinski et al investigated the collisional sensitivity oftime-resolved fs-CARS by measuring the N2 spectra forvarious pressures ranging from atmosphere to 50 bar both forpure gas and for gas mixtures (i.e. with different collisionalpartners) [46]. It was observed that the time-resolved signal atearly delays did not exhibit any dependence on the systempressure, as shown in figure 23. For pressures of 1–5 bar,rotational recurrences are observed at longer delays, becauseof rephasing of a subset of Q-branch transitions of N2 [167].The overall signal initially decays at the frequency-spread-dephasing timescale [168] (see section 5.2 for more details),

but the signal decay at a longer probe delay is governed bythe molecular-collision timescales. However, the timescalesof such long-time recurrences become shorter with pressureand the recurrences disappear for pressures above 20 bar.Clearly, the CARS signal is collision independent within thefirst 10–15 ps for pressures up to 40 bar. The modeling ofhigh-pressure data with the exponential-gap energy-corrected-sudden (ECS-E) scaling law required inclusion of the line-mixing effect, as suggested by Knopp et al [167, 169].Fs-CARS has also been used to demonstrate collisional-interference-free measurement of temperature in high-pres-sure cells [170]. Such collision-independent fs-CARS can betransitioned to diagnostics of high-pressure and high-temp-erature plasmas containing ions and neutrals.

4.5. Kilohertz- to megahertz-rate sources

Until the end of the previous century, time-resolved fs-CARShad been developed using a multipass, dye-amplified laserpumped by a Nd:YAG laser operating at tens of hertz [151].However, this slow repetition rate is not suitable for highly-transient reacting flows or plasmas, where the temperatureand species concentration vary rapidly in both the spatial andtemporal domains. With the availability of regenerative Ti:Sapphire amplifiers that generate high-energy pulses (up to10 mJ/pulse) at kilohertz (or higher) repetition rates, high-speed diagnostics have been demonstrated in reacting flows[86] and even in unsteady flow conditions [171]. In the firstdemonstration of 1 kHz, single-shot fs-CARS thermometry byRoy et al [86], time-resolved N2 spectra were obtained in anatmospheric-pressure, near-adiabatic H2-air flame stabilizedover a Hencken flame. The single-shot spectra and associatedprobability-density functions (PDFs) of the extracted temp-erature from 1000 spectra for equivalence ratios Φ of 0.5 and1.0 are shown in figure 24. For the temperature range of1000–2400 K, the accuracy of the measurements is found tobe ∼1%–6% of the adiabatic flame temperature and the pre-cisions are ∼1.5% of the mean temperature. The use of high-energy fs lasers has also enabled 1 kHz rate, 1D (line) [87]and 2D (planar) [172] CARS imaging, as will be discussed insection 4.6.

Recent years have seen the rapid improvement of BMLtechnology to produce high-intense pulses at repetition rates

Figure 22. Time-resolved NRB and resonant-CARS signal. NRBfalls off rapidly once all pulses move out of the interaction region.Hence, resonant-CARS signal three orders-of-magnitude smallerNRB can be retrieved with delayed probe pulse. Reproduced withpermission from [158].

Figure 23. (a) Time-resolved fs-CARS spectrum of N2 gas atpressures from 1 to 40 bar. (b) Plots from ESC-E model with line-mixing effect included. [46] John Wiley & Sons (Copyright © 2013John Wiley & Sons, Ltd).

20


up to megahertz [72]. The high-energy BML has enabled 3Dmeasurements of LIF in reacting flows and also extended to4D with temporal evolution recorded [111, 112]. Recentdevelopments have pushed BML towards generating pulseswith shorter durations that have the ability to produce intensepulses up to megahertz repetition rates. The short-pulse BMLwill pave the way for high-speed (>100 kHz) diagnostics ofturbulent reacting and nonreacting flows [72, 110]. Suchcapabilities with high-energy and high-bandwidth laserswould be extremely useful for investigating transient andturbulent flows. These capabilities could be further extendedto higher-dimensional CARS measurements, which will bediscussed in detail in section 4.6.

4.6. Extending CARS measurements to multiple spatialdimensions: 1D (line) and 2D (plane) measurements

As noted earlier, CARS is an intensity-driven nonlinear pro-cess; hence, fs lasers are well suited for the generation ofstronger Raman coherences and CARS signal. With theavailability of fs lasers that can produce a few mJ/pulse, ithas been possible to create laser sheets for performinginstantaneous multidimensional imaging based on CARSspectroscopy. 1D and 2D measurements can be extremelyvaluable in understanding the spatio-temporal instabilities inpractical combustion devices with turbulent flows and inaiding the validation of complex turbulent plasma and flowmodels.

For 1D-CARS, laser sheets are produced by a combina-tion of cylindrical and spherical lenses, arranged in a BOX-CARS setup such that all three beams intersect on a line at theburner and generate the line-CARS signal [87], as shown infigure 25(A). Note that a chirped-probe beam was used toobtain the spectrally- and spatially-resolved line image pre-sented in figure 25(B) (details of the chirped-probe CARSwill be discussed in section 5. The CARS line imaging of N2

is presented for room air and a H2–air flame in (a) and (b),respectively. Single-shot temperature measurements at 1 kHzrepetition rate are reported in a flat-flame field up to 2000 K,suggesting the possibility of realizing fs-CARS line imagingbased on high-speed thermometry in turbulent flames.Kliewer et al demonstrated line-temperature and O2:N2 con-centration ratio employing ps-laser-based dual-broadbandpure-rotational CARS [173]. The reported accuracy and pre-cision for a temperature range of 410–1200 K were compar-able to those of ns-laser-based dual-broadband pure-rotationalCARS and 1D vibrational CARS.

Recently, Bohlin and Kliewer demonstrated that thedimensionality of detection for the CARS signal can beextended from 1D imaging to 2D imaging [174], and the laserpowers were sufficient to obtain single-shot 2D CARS images[175]. Planar CARS at a 1 kHz rate was demonstrated in aheated jet by Miller et al [172]. These 2D-imaging effortsemployed a variant of ultrafast CARS that is referred to ashybrid fs/ps-CARS, where a single broadband, transform-limited fs pulse acting as both pump and Stokes pulses excitemany rotational or vibrational Raman transitions and a

Figure 24. Single-shot fs-CARS experimental spectra with least-squares theoretical fit in H2-air flame stabilized over Hencken flame for (a)Φ=0.5 and (b) Φ=1.0; their corresponding PDFs of extracted temperature from 1000 consecutive shots shown in (c) and (d), respectively.Reproduced with permission from [86].

21


narrowband (typically ps duration) probe interrogates thebroadband Raman coherence to generate the spectrally-resolved CARS signal [176, 177]. Using such a two-beamCARS configuration, the planar temperature map of a pre-mixed laminar CH4–air flame has been obtained from N2 andO2 rotational CARS, as shown in figure 26 [175]. The abilityto perform accurate 2D temperature measurements for tem-peratures in the range 450–2000 K in combination with sin-gle-shot data acquisition at high-repetition rates would aid theunderstanding of the flow physics of plasmas and flames,even in turbulent environments.

In the following section, we present in-depth discussionon how the unique capabilities of the ultrafast CARSdescribed above have been employed to measure temperatureand species concentrations in reacting plasmas and flows.

5. Interference-free temperature and species-concentration measurements

As noted in the previous section, the coherent nature of theCARS signal generated from a small, targeted, nonlinear

interaction volume eliminates the large background fromfluorescence, chemiluminescence and laser scatters that couldinterfere with the signal. Furthermore, as discussed insection 4.2, it was recognized during the very early devel-opment of CARS that employing short-pulse lasers was veryuseful for suppressing the NRB by the time-delayed inter-rogation [157, 178, 179]. Hence, ps-CARS spectroscopy wasextensively used for spectrally-resolved probing of thedephasing dynamics in solid and condensed-phase systems[180]. Tunkin and co-workers exploited the NRB-suppressionadvantage associated with ps-CARS for the spectroscopicinterrogation of gas-phase molecules [181]. However, thechallenge remained from the fact that most of the ps-CARSsystems employed synchronously pumped dye lasers runningat limited repetition rates (10–30 Hz) and, hence, were notsuitable for high-speed dynamics in rapidly changing reactingproducts in plasmas and reacting flows.

It should be noted that gas-phase temperature measure-ments rely on the temperature-dependent initial Boltzmanndistribution in the microscopic (rotational or rovibrational)states of the molecules. At an equilibrium temperature T, thenumber density of a target molecule ni in a molecular state

Figure 25. (A) Planar BOXCARS configuration for 1D-CARS imaging. (B) Sample of (a) room-air image obtained with 16-shot average and(b) single-laser-shot line image of high-temperature flame (equivalence ratio Φ=0.46). Reproduced with permission from [87].

22


i with energy iw is given by [28]

nN

Qg k TExp . 5.1i i i Bw= -[ ] ( )/

Here, N is the number density of the target molecule in theprobe volume, gi the degeneracy of the molecular level, andkB the Boltzmann’s constant. The partition function Q isgiven by

Q g k TExp . 5.2i

i i Bå w= -[ ] ( )/

Hence, to obtain an accurate temperature in a rapidly evolvingreacting flow, it is important to obtain single-shot measurementdata that can simultaneously excite multiple rotational orrovibrational states for parametrically generating the CARSspectra. In the ns excitation regime, the narrow bandwidth ofpulses can couple only one or a few of those ground-statemolecular states that are needed for inferring the thermometricinformation. Alden et al demonstrated that employing a

broadband modeless dye laser pumped by Nd:YAG laser withns duration can simultaneously capture the full spectrum of theBoltzmann distribution in each of the laser shots to achievesingle-shot CARS thermometry [182]. For the next two dec-ades, numerous works were reported with different variants ofthe broadband ns-CARS for thermometric measurements[3, 28]. However, the NRB in the ns-CARS significantly limitsthe accuracy and degrades the sensitivity of temperature mea-surement, in addition to the other limitations of the ns-CARS asdiscussed earlier. Roy et al proposed the use of ps-laser-basedCARS to demonstrate NRB-interference-free, single-shot, gas-phase thermometry [158]. Those results were followed bymany other investigations in reacting plasmas and flows, whichwill be discussed in detail in the following subsections.

Although ps-CARS could provide the time-resolved,NRB-free CARS for extracting thermometric information, themeasured signal is still strongly dependent on collisions[183]. Hence, accurate modeling of collisional parameters isrequired for ps-CARS thermometry; but those parameters are

Figure 26. 2D CARS image of specific S-branch transitions of N2 and O2 molecules in premixed laminar CH4/air flame and derivedtemperature contour. (a) Spatially-resolved planar CARS as dispersed and detected on CCD. Spectrum consists of many N2 and O2 S-branchlines. (b) Two S-branch lines of N2 lines corresponding to J=7 and J=28. (c) 2D temperature map ranging from 450–2000 K derived fromN2 CARS signal, which is a zoomed-in view of the 2.4×9.1 mm area in the flame shown in (a). (d) Stick spectrum of rotational CARS of N2

extracted from pixel-to-pixel value of each transition at three different spatial locations, as marked in (c). Reprinted with permission from[175]. Copyright (2014) American Chemical Society.

23


not well known in most of the highly reacting flows. Haydenand Chandler were the first to show the potential application offs-CARS for gas-phase diagnostics by obtaining time-resolvedCARS of gas-phase benzene and 1,3,5-hexatriene, demon-strating a significant Raman-shifted signal with modest laserenergies (∼10μJ) [151]. The collisional independence of thesignal was demonstrated by Lang et al [166]. The low energyand high-peak intensity afforded by the fs-laser pulse are wellsuited for nonlinear techniques such as CARS. Hence, from theearly twenty-first century, fs-CARS has been employed in awide variety of studies in reacting flames and plasmas.

In the following subsections, we discuss different CARStechniques utilizing the distinct advantages of the ps and fslasers to obtain interference-free temperature and species-concentration measurements.

5.1. Pure-rotational CARS (ps, fs, fs/ps)

At a given temperature T, the Boltzmann distribution (5.1)determines the rotational population distribution nJ corresp-onding to the rotational number J within a given vibrationalmanifold, and can be written as [28]

nN

Qg J BJ J hc k T2 1 Exp 1 , 5.3J I

rB= + - +( ) [ ( ) ] ( )/

where gI is the nuclear-spin degeneracy for nonzero nuclearspins, B the rotational constant of a given molecule, andQr=kBT/(hcB) the partition function for rotation. Note thatthe Boltzmann distribution is an accurate signature of temp-erature in the equilibrium conditions. However, in none-quilibrium conditions, the measured rotational temperaturecan be different from the vibrational temperature (to be dis-cussed in section 5.2), because the rotational states equilibratemuch faster than the vibrational states, and hence it isassumed that rotational temperature is also representative ofthe translational temperature in highly-transient plasmas (seesection 5.4 for more details).

The rotational population distribution is spectro-scopically measured and the rotational temperature isextracted using equation (5.3). Typically pure-rotationalCARS (RCARS) signal is an order-of-magnitude strongerthan that of vibrational CARS, because the Raman cross-section of pure-rotational transitions is generally higher thanthat of the vibrational transition lines [28]. RCARS has beenextensively used for measurements of gas-phase temperaturesemploying ns lasers [3]. Seeger et al reported the firstdemonstration of NRB-free RCARS thermometry employingps lasers to measure the time-resolved N2 in air and in alaminar CH4 diffusion flame [183]. However, it was observedthat the spectrum changes significantly within a short periodof time when the probe pulse is delayed with respect to thetime-overlapped pump and Stokes pulse, because of mole-cular collisions; hence, without a detailed time-dependentmodel, it was difficult to extract temperature from the time-delayed spectrum. In another study, it has also been demon-strated that the interferences from the broadband signal pro-duced by other FWM pathways within the species of interestand also those from other molecules can be reduced by tuning

the time delay between the probe and time-overlapped pumpand Stokes pulses [184]. Time-resolved dual-broadbandps-RCARS has been demonstrated for thermometry andmajor-species-concentration measurements in a laminar,nonpremixed sooty flame using a polarization-based inter-ference-suppression technique [185]. ps-RCARS has alsobeen extended to obtain 1D thermometry and the N2/O2 ratioin flames [173] and to measure the self-broadening of theJ-dependent N2 Raman linewidths directly for a temperaturerange of 294–1466 K [186]. Similar state-specific linewidthmeasurements have been made using the pure-rotational(S-branch) transitions of self- and N2-broadened C2H2 [187]and CO2 [188].

The temperature distributions within plasmas are criti-cally important to aid in bracketing the total thermal effects ofthe pulsed discharges as well as to better determine the pro-duction methods of the atomic species and to help supportdiffusion-rate calculations. Zuzeek et al have employed ps-RCARS for thermometry in low-temperature plasma-assistedcombustion in a repetitive ns-pulse discharge (at a 40 kHzpulse-repetition rate and 10 Hz burst repetition rate) in anethane-air flame operated under fuel-lean conditions andexperimentally observed the higher heating rate in the fuel–airplasma compared to the air plasma [189]. Ps-RCARS has alsobeen used in a similar nonequilibrium plasma system toobtain the time-resolved rotational temperature and the O2

mole fraction [190]. A comparison of the rotational tem-peratures obtained from ps-RCARS in a pulsed-ns-dischargeplasma in air and in a flame with different equivalence ratiosis presented in figure 27. It is observed that the temperaturedeclines after a subsequent burst of pulses hitting the flamefor the Φ�0.5 condition, and from kinetic-modeling

Figure 27. Comparison of time-resolved ps-RCARS temperatures inrepetitively pulsed plasma at 40 Torr in air and in hydrogen–airflame (for different equivalence ratios). Solid lines are kinetic-modelpredictions. Reprinted from [190], Copyright (2010), with permis-sion from Elsevier.

24


calculation, it is clear that the removal of radicals in a pulsed-discharge plasma completely blocks subsequent exothermicreactions, which renders ignition impossible. In a recentstudy, time- and spatially-resolved temperature measurementhave been demonstrated in ns-pulse discharges in air andH2–air mixtures employing ps-RCARS of N2 to study kineticsof energy thermalization [85]. The authors have observed twodistinct thermalization regimes: (1) the rapid heating wasfound to be caused by collisional quenching of electronicallyexcited N2 and O2, and (2) the slow heating was determinedto be caused by N2 vibrational relaxation in air and by che-mical energy release during the partial oxidation of hydrogenin the H2–air flame.

Broadband ps-laser-based CARS measurements led theway for the removal of NRB interference in the thermometricmeasurements, as described above. However, the signal was stillprone to collisional quenching and broadening; hence, the acc-uracy of thermometry was strongly dependent on accuratemodeling of the collision [183]. In the meantime, withthe development of fs lasers that have hundreds of cm−1

available bandwidth for simultaneous excitation of the Ramantransitions in multiple species, many research groups divertedtheir focus toward fs-CARS–based single-shot thermometric andspecies detection measurements [3, 153]. Most of the fs-CARSdevelopments were focused on time-resolved rovibrational orvibrational CARS (discussed in the next subsection). Until thebeginning of this century, it was believed that because of thecoarse spectral resolution aided by the broadband fs probe-pulsestandard fs-RCARS was not suitable for thermometry. Inthe early twenty-first century, Lang and Motzkus were the firstto demonstrate fs-RCARS-based thermometry on CO2 [166].

In another diagnostics development, more than threedecades ago, Kaiser and co-workers had observed that withshort-pulse Raman excitation and a prolonged interrogation, anarrow spectrum can be generated with high spectral resolu-tion [178]. Recently, such a hybrid ps/fs-CARS scheme hasbeen proposed where fs-duration pump and Stokes excitestrong Raman coherence and a ps-duration probe pulseinterrogates the coherence for generating CARS signal withhigh spectral resolution for thermometry, as demonstrated by

Prince et al [176]. A hybrid-CARS scheme has the combinedadvantages of time-resolved spectroscopy with fs Ramanexcitation and frequency-resolved spectroscopy with ps-probepulse. Slipchenko et al have demonstrated pure gas-phasethermometry using the hybrid fs/ps-RCARS scheme [177]. Inthe past decade, a wide range of CARS measurements havebeen reported with spectacular thermometry and imagingcapabilities, as discussed below.

Miller et al demonstrated the first pure-RCARSemploying the hybrid configuration to measure single-laser-shot gas-phase temperature measurements in a stoichiometricH2–air flame [191]. A typical energy-level diagram and pulsesequence for hybrid RCARS is shown in figure 28. Thecollision-independent, NRB-free, single-laser-shot, hybrid-RCARS signal of N2 is presented in figure 29 for varioustemperatures. The reported accuracies of the best-fit temper-ature from a probe delay of 13.5 ps exhibit accuracies of ∼1%without the need of NRB or collisional corrections. J-level-dependent dephasing can also be directly measured fromscanning the probe delays, while recording the CARS signal.The high spectral resolution capability of hybrid RCARS isextended to measure pressure dependence of the self-broa-dened S-branch linewidths of N2 and O2 at elevated pressuresup to 20 atm in [192]. From a systematic study of the RCARSof N2 and O2 at high pressure, it is concluded that the effect ofcollision can be ignored by delaying the probe up to ∼6.5 pswhere the thermometric errors can be maintained below 5%for pressures up to 15 atm, where, however, for pressureshigher than 15 atm, the collisional effects cannot be ignored[170]. Using a time-asymmetric probe pulse for hybridRCARS, Stauffer et al have shown that NRB-interference-free pure-RCARS can be obtained at much shorter probedelays, where the collisional dephasing is negligible; theyreported single-shot accuracy (<1% error) and precision(∼2.5%) for N2 single-shot thermometry [193]. Employing atwo-beam variant of hybrid fs/ps-CARS configuration [177]that allows tunability of spatial resolution by adjusting thecrossing angle of the beams, Kliewer and co-workers havedemonstrated single-shot 1D [194] and 2D [174, 175]RCARS thermometry, which is discussed in detail in

Figure 28. (a) Typical energy-level diagram for fs/ps-hybrid-RCARS configuration with fs pump (ω1), Stokes (ω2), probe (ω3), and RCARS(ωCARS) beams. (b) Timing sequence of overlapped pump and Stokes pulses, and time-delayed ps-probe pulse generated by a 4f pulse shaper.Reproduced with permission from [191].

25


section 4.6. Kearney and Scoglietti demonstrated the use of anefficient second-harmonic bandwidth compression technique,which generates a high-energy ps-probe beam that allowssingle-shot thermometry and O2/N2 ratio measurement in aflame with temperatures up to 2000 K [195]. In figure 30, theobserved temperature and O2-concentration histogram fromthe single-shot spectra are presented for two equivalenceratios in a near-adiabatic H2–air flame. Another variant of theprobe pulse, using two etalons has been employed to compare

the accuracy and precision of hybrid fs/ps-RCARS thermo-metry for two probe durations (1.5 and 7 ps) [196]. Highlyaccurate Raman linewidth measurement of N2 S-branch linesis demonstrated from a single-shot hybrid-RCARS config-uration by splitting the probe beam into four spatially-resolved beams with independently adjustable delays; thatallows simultaneous recording of the RCARS signal at fourdiscrete probe delays [197]. A spectral-focusing-basedRCARS approach has also been demonstrated, which showsbandwidth optimization of the excitation beams by introdu-cing a delay between the pump and Stokes pulses [198]; thismethod is shown to be useful when chirp is present in theexperiment or the laser source is not perfectly transformlimited [199]. Hybrid-RCARS N2 gas in a pressurized gas cellhas been employed to measure pressures with a reportedsingle-shot accuracy of 0.1%–6.5% [200]. Recently, a variantof ultrafast RCARS was employed to study the molecularrotation of N2 using an optical centrifuge [201], and from thedirect measurement it was shown that the dephasing of rota-tional lines with J value as high as J=66 has an order-of-magnitude slower coherence-dephasing rate compared to thatof low-J transitions [202].

5.2. Vibrational CARS (ps, fs, fs/ps)

As mentioned in the previous section, the rotational andvibrational temperatures can be different in nonequilibriumplasmas. The difference between the two temperaturesdetermines the degree of nonequlibrium of the plasmas. Thevibrational CARS (VCARS) spectrum is obtained by inter-rogating the initial population distribution of the rovibrationalstates (rotational manifolds across different vibrational levels)via Raman excitation. The vibrational-population distributionat a given vibrational state v at temperature T modifies the

Figure 29. Open symbols are single-shot fs/ps-hybrid-RCARSspectra of N2 at (a) 304 K, (b) 497 K, and (c) 705 K. Solid lines arebest-fit simulation, and residual is shown shifted by −0.1.Reproduced with permission from [191].

Figure 30. Histogram of temperature and O2 concentration plottedfor F=0.48 and 0.7 from hybrid-RCARS data. For each histogram,1000 single-laser-shot spectra are shown. Reproduced with permis-sion from [195].

26


general Boltzmann equation (5.1) to [28]

n N e 1 e . 5.4vvhc k T hc k Te eB B= -w w- -[ ] ( )( ) ( )/ /

Since the vibrational state separations within the same elec-tronic manifold in gas-phase molecules are much larger (e.g.ωe =ν/c∼2360 cm−1 for N2) than the rotational separa-tions, only a small fraction populates the vibrational hot bandsat room temperature; note that the scaling parameter in theabove exponential hc/kB=1.44 K cm−1. Both the low- andhigh-temperature regimes can be measured if rotational lineswithin the vibrational manifolds can be resolved. Ns-laser-based high-resolution VCARS has been routinely used inmany reacting-fuel and plasma studies; however, the ns-VCARS—like in ns-RCARS, as discussed above—is plaguedby significant NRB [3].

Roy et al were the first to demonstrate the ps-laser-basedVCARS in the gas phase [158, 203]; this work has demon-strated three orders-of-magnitude reduction of NRB, as dis-cussed in section 4.2. Ps-VCARS has been employed inhydrocarbon–air flames operating at different equivalenceratios; observed N2 spectra exhibited up to three vibrationalbands [204]. By temporally delaying the probe pulse in ps-VCARS, the Raman-coherence lifetimes of the rotationallyresolved Q-branch transitions of H2 have been directly mea-sured [205]; also, the temperature has been extracted [206].The high-speed capabilities of ps-VCARS have been sig-nificantly enhanced with the advent of the BML [72]. Forexample, gas-phase thermometry at a 100 kHz repetition ratehas been demonstrated using ps-VCARS of H2 in an H2–airflame stabilized over a Hencken burner [110]. It has also beenshown that the ps duration is optimal for coupling the laserpulse into optical fibers, allowing the realization of fiber-based CARS [207]; single-shot, fiber-based ps-VCARS hasalso been demonstrated for measuring temperature [208]. Thehigh-speed temperature measurement capabilities can betransitioned to highly-transient plasmas, and the fiber CARScan be implemented for harsh environments.

Montello et al have employed rotationally resolved ps-VCARS to measure the vibrational and rotational/

translational temperatures in the subsonic plenum andsupersonic flow of a highly nonequilibrium Mach 5 flow witha sustained discharge [209]. They have also reported theeffect of the injection of CO2, NO, and H2 in the downstreamof the pulser-sustainer discharge. In figure 31, temperaturesextracted from the N2 Q-branch VCARS spectrum is pre-sented and the flow is highly nonequilibrium with vibrationaltemperature three times higher than the rotational temper-ature. In contrast, when only 1 Torr (partial pressure) of CO2

is injected, almost all the vibrational energy is removed fromN2, resulting in a vibrational temperature Tvib=815 K and arotational/translational temperature Trot=630 K. In a point-to-plane ns streamer discharge, similar to those used in tran-sient-plasma ignition, ps-VCARS has been employed tomeasure Trot and Tvib to study the vibrational loading [210]. Itwas observed that the direct electron impact on vibrationalexcitation of N2 is relatively inefficient in this discharge in air,but in fuel–air mixtures the electron impact accounts fornearly half of the vibrational excitation of N2. Rapid localizedheating of ns-pulse discharge has been studied employing ps-VCARS in a diffuse filament between a pair of sphericalelectrodes; measurements were made both in N2 and air at100 Torr [84]. In N2, a temperature rise up to ∼200 K hasbeen reported within ∼1 μs, i.e. on timescales shorter thanacoustic timescales. The vibrational loading in repetitivelypulsed plane-to-plane plasma has been studied employingVCARS and it is reported that up to ∼50% of the coupleddischarge power is found to load vibrations [83].

On the fs-CARS development, after Hayden andChandler’s first gas-phase fs-VCARS demonstration [151],many studies have been reported on fs-VCARS for gas-phasethermometry using several species, including hydrogen [159],nitrogen [165], etc. Unlike ps-CARS, since fs-CARS isinherently collision independent, the fs-VCARS signal of N2

could be obtained for pressures up to 5 bar and used forstudying collision-induced rotational-energy transfer rates[167, 169]. Time-resolved fs-VCARS was also employed forhigh-temperature thermometry in a H2–air diffusion flame,where the decay rate of the initial Raman coherence was usedas a measure of temperature [155]. Since at different tem-peratures a different number of rovibrational levels are exci-ted, the collective CARS signal obtained from those levelsdephases at different rates; the collective dephasing has beennamed ‘frequency-spread dephasing’ (FSD) [154, 168].Measuring those decay rates from time-resolved VCARSyields the temperature of the target N2 gas. Furthermore, thebroadband fs lasers (bandwidth ∼350 cm−1 for 100 fs pulse)can simultaneously excite multiple molecules at the targetvolume; Roy et al have investigated the polarization beatingof N2 in the presence of CO in the FSD-based fs-VCARSthermometry [156]. It has been reported that the presence ofCO does not impact the accuracy of N2-based thermometricmeasurement, particularly for the CO concentrations presentin typical reacting flows. Furthermore, the depth of themodulation produced from polarization beating could be usedfor extracting other major species with concentration >5%.Similarly, the presence of CO2 on O2 fs-VCARS thermometryhas also been investigated in a heated cell and it has been

Figure 31. N2 Q-branch CARS spectra of pulser-sustainer dischargeat 300 Torr pressure in pure nitrogen (black dash-dot line) and afterthe injection of CO2 with partial pressure of 1 Torr (blue solid line).[209] (2012) (© Springer-Verlag 2012). With permission ofSpringer.

27


observed that the thermometric measurement is affected whenCO2 concentration reaches over 10%; it is concluded thatcareful modeling of fs-CARS is required to achieve thethermometric accuracy for such mixtures [211]. Bitter andMilner have presented another strategy for single-shot mea-surement of species concentration by separating the VCARSsignals from a mixture of O2 and CO2 by exploiting a ‘silencewindow’ of one of the species in time domain [212]. Thecollisional effects on N2 fs-VCARS is studied at high pres-sures (up to 50 bars) in the presence of different collisionalpartners [46], as discussed earlier in section 4.4.

It has been shown that the thermometric information fromN2 VCARS can be extracted by mapping the FSD of the Ramancoherence using a chirped-probe pulse [86], as discussed earlierin figure 24. The measurement accuracy and precision formeasuring a temperature range of 300–2400K in a near-adia-batic laboratory flame is found to be ∼1%–6% and ∼1.5%–3%,respectively. This method is applied to extract the temperaturefrom a driven flame at 1 kHz-data-acquisition rate [171]. Adetailed model for the chirped-probe fs-CARS is developed, andthe results are compared with the experiment; also, it is reportedthat the mean temperature accuracy is within ∼3% with respectto the adiabatic flame temperature [213]. Chirp-probe fs-VCARS is also used to measure species concentrations in gasmixtures of CO–N2 and Ar–N2 at atmospheric pressure [214].

Furthermore, to selectively measure species in a mixture ofmultiple molecules, a spectral-focusing method was employedby Wrzesinski et al [198]. In this approach, the broadband fspump and Stokes pulses were chirped and delayed with respectto each other, which allowed the pump and Stokes pair toselectively and controllably excite the Raman transition of oneof the molecules. In a 10-bar, 1:1 binary mixture of O2 andCO2, selective excitation of each of these species wasdemonstrated by varying the delay between the pump andStokes pulse. Keeping the delay of the pump and Stokes pulsefixed, the coherence dephasing of either of the species wasmeasured from the time-resolved VCARS signal using adelayed probe without interference from other species.

The hybrid fs/ps-CARS approach (described earlier insection 5.1) is also employed for spectrally-resolved VCARSof N2 [215]. Using a pulsed high-temperature flame, high-speed real-time N2 thermometry is demonstrated, as shown infigure 32. The narrowband time-asymmetric ps-duration probeis prepared using a 4f pulse shaper and filters; using such aprobe hybrid, VCARS of N2 is measured in a H2–air stabilizedover a Hencken flame and it is concluded that using a sharplyraised probe pulse, the NRB can be assumed to be zero—eliminating the need for correction procedures [216]. Enhancedsensitivity of flame thermometry has been reported employinghybrid VCARS of N2 for a temperature range 300–2400K[217]. Hybrid VCARS has also been used to measure the molefraction of CH4 in a gas binary mixture with N2 [218].Scherman et al have achieved a high spectral resolution of∼0.7 cm−1 in the hybrid-VCARS measurement of N2 using afiber-based fs-laser and deriving the ps probe from the fs beamusing a volume Bragg grating [219]. Stauffer et al have pre-sented a detailed theoretical and computational analysis ofhybrid VCARS with time- and frequency-domain models for

both gas-phase thermometry and condensed-phase excited-state dynamics to show the robustness in quantitative analysisusing various probe-pulse shapes [220]. Marrocco has pre-sented theoretical analysis of the effects of different pulseshaped in hybrid CARS [221]; also, analytical results for dif-ferent pulse shapes are presented [222].

In a recent study, the chirped-probe CARS and hybrid-VCARS methods are compared for temperature measurementin near-adiabatic flames [223]. It is reported that chirped-probe-pulse CARS requires more optimization of modelparameters to extract the temperature compared to the hybridCARS, and hence the computational cost is significantly lessfor the hybrid-CARS configuration. Another variant of hybridps/fs-VCARS has been reported, where both pump and probeare ps-duration pulses and Stokes is a broadband fs pulse, forobserving both time and frequency-resolved CARS spectra ofvarious C-H stretch regions in condensed-phase molecules[224, 225]. Hybrid CARS and other VCARS configurationshave also been used in completely different applications, e.g.for the standoff detection of bacterial spores [226–229], andin high-wavenumber regimes in liquids [230–232].

5.3. Broadband CARS (single-beam fs, dual-pump fs/ps, ultra-broadband fs/ps)

The availability of ultrashort lasers, with their durationsshorter than the typical molecular vibrational period andbandwidths of a few thousand cm−1, has enabled single-beamVCARS spectroscopy for microscopic and standoff detections[233–236]. Typical broadband fs-duration lasers providesufficient bandwidth (∼200–300 cm−1), with correspondingtransform-limited durations (50–70 fs) to access low-fre-quency Raman modes (0–600 cm−1) via impulsive excitation.However, frequency-conversion approaches have been

Figure 32. Time series of temperature obtained from fundamental-to-hot-band-area ratio correlation for 500 Hz hybrid-CARS spectra (a)for different fuel-to-air ratio Φ and (b) for a pulsed flame.Reproduced with permission from [215].

28


developed that allow the formation of ultra-broadband ultra-fast pulses with bandwidths of the order of a few thousandcm−1. Employing phase and polarization shaping of broad-band continuum of ultrashort pulses of duration ∼7 fs with aspectral bandwidth of ∼3500 cm−1, as introduced by Silber-berg and co-workers [237], Roy et al have shown how abroadband pump and Stokes and a narrowband probe beampresent in the single beam can be utilized for N2 CARSspectroscopy [238]; the spectra of the three beams and thegenerated CARS signal are shown in figure 33. This single-beam configuration has also been employed for the simulta-neous excitation of CO2 Fermi dyads at ∼1285 and∼1388 cm−1, and the variation of CARS signal intensities forthe two lines under different pressure conditions have beenanalyzed and the concentration of CO2 is measured in amixture of CO2–Ar in a cell [239]. Singe-shot N2 thermo-metry is demonstrated employing a single-beam configurationwith three different delayed probes using glass cover slips tomeasure the Raman dephasing and extract temperature from adephasing-temperature calibration [240].

Bohlin and Kliewer demonstrated the utilization of theultra-broadband spectrum of 3500 cm−1 associated with a 7 fspulse in a hybrid-CARS configuration—which is also calledas multiplex CARS—to acquire the RCARS and VCARSdata simultaneously from multiple gas-phase species, such asN2, H2, CO2, O2, and CH4 [241]. The spectrally-resolved,wideband, multiplexed VCARS of a few species obtainedfrom a two-beam configuration is presented in figure 34. Thesame group extended the ultra-broadband multiplex CARS toobtain simultaneous single-shot rotational and vibrationalCARS of H2 [242]. Covering a spectral region spanning4200 cm−1, they have demonstrated planar thermometricmeasurements with high spectral and spatial resolution in aH2-diffusion flame. This approach has further been employedto obtain simultaneous CARS imaging and wideband che-mical detection in a fuel-rich premixed hydrocarbon flat flame(ethylene/air Φ=2.35) [243]. The authors of the work have

demonstrated robust 1D-CARS thermometry from the spec-trally-resolved rotational S-branch transitions of N2 for atemperature range of 300–1600 K.

5.4. Simultaneous measurements of vibrational and rotationaltemperatures for the characterization of nonequilibriumenvironments

The high-speed, spatially-resolved, multidimensional, single-shot CARS capability for the simultaneous measurement ofvibrational and rotational temperatures, as discussed in theprevious subsections, would be extremely useful for highly-transient plasmas to understand the degree of nonequilibrium inplasmas and the process of energy transfer to vibrational androtational/translational states of the plasmas (also discussed insection 5.2). In the highly-transient nonequilibrium plasmas,high-spatial and -temporal resolution thermometry and species-concentration measurement are critical for getting quantitativeinsight into the plasma kinetics, as discussed earlier insections 2.2 and 2.3. Montello et al have employed ps-CARS toobtain the nitrogen Q-branch spectrum in a nonequilibriumsubsonic plenum and supersonic wind tunnel excited by a nspulser/DC sustainer electric discharge with enhanced spatialresolution [209]; the rotational/translational and vibrationaltemperature are derived from the rotationally resolved N2

VCARS spectrum. It is concluded that while the pulsed-dis-charge results in significant vibrational loading, the rise inrotational/translational temperature remains nominal. Transient-plasma ignition conditions are evaluated by measuring the N2

vibrational CARS in a discharge afterglow in C2H4/air,CH4/air, and C3H8/air mixtures; this measurement allows forthe determination of the temporal evolution of rotational andvibrational temperatures [210]. The vibrational energy loadingand relaxation kinetics of N2 in a pulsed nonequilibrium plasmain both pin-to-pin and plane-to-plane geometry have been stu-died using ps-VCARS, and rotational/translational temperatureis extracted from the rotationally resolved signal [83]. In aplasma-assisted combustion, simultaneous LIF of intermediatespecies OH and ps-VCARS-based vibrational and rotationalthermometry has been presented [244]. Note that all of thesimultaneous measurements of the rotational and vibrationaltemperature are derived from spectrally-resolved VCARS only.

Although ps-CARS can provide sufficient resolution tomeasure rotational and vibrational temperature simultaneously,such measurements are limited to low repetition rates, as dis-cussed earlier. Extending the ps-CARS measurements to fs-laser-based measurements, all of the broadband advantages forRaman excitations are realized. Employing fs/ps-hybridCARS, simultaneous rotational and vibrational CARS ofmultiple species (N2, O2, CH4, H2), mole fractions of thecombustion species has been measured simultaneously [245].Dedic et al have measured N2 vibrational and rotationaltemperature in a ns-laser-induced spark generated in a nitrogenjet [246]; and they have shown that the rotational temperaturevaries from 540–542 K, and the vibrational temperature variesfrom 1100–1600 K when the measurement time was delayedbetween 2–5 μs after the shock. The simultaneous measure-ment of RCARS and VCARS in an ns-laser spark of N2 jet are

Figure 33. Spectrum of the 7 fs beam (gray) and the N2 CARS signal(blue). Extracted spectrum of the probe beam using the phase andpolarization masks is shown in the left inset. Temporal profiles of theexcitation and probe part of the beam along with their overlap arepresented in the right inset. Reprinted with permission from [238],Copyright (2009), AIP Publishing LLC.

29


reported; also, the shadowgraph of the shock wave is recorded[247]. Many rotational lines and vibrational hot bands in theCARS spectra are reported to appear for longer measurementdelays after the laser-induced shock, which demonstrated theincrease in both rotational and vibrational temperature, parti-cularly after 1.1 μs. However, the signal is reported to declineafter 2 μs primarily because of the expansion of the plasmakernel. The dual-pump ps/fs-hybrid CARS has been employedin a detonation environment to measure simultaneously thevibrational and rotational temperature on a shot-to-shot basis[248]. Dedic, Meyer and Michael have demonstrated single-shot rotational and vibrational temperature in a DBD plasmausing the hybrid-CARS signal from S- and Q-branch of N2, asshown in figure 35 [249]. The authors have reported thatbecause of the high concentration of N2, the plasma wasunsteady and the location of the large filament structure waschanging with time. Clearly, the rotational/translationaltemperature almost remains invariant, whereas the vibrationalloading is found to differ significantly from shot to shot. Thesemeasurements in the nonequilibrium plasma are achieved witha spatial resolution of ∼650 μm and a sampling rate of <10 ps.This work outlines the potential of linear or planar measure-ments at higher repetition rates in plasma environments.

6. Prospects for collision-free measurements ofother chemical species

CARS has long been used as a preferred approach for gas-phase thermometry, but it is typically limited to major flameor plasma species (such as N2 and O2), because the CARSsignal intensity is proportional to the square of the numberdensity of the target species. Hence, the SNR is low for thedetection of minor species [43]. Since minor species also play

an important role in spatial, temporal, and chemical dynamicsin a reactive environment, it is important to develop suitabledetection techniques. A variant of CARS known as ERE-CARS (briefly discussed in section 4.3), where the probecouples resonantly to the electronic transition in the measuredspecies, has been demonstrated using ns-duration lasers to

Figure 34. Multiplex CARS with high spectral resolution measurement of vibrational manifold for four different species (a) CH4, (b) N2, (c)CO2 Q-branch, and (d) CO2 S-branch. Experimental setup was the same for all of the experiments. Reprinted with permission from [241].Copyright (2014), AIP Publishing LLC.

Figure 35. (a) Histogram of the extracted single-shot vibrationaltemperature Tvib and rotational temperature Trot measured at thecenter of the DBD. Pair of single-shot spectra plotted in (b) and (c)with the corresponding best-fit temperatures Trot=380 K,Tvib=2850 K, respectively; similarly (d) and (e) correspond toTrot=390 K, and Tvib=3460 K, respectively. Reproduced withpermission from [249].

30


enhance the CARS signal by a few orders-of-magnitude and,hence, is suitable for the detection of minor species such asNO [250]. Even with the use of pulses longer than quenchingtimescale, ERE-CARS has been demonstrated to be quench-ing independent [251]. In ERE-CARS, the simultaneousexploitation of both vibrational and electronic resonancesduring the three-color excitation process results in a highdegree of chemical specificity, allowing for detection that isfree of interference from other species with similar electronictransitions [252]. From careful theoretical analysis it wasdemonstrated that ERE-CARS is dependent on the coherence-dephasing rate but independent of the electronic quenchingrate, and such dephasing dependence can be overcome byoperating near the saturation condition for probe coupling[163, 253]. The saturation conditions for the probe pulse inERE-CARS have been analyzed theoretically for pulsedurations ranging from ns to fs [164]. It is observed that forpulse durations longer that collisional timescales, the thresh-old intensity for saturation is independent of the duration ofthe pulse; however, for pulse durations shorter than collisionaltimescales, the threshold intensity of saturation is inverselyproportional to the square of the pulse duration. The advan-tage of electronic resonance was also realized in other con-figurations, such as three-photon resonant CARS [254] anddegenerate four-wave mixing (DFWM) [43].

In spite of these advantages associated with the electronicresonances, these techniques are typically limited to fre-quency-domain approaches that exploit ns-duration laserpulses; also this technique is often limited to acquisition ratesof ∼10 Hz based on the repetition rates of the high pulseenergy ns lasers.

6.1. FREE-CARS for minor-species detection

Recently, Stauffer et al demonstrated an fs-laser-based, non-invasive optical approach that allows simultaneous gas-phasethermometry and minor-species-concentration measurementsunder single-laser-shot conditions [255]. In this approach,referred to as FREE-CARS, two independently tunable, UVpulses are employed to induce resonant coupling of electronictransitions of minor species. The FREE-CARS approachprovides highly sensitive, spectrally-resolved signature ofmultiple rovibronic transitions of minor combustion species(OH, NO). This technique allows for the simultaneous accessto molecular transitions spanning 200–300 cm−1 (anapproximately ten-fold increase over prior DFWM approa-ches) at acquisition rates of 1 kHz (100-fold increase).Implementation of the broadband fs-duration pulses in FREE-CARS also allows simultaneous temperature and species-concentration measurements that are quenching independent[96]. The versatility of fs-duration laser pulses also allowseasy access to multiple tunable pulses that cover a broadrange of wavelengths; thus, these approaches can exploitmultiple electronic resonances to increase detection sensitiv-ity dramatically. The experimental setup used for the FREE-CARS measurement is shown in figure 36(a). With the abilityto scan the timing between the pulses (as shown infigure 36(b)), many rovibrational states in the ground and

excited electronic states can be mapped. The resonant cou-pling of the two UV lasers in the FREE-CARS configurationand the diagram associated with the most significant term inthe FWM expansion are presented in figure 36(c).

Spectrally and temporally resolved FREE-CARS signalof NO acquired from a mixture of NO with Ar is plotted infigure 37; figure 37(a) is the contour plot of the FREE-CARSsignal; figure 37(b) is the typical time-resolved FREE-CARSsignal obtained by integrating over the spectrum; figure 37(c)is the spectrally-resolved signal obtained by integrating thesignal over the whole range of the delay τ23 [255]. All ofthese data were taken with the pump-Stokes delay τ12 fixed at500 fs. While the electronic dephasing is likely responsiblefor the observed decay of the signal, the high spectral reso-lution is afforded by the signal integration over the electronicdephasing time of several ps.

The high-resolution measurement capability of thistechnique has been employed to determine simultaneously thetemperature and number density of OH from single-laser-shotFREE-CARS in a laminar flame, as shown in figure 38 [96].The highly sensitive single-shot temperature data wereobtained with excellent temperature accuracy (2.2% com-pared to adiabatic flame temperature) and precision (average2% standard deviation) and compared very well with previousresults [215]. Similarly, the accuracy and precision of thenumber-density measurements are reported to be reasonable,with high detection sensitivities (100–300 ppm) at flametemperature.

Molecular wavepacket dynamics and their control hasbeen demonstrated employing fs lasers resonant to moleculartransitions in various diatomic molecules employing pump-probe configurations [256–259].

Figure 36. Schematic of experimental FREE-CARS setup. (b)Timing sequence for three input (ω1, ω2, ω3) laser pulses andgenerated FREE-CARS signal ωCARS. (c) Diagram associated withthe most significant FWM term. Reproduced with permissionfrom [96].

31


6.2. Other FWM techniques

Many FWM techniques have been extensively used for gas-phase diagnostics [43] and in plasmas [53, 56]. In a recentwork, an fs time-resolved parametric FWM (fs-PFWM)technique is developed [260]. In PFWM, a molecular elec-tronic state is excited by a resonant two-photon transitionusing an fs-laser, which is followed by a second fs-durationprobe pulse (ω

2) inducing the coherent and directed outgoing

signal pulse (ωPFWM

), allowing much faster measurements to bemade compared to the fluorescence lifetime. Note that theexcitation scheme of PFWM is similar to TALIF; but unlikeTALIF that generates signal in all 4π steradian, the PFWMsignal is directed because of the coherent probing of the two-photon coherence. By precisely scanning the temporal delaybetween pulses 1 and 2, the time-dependent spectral signature

of the two-photon-excited species can be mapped using thisapproach. Figure 39, for example, depicts the fs-PWFM sig-nal observed for room-temperature nitric oxide (NO, 10%mole fraction) in a mixture containing both N2 and CO gases.This time-dependent signal exhibits a reproducible oscillatorystructure that results from interferences between multiplerovibrational levels in the first electronically excited state ofNO, all of which are simultaneously excited via two-photontransitions within the broadband spectrum of the pump pulse(inset of figure 39). The simulated time-domain signaldepicted in figure 39 results from simulations of the time-domain third-order polarization that account for the calculated

Figure 37. (a) Spectrally and temporally resolved FREE-CARSspectra of NO as the contour plot with fixed τ12=500 fs. (b)Spectrally integrated, temporally resolved, normalized FREE-CARSsignal as a function of pump-probe delay τ23. (c) Spectrally-resolved, normalized FREE-CARS signal integrated over the wholeτ23 range. Reprinted with permission from [255]. Copyright (2016),AIP Publishing LLC.

Figure 38. Single-shot OH spectra (solid curves) plotted for differentequivalence ratios (a) f=0.45, (b) f=1.0. Simulated spectra(dashed) and residual (brown line offset for clarity) are shown. In theinset, histograms of the best-fit temperature for 1000 laser shots areshown. Reproduced with permission from [96].

Figure 39. Experimental and simulated fs-PFWM observed for nitricoxide (NO) gas in the presence of N2 and CO2 bath gases. Inset:Calculated room-temperature Boltzmann-weighted NO TPA spec-trum. Squared-magnitude of pump-pulse intensity profile includedfor reference. [260] John Wiley & Sons (Copyright © 2016 JohnWiley & Sons, Ltd).

32


two-photon-absorption cross-sections from ground-state NO,assuming a room-temperature Boltzmann distribution of theinitial states.

Although all of the resonant nonlinear techniques dis-cussed in this section have been realized for mixtures in gascells or reacting flows, they can be easily extended to measureminor species in reacting plasmas too. These techniques willbe particularly useful for measuring the time-resolved speciesconcentration of reaction intermediates in plasmas, e.g. NO,OH, etc, that are crucial to understanding the kinetics.

7. Concluding remarks and future opportunities

We have reviewed the rapid growth and advances in thedevelopment of ultrafast-laser-based spectroscopy and ima-ging for reacting plasmas and flames over the past two dec-ades. We have discussed spectroscopic measurements withexpanded capabilities for obtaining spatially and temporallyresolved measurements of the key performance parametersrelated to highly-transient reacting flows such as temperature,number densities of atomic and molecular species, electricfields, and velocimetry. With orders-of-magnitude higherintensities compared to traditional ns lasers, ultrashort-pulselasers have enabled 1D and 2D single-shot measurements atrepetition rates of 1 kHz or higher, which was unthinkableonly a few years ago. We have also reviewed the use of fs-laser-based measurements for circumventing various inter-ferences in spectroscopic approaches that are common in ns-laser-based measurements. We have discussed the ability ofthese new measurement techniques to enable detailed under-standing of plasmas and flows and aid the development ofpredictive models for reacting plasmas and flames.

Highly-transient and nonequilibrium plasmas have beenwidespread in many plasma applications in a variety ofconfigurations such as DBDs, pin-to-pin discharges, andplasma jets sustained by AC, DC, or short (ns) electricalpulses. However, until recently, plasma characterization hasmainly employed long-duration ns lasers, which severelylimited capabilities in terms of the quantitative understandingof plasmas. We have discussed in detail the current state-of-the-art plasma diagnostics for electric-field measurementsusing both linear (such as emission spectroscopy) and non-linear (such as FWM, SHG, and CARS) methods, electrontemperature and density measurements (using Thomsonscattering), thermometry (using CARS), number density ofatomic species such as H, O, and N atoms and radical speciessuch as OH and NO (using LIF and PLIF). The technicalchallenges and future prospects have been outlined for each ofthese measurements.

Although ns-laser-pumped LIF and PLIF measurementshave been extremely useful in measuring minor species withvery high sensitivity, they have not been suitable for mea-suring higher concentrations of intermediate atomic andmolecular species, because of large absorptions. With TALIFmeasurements, the two-photon excitation of atomic specieswith UV beams has reduced absorption during the propaga-tion, but since the nonlinear process is intensity dependent, a

much higher fluence has been required to obtain the ns-TALIF signal. The higher fluence causes photolytic dis-sociation of other molecules, which has resulted in extremelylarge photolytic interference in the signal. In contrast, theexcitation of atomic and molecular species with short-dura-tion pulses has essentially eliminated the photolytic dis-sociation. It has been shown that the fs-laser with its highintensity and ultrashort-duration efficiently excites the mea-sured species, and the TALIF signal has been generatedwithout photolytic interferences. The comparison of ns-TALIF and fs-TALIF has been discussed and recentadvancements in accurate fs-TALIF measurements of keyintermediates such as H and O atoms and reactive radicals OHand NO, along with the single-shot and 2D measurementcapabilities have been reviewed.

For thermometric measurements in reacting plasmas andflames, ns-CARS has been the most popular nonintrusivetechnique. However, ns-CARS has suffered from the fol-lowing drawbacks: the signal, accompanied by strong NRB,has limited the accuracy and sensitivity; also the thermometricaccuracy has been strongly dependent on the collisionalmodels, which limited its usefulness to steady flows withknown collisional environments. In this review, we havediscussed the impact of fs lasers in fueling the developmentsin ultrafast CARS research with many advanced capabilitiesthat were unthinkable only a few years ago. Very rapidgrowth has taken place over the past two decades in fs-CARS-based thermometry and species-concentration mea-surement capabilities, including: (a) NRB-interference-freemeasurements, (b) time-resolved single-shot measurementswith kilohertz or higher repetition rate, (c) spatially-resolvedmeasurements, (d) 1D and 2D measurements, and (e) simul-taneous measurements of multiple species. The paths to rea-lization of each of these capabilities have been extensivelyreviewed. Also, we have discussed the new avenues that haveopened up with these capabilities, such as time-resolvedmeasurements at high pressure (50 bar or higher) afforded bycollision-independent fs-CARS and thermometry using fiber-coupled CARS in harsh environments.

Many different configurations that have been employedfor fs-CARS-based interference-free and collision-indepen-dent gas-phase thermometry and species-concentration mea-surements in the literature have been extensively discussed.Myriad recent developments in rotational- and vibrational-temperature measurements employing ps lasers, chirped fspulses, spectral focusing, single-beam, and hybrid fs/ps-CARS have been reviewed both for temporally and spec-trally-resolved detections. These developments have beenreported ranging from equilibrium to highly nonequilibriumconditions with test targets such as gas cells, gas jets, adia-batic and turbulent flames, hypersonic environments, high-pressure plasmas and flames. The ultra-high bandwidth(∼3500 cm−1) afforded by ultrashort, 7 fs pulses has allowedthe realization of single-beam CARS thermometry andsimultaneous multiple species measurements (N2, H2, O2,CH4, CO2, etc). Simultaneous 2D single-shot, temperatureand concentration measurements have significantly increasedthe prospect of employing fs-CARS tools for understanding

33


combustion mechanisms in turbulent flows and practicalcombustors. For highly-transient nonequilibrium plasmas, therotational/translational temperature can be very differentfrom the vibrational temperature, and the degree of none-quilibrium dictates the plasma kinetics. Hence, simultaneousmeasurement of rotational and vibrational temperatures withhigh temporal and spatial resolution has been critical tounderstanding energy transfers and complex plasma kinetics;such measurements will help in validating the plasma models.A few recent studies from the literature have been discussedwhere simultaneous measurements of the rotational andvibrational CARS signal in nonequilibrium plasmas wereperformed to obtain single-shot rotational/translational andvibrational temperatures. A few resonant variants of CARSand FWM have been presented, which would be suitable fortime-resolved concentration measurement of reaction inter-mediates in the plasmas.

7.1. Future opportunities

Fs-TALIF can be useful for measuring many other chemicalspecies and reaction intermediates. While fs-TALIF excitationis inherently collision independent, the measurement timelimits the capability of this technique for obtaining the time-resolved fluorescence signal. Advanced gated detection withgate duration shorter than the quenching timescale could aidthe direct measurement of the quenching rates in transientconditions. Such gated detection may be helpful for makingmeasurements in high-pressure conditions to obtain scalinglaws of the reaction intermediates. With the availability of thehigh-repetition rate (kilohertz to megahertz) high-powerlasers, TALIF measurements may be extended to volumetricmeasurements. Such high-dimensional measurements couldaid in the capture of the complete dynamics associated withturbulent flows; e.g. in highly-transient plasmas or in high-pressure plasma conditions. Many future opportunities existfor fast, interference-free, higher-dimensional ultrafast diag-nostics of plasmas using CARS, FWM, etc. These ultrafastdiagnostics could be able to perform plasma characterizationand imaging in extreme conditions such as those under theSun-like extreme mix of temperature (hundreds of millions ofKelvin), pressure (hundreds of millibars to gigabars), andnumber density; direct measurement of these parameterswould be critical for the realization of self-sustainable fusionignition in a controllable way for extracting useful energy.

Acknowledgments

We gratefully acknowledge a host of outstanding colleaguesand collaborators with whom we have interacted over thecourse of the past decade in the area of TALIF and CARSspectroscopy. Their tremendous expertize and enthusiasmhave contributed immeasurably to many of the effortsdescribed in this review. Of particular note are the contribu-tions of late Prof Walter Lempert, Prof Terrence Meyer, DrHans Stauffer, Dr Paul Hsu, Dr Naibo Jiang, Dr Paul Wrze-sinski, Dr Joseph Miller, Dr Daniel Richardson, Dr Benjamin

Hall and Dr Jake Schmidt. Special thanks to Dr Hans Staufferfor thorough and detailed proof reading of the manuscript andexcellent discussions on ultrafast CARS, and to Dr NaiboJiang for helpful discussions on plasma diagnostics. Theauthors gratefully acknowledge the funding provided by theAir Force Research Laboratory under Contract Nos. FA8650-15-D-2518, by the Air Force Office of Scientific Research (DrEnrique Parra and Dr Chiping Li, Program Officers), and theeditorial assistance provided by Ms Marian Whitaker.

References

[1] Bruggeman P J and Czarnetzki U 2016 Retrospective on ‘The2012 Plasma Roadmap’ J. Phys. D: Appl. Phys. 49 431001

[2] Samukawa S et al 2012 The 2012 plasma roadmap J. Phys.D: Appl. Phys. 45 253001

[3] Roy S, Gord J R and Patnaik A K 2010 Recent advances incoherent anti-Stokes Raman scattering: Fundamentaldevelopments and applications in reacting flows Prog.Energ. Combust. Sci. 36 280–306

[4] Ochkin V N 2009 Spectroscopy of Low temperature Plasmas(Weinheim: Wiley-VCH Verlag GmbH & Co. KGaA)

[5] Hurricane O A et al 2014 Fuel gain exceeding unity in aninertially confined fusion implosion Nature 506 343–8

[6] Ju Y G and Sun W T 2015 Plasma assisted combustion:Dynamics and chemistry Prog. Energ. Combust. Sci. 48 21–83

[7] Penkov O V, Khadem M, Lim W S and Kim D E 2015 Areview of recent applications of atmospheric pressure plasmajets for materials processing J. Coat. Technol. Res. 12 225–35

[8] von Woedtke T, Reuter S, Masur K and Weltmann K D 2013Plasmas for medicine Phys. Rep. 530 291–320

[9] National Research Council 1996 Database Needs forModeling and Simulation of Plasma Processing(Washington, DC: National Academies Press) (https://doi.org/10.17226/5434)

[10] Ryo O 2016 Optical diagnostics of reactive species inatmospheric-pressure nonthermal plasma J. Phys. D: Appl.Phys. 49 083001

[11] Benard N and Moreau E 2014 Electrical and mechanicalcharacteristics of surface AC dielectric barrier dischargeplasma actuators applied to airflow control Exp. Fluids 55 43

[12] Kotsonis M 2015 Diagnostics for characterisation of plasmaactuators Meas. Sci. Technol. 26 30

[13] Leonov S B, Adamovich I V and Soloviev V R 2016Dynamics of near-surface electric discharges andmechanisms of their interaction with the airflow PlasmaSources Sci. Technol. 25 063001

[14] Popov N A 2016 Kinetics of plasma-assisted combustion:effect of non-equilibrium excitation on the ignition andoxidation of combustible mixtures Plasma Sources Sci.Technol. 25 043002

[15] Starikovskaia S M 2014 Plasma-assisted ignition andcombustion: nanosecond discharges and development ofkinetic mechanisms J. Phys. D: Appl. Phys. 47 353001

[16] Bruggeman P and Leys C 2009 Non-thermal plasmas in andin contact with liquids J. Phys. D: Appl. Phys. 42 053001

[17] Bruggeman P J et al 2016 Plasma-liquid interactions: areview and roadmap Plasma Sources Sci. Technol. 25 59

[18] Kong M G, Kroesen G, Morfill G, Nosenko T, Shimizu T,van Dijk J and Zimmermann J L 2009 Plasma medicine: anintroductory review New J. Phys. 11 35

[19] Graves D B 2012 The emerging role of reactive oxygen andnitrogen species in redox biology and some implications forplasma applications to medicine and biology J. Phys. D:Appl. Phys. 45 42

34


https://doi.org/10.1088/0022-3727/49/43/431001

https://doi.org/10.1088/0022-3727/45/25/253001

https://doi.org/10.1016/j.pecs.2009.11.001



https://doi.org/10.1038/nature13008






https://doi.org/10.1007/s11998-014-9638-z

https://doi.org/10.1007/s11998-014-9638-z

https://doi.org/10.1007/s11998-014-9638-z

https://doi.org/10.1016/j.physrep.2013.05.005



https://doi.org/10.17226/5434

https://doi.org/10.17226/5434

https://doi.org/10.1088/0022-3727/49/8/083001

https://doi.org/10.1007/s00348-014-1846-x

https://doi.org/10.1088/0957-0233/26/9/092001

https://doi.org/10.1088/0963-0252/25/6/063001

https://doi.org/10.1088/0963-0252/25/4/043002

https://doi.org/10.1088/0022-3727/47/35/353001

https://doi.org/10.1088/0022-3727/42/5/053001

https://doi.org/10.1088/0963-0252/25/5/053002

https://doi.org/10.1088/1367-2630/11/11/115012

https://doi.org/10.1088/0022-3727/45/26/263001

[20] Becker K H, Schoenbach K H and Eden J G 2006Microplasmas and applications J. Phys. D: Appl. Phys.39 R55

[21] Tardiveau P, Magne L, Marode E, Ouaras K, Jeanney P andBournonville B 2016 Sub-nanosecond time resolved lightemission study for diffuse discharges in air under steep highvoltage pulses Plasma Sources Sci. Technol. 25 054005

[22] Höft H, Becker M M, Loffhagen D and Kettlitz M 2016 Onthe influence of high voltage slope steepness on breakdownand development of pulsed dielectric barrier dischargePlasma Sources Sci. Technol. 25 064002

[23] Stepanyan S A, Starikovskiy A Y, Popov N A andStarikovskaia S M 2014 A nanosecond surface dielectricbarrier discharge in air at high pressures and differentpolarities of applied pulses: transition to filamentary modePlasma Sources Sci. Technol. 23 045003

[24] Nishihara M, Takashima K, Jiang N, Lempert W R,Adamovich I V, Rich J W, Doraiswamy S and Candler G V2012 Development of a Mach 5 nonequilibrium-flow windtunnel AIAA J. 50 2255–67

[25] Lu X, Laroussi M and Puech V 2012 On atmospheric-pressure non-equilibrium plasma jets and plasma bulletsPlasma Sources Sci. Technol. 21 034005

[26] Brandenburg R, Bruggeman P J and Starikovskaia S M 2017Fast pulsed discharges Plasma Sources Sci. Technol. 26020201

[27] Dobele H F, Mosbach T, Niemi K andSchulz-von der Gathen V 2005 Laser-induced fluorescencemeasurements of absolute atomic densities: concepts andlimitations Plasma Sources Sci. Technol. 14 S31–41

[28] Eckbreth A C 1996 Laser Diagnostics for CombustionTemperature and Species (Amsterdam: Gordon and Breach)

[29] Schulz C and Sick V 2005 Tracer-LIF diagnostics:quantitative measurement of fuel concentration, temperatureand fuel/air ratio in practical combustion systems Prog.Energ. Combust. Sci. 31 75–121

[30] Schmidt J B, Jiang N and Ganguly B N 2014 Nitric oxidePLIF measurement in a point-to-plane pulsed discharge invitiated air of a propane/air flame Plasma Sources Sci.Technol. 23 2820–6

[31] Verreycken T, Sadeghi N and Bruggeman P J 2014 Time-resolved absolute OH density of a nanosecond pulseddischarge in atmospheric pressure He-H2O: absolutecalibration, collisional quenching and the importance ofcharged species in OH production Plasma Sources Sci.Technol. 23 045005

[32] Gessel A F H V, Hrycak B, Jasiński M, Mizeraczyk J,Mullen J J A M V D and Bruggeman P J 2013 Temperatureand NO density measurements by LIF and OES on anatmospheric pressure plasma jet J. Phys. D: Appl. Phys. 46095201

[33] Uddi M, Jiang N, Adamovich I V and Lempert W R 2009Nitric oxide density measurements in air and air/fuelnanosecond pulse discharges by laser induced fluorescenceJ. Phys. D: Appl. Phys. 42 075205

[34] Yin Z, Adamovich I V and Lempert W R 2013 OH radicaland temperature measurements during ignition of H2-airmixtures excited by a repetitively pulsed nanoseconddischarge Proc. Combust. Inst. 34 3249–58

[35] Schmidt J and Ganguly B 2013 Effect of pulsed, sub-breakdown applied electric field on propane/air flamethrough simultaneous OH/acetone PLIF Combust. Flame160 2820–6

[36] Bokor J, Freeman R R, White J C and Storz R H 1981 Two-photon excitation of the n=3 level in H and D atoms Phys.Rev. A 24 612

[37] Goldsmith J E M 1987 Photochemical effects in two-photon-excited fluorescence detection of atomic oxygen in flamesAppl. Opt. 26 3566–72

[38] Wysong I J, Jeffries J B and Crosley D R 1989 Laser-inducedfluorescence of O(3p3P), O2, and NO near 226 nm:photolytic interferences and simultaneous excitation inflames Opt. Lett. 14 767–9

[39] Meier U, Kohsehoinghaus K and Just T 1986 H-atom andO-atom detection for combustion applications—study ofquenching and laser photolysis effects Chem. Phys. Lett. 126567–73

[40] van Oostendorp D L, Levinsky H B, van der Meij C E,Jacobs R A A M and Borghols W T A 1993 Avoidance ofthe photochemical production of oxygen atoms in one-dimensional, two-photon laser-induced fluorescenceimaging Appl. Opt. 32 4636–40

[41] Frank J H, Chen X, Patterson B D and Settersten T B 2004Comparison of nanosecond and picosecond excitation fortwo-photon laser-induced fluorescence imaging of atomicoxygen in flames Appl. Opt. 43 2588–97

[42] Lempert W R and Adamovich I V 2014 Coherent anti-StokesRaman scattering and spontaneous Raman scatteringdiagnostics of nonequilibrium plasmas and flows J. Phys. D:Appl. Phys. 47 433001

[43] Kiefer J and Ewart P 2011 Laser diagnostics and minorspecies detection in combustion using resonant four-wavemixing Prog. Energ. Combust. Sci. 37 525–64

[44] Schmidt J B, Sands B, Scofield J, Gord J R and Roy S 2017Comparison of femtosecond- and nanosecond-two-photon-absorption laser-induced fluorescence (TALIF) of atomicoxygen in atmospheric-pressure plasmas Plasma SourcesSci. Technol. 26 055004

[45] Richardson D R, Roy S and Gord J R 2017 Femtosecond,two-photon, planar laser-induced fluorescence of carbonmonoxide in flames Opt. Lett. 42 875–8

[46] Wrzesinski P J, Stauffer H U, Kulatilaka W D, Gord J R andRoy S 2013 Time-resolved femtosecond CARS from 10 to50 bar: collisional sensitivity J. Raman Spectrosc. 441344–8

[47] Chanrion O and Neubert T 2008 A PIC-MCC code forsimulation of streamer propagation in air J. Comput. Phys.227 7222

[48] Hagelaar G J M and Pitchford L C 2005 Solving theBoltzmann equation to obtain electron transport coefficientsand rate coefficients for fluid models Plasma Sources Sci.Technol. 14 722

[49] Paris P, Aints M, Laan M and Valk F 2004 Measurement ofintensity ratio of nitrogen bands as a function of fieldstrength J. Phys. D: Appl. Phys. 37 1179–84

[50] Hoder T, Loffhagen D, Vorac J, Becker M M andBrandenburg R 2016 Analysis of the electric fielddevelopment and the relaxation of electron velocitydistribution function for nanosecond breakdown in airPlasma Sources Sci. Technol. 25 15

[51] Ise T, Tsukizaki R, Togo H, Koizumi H and Kuninaka H2012 Electric field measurement in microwave discharge ionthruster with electro-optic probe Rev. Sci. Instrum. 83 6

[52] Bourdon A, Darny T, Pechereau F, Pouvesle J M, Viegas P,Iseni S and Robert E 2016 Numerical and experimentalstudy of the dynamics of a μs helium plasma gun dischargewith various amounts of N2 admixture Plasma Sources Sci.Technol. 25 17

[53] Gavrilenko V P, Kupriyanova E B, Okolokulak D P,Ochkin V N, Savinov S Y, Tskhai S N and Yarashev A N1992 Generation of coherent IR light on a dipole-forbiddenmolecular transition with biharmonic pumping in a staticfield JETP Lett. 56 1–4

[54] Ito T, Kobayashi K, Mueller S, Luggenhölscher D,Czarnetzki U and Hamaguchi S 2009 Electric fieldmeasurement in an atmospheric or higher pressure gas bycoherent Raman scattering of nitrogen J. Phys. D: Appl.Phys. 42 092003

35


https://doi.org/10.1088/0022-3727/39/3/R01

https://doi.org/10.1088/0963-0252/25/5/054005

https://doi.org/10.1088/0963-0252/25/6/064002

https://doi.org/10.1088/0963-0252/23/4/045003

https://doi.org/10.2514/1.J051605

https://doi.org/10.2514/1.J051605

https://doi.org/10.2514/1.J051605

https://doi.org/10.1088/0963-0252/21/3/034005

https://doi.org/10.1088/1361-6595/aa5205

https://doi.org/10.1088/1361-6595/aa5205

https://doi.org/10.1088/0963-0252/14/2/S05

https://doi.org/10.1088/0963-0252/14/2/S05

https://doi.org/10.1088/0963-0252/14/2/S05




https://doi.org/10.1088/0963-0252/23/6/065005

https://doi.org/10.1088/0963-0252/23/6/065005

https://doi.org/10.1088/0963-0252/23/6/065005

https://doi.org/10.1088/0963-0252/23/4/045005

https://doi.org/10.1088/0022-3727/46/9/095201

https://doi.org/10.1088/0022-3727/46/9/095201

https://doi.org/10.1088/0022-3727/42/7/075205

https://doi.org/10.1016/j.proci.2012.07.015



https://doi.org/10.1016/j.combustflame.2013.06.031



https://doi.org/10.1103/PhysRevA.24.612

https://doi.org/10.1364/AO.26.003566

https://doi.org/10.1364/AO.26.003566

https://doi.org/10.1364/AO.26.003566

https://doi.org/10.1364/OL.14.000767

https://doi.org/10.1364/OL.14.000767

https://doi.org/10.1364/OL.14.000767

https://doi.org/10.1016/S0009-2614(86)80175-0

https://doi.org/10.1016/S0009-2614(86)80175-0

https://doi.org/10.1016/S0009-2614(86)80175-0

https://doi.org/10.1016/S0009-2614(86)80175-0

https://doi.org/10.1364/AO.32.004636

https://doi.org/10.1364/AO.32.004636

https://doi.org/10.1364/AO.32.004636

https://doi.org/10.1364/AO.43.002588

https://doi.org/10.1364/AO.43.002588

https://doi.org/10.1364/AO.43.002588

https://doi.org/10.1088/0022-3727/47/43/433001




https://doi.org/10.1088/1361-6595/aa61be

https://doi.org/10.1364/OL.42.000875

https://doi.org/10.1364/OL.42.000875

https://doi.org/10.1364/OL.42.000875

https://doi.org/10.1002/jrs.4287




https://doi.org/10.1016/j.jcp.2008.04.016

https://doi.org/10.1088/0963-0252/14/4/011

https://doi.org/10.1088/0022-3727/37/8/005

https://doi.org/10.1088/0022-3727/37/8/005

https://doi.org/10.1088/0022-3727/37/8/005

https://doi.org/10.1088/0963-0252/25/2/025017

https://doi.org/10.1063/1.4770116

https://doi.org/10.1088/0963-0252/25/3/035002

https://doi.org/10.1088/0022-3727/42/9/092003

[55] Ito T, Kobayashi K, Czarnetzki U and Hamaguchi S 2010Rapid formation of electric field profiles in repetitivelypulsed high-voltage high-pressure nanosecond dischargesJ. Phys. D: Appl. Phys. 43 062001

[56] Lempert W R, Kearney S P and Barnat E V 2011 Diagnosticstudy of four-wave-mixing-based electric-fieldmeasurements in high pressure nitrogen plasma Appl. Opt.50 5688

[57] Muller S, Ito T, Kobayashi K, Luggenholscher D,Czarnetzki U and Hamaguchi S 2010 14th Int. Symp. onLaser-Aided Plasma Diagnostics (Bristol: IOP Publishing)

[58] Ito T, Kanazawa T and Hamaguchi S 2011 Rapid breakdownmechanisms of open air nanosecond dielectric barrierdischarges Phys. Rev. Lett. 107 065002

[59] Bohm P, Kettlitz M, Brandenburg R, Hoft H andCzarnetzki U 2016 Determination of the electric fieldstrength of filamentary DBDs by CARS-based four-wavemixing Plasma Sources Sci. Technol. 25 8

[60] Goldberg B M, Böhm P S, Czarnetzki U, Adamovich I V andLempert W R 2015 Electric field vector measurements in asurface ionization wave discharge Plasma Sources Sci.Technol. 24 055017

[61] Goldberg B M, Shkurenkov I, Adamovich I V andLempert W R 2016 Electric field in an AC dielectric barrierdischarge overlapped with a nanosecond pulse dischargePlasma Sources Sci. Technol. 25 045008

[62] Simeni M S, Goldberg B M, Zhang C, Frederickson K,Lempert W R and Adamovich I V 2017 Electric fieldmeasurements in a nanosecond pulse discharge inatmospheric air J. Phys. D 50 184002

[63] Goldberg B M, Shkurenkov I, Adamovich I V andLempert W R 2016 Electric field vector measurements in anAC dielectric barrier discharge overlapped with ananosecond pulse discharge Plasma Sources Sci. Technol.25 045008

[64] Goldberg B M, Shkurenkov I, O’Byrne S,Adamovich I V and Lempert W R 2015 Electric fieldmeasurements in a dielectric barrier nanosecond pulsedischarge with sub-nanosecond time resolution PlasmaSources Sci. Technol. 24 035010

[65] Dogariu A, Goldberg B M, O’Byrne S and Miles R B 2017Species-Independent Femtosecond Localized Electric FieldMeasurement Phys. Rev. Appl. 7 024024

[66] Callegari T, Bernecker B and Boeuf J P 2014 Patternformation and dynamics of plasma filaments in dielectricbarrier discharges Plasma Sources Sci.Technol. 23 16

[67] La Fontaine B et al 2000 Guiding large-scale spark dischargeswith ultrashort pulse laser filaments J. Appl. Phys. 88 610–5

[68] Crintea D L, Luggenholscher D, Kadetov V A,Isenberg C and Czarnetzki U 2008 Phase resolvedmeasurement of anisotropic electron velocity distributionfunctions in a radio-frequency discharge J. Phys. D: Appl.Phys. 41 082003

[69] van Gessel A F H, Carbone E A D, Bruggeman P J andvan der Mullen J 2012 Laser scattering on an atmosphericpressure plasma jet: disentangling Rayleigh, Raman andThomson scattering Plasma Sources Sci. Technol. 21015003

[70] van Gessel B, Brandenburg R and Bruggeman P 2013Electron properties and air mixing in radio frequency drivenargon plasma jets at atmospheric pressure Appl. Phys. Lett.103 064103

[71] Hubner S, Hofmann S, van Veldhuizen E M andBruggeman P J 2013 Electron densities and energies of aguided argon streamer in argon and air environments PlasmaSources Sci. Technol. 22 065011

[72] Roy S, Miller J D, Slipchenko M N, Hsu P S, Mance J G,Meyer T R and Gord J R 2014 100-ps-pulse-duration, 100-J

burst-mode laser for kHz-MHz flow diagnostics Opt. Lett.39 6462–5

[73] Popov N A 2001 Investigation of the mechanism for rapidheating of nitrogen and air in gas discharges Plasma Phys.Rep. 27 886–96

[74] Aleksandrov N L, Kindysheva S V, Nudnova M M andStarikovskiy A Y 2010 Mechanism of ultra-fast heating in anon-equilibrium weakly ionized air discharge plasma in highelectric fields J. Phys. D: Appl. Phys. 43 255201

[75] Popov N A 2011 Fast gas heating in a nitrogen-oxygendischarge plasma: I. Kinetic mechanism J. Phys. D: Appl.Phys. 44 285201

[76] Mintoussov E I, Pendleton S J, Gerbault F G, Popov N A andStarikovskaia S M 2011 Fast gas heating in nitrogen-oxygendischarge plasma: II. Energy exchange in the afterglow of avolume nanosecond discharge at moderate pressures J. Phys.D: Appl. Phys. 44 285202

[77] Shkurenkov I and Adamovich I V 2016 Energy balance innanosecond pulse discharges in nitrogen and air PlasmaSources Sci. Technol. 25 015021

[78] Roupassov D V, Nikipelov A A, Nudnova M M andStarikovskii A Y 2009 Flow separation control by plasmaactuator with nanosecond pulsed-periodic discharge AIAA J.47 168–85

[79] Little J, Takashima K, Nishihara M, Adamovich I andSamimy M 2012 Separation control with nanosecond-pulse-driven dielectric barrier discharge plasma actuators AIAA J.50 350–65

[80] Takashima K, Zuzeek Y, Lempert W R and Adamovich I V2011 Characterization of a surface dielectric barrierdischarge plasma sustained by repetitive nanosecond pulsesPlasma Sources Sci. Technol. 20 055009

[81] Xu D A, Lacoste D A, Rusterholtz D L, Elias P Q,Stancu G D and Laux C O 2011 Experimental study of thehydrodynamic expansion following a nanosecond repetitivelypulsed discharge in air Appl. Phys. Lett. 99 121502

[82] Rusterholtz D L, Lacoste D A, Stancu G D, Pai D Z andLaux C O 2013 Ultrafast heating and oxygen dissociation inatmospheric pressure air by nanosecond repetitively pulseddischarges J. Phys. D: Appl. Phys. 46 464010

[83] Montello A, Yin Z, Burnette D, Adamovich I V andLempert W R 2013 Picosecond CARS measurements ofnitrogen vibrational loading and rotational/translationaltemperature in non-equilibrium discharges J. Phys. D: Appl.Phys. 46 464002

[84] Montello A, Burnette D, Nishihara M, Lempert W R andAdamovich I V 2013 Dynamics of rapid localized heating innanosecond pulse discharges for high speed flow controlJ. Fluid Sci. Technol. 8 147–59

[85] Lanier S, Shkurenkov I, Adamovich I V and Lempert W R2015 Two-stage energy thermalization mechanism innanosecond pulse discharges in air and hydrogen–airmixtures Plasma Sources Sci. Technol. 24 025005

[86] Roy S, Kulatilaka W D, Richardson D R, Lucht R P andGord J R 2009 Gas-phase single-shot thermometry at 1 kHzusing fs-CARS spectroscopy Opt. Lett. 34 3857–9

[87] Kulatilaka W D, Stauffer H U, Gord J R and Roy S 2011One-dimensional single-shot thermometry in flames usingfemtosecond-CARS line imaging Opt. Lett. 36 4182–4

[88] Hsu P S, Gragston M, Wu Y, Zhang Z, Patnaik A K, Kiefer J,Roy S and Gord J R 2016 Sensitivity, stability, and precisionof quantitative Ns-LIBS-based fuel-air-ratio measurementsfor methane-air flames at 1–11 bar Appl. Opt. 55 8042–8

[89] Patnaik A K, Hsu P S, Wu Y, Gragston M, Zhang Z,Gord J R and Roy S 2017 Time-resolved correlatedmeasurement of laser-induced-breakdown spectroscopy andelectron number density: application to high-pressurehydrocarbon flames Proc. SPIE 10094 100941G

36


https://doi.org/10.1088/0022-3727/43/6/062001

https://doi.org/10.1364/AO.50.005688

https://doi.org/10.1103/PhysRevLett.107.065002

https://doi.org/10.1088/0963-0252/25/5/054002

https://doi.org/10.1088/0963-0252/24/5/055017

https://doi.org/10.1088/0963-0252/25/4/045008

https://doi.org/10.1088/1361-6463/aa6668

https://doi.org/10.1088/0963-0252/25/4/045008

https://doi.org/10.1088/0963-0252/24/3/035010

https://doi.org/10.1103/PhysRevApplied.7.024024

https://doi.org/10.1088/0963-0252/23/5/054003

https://doi.org/10.1063/1.373710

https://doi.org/10.1063/1.373710

https://doi.org/10.1063/1.373710

https://doi.org/10.1088/0022-3727/41/8/082003

https://doi.org/10.1088/0963-0252/21/1/015003

https://doi.org/10.1088/0963-0252/21/1/015003

https://doi.org/10.1063/1.4817936

https://doi.org/10.1088/0963-0252/22/6/065011

https://doi.org/10.1364/OL.39.006462

https://doi.org/10.1364/OL.39.006462

https://doi.org/10.1364/OL.39.006462

https://doi.org/10.1134/1.1409722

https://doi.org/10.1134/1.1409722

https://doi.org/10.1134/1.1409722

https://doi.org/10.1088/0022-3727/43/25/255201

https://doi.org/10.1088/0022-3727/44/28/285201

https://doi.org/10.1088/0022-3727/44/28/285202

https://doi.org/10.1088/0963-0252/25/1/015021

https://doi.org/10.2514/1.38113

https://doi.org/10.2514/1.38113

https://doi.org/10.2514/1.38113

https://doi.org/10.2514/1.J051114

https://doi.org/10.2514/1.J051114

https://doi.org/10.2514/1.J051114

https://doi.org/10.1088/0963-0252/20/5/055009

https://doi.org/10.1063/1.3641413

https://doi.org/10.1088/0022-3727/46/46/464010

https://doi.org/10.1088/0022-3727/46/46/464002

https://doi.org/10.1299/jfst.8.147



https://doi.org/10.1088/0963-0252/24/2/025005

https://doi.org/10.1364/OL.34.003857

https://doi.org/10.1364/OL.34.003857

https://doi.org/10.1364/OL.34.003857

https://doi.org/10.1364/OL.36.004182

https://doi.org/10.1364/OL.36.004182

https://doi.org/10.1364/OL.36.004182

https://doi.org/10.1364/AO.55.008042

https://doi.org/10.1364/AO.55.008042

https://doi.org/10.1364/AO.55.008042

https://doi.org/10.1117/12.2252546

[90] Verreycken T and Bruggeman P J 2014 OH densitymeasurements in nanosecond pulsed discharges inatmospheric pressure N2–H2O mixtures Plasma Sources Sci.Technol. 23 015009

[91] Verreycken T, van der Horst R M, Sadeghi N andBruggeman P J 2013 Absolute calibration of OH density in ananosecond pulsed plasma filament in atmospheric pressureHe-H2O: comparison of independent calibration methodsJ. Phys. D: Appl. Phys. 46 464004

[92] Yonemori S, Nakagawa Y, Ono R and Oda T 2012Measurement of OH density and air-helium mixture ratio inan atmospheric-pressure helium plasma jet J. Phys. D: Appl.Phys. 45 225202

[93] Winters C, Petrishchev V, Yin Z, Lempert W R andAdamovich I V 2015 Surface charge dynamics and OH andH number density distributions in near-surface nanosecondpulse discharges at a liquid/vapor interface J. Phys. D: Appl.Phys. 48 424002

[94] Schmidt J B, Sands B L, Kulatilaka W D, Roy S,Scofield J and Gord J R 2015 Femtosecond, two-photonlaser-induced-fluorescence imaging of atomic oxygen in anatmospheric-pressure plasma jet Plasma Sources Sci.Technol. 24 032004

[95] Schmidt J B, Roy S, Kulatilaka W D, Shkurenkov I,Adamovich I V, Lempert W R and Gord J R 2017Femtosecond, two-photon-absorption, laser-induced-fluorescence (fs-TALIF) imaging of atomic hydrogen andoxygen in non-equilibrium plasmas J. Phys. D: Appl. Phys.50 015204

[96] Wrzesinski P J, Stauffer H U, Schmidt J B, Roy S andGord J R 2016 Single-shot thermometry and OH detectionvia femtosecond fully resonant electronically enhancedCARS (FREE-CARS) Opt. Lett. 41 2021–4

[97] Brockhinke A and Kohse-Hoinghaus K 2002 Energy transferin combustion diagnostics: experiment and modelingFaraday Discuss. 119 275–86

[98] Hansch T W 1977 High-resolution spectroscopy of atoms andmolecules Phys. Today 30 34

[99] Bischel W K, Perry B E and Crosley D R 1981 Two-photonlaser-induced fluorescence in oxygen and nitrogen atomsChem. Phys. Lett. 82 85

[100] Aldén M, Edner H, Grafström P and Svanberg S 1982 Two-photon excitation of atomic oxygen in a flame Opt.Commun. 42 244–6

[101] DiMauro L F, Gottscho R A and Miller T A 1984 Two‐photon laser‐induced fluorescence monitoring of O atoms ina plasma etching environment J. Appl. Phys. 56 2007–11

[102] Amorim J, Baravian G, Touzeau M and Jolly J 1994 Two‐photon laser induced fluorescence and amplifiedspontaneous emission atom concentration measurements inO2 and H2 discharges J. Appl. Phys. 76 1487–93

[103] Tserepi A D, Dunlop J R, Preppernau B L and Miller T A1992 Absolute H‐atom concentration profiles in continuousand pulsed rf discharges J. Appl. Phys. 72 2638–43

[104] Boogaarts M G H, Mazouffre S, Brinkman G J,van der Heijden H W P, Vankan P, van der Mullen J A M,Schram D C and Dobele H F 2002 Quantitative two-photonlaser-induced fluorescence measurements of atomichydrogen densities, temperatures, and velocities in anexpanding thermal plasma Rev. Sci. Instrum. 73 73–86

[105] Settersten T B, Dreizler A, Patterson B D, Schrader P E andFarrow R L 2003 Photolytic interference affecting two-photon laser-induced fluorescence detection of atomicoxygen in hydrocarbon flames Appl. Phys. B 76 479–82

[106] Patnaik A K, Roy S and Gord J R 2014 Ultrafast saturation ofresonant optical processes Phys. Rev. A 90 063813

[107] Patnaik A K, Roy S and Gord J R 2015 Lineshapedependence of ultrafast saturation of absorption andfluorescence Phys. Scr. T165 014031

[108] Stauffer H U, Kulatilaka W D, Gord J R and Roy S 2011Laser-induced fluorescence detection of hydroxyl (OH)radical by femtosecond excitation Opt. Lett. 36 1776–8

[109] Kulatilaka W D, Gord J R, Katta V R and Roy S 2012Photolytic-interference-free, femtosecond two-photonfluorescence imaging of atomic hydrogen Opt. Lett. 373051–3

[110] Roy S, Hsu P S, Jiang N, Slipchenko M N and Gord J R 2015100-kHz-rate gas-phase thermometry using 100-ps pulsesfrom a burst-mode laser Opt. Lett. 40 5125–8

[111] Meyer T R, Halls B R, Jiang N, Slipchenko M N, Roy S andGord J R 2016 High-speed, three-dimensional tomographiclaser-induced incandescence imaging of soot volumefraction in turbulent flames Opt. Express 24 29547–55

[112] Halls B R, Gord J R, Meyer T R, Thul D J, Slipchenko M andRoy S 2017 20-kHz-rate three-dimensional tomographicimaging of the concentration field in a turbulent jet Proc.Combust. Inst. 36 4611–8

[113] Tatarova E, Dias F M, Felizardo E, Henriques J, Pinheiro M J,Ferreira C M and Gordiets B 2010 Microwave air plasmasource at atmospheric pressure: Experiment and theoryJ. Appl. Phys. 108 123305

[114] Tserepi A D, Dunlop J R, Preppernau B L and Miller T A1992 Absolute H-atom concentration profiles in continuousand pulsed rf discharge J. Appl. Phys. 72 2638–43

[115] Miyazaki K, Kajiwara T, Uchino K, Muraoka K, Okada T andMaeda M 1996 Laser-induced dissociation of moleculesduring measurements of hydrogen atoms in processingplasmas using two-photon laser-induced fluorescence J. Vac.Sci. Technol. A 14 125–31

[116] Wysong I J and Pobst J A 1998 Quantitative two-photonlaser-induced fluorescence of hydrogen atoms in a 1 kWarcjet thruster Appl. Phys. B 67 193–205

[117] van der Heijden H W P, Boogaarts M G H, Mazouffre S,van der Mullen J A M and Schram D C 2000 Time-resolvedexperimental and computational study of two-photon laser-induced fluorescence in a hydrogen plasma Phys. Rev. E 614402–9

[118] Hertl M, Jolly J and Baravian G 2002 Detection of hydrogenatoms in SiH4-H2 radio-frequency plasmas using two-photonlaser-induced fluorescence J. Appl. Phys. 92 710–5

[119] Duan X R, Lange H and Meyer-Plath A 2003 Absolutedensity distribution of H atoms in a large-scale microwaveplasma reactor Plasma Sources Sci. Technol. 12 554–60

[120] Lamara T, Hugon R and Bougdira J 2006 Influence of gastemperature on the loss mechanisms of H-atoms in apulsed microwave discharge identified by time-resolvedLIF measurements Plasma Sources Sci. Technol. 15526–32

[121] Juchmann W, Luque J and Jeffries J B 2005 Two-photonlaser-induced fluorescence of atomic hydrogen in adiamond-depositing dc arcjet Appl. Opt. 44 6644–52

[122] Gabriel O, Schram D C and Engeln R 2008 Formation andrelaxation of rovibrationally excited H2 molecules due toplasma-surface interaction Phys. Rev. E 78 016407

[123] Galante M E, Magee R M and Scime E E 2014 Two photonabsorption laser induced fluorescence measurements ofneutral density in a helicon plasma Phys. Plasmas 21055704

[124] Yin Z, Eckert Z, Adamovich I V and Lempert W R 2015Time-resolved radical species and temperature distributionsin an Ar–O2–H2 mixture excited by a nanosecond pulsedischarge Proc. Combust. Inst. 35 3455–62

[125] Hall C A, Kulatilaka W D, Gord J R and Pitz R W 2014Quantitative atomic hydrogen measurements in premixedhydrogen tubular flames Combust. Flame 161 2924–32

[126] Kulatilaka W D, Gord J R and Roy S 2014 Femtosecond two-photon LIF imaging of atomic species using a frequency-quadrupled Ti:sapphire laser Appl. Phys. B 116 7–13

37


https://doi.org/10.1088/0963-0252/23/1/015009

https://doi.org/10.1088/0022-3727/46/46/464004

https://doi.org/10.1088/0022-3727/45/22/225202

https://doi.org/10.1088/0022-3727/48/42/424002

https://doi.org/10.1088/0963-0252/24/3/032004

https://doi.org/10.1088/1361-6463/50/1/015204

https://doi.org/10.1364/OL.41.002021

https://doi.org/10.1364/OL.41.002021

https://doi.org/10.1364/OL.41.002021

https://doi.org/10.1039/b102058g



https://doi.org/10.1063/1.3037544

https://doi.org/10.1016/0009-2614(81)85112-3

https://doi.org/10.1016/0030-4018(82)90025-6

https://doi.org/10.1016/0030-4018(82)90025-6

https://doi.org/10.1016/0030-4018(82)90025-6

https://doi.org/10.1063/1.334242

https://doi.org/10.1063/1.334242

https://doi.org/10.1063/1.334242

https://doi.org/10.1063/1.357723

https://doi.org/10.1063/1.357723

https://doi.org/10.1063/1.357723

https://doi.org/10.1063/1.351564

https://doi.org/10.1063/1.351564

https://doi.org/10.1063/1.351564

https://doi.org/10.1063/1.1425777

https://doi.org/10.1063/1.1425777

https://doi.org/10.1063/1.1425777

https://doi.org/10.1007/s00340-003-1154-2

https://doi.org/10.1007/s00340-003-1154-2

https://doi.org/10.1007/s00340-003-1154-2


https://doi.org/10.1088/0031-8949/2015/T165/014031

https://doi.org/10.1364/OL.36.001776

https://doi.org/10.1364/OL.36.001776

https://doi.org/10.1364/OL.36.001776

https://doi.org/10.1364/OL.37.003051

https://doi.org/10.1364/OL.37.003051

https://doi.org/10.1364/OL.37.003051

https://doi.org/10.1364/OL.37.003051

https://doi.org/10.1364/OL.40.005125

https://doi.org/10.1364/OL.40.005125

https://doi.org/10.1364/OL.40.005125

https://doi.org/10.1364/OE.24.029547

https://doi.org/10.1364/OE.24.029547

https://doi.org/10.1364/OE.24.029547




https://doi.org/10.1063/1.3525245

https://doi.org/10.1063/1.351564

https://doi.org/10.1063/1.351564

https://doi.org/10.1063/1.351564

https://doi.org/10.1116/1.579907

https://doi.org/10.1116/1.579907

https://doi.org/10.1116/1.579907

https://doi.org/10.1007/s003400050493

https://doi.org/10.1007/s003400050493

https://doi.org/10.1007/s003400050493

https://doi.org/10.1103/PhysRevE.61.4402




https://doi.org/10.1063/1.1486033

https://doi.org/10.1063/1.1486033

https://doi.org/10.1063/1.1486033

https://doi.org/10.1088/0963-0252/12/4/307

https://doi.org/10.1088/0963-0252/12/4/307

https://doi.org/10.1088/0963-0252/12/4/307

https://doi.org/10.1088/0963-0252/15/3/031

https://doi.org/10.1088/0963-0252/15/3/031

https://doi.org/10.1088/0963-0252/15/3/031

https://doi.org/10.1088/0963-0252/15/3/031

https://doi.org/10.1364/AO.44.006644

https://doi.org/10.1364/AO.44.006644

https://doi.org/10.1364/AO.44.006644


https://doi.org/10.1063/1.4873900

https://doi.org/10.1063/1.4873900







https://doi.org/10.1007/s00340-014-5845-7

https://doi.org/10.1007/s00340-014-5845-7

https://doi.org/10.1007/s00340-014-5845-7

[127] Mcllrath T J, Hudson R, Aikin A and Wilkerson T D 1979Two-photon lidar technique for remote sensing of atomicoxygen Appl. Opt. 18 316

[128] Tserepi A D, Wurzberg E and Miller T A 1997 Two-photon-excited stimulated emission from atomic oxygen in rfplasmas: detection and estimation of its threshold Chem.Phys. Lett. 265 297–302

[129] Niemi K, Schulz-von der Gathen V and Dobele H F 2005Absolute atomic oxygen density measurements by two-photon absorption laser-induced fluorescence spectroscopyin an RF-excited atmospheric pressure plasma jet PlasmaSources Sci. Technol. 14 375–86

[130] Ono R, Takezawa K and Oda T 2009 Two-photon absorptionlaser-induced fluorescence of atomic oxygen in theafterglow of pulsed positive corona discharge J. Appl. Phys.106 043302

[131] Uddi M, Jiang N, Mintusov E, Adamovich I V andLempert W R 2009 Atomic oxygen measurements in air andair/fuel nanosecond pulse discharges by two photon laserinduced fluorescence Proc. Combust. Inst. 32 929–36

[132] Pendleton S J, Bowman S, Carter C, Gundersen M A andLempert W 2013 The production and evolution of atomicoxygen in afterglow of streamer discharge in atmosphericpressure fuel/air mixtures J. Phys. D: Appl. Phys. 46305202

[133] Bowman S S, Adamovich I V and Lempert W R 2014Experimental and modeling analysis of singlet delta oxygenkinetics in a repetitively pulsed nanosecond dischargePlasma Sources Sci. Technol. 23 035009

[134] Burnette D, Montello A, Adamovich I V and Lempert W R2014 Nitric oxide kinetics in the afterglow of a diffuseplasma filament Plasma Sources Sci. Technol. 23 045007

[135] Lanier S, Bowman S, Burnette D, Adamovich I V andLempert W R 2014 Time-resolved temperature and O atommeasurements in nanosecond pulse discharges incombustible mixtures J. Phys. D: Appl. Phys. 47 445204

[136] Klochko A V, Lemainque J, Booth J P and Starikovskaia S M2015 TALIF measurements of oxygen atom density in theafterglow of a capillary nanosecond discharge PlasmaSources Sci. Technol. 24 025010

[137] Kulatilaka W D, Roy S, Jiang N B and Gord J R 2016Photolytic-interference-free, femtosecond, two-photon laser-induced fluorescence imaging of atomic oxygen in flamesAppl. Phys. B 122 26

[138] Mazouffre S, Bakker I, Vankan P, Engeln R and Schram D C2002 Two-photon laser induced fluorescence spectroscopyperformed on free nitrogen plasma jets Plasma Sources Sci.Technol. 11 439–47

[139] Rouffet B, Gaboriau F and Sarrette J P 2010 Pressuredependence of the nitrogen atom recombination probabilityin late afterglows J. Phys. D: Appl. Phys. 43 185203

[140] Teramoto Y, Ono R and Oda T 2012 Production mechanismof atomic nitrogen in atmospheric pressure pulsed coronadischarge measured using two-photon absorption laser-induced fluorescence J. Appl. Phys. 111 113302

[141] Burnette D, Shkurenkov I, Adamovich I V and Lempert W R2016 Kinetics of NO formation and decay in nanosecondpulse discharges in Air, H2-Air, and C2H4-Air mixturesPlasma Sources Sci. Technol. 25 025012

[142] Choi I, Yin Z, Adamovich I V and Lempert W R 2011Hydroxyl radical kinetics in repetitively pulsed hydrogen-airnanosecond plasmas IEEE Trans. Plasma Sci. 39 3288

[143] Verreycken T, Mensink R, Horst R V D, Sadeghi N andBruggeman P J 2013 Absolute OH density measurements inthe effluent of a cold atmospheric-pressure Ar–H2O RFplasma jet in air Plasma Sources Sci. Technol. 22 055014

[144] Voráč J, Dvořák P, Procházka V, Ehlbeck J and Reuter S2013 Measurement of hydroxyl radical (OH) concentration

in an argon RF plasma jet by laser-induced fluorescencePlasma Sources Sci. Technol. 22 025016

[145] Renfro M W, Guttenfelder W A, King G B andLaurendeau N M 2000 Scalar time-series measurements inturbulent CH4/H2/N2 nonpremixed flames: OH Combust.Flame 123 389–401

[146] Settersten T B, Dreizler A and Farrow R L 2002 Temperature-and species-dependent quenching of CO B probed by two-photon laser-induced fluorescence using a picosecond laserJ. Chem. Phys. 117 3173–9

[147] Stauffer H U, Roy S, Kulatilaka W D and Gord J R 2012Detailed calculation of hydroxyl (OH) radical two-photonabsorption via broadband ultrafast excitation J. Opt. Soc.Am. B 29 40–52

[148] Morgen M, Price W, Hunziker L, Ludowise P,Blackwell M and Chen Y 1993 Femtosecond Raman-induced polarization spectroscopy studies of rotationalcoherence in O2, N2 and CO2 Chem. Phys. Lett. 209 1–9

[149] Lucht R P, Kinnius P J, Roy S and Gord J R 2007 Theory offemtosecond coherent anti-Stokes Raman scatteringspectroscopy of gas-phase transitions J. Chem. Phys. 127044316

[150] Patnaik A K, Roy S and Gord J R 2013 Saturation ofvibrational coherent anti-Stokes Raman scattering mediatedby saturation of the rotational Raman transition Phys. Rev. A87 043801

[151] Hayden C C and Chandler D W 1995 Femtosecond time-resolved studies of coherent vibrational Raman scattering inlarge gas-phase molecules J. Chem. Phys. 103 10465–72

[152] Zheltikov A M 2000 Coherent anti-stokes Raman scattering:from proof-of-the-principle experiments to femtosecondCARS and higher order wave-mixing generalizationsJ. Raman Spectrosc. 31 653–67

[153] Gord J R, Meyer T R and Roy S 2008 Applications ofultrafast lasers for optical measurements in combustingflows Annu. Rev. Phys. Chem. 1 663–87

[154] Lucht R P, Roy S, Meyer T R and Gord J R 2009Femtosecond coherent anti-Stokes Raman scatteringmeasurement of gas temperatures from frequency-spreaddephasing of the Raman coherence Appl. Phys. Lett. 89251112

[155] Roy S, Kinnius P J, Lucht R P and Gord J R 2008Temperature measurements in reacting flows by time-resolved femtosecond coherent anti-Stokes Raman scattering(fs-CARS) spectroscopy Opt. Commun. 281 319–25

[156] Roy S, Richardson D, Kinnius P J, Lucht R P and Gord J R2009 Effects of N2-CO polarization beating on femtosecondcoherent anti-Stokes Raman scattering spectroscopy of N2

Appl. Phys. Lett. 94 144101[157] Kamga F M and Sceats M G 1980 Pulse-sequenced coherent

anti-Stokes Raman scattering spectroscopy: a method forsuppression of the nonresonant background Opt. Lett. 5126–8

[158] Roy S, Meyer T R and Gord J R 2005 Time-resolveddynamics of resonant and nonresonant broadbandpicosecond coherent anti-Stokes Raman scattering signalsAppl. Phys. Lett. 87 264103

[159] Lang T, Kompa K L and Motzkus M 1999 FemtosecondCARS on H2 Chem. Phys. Lett. 310 65–72

[160] Rahn L A, Farrow R L, Koszykowski M L and Mattern P L1980 Observation of an optical Stark effect on vibrationaland rotational transitions Phys. Rev. Lett. 45 620–3

[161] Zhang Y, Zhen Y R, Neumann O, Day J K, Nordlander P andHalas N J 2014 Coherent anti-stokes Raman scattering withsingle-molecule sensitivity using a plasmonic Fanoresonance Nat. Commun. 5 4424

[162] Yonemaru Y, Palonpon A F, Kawano S, Smith N I,Kawata S and Fujita K 2015 Super-spatial- and -spectral-

38


https://doi.org/10.1364/AO.18.000316

https://doi.org/10.1016/S0009-2614(97)84252-2

https://doi.org/10.1016/S0009-2614(97)84252-2

https://doi.org/10.1016/S0009-2614(97)84252-2

https://doi.org/10.1088/0963-0252/14/2/021

https://doi.org/10.1088/0963-0252/14/2/021

https://doi.org/10.1088/0963-0252/14/2/021

https://doi.org/10.1063/1.3190530




https://doi.org/10.1088/0022-3727/46/30/305202

https://doi.org/10.1088/0022-3727/46/30/305202

https://doi.org/10.1088/0963-0252/23/3/035009

https://doi.org/10.1088/0963-0252/23/4/045007

https://doi.org/10.1088/0022-3727/47/44/445204

https://doi.org/10.1088/0963-0252/24/2/025010

https://doi.org/10.1007/s00340-016-6330-2

https://doi.org/10.1088/0963-0252/11/4/311

https://doi.org/10.1088/0963-0252/11/4/311

https://doi.org/10.1088/0963-0252/11/4/311

https://doi.org/10.1088/0022-3727/43/18/185203

https://doi.org/10.1063/1.4722317

https://doi.org/10.1088/0963-0252/25/2/025012

https://doi.org/10.1109/TPS.2011.2163736

https://doi.org/10.1088/0963-0252/22/5/055014

https://doi.org/10.1088/0963-0252/22/2/025016

https://doi.org/10.1016/S0010-2180(00)00184-X

https://doi.org/10.1016/S0010-2180(00)00184-X

https://doi.org/10.1016/S0010-2180(00)00184-X

https://doi.org/10.1063/1.1493198

https://doi.org/10.1063/1.1493198

https://doi.org/10.1063/1.1493198

https://doi.org/10.1364/JOSAB.29.000040



https://doi.org/10.1016/0009-2614(93)87192-6

https://doi.org/10.1016/0009-2614(93)87192-6

https://doi.org/10.1016/0009-2614(93)87192-6

https://doi.org/10.1063/1.2751184

https://doi.org/10.1063/1.2751184


https://doi.org/10.1063/1.469896

https://doi.org/10.1063/1.469896

https://doi.org/10.1063/1.469896

https://doi.org/10.1002/1097-4555(200008/09)31:8/9<653::AID-JRS597>3.0.CO;2-W



https://doi.org/10.1146/annurev.anchem.1.031207.112957



https://doi.org/10.1063/1.2410237

https://doi.org/10.1063/1.2410237

https://doi.org/10.1016/j.optcom.2007.09.040



https://doi.org/10.1063/1.3116087

https://doi.org/10.1364/OL.5.000126

https://doi.org/10.1364/OL.5.000126

https://doi.org/10.1364/OL.5.000126

https://doi.org/10.1364/OL.5.000126

https://doi.org/10.1063/1.2159576

https://doi.org/10.1016/S0009-2614(99)00787-3

https://doi.org/10.1016/S0009-2614(99)00787-3

https://doi.org/10.1016/S0009-2614(99)00787-3




https://doi.org/10.1038/ncomms5424

resolution in vibrational imaging via saturated coherentanti-stokes Raman scattering Phys. Rev. Appl. 4 014010

[163] Patnaik A K, Roy S, Lucht R P and Gord J R 2008 Collisionaleffects on molecular dynamics in electronic-resonance-enhanced CARS J. Mod. Opt. 55 3263–72

[164] Patnaik A K, Roy S and Gord J R 2016 Ultrafast saturation ofelectronic-resonance-enhanced coherent anti-Stokes Ramanscattering and comparison for pulse durations in thenanosecond to femtosecond regime Phys. Rev. A 93 023812

[165] Beaud P, Frey H M, Lang T and Motzkus M 2001 Flamethermometry by femtosecond CARS Chem. Phys. Lett. 344407–12

[166] Lang T, Motzkus M, Frey H M and Beaud P 2001 Highresolution femtosecond coherent anti-Stokes Ramanscattering: Determination of rotational constants, molecularanharmonicity, collisional line shifts, and temperatureJ. Chem. Phys. 115 5418–26

[167] Knopp G, Beaud P, Radi P, Tulej M, Bougie B,Cannavo D and Gerber T , 2002 Pressure-dependent N2

Q-branch fs-CARS measurements J. Raman Spectrosc. 33861–5

[168] Lucht R P, Roy S, Meyer T R and Gord J R 2006Femtosecond coherent anti-Stokes Raman scatteringmeasurement of gas temperatures from frequency-spreaddephasing of the Raman coherence Appl. Phys. Lett. 89251112

[169] Knopp G, Radi P, Tulej M, Gerber T and Beaud P 2003Collision induced rotational energy transfer probed by time-resolved coherent anti-Stokes Raman scattering J. Chem.Phys. 118 8223–33

[170] Miller J D, Dedic C E, Roy S, Gord J R and Meyer T R 2012Interference-free gas-phase thermometry at elevated pressureusing hybrid femtosecond/picosecond rotational coherentanti-Stokes Raman scattering Opt. Express 20 5003–10

[171] Richardson D R, Lucht R P, Roy S, Kulatilaka W D andGord J R 2011 Single-laser-shot femtosecond coherent anti-Stokes Raman scattering thermometry at 1000 Hz inunsteady flames Proc. Combust. Inst. 33 839–45

[172] Miller J D, Slipchenko M N, Mance J G, Roy S and Gord J R2016 1 kHz two-dimensional coherent anti-Stokes Ramanscattering (2D-CARS) for gas-phase thermometry Opt.Express 24 24971–9

[173] Kliewer C J, Gao Y, Seeger T, Patterson B D, Farrow R L andSettersten T B 2011 Quantitative one-dimensional imagingusing picosecond dual-broadband pure-rotational coherentanti-Stokes Raman spectroscopy Appl. Opt. 50 1770–8

[174] Bohlin A and Kliewer C J 2013 Communication: two-dimensional gas-phase coherent anti-Stokes Ramanspectroscopy (2D-CARS): simultaneous planar imaging andmultiplex spectroscopy in a single laser shot J. Chem. Phys.138 221101

[175] Bohlin A and Kliewer C J 2014 Diagnostic imaging in flameswith instantaneous planar coherent Raman spectroscopyJ. Phys. Chem. Lett. 5 1243–8

[176] Prince B D, Chakraborty A, Prince B M and Stauffer H U2006 Development of simultaneous frequency- and time-resolved coherent anti-Stokes Raman scattering for ultrafastdetection of molecular Raman spectra J. Chem. Phys. 125044502

[177] Slipchenko M N, Prince B D, Ducatman S C and Stauffer H U2009 Development of a simultaneously frequency- and time-resolved Raman-induced Kerr effect probe J. Phys. Chem. A113 135–40

[178] Zinth W, Nuss M C and Kaiser W 1983 Narrowing of spectrallines beyond the natural or dephasing linewidth Opt.Commun. 44 262–6

[179] Laubereau A and Kaiser W 1978 Vibrational dynamics ofliquids and solids investigated by picosecond light-pulsesRev. Mod. Phys. 50 607–65

[180] Nuss M C, Zinth W and Kaiser W 1985 Prolonged-excitationcoherent Raman spectroscopy with spectral resolutionbeyond the transition linewidth using two tunablepicosecond dye lasers J. Opt. Soc. Am. B 2 322–6

[181] Akhmanov S A, Koroteev N I, Magnitskii S A, Morozov V B,Tarasevich A P and Tunkin V G 1985 Time-domaincoherent active Raman-spectroscopy of a free-nitrogen jetJ. Opt. Soc. Am. B 2 640–8

[182] Alden M, Bengtsson P-E and Edner H 1986 Rotational CARSgeneration through a multiple four-color interaction Appl.Opt. 25 4493–500

[183] Seeger T, Kiefer J, Leipertz A, Patterson B D,Kliewer C J and Settersten T B 2009 Picosecond time-resolved pure-rotational coherent anti-Stokes Ramanspectroscopy for N2 thermometry Opt. Lett. 34 3755–7

[184] Seeger T, Kiefer J, Gao Y, Patterson B D, Kliewer C J andSettersten T B 2010 Suppression of Raman-resonantinterferences in rotational coherent anti-Stokes Ramanspectroscopy using time-delayed picosecond probe pulsesOpt. Lett. 35 2040–2

[185] Kliewer C J, Gao Y, Seeger T, Kiefer J, Patterson B D andSettersten T B 2011 Picosecond time-resolved pure-rotational coherent anti-Stokes Raman spectroscopy insooting flames Proc. Combust. Inst. 33 831–8

[186] Kliewer C J, Bohlin A, Nordstrom E, Patterson B D,Bengtsson P E and Settersten T B 2012 Time-domainmeasurements of S-branch N2–N2 Raman linewidths usingpicosecond pure rotational coherent anti-Stokes Ramanspectroscopy Appl. Phys. B 108 419–26

[187] Hsu P S, Stauffer H U, Jiang N, Gord J R and Roy S 2013Direct measurements of collisional Raman line broadeningin the S-branch transitions of acetylene (C2H2) J. Chem.Phys. 139 154201

[188] Roy S, Hsu P S, Jiang N B, Gord J R, Kulatilaka W D,Stauffer H U and Gord J R 2013 Direct measurements ofcollisionally broadened Raman linewidths of CO2 S-branchtransitions J. Chem. Phys. 138 154201

[189] Zuzeek Y, Choi I, Uddi M, Adamovich I V and Lempert W R2010 Pure rotational CARS thermometry studies of low-temperature oxidation kinetics in air and ethene–airnanosecond pulse discharge plasmas J. Phys. D: Appl. Phys.43 124001

[190] Zuzeek Y, Bowman S, Choi I, Adamovich I V andLempert W R 2011 Pure rotational CARS studies of thermalenergy release and ignition in nanosecond repetitivelypulsed hydrogen-air plasmas Proc. Combust. Inst. 333225–32

[191] Miller J D, Roy S, Slipchenko M N, Gord J R and Meyer T R2011 Single-shot gas-phase thermometry using pure-rotational hybrid femtosecond/picosecond coherent anti-Stokes Raman scattering Opt. Express 19 15627–40

[192] Miller J D, Roy S, Gord J R and Meyer T R 2011Communication: time-domain measurement of high-pressure N2 and O2 self-broadened linewidths using hybridfemtosecond/picosecond coherent anti-Stokes Ramanscattering J. Chem. Phys. 135 201104

[193] Stauffer H U, Miller J D, Roy S, Gord J R and Meyer T R2012 Communication: hybrid femtosecond/picosecondrotational coherent anti-Stokes Raman scatteringthermometry using a narrowband time-asymmetric probepulse J. Chem. Phys. 136 111101

[194] Bohlin A, Patterson B D and Kliewer C J 2013Communication: Simplified two-beam rotational CARSsignal generation demonstrated in 1D J. Chem. Phys. 138081102

[195] Kearney S P and Scoglietti D J 2013 Hybrid femtosecond/picosecond rotational coherent anti-Stokes Raman scatteringat flame temperatures using a second-harmonic bandwidth-compressed probe Opt. Lett. 38 833–5

39


https://doi.org/10.1103/PhysRevApplied.4.014010

https://doi.org/10.1080/09500340802450581

https://doi.org/10.1080/09500340802450581

https://doi.org/10.1080/09500340802450581


https://doi.org/10.1016/S0009-2614(01)00819-3

https://doi.org/10.1016/S0009-2614(01)00819-3

https://doi.org/10.1016/S0009-2614(01)00819-3

https://doi.org/10.1016/S0009-2614(01)00819-3

https://doi.org/10.1063/1.1397325

https://doi.org/10.1063/1.1397325

https://doi.org/10.1063/1.1397325





https://doi.org/10.1063/1.2410237

https://doi.org/10.1063/1.2410237

https://doi.org/10.1063/1.1566437

https://doi.org/10.1063/1.1566437

https://doi.org/10.1063/1.1566437

https://doi.org/10.1364/OE.20.005003

https://doi.org/10.1364/OE.20.005003

https://doi.org/10.1364/OE.20.005003




https://doi.org/10.1364/OE.24.024971

https://doi.org/10.1364/OE.24.024971

https://doi.org/10.1364/OE.24.024971

https://doi.org/10.1364/AO.50.001770

https://doi.org/10.1364/AO.50.001770

https://doi.org/10.1364/AO.50.001770

https://doi.org/10.1063/1.4810876

https://doi.org/10.1021/jz500384y



https://doi.org/10.1063/1.2219439

https://doi.org/10.1063/1.2219439

https://doi.org/10.1021/jp804283s



https://doi.org/10.1016/0030-4018(83)90134-7

https://doi.org/10.1016/0030-4018(83)90134-7

https://doi.org/10.1016/0030-4018(83)90134-7

https://doi.org/10.1103/RevModPhys.50.607









https://doi.org/10.1364/AO.25.004493

https://doi.org/10.1364/AO.25.004493

https://doi.org/10.1364/AO.25.004493

https://doi.org/10.1364/OL.34.003755

https://doi.org/10.1364/OL.34.003755

https://doi.org/10.1364/OL.34.003755

https://doi.org/10.1364/OL.35.002040

https://doi.org/10.1364/OL.35.002040

https://doi.org/10.1364/OL.35.002040




https://doi.org/10.1007/s00340-012-5037-2

https://doi.org/10.1007/s00340-012-5037-2

https://doi.org/10.1007/s00340-012-5037-2

https://doi.org/10.1063/1.4824859

https://doi.org/10.1063/1.4774093

https://doi.org/10.1088/0022-3727/43/12/124001





https://doi.org/10.1364/OE.19.015627

https://doi.org/10.1364/OE.19.015627

https://doi.org/10.1364/OE.19.015627

https://doi.org/10.1063/1.3665932

https://doi.org/10.1063/1.3693669

https://doi.org/10.1063/1.4793556

https://doi.org/10.1063/1.4793556

https://doi.org/10.1364/OL.38.000833

https://doi.org/10.1364/OL.38.000833

https://doi.org/10.1364/OL.38.000833

[196] Kearney S P, Scoglietti D J and Kliewer C J 2013 Hybridfemtosecond/picosecond rotational coherent anti-StokesRaman scattering temperature and concentrationmeasurements using two different picosecond-durationprobes Opt. Express 21 12327–39

[197] Patterson B D, Gao Y, Seeger T and Kliewer C J 2013 Split-probe hybrid femtosecond/picosecond rotational CARS fortime-domain measurement of S-branch Raman linewidthswithin a single laser shot Opt. Lett. 38 4566–9

[198] Wrzesinski P J, Roy S and Gord J R 2012 Interference-freecoherence dynamics of gas-phase molecules using spectralfocusing Opt. Express 20 23390–7

[199] Kearney S P 2014 Bandwidth optimization of femtosecondpure-rotational coherent anti-Stokes Raman scattering bypump/Stokes spectral focusing Appl. Opt. 53 6579–85

[200] Kearney S P and Danehy P M 2015 Using hybrid fs/psrotational CARS 53rd AIAA Aerospace Sciences Meeting(American Institute of Aeronautics and Astronautics)

[201] Korobenko A, Milner A A and Milner V 2014 Directobservation, study, and control of molecular superrotorsPhys. Rev. Lett. 112 113004

[202] Milner A A, Korobenko A, Hepburn J W and Milner V 2014Effects of ultrafast molecular rotation on collisionaldecoherence Phys. Rev. Lett. 113 043005

[203] Roy S, Meyer T R and Gord J R 2005 Broadband coherentanti-Stokes Raman scattering spectroscopy of nitrogen usinga picosecond modeless dye laser Opt. Lett. 30 3222–4

[204] Meyer T R, Roy S and Gord J R 2007 Improving signal-to-interference ratio in rich hydrocarbon–air flames usingpicosecond coherent anti-Stokes Raman scattering Appl.Spectrosc. 61 1135–40

[205] Kulatilaka W D, Hsu P S, Stauffer H U, Gord J R and Roy S2010 Direct measurement of rotationally resolved H2

Q-branch Raman coherence lifetimes using time-resolvedpicosecond coherent anti-Stokes Raman scattering Appl.Phys. Lett. 97 081112

[206] Stauffer H U, Kulatilaka W D, Hsu P S, Gord J R and Roy S2011 Gas-phase thermometry using delayed-probe-pulsepicosecond coherent anti-Stokes Raman scattering spectra ofH-2 Appl. Opt. 50 A38–48

[207] Gord J R, Hsu P S, Patnaik A K, Meyer T R and Roy S 2009Gas-phase temperature measurements in reacting flowsusing fiber-coupled picosecond coherent anti-Stokes Ramanscattering spectroscopy 47th AIAA Aerospace ScienceMeeting (Orlando, FL)

[208] Hsu P S, Kulatilaka W D, Gord J R and Roy S 2013 Single-shot thermometry using fiber-based picosecond coherentanti-stokes Raman scattering (CARS) spectroscopyJ. Raman Spectrosc. 44 1330–5

[209] Montello A, Nishihara M, Rich J W, Adamovich I V andLempert W R 2012 Picosecond CARS measurements ofnitrogen rotational/translational and vibrational temperaturein a nonequilibrium Mach 5 flow Exp. Fluids 54 1422

[210] Pendleton S J, Montello A, Carter C, Lempert W andGundersen M A 2012 Vibrational and rotational CARSmeasurements of nitrogen in afterglow of streamer dischargein atmospheric pressure fuel/air mixtures J. Phys. D: Appl.Phys. 45 495401

[211] Kulatilaka W D, Gord J R and Roy S 2011 Effects of O2–CO2

polarization beating on femtosecond coherent anti-StokesRaman scattering (fs-CARS) spectroscopy of O2 Appl. Phys.B 102 141–7

[212] Bitter M and Milner V 2013 Coherent anti-Stokes Ramanspectroscopy in the presence of strong resonant signal frombackground molecules Opt. Lett. 38 2050–2

[213] Richardson D R, Lucht R P, Kulatilaka W D, Roy S andGord J R 2011 Theoretical modeling of single-laser-shot,chirped-probe-pulse femtosecond coherent anti-StokesRaman scattering thermometry Appl. Phys. B 104 699–714

[214] Richardson D R, Lucht R P, Kulatilaka W D, Roy S andGord J R 2013 Chirped-probe-pulse femtosecond coherentanti-Stokes Raman scattering concentration measurementsJ. Opt. Soc. Am. B 30 188–96

[215] Miller J D, Slipchenko M N, Meyer T R, Stauffer H U andGord J R 2010 Hybrid femtosecond/picosecond coherentanti-Stokes Raman scattering for high-speed gas-phasethermometry Opt. Lett. 35 2430–2

[216] Miller J D, Slipchenko M N and Meyer T R 2011 Probe-pulseoptimization for nonresonant suppression in hybrid fs/pscoherent anti-Stokes Raman scattering at high temperatureOpt. Express 19 13326–33

[217] Miller J D, Dedic C E and Meyer T R 2015 Vibrationalfemtosecond/picosecond coherent anti-Stokes Ramanscattering with enhanced temperature sensitivity for flamethermometry from 300 to 2400 K J. Raman Spectrosc. 46702–7

[218] Engel S R, Miller J D, Dedic C E, Seeger T, Leipertz A andMeyer T R 2013 Hybrid femtosecond/picosecond coherentanti-Stokes Raman scattering for high-speed CH4/N2

measurements in binary gas mixtures J. Raman Spectrosc.44 1336–43

[219] Scherman M, Nafa M, Schmid T, Godard A, Bresson A,Attal-Tretout B and Joubert P 2016 Rovibrational hybrid fs/ps CARS using a volume Bragg grating for N2 thermometryOpt. Lett. 41 488–91

[220] Stauffer H U, Miller J D, Slipchenko M N, Meyer T R,Prince B D, Roy S and Gord J R 2014 Time-and frequency-dependent model of time-resolved coherent anti-StokesRaman scattering (CARS) with a picosecond-duration probepulse J. Chem. Phys. 140 024316

[221] Marrocco M 2014 Time-domain coherent anti-Stokes Ramanscattering in terms of the time-delayed Yuratich equationOpt. Lett. 39 4831

[222] Marrocco M, Nordstrom E and Bengtsson P-E 2014Sensitivity of coherent anti-Stokes Raman lineshape to timeasymmetry of laser pulses Opt. Lett. 39 5748

[223] Richardson D R, Stauffer H U, Roy S and Gord J R 2017Comparison of chirped-probe-pulse and hybridfemtosecond/picosecond coherent anti-Stokes Ramanscattering for combustion thermometry Appl. Opt. 56 E37

[224] Lütgens M, Chatzipapadopoulos S, Friedriszik F andLochbrunner S 2014 Analyzing ultrafast multiplex coherentanti-Stokes Raman spectra with picosecond probingJ. Raman Spectrosc. 45 359–68

[225] Lütgens M, Chatzipapadopoulos S and Lochbrunner S 2012Coherent anti-Stokes Raman scattering with broadbandexcitation and narrowband probe Opt. Express 20 6478–87

[226] Pestov D et al 2007 Optimizing the laser-pulse configurationfor coherent Raman spectroscopy Science 316 265–8

[227] Petrov G I, Arora R, Yakovlev V V, Wang X,Sokolov A V and Scully M O 2007 Comparison of coherentand spontaneous Raman microspectroscopies fornoninvasive detection of single bacterial endospores Proc.Natl Acad. Sci. USA 104 7776–9

[228] Ogilvie J P, Cui M, Pestov D, Sokolov A V and Scully M O2008 Time-delayed coherent Raman spectroseopy Mol.Phys. 106 587–94

[229] Arora R, Petrov G I, Yakovlev V V and Scully M O 2012Detecting anthrax in the mail by coherent Ramanmicrospectroscopy Proc. Natl Acad. Sci. USA 109 1151–3

[230] Du X, He X, Song Y, Wang Y, Liu Y, Jiang L, Zhang M andYang Y 2012 Quantum beats of vibrational modes in boththe low and high wavenumber regions studied by fs-CARSOpt. Laser Technol. 44 1324–7

[231] He P, Wang H, Fan R, Chen D, Xia Y, Yu X, Wang J andJiang Y 2012 Terahertz beats of vibrational modes inmethanol and ethanol selectively excited by tr-CARStechnique Sci. China Phys. Mech. Astro. 55 2351–6

40


https://doi.org/10.1364/OE.21.012327

https://doi.org/10.1364/OE.21.012327

https://doi.org/10.1364/OE.21.012327

https://doi.org/10.1364/OL.38.004566

https://doi.org/10.1364/OL.38.004566

https://doi.org/10.1364/OL.38.004566

https://doi.org/10.1364/OE.20.023390

https://doi.org/10.1364/OE.20.023390

https://doi.org/10.1364/OE.20.023390

https://doi.org/10.1364/AO.53.006579

https://doi.org/10.1364/AO.53.006579

https://doi.org/10.1364/AO.53.006579



https://doi.org/10.1364/OL.30.003222

https://doi.org/10.1364/OL.30.003222

https://doi.org/10.1364/OL.30.003222

https://doi.org/10.1366/000370207782596996

https://doi.org/10.1366/000370207782596996

https://doi.org/10.1366/000370207782596996

https://doi.org/10.1063/1.3483871

https://doi.org/10.1364/AO.50.000A38

https://doi.org/10.1364/AO.50.000A38

https://doi.org/10.1364/AO.50.000A38




https://doi.org/10.1007/s00348-012-1422-1

https://doi.org/10.1088/0022-3727/45/49/495401

https://doi.org/10.1007/s00340-010-4188-2

https://doi.org/10.1007/s00340-010-4188-2

https://doi.org/10.1007/s00340-010-4188-2

https://doi.org/10.1364/OL.38.002050

https://doi.org/10.1364/OL.38.002050

https://doi.org/10.1364/OL.38.002050

https://doi.org/10.1007/s00340-011-4489-0

https://doi.org/10.1007/s00340-011-4489-0

https://doi.org/10.1007/s00340-011-4489-0




https://doi.org/10.1364/OL.35.002430

https://doi.org/10.1364/OL.35.002430

https://doi.org/10.1364/OL.35.002430

https://doi.org/10.1364/OE.19.013326

https://doi.org/10.1364/OE.19.013326

https://doi.org/10.1364/OE.19.013326








https://doi.org/10.1364/OL.41.000488

https://doi.org/10.1364/OL.41.000488

https://doi.org/10.1364/OL.41.000488

https://doi.org/10.1063/1.4860475

https://doi.org/10.1364/OL.39.004831

https://doi.org/10.1364/OL.39.005748

https://doi.org/10.1364/AO.56.000E37




https://doi.org/10.1364/OE.20.006478

https://doi.org/10.1364/OE.20.006478

https://doi.org/10.1364/OE.20.006478

https://doi.org/10.1126/science.1139055



https://doi.org/10.1073/pnas.0702107104



https://doi.org/10.1080/00268970801961005

https://doi.org/10.1080/00268970801961005

https://doi.org/10.1080/00268970801961005




https://doi.org/10.1016/j.optlastec.2011.12.036



https://doi.org/10.1007/s11433-012-4937-9

https://doi.org/10.1007/s11433-012-4937-9

https://doi.org/10.1007/s11433-012-4937-9

[232] Niu K and Lee S Y 2012 Analysis of time resolvedfemtosecond and femtosecond/picosecond coherent anti-Stokes Raman spectroscopy: application to toluene andRhodamine 6G J. Chem. Phys. 136 064504

[233] Dudovich N, Oron D and Silberberg Y 2002 Single-pulsecoherently controlled nonlinear Raman spectroscopy andmicroscopy Nature 418 512–4

[234] Katz O, Natan A, Silberberg Y and Rosenwaks S 2008Standoff detection of trace amounts of solids by nonlinearRaman spectroscopy using shaped femtosecond pulses Appl.Phys. Lett. 92 171116

[235] Li H, Harris D A, Xu B, Wrzesinski P J, Lozovoy V V andDantus M 2008 Coherent mode-selective Raman excitationtowards standoff detection Opt. Express 16 5499–504

[236] Li H, Harris D A, Xu B, Wrzesinski P J, Lozovoy V V andDantus M 2009 Standoff and arms-length detection ofchemicals with single-beam coherent anti-Stokes Ramanscattering Appl. Opt. 48 B17–22

[237] Oron D, Dudovich N and Silberberg Y 2003 Femtosecondphase-and-polarization control for background-free coherentanti-stokes Raman spectroscopy Phys. Rev. Lett. 90 213902

[238] Roy S, Wrzesinski P, Pestov D, Gunaratne T, Dantus M andGord J R 2009 Single-beam coherent anti-Stokes Ramanscattering spectroscopy of N2 using a shaped 7 fs laser pulseAppl. Phys. Lett. 95 074102

[239] Roy S, Wrzesinski P J, Pestov D, Dantus M and Gord J R2010 Single-beam coherent anti-Stokes Raman scattering(CARS) spectroscopy of gas-phase CO2 via phase andpolarization shaping of a broadband continuum J. RamanSpectrosc. 41 1194–9

[240] Yue O, Bremer M T, Pestov D, Gord J R, Roy S andDantus M 2012 Single-shot gas-phase thermometry by time-to-frequency mapping of coherence dephasing J. Phys.Chem. A 116 8138–41

[241] Bohlin A and Kliewer C J 2014 Two-beam ultrabroadbandCARS for high resolution gas-phase multiplex imagingAppl. Phys. Lett. 104 031107

[242] Bohlin A and Kliewer C J 2014 Single-shot hyperspectralcoherent Raman planar imaging in the range 0–4200 cm−1

Appl. Phys. Lett. 105 161111[243] Bohlin A and Kliewer C J 2015 Direct coherent Raman

temperature imaging and wideband chemical detection in ahydrocarbon flat flame J. Phys. Chem. Lett. 6 643–9

[244] Yin Z, Montello A, Carter C D, Lempert W R andAdamovich I V 2013 Measurements of temperature andhydroxyl radical generation/decay in lean fuel–air mixturesexcited by a repetitively pulsed nanosecond dischargeCombust. Flame 160 1594–608

[245] Dedic C E, Miller J D, Meyer T R and Gord J R 2013 51stAIAA Aerospace Sciences Meeting (Dallas, TX)

[246] Dedic C E, Michael J B and Meyer T R 2015 15th AIAAAviation Technology, Integration, and Operations Conf.(Dallas, TX)

[247] Dedic C E, Meyer T R and Michael J B 2016 Hybrid fs/pscoherent anti-Stokes Raman scattering for non-equilibirumenvironments Laser Applications to Chemical, Security, and

Environmental Analysis (Heidelberg, Germany: OSA)p LW5G.8

[248] Dedic C E, Michael J B and Meyer T R 2017 Investigation ofenergy distributions behind a microscale gas-phasedetonation tube using hybrid fs/ps coherent anti-StokesRaman scattering 55th AIAA Aerospace Sciences Meeting(Grapevine, TX) (ARC) 2017-0030

[249] Dedic C E, Meyer T R and Michael J B 2017 Single-shot fs/ps CARS measurement of vibrational/rotationalnonequilibrium Optica 4 563

[250] Hanna S F, Kulatilaka W D, Arp Z, Opatrný T, Scully M O,Kuehner J P and Lucht R P 2003 Electronic-resonance-enhanced coherent anti-Stokes Raman spectroscopy of nitricoxide Appl. Phys. Lett. 83 1887–9

[251] Roy S, Kulatilaka W D, Naik S V, Laurendeau N M,Lucht R P and Gord J R 2006 Effects of quenching onelectronic-resonance-enhanced coherent anti-Stokes Ramanscattering of nitric oxide Appl. Phys. Lett. 89 104105

[252] Kuehner J P, Naik S V, Kulatilaka W D, Chai N,Laurendeau N M, Lucht R P, Scully M O, Roy S,Patnaik A K and Gord J R 2008 Perturbative theory andmodeling of electronic-resonance-enhanced coherent anti-Stokes Raman scattering, spectroscopy of nitric oxideJ. Chem. Phys. 128 174308

[253] Patnaik A K, Roy S, Gord J R, Lucht R P and Settersten T B2009 Effects of collisions on electronic-resonance-enhancedcoherent anti-Stokes Raman scattering of nitric oxideJ. Chem. Phys. 130 214304

[254] Druet S A J, Attal B, Gustafson T K and Taran J P 1978Electronic resonance enhancement of coherent anti-StokesRaman scattering Phys. Rev. A 18 1529–57

[255] Stauffer H U, Roy S, Schmidt J B, Wrzesinski P J andGord J R 2016 Two-color vibrational, femtosecond, fullyresonant electronically enhanced CARS (FREE-CARS) ofgas-phase nitric oxide J. Chem. Phys. 145 124308

[256] Bowman R M, Dantus M and Zewail A H 1989 Femtosecondtransition-state spectroscopy of iodine: from strongly boundto repulsive surface dynamics Chem. Phys. Lett. 161297–302

[257] Baumert T, Grosser M, Thalweiser R and Gerber G 1991Femtosecond time-resolved molecular multiphotonionization: the Na2 system Phys. Rev. Lett. 67 3753–6

[258] Papanikolas J M, Williams R M, Kleiber P D, Hart J, Brink C,Price S D and Leone S R 1997 Compositional control ofrovibrational wave packets in the E(1Σg

+) ‘shelf’ state of Li2via quantum-state-resolved intermediate state selectionJ. Chem. Phys. 106 8310–23

[259] Yuan L, Ariunbold G O, Murawski R K, Pestov D, Wang X,Patnaik A K, Sautenkov V A, Sokolov A V,Rostovtsev Y V and Scully M O 2010 Femtosecond wave-packet dynamics in cesium dimers studied throughcontrolled stimulated emission Phys. Rev. A 81 053405

[260] Stauffer H U, Wrzesinski P J, Roy S, Kulatilaka W D andGord J R 2016 Collision-independent detection of moleculartwo-photon excitation by time-resolved parametric four-wave mixing J. Raman Spectrosc. 47 1124–9

41


https://doi.org/10.1063/1.3682470




https://doi.org/10.1063/1.2918014

https://doi.org/10.1364/OE.16.005499

https://doi.org/10.1364/OE.16.005499

https://doi.org/10.1364/OE.16.005499

https://doi.org/10.1364/AO.48.000B17

https://doi.org/10.1364/AO.48.000B17

https://doi.org/10.1364/AO.48.000B17


https://doi.org/10.1063/1.3207829




https://doi.org/10.1021/jp3010103

https://doi.org/10.1021/jp3010103

https://doi.org/10.1021/jp3010103

https://doi.org/10.1063/1.4862980

https://doi.org/10.1063/1.4900477

https://doi.org/10.1021/acs.jpclett.5b00014






https://doi.org/10.1364/LACSEA.2016.LW5G.8

https://doi.org/10.2514/6.2017-0030

https://doi.org/10.1364/OPTICA.4.000563

https://doi.org/10.1063/1.1604947

https://doi.org/10.1063/1.1604947

https://doi.org/10.1063/1.1604947

https://doi.org/10.1063/1.2338014

https://doi.org/10.1063/1.2909554

https://doi.org/10.1063/1.3137106




https://doi.org/10.1063/1.4962834

https://doi.org/10.1016/0009-2614(89)85088-2

https://doi.org/10.1016/0009-2614(89)85088-2

https://doi.org/10.1016/0009-2614(89)85088-2

https://doi.org/10.1016/0009-2614(89)85088-2




https://doi.org/10.1063/1.473894

https://doi.org/10.1063/1.473894

https://doi.org/10.1063/1.473894





recent advances in ultrafast-laser-based spectroscopy and ... · topical review recent advances in...

Documents