photonics for explosives detection' in: encyclopedia of

1

Photonics for Explosives DetectionSoma Venugopal Rao, Shaik Abdul Kalam, and Moram Sree Satya Bharathi

Advanced Centre of Research in High Energy Materials (ACRHEM), University of Hyderabad, Hyderabad, India

Abstract

In this article, we present an overview of the various photonic aspects involved in different tech-niques for explosives detection on field and in the lab. We confine this synopsis to only laser-basedtechniques for detecting explosive molecules in point or proximal setup (laser source and detectorsare in the proximity of sample) and in standoff mode (laser and detectors are at certain distance fromthe sample). The techniques considered in this overview are (a) laser-induced breakdown spec-troscopy (LIBS), (b) Raman spectroscopy and its variants [surface-enhanced Raman spectroscopy(SERS), coherent anti-Stokes Raman spectroscopy (CARS), and spatial offset Raman spectroscopy(SORS)], (c) terahertz (THz) spectroscopy, and (d) photoacoustic spectroscopy (PAS). Various pho-tonic aspects related to these techniques such as (i) laser sources used and the future requirements,(ii) detectors employed at present and improvements required, (c) design and advances in varietyof optics used for illuminating, collimating, collecting, focusing, etc., and (d) integration of all thesecomponents for the creation of efficient portable devices for explosives detection in the laboratoryand field are discussed in detail. We also present results obtained through some of our efforts towardtrace and standoff explosives detection using SERS and femtosecond LIBS techniques, respectively.

Keywords explosives detection; Raman; LIBS; SERS; SORS; photoacoustic spectroscopy; terahertzspectroscopy

1 Introduction 22 Laser-Induced Breakdown Spectroscopy (LIBS) 33 Raman Spectroscopy and Coherent Anti-Stokes Raman Spectroscopy (CARS) 114 Terahertz (THz) Spectroscopy 195 Photoacoustic Spectroscopy (PAS) 226 Concluding Remarks 24

Abbreviations 25Glossary 25Related Articles 25References 26

Encyclopedia of Applied Physics.© 2019 Wiley-VCH Verlag GmbH & Co. KGaA.DOI: http://dx.doi.org/10.1002/3527600434.eap826

2 Photonics for Explosives Detection

1 Introduction

Field detection of explosives in traces (orexplosive mixtures) either in near regimeor from a distance, rapidly/precisely, isessential for military, aviation, and securitysectors [1–9]. Detection and identifica-tion of explosives and their associatedmixtures/compounds prior to any poten-tial blast is important as they can inducelot of damage to public and private prop-erties as well as incur hundreds of humancausalities. Identification of explosives fromnon-explosives is as well important to safe-guard citizens by alerting and evacuatingthem from the suspected area. There areseveral scenarios wherein detection of explo-sives is important: (i) in the airport whilescreening the baggage or personnel, (ii) inthe form of land mines buried in the fieldwhile the military personnel are moving,(iii) in the public places (e.g. malls) in termsof hazardous vapors/bulk explosives, (iv)moving containers/trucks and other vehiclesat check posts or entry points, (v) post-blastcase wherein only very small amounts ofleft out (unreacted) explosives need to beidentified, and (vi) concealed explosives inopaque containers, wherein only traces willbe at best available, that are left out whilehandling these sticky materials. In somecases, the samples can be brought to the laband can be evaluated, whereas in some othercases the instruments need to be taken tothe field for real-time assessment. Thoughthe existing lab-based explosive detectiontechniques are sensitive enough to detecttraces (approximately few picograms) theylack the selectivity and cannot be accommo-dated in real-time investigations. Moreover,chromatographic techniques such as gaschromatography mandate sampling pro-cedure and are also time consuming, thusprecluding their applications in in-situ inves-tigations [10]. There are some extreme caseswherein neither the people nor the instru-ments can be taken close to the sample orsampling area, and sample collection is notpossible. In such cases, standoff detection is

required (where the sources and detectors areat a distance away from the sampling region).There are several techniques reported inthe literature and being used worldwidefor detection of explosives in various forms[11–17]. The following laser-based spectro-scopic techniques [18–20], exclusively usedfor explosives detection, are discussed indetail in terms of the photonic components(input lasers, detectors, optics, spectrom-eters, etc.) being used at present and theimprovements required:1. Laser-induced breakdown spectroscopy

(LIBS) using femtosecond (fs) andnanosecond (ns) pulses in the proximal(near-field) and standoff mode.

2. Raman spectroscopy (normal and in thestandoff mode).

3. Surface-enhanced Raman spectroscopy(SERS) for trace explosives detection,coherent anti-Stokes Raman spectroscopy(CARS) for standoff explosives detection,and spatial offset Raman spectroscopy(SORS) for detecting explosives insideopaque bottles (SERS, CARS, and SORSare lab based).

4. Terahertz (THz) spectroscopy in thenear field and for imaging applications,especially for explosives detection.

5. Photoacoustic spectroscopy (PAS) in thenear field and standoff mode for explosivesdetection.

Laser-based techniques are attractive forexplosives detection as they (i) offer rapidreal-time in situ investigation of sample,(ii) require only an optical accessibilityof sample, (iii) utilize optics free path forinvestigation, and (iv) offer the capability ofremote and standoff detection. Further, laserspectroscopy-based methods are ubiquitous,versatile, and sensitive [19]. Moore [21] hasrecently discussed about the instrumenta-tion aspects required for trace detectionof high explosives. The most importantphotonic components in laser spectro-scopic experiments are (a) laser sources,(b) focusing/collection optics and fibers,(c) spectrometer, and (d) intensified charge

Photonics for Explosives Detection 3

coupled device (ICCD) camera. Similarly,any standoff detection technique towardeffective detection of explosives must meettwo basic requirements: (i) capacity to detectthe response generated from only a smallamount of material located at a distance ofseveral meters (high sensitivity) and (ii) theability to provide easily distinguishableresponses for different materials (high speci-ficity). LIBS and Raman spectroscopy areprobably the only two analytical techniquesthat share similar instrumentation and, at thesame time, generate complementary data.

2 Laser-Induced BreakdownSpectroscopy (LIBS)

LIBS, over the past three decades, has evolvedas one of the promising laser spectroscopictechniques in this direction with enhanceddetection capabilities, particularly attractivefor the detection of explosives and relatedhazardous materials due to its standoff detec-tion capability, requirement of minisculequantities of material, and rapid detectionand analysis [22, 23]. Apart from explosivedetection, the development of LIBS hasbenefited many fields such as environmentalstudies, nuclear waste management, biology,medicine, space applications, defense, andagriculture [24, 25]. Recently, LIBS rowers’instrument has been developed for MARS bythe NASA (National Aeronautics and SpaceAdministration), USA, and Chandrayan-2by ISRO (Indian Space Research Organi-zation), India, to analyze the constituentspresent in mars and moon atmosphere,respectively [26–28]. LIBS technique, anefficient tool for multielemental analysis,has its own advantages compared to otherconventional methods such as inductivelycoupled plasma mass spectroscopy (ICP MS),atomic absorption spectroscopy (AAS), andatomic emission spectroscopy (AES). LIBStechnique offers several advantages includ-ing (i) no or minimal sample preparationrequired, (ii) concurrent multiple elementanalysis for almost all the elements in the

periodic table, (iii) real-time response withraw spectra available in less than a second,(iv) no limitation on the state of the sample (itcan be in any form – solid, liquid, or gas), and(v) high sensitivity (parts per billion/tosub parts per billions can be detected, inprinciple). The technique also permits thedetection of small molecular fragments suchas CN and C2. In fact, the ratios between thespectral emissions of these species and of theelements provide the possibility to correlatethe emission lines to the molecular structureand to identify and discriminate the originalcompounds such as plastics and explosives[29–32]. This technique involves the interac-tion of an intense laser pulse with the sample,leading to the generation of dense plasma,termed as laser-induced plasma (LIP) orlaser-produced plasma (LPP). Optical emis-sions from plasma, containing specific signa-tures of the constituent atoms from the mate-rial, are collected using a detector and fedto a spectrometer, resulting in laser-inducedbreakdown spectrum commonly known asLIBS signal. However, the transient evolutionof LIP is complex as it involves various reac-tions within plasma constituents as well aswith elements from surrounding ambianceinto which plasma expands. ICCD camerashave become an integral part of the LIBSinstrument as these cameras offer high gainand substantially increase the captured LIBSsignal, thus improving the signal-to-noiseratio (SNR). A gated ICCD is useful in collect-ing the plasma emissions (signal) with respectto the input laser trigger or with respect tothe plasma formation and thus eliminatingthe contribution from bright continuum gen-erated in the process of plasma formation. Inother words, it offers delay (when signal to becollected) and width (how far signal has to becollected) to capture optical emissions ema-nating from plasma. Usage of ICCD enablestwo types of LIBS spectra: (i) time-integratedspectra and (ii) time-resolved spectra. Signalcan be recorded in particular time intervalsso as to understand the evolution of plasmaconstituents as it expands. This technique iscalled as kinetic series where time-resolved


emissions are acquired by maximizingsignal-to-background ratio.

Typical LIBS experiments are performed inthe point or proximity [24] (near-field), andthe detection is usually achieved by collectingthe light using lenses configuration and/oroptical fibers. LIBS technique has been usedfor qualitative as well as quantitate measure-ments. Proximal or point detection with LIBStechnique has been utilized by numerousindustrial as well as scientific communitiesfor various analyses [33, 34]. Unlike thosetechniques, standoff LIBS experiments areexecuted for rapid analysis of distant sam-ples, where optical emissions from plasma ofa distant target are coupled using an array oflenses or a telescope attached with fibers toa spectrometer, for the investigation. In thecase of collection optics, the following areimportant: (i) transmission window of thelenses, (ii) transmission as well as reflectionwavelength region of the optics in telescope,and (iii) the fibers used should transmit thelight in the (ultraviolet) UV–visible spectralregion (ideally from 240 to 1100 nm, whichmost of the fibers do not work). Especially forexplosive detection, this is even more signifi-cant since the CI peak at 247.8 nm appears inthe UV region. In few cases, the dichroicmirrors (reflect particular wavelengthand transmit complementary wavelengthregion) and UV-fused silica lenses (transmitUV–visible radiation) are used as deliv-ering optics enabling simultaneous plasmaemission collection. Spectrographs with widewavelength range (for robust multi-elementalanalysis in a single shot), with high resolu-tion and high-throughput, are required forLIBS experiments. However, there exists atrade-off between resolution and through-put, i.e. the increase in slit size increasesthe throughput at the cost of resolution.Especially for trace explosives detection, effi-cient collection optics, spectrometers, fibers(for transmitting the signal), and ICCDsare required, which facilitate in collectingbetter signal and simultaneously enablingto record large number of spectra. Further-more, advanced chemometric algorithms are

required for identification/classification ofexplosives by LIBS data [35].

The major challenge in identifying explo-sives is the presence of common elements (C,N, O, and H) in all these materials, which arealso present in atmosphere, and invariablyon-field detection needs to be performedin open atmosphere. Several reports, citedin the literature, have discussed variousissues of explosives detection in bulk as wellas traces on various organic and inorganicsubstrates including utilization of variousalgorithms for classification/identification instandoff configurations [36–49]. Apart fromidentifying explosives from non-explosives,distinguishing each among themselves isanother mammoth challenge. It requirescareful analysis of the data to concludewhether the given material is an explosive ornot and importantly the exact identificationof the explosive. LIBS technique has beensuccessfully implemented/demonstratedin understanding of fundamental science,conservation of arts, industry, etc. The suc-cess of LIBS devices in Chadrayaan andChemCam is an indication of the technique’smaturity for various applications. Apart fromapplication of LIBS in standoff analyses, sev-eral LIBS-based handheld devices utilizingcompact ns lasers, for remote detection,are being designed and readily supplied byvarious companies in market. Few such linksare provided, from the reader’s point of view,in these references [50–53].

Two separate groups, led by Miziolekand coworkers [36–40, 54] and Lasernaand coworkers [24, 41, 43–47], have per-formed several pioneering experimentsusing ns laser and proved the potential ofns-LIBS technique for standoff detectionand discrimination of explosives. In 2005,De Lucia et al. reported the versatility ofns-LIBS for detection of hazardous materialsand reported a prototype of man-portablelaser-induced breakdown spectroscopy(MP-LIBS) backpack system (as shown inthe Figure 1a) that was tested by recordingLIBS spectrum of a bacteria [36]. In 2006,


(a) (b)

Figure 1 (a) Photograph showing a simulation of the MP-LIBS man-portable/backpack sensor in use byDr. Richard Hark inside a Tyvek Biohazard Protective suit. Laser is in the handle. The computer, power supply,battery, and the high-resolution broadband spectrometer are in the backpack. HUD connected to theeyeglasses. Insert: LIBS 3000 spectrometer (opened). Copyright @IEEE. Reproduced with permission from IEEESENSORS JOURNAL, VOL. 5, NO. 4, AUGUST 2005. (b) Photograph of the double-pulse ST-LIBS systemdeveloped by ARL in collaboration with Applied Photonics, Ltd. and Ocean Optics, Inc. Source: Copyright@Elsevier. Reproduced with permission from Spectrochimica Acta Part B 62 (2007) 1405–1411.

both these groups together reported a stand-off laser-induced breakdown spectroscopy(ST-LIBS) sensor for detection of explosiveresidues on solid surfaces, which uses aBrilliant Q-switched Nd:YAG laser operatingat 1064 nm and producing 350 mJ singlepulses at 20 Hz [41]. Later on, a double-pulseST-LIBS system was developed (as shownin Figure 1b) by De Lucia et al. for detectinga variety of hazardous materials includingRDX, Composition-B, explosive residues,biological species such as the anthrax surro-gate Bacillus subtilis, and chemical warfaresimulants at 20 m with improved sensitivityand selectivity [37]. Furthermore, they haveutilized multivariate analysis approachessuch as partial least squares discriminantanalysis (PLS-DA) to discriminate explo-sive residues. Gottfried et al. demonstratedthe possibility of discrimination of variousresidues on organic and inorganic substrates,which is a cumbersome task [39]. More-over, their preliminary studies with fs pulsesshowed that fs-LIBS has a few advantagescompared to ns-LIBS for explosive residuedetection [40, 55]. The second group led byLaserna also demonstrated progress in stand-off detection of explosives using the ns-LIBStechnique by analyzing various explosive and

organic residues, detecting explosives behindbarriers [24, 43, 44]. They have successfullycombined the techniques of LIBS and Raman(fusion) for identification of explosives [45].They also have demonstrated the possibilityof detecting explosive molecules in finger-prints (on solid surfaces) using LIBS andsupervised learning method [46, 47]. A fewof the recent significant works including thedeployment of different laser sources leadingto hybrid techniques and advanced multi-variate techniques for explosive detection innear and far-field from various other groupsas well have been summarized in Table 1.

Our group (at ACRHEM, University ofHyderabad, India) has been working onthe LIBS technique for explosives detec-tion since 2009. Over the past decade,we have studied various aspects in detec-tion/discrimination and identification ofpure explosive molecules using ns andfs pulses. In our earlier studies, the effectof pulse duration [56] and the effect ofambiance (gas) surrounding the LPP [65]on CN and C2 molecular band emissionsfrom explosives was investigated. Simulta-neously, advanced chemometric techniqueswere developed and combined with LIBSdata of pharmaceutical tablets for classifying


Table 1 Femtosecond and nanosecond LIBS studies for the detection of explosives.

Experimental details Technique Samples Comments Refs.

Two Nd-YAG lasers (UltraBig Sky Laser and Big SkyCFR200) to providedifferent pulse energies

Man-portableLIBS andproximal LIBS

Chemical and biologicalagents, landminedetection, and firstportable MP-LIBS systemby ARL

[36]

Quantel Brilliant Twinslaser (1064 nm, 10 Hz,335 mJ/pulse, 5 ns/pulsewidth),Schmidt–Cassegraintelescope by Meade(LX200GPS) fitted withUV-coated optics,DP-ST-LIBS system

Double pulseST-LIBS at 20 m

Various metallic andplastic materials, bulkexplosives RDX andComposition-B, explosiveresidues, biological speciessuch as the anthraxsurrogate Bacillus subtilis,and chemical warfaresimulants and nerve agents

Classification ofaluminum andRDX, oil, dustand fingerprintresidues onaluminumusing PCA

[37]

Double pulse LIBS, twoNd:YAG lasers(Continuum Surelite),

Near-field LIBSonly

Residues of RDX andComp-B on Al

[54]

Quantel Brilliant Twins(1064 nm, 10 Hz,335 mJ/pulse, 5 ns/pulsewidth) provide a collineardouble pulse generator,designed for 30 moperation

DP-LIBS andPLS-DA

Explosive residues on Al,Arizona road dust, and oilresidue

Multivariateanalysis

[38]

Quantel Brilliant Twins(1064 nm, 10 Hz,335 mJ/pulse, 5 ns/pulsewidth) provide a collineardouble pulse generator,designed for 30 moperation 20 m usingstandoff laser-inducedbreakdown spectroscopy(ST-LIBS)

Double pulsetechnique andPLS-DA

Biological warfare agentsurrogates Bacillus subtilis(BG) and ovalbumin andlinear correlation

(2% falsenegatives and 0%false positives)(0% falsenegatives and 1%false positives)

[48]

Quantel Brilliant Twins(1064 nm, 10 Hz,335 mJ/pulse, 5 ns/pulsewidth) provide a collineardouble pulse generator,25 m standoff spectra

Double pulse,PLS-DAdifferent models

RDX residue, oil residue,Arizona dust residue onplastic, wood, cardboard,etc.

IndividualPLS-DA modelsperformed

[39]

Ti:Sapphire amplifiersystem (Coherent,Hidra-25), Nd:YAG (BigSky, CFR400) laser at1064 nm

Near LIBS Bulk explosives (RDX, C-4,and Composition-B), RDXresidue on aluminumsubstrate

No classification [40]

Quantel Brilliant TwinsQ-switched Nd:YAG laser(1064 nm, 350 mJ, 20 Hz)

Single pulsestandoff setup.Algorithm basedon the peakintensity ratiosused

Classification betweenorganic samples andinorganic samplesincluding explosives

Decision-makingstrategy forstandoff LIBspectral analysisof energeticmaterials.

[41]


Table 1 (Continued)


Two Q-Switched Nd:YAG lasers(Brilliant B, Quantel, 5 ns, 800 mJ,10 Hz), up to 30 m

ns DP-LIBS forstandoff

PMMA and a variety ofglasses. Detection oforganic (DNT, TNT, C4,and H15) and inorganicexplosive (sodiumchlorate)

Mimickingconditions ofinside vehicles,industrialwarehouses, andbuildings

[43]

A Quantel Brilliant TwinsQ-switched Nd:YAG laser(532 nm, 10 Hz, 400 mJ/pulse, 5.5 ns/pulse width)

Raman-LIBSintegration

DNT/TNT,NaClO3/KClO3DNT/Nylon,NaClO3/NaCl

[45]

A high-power double-pulseNd:YAG laser system (10 Hz,850 mJ/pulse, 5.5 ns/pulse width)

ns DP-LIBS forstandoff

Detect chloratite, DNT,TNT, RDX, and PETNresidues that have beendeposited on the surface ofaluminum and glasssubstrates

Explosivehumanfingerprintimaging

[46]

A Q-switched 1064 nm Nd:YAGtwins laser system (10 Hz,850 mJ/pulse, 5.5 ns/pulse width)

Standoff LIBS,supervisedlearningmethods

Standoff LIBS spectra at30 m corresponding tounquantified residues ofRDX and fuel oil

— [47]

Legend Ti:Sapphire fs amplifier(2W, Coherent Co.), INNOLASns laser (1.2 J at 1064 nm, 10 Hz)

Near LIBS,time-resolvedstudies

LIBS spectra of NTO,RDX, and HMX using nsand fs pulses recorded inambient atmosphere

Nodiscriminationor classification

[56]

Legend Ti:Sapphire fs amplifier(2W, Coherent Co.)

Near fs-LIBS LIBS spectra of NTO,RDX, and HMX using fspulses in air, argon, andnitrogen

No classificationordiscrimination

[56]


Near-field LIBS(fs)

A set of nitropyrazolesvarying with number ofnitro groups


Ti:Sapphire fs amplifier and nslaser systems (∼1.2 J at 1064 nm,10 Hz)

Near LIBS,time-resolvedstudies

A set of nitroimidazolesvarying with number ofnitro groups


Ti:Sapphire fs laser system(amplitude, ∼3 mJ, 1 kHz, ∼50 fs,800 nm), ns laser systems(INNOLAS, ∼1.2 J at 1064 nm,10 Hz)

Near-field LIBS(fs and ns)

Explosive molecules(structural and functionalisomers oftriazole-substitutednitroarene derivatives)


SpitLight 1200, InnoLas LaserGmbH, Germany (1.2 J at1064 nm, 10 Hz), nongatedCzerny-Turner spectrometer(MAYA 2000, Ocean Optics,USA)

Near-field LIBSusing ns pulses

Isomers of pyrazoles(1-nitro-pyrazole,3-nitro-1H-pyrazole, and4-nitro-1H-pyrazole)

Classificationthrough ANNand PCA

[59]

(continued overleaf )


Table 1 (Continued)


Ti-sapphire laser with a chirpedpulse amplifier system,Q-switched Nd: YVO4 laser

Remote filamentLIBS

DNT coated on Al and Cu No classification [60]

Filament LIBS Chemical and biologicalagents, barley, corn, andwheat grains, aluminum(50 m away)


Nanosecond and femtosecondLIBS

Filament libs Al, Brass (30 m), Militarygrade TNT


Remote fs filament libs (2 mfocusing and 12 m collection)

Femtosecondfilaments

Copper, graphite, andorganic sample



fs-LIBS in argonambiance

ANTA, DADNE, HMX,NTO, and RDX using fspulses in argonatmosphere

Classificationwith intensityratios and kNNtechnique

[64]

and identifying them [66]. Our detailedstudies revealed that the fs-LIBS spectraof explosives are dominated with molecu-lar emissions, whereas the ns spectra aredominated with atomic emissions (com-pared to molecular emissions). This could beattributed to the difference in fs and ns abla-tion mechanism in turn to the pulse duration.fs pulse is extremely short lived when com-pared to ns pulse. Thus, during fs ablation,ideally there is no plasma–pulse interaction,and no reheating of plasma. Thus, fragmen-tation is dominated and results in molecularradicals, whereas in the case of ns ablation,plasma–plume reheating exists, and thefragments are broken down into the atomicconstituents. However, at later time intervals(after the ns pulse ends), as the plasma coolsdown, molecular emissions dominate overatomic emissions (due to the recombinationof atomic species) even in ns-LIBS. Fur-ther, we have performed several correlationstudies wherein the LIBS spectra fromvarious explosive molecules were collectedusing ns and fs pulses and correlated withthe structures of the molecules, numberof nitro groups, explosive properties of theexplosive molecules, etc. in the near field [35,57, 58]. Sreedhar et al. have utilized intensityratios from ns-LIBS data for classification

of inorganic nitrogen-rich high energymaterials (HEMs) and oxidizers (ammoniumperchlorate, boron potassium nitrate, andammonium nitrate) [67]. Myakalwar et al.classified isomers of pyrazoles (energeticmaterials) using artificial neural networksand principal component analysis (PCA)[59] using ns pulses in proximity. Severalrecent review articles [23, 68] describe andexplain the importance of fs-LIBS in compar-ison with ns-LIBS technique. fs pulses offerseveral advantages in LIBS and its applica-tions such as (i) low breakdown threshold,(ii) precise interrogation with the material,(iii) efficient ablation, (iv) lower continuumemission, (v) small ablated mass and sampledamage, (vi) minimal heat affected zones(vii,) absence of fractionation vaporizationand (viii) improved spatial resolution for3-D mapping applications. Furthermore,several recent studies have also shown thatfs lasers may have advantages over ns lasersfor LIBS analysis in terms of SNR, and ithas been proved that the substrate effectsare minimal in the fs-LIBS case when com-pared with ns-LIBS. Additionally, since fspulses can travel through the atmosphereas a self-propagating transient waveguide,they have advantages over conventionalstandoff LIBS approaches [69]. Mirell et al.


Oscilloscope

D2

D1

M2 A1 A2

Flipmirror

M1

BP

BP

HWP

HWP

A3

M3

Standoff (configuration 1)

Remote (configuration 2)

M4

A4

L2

L2 10 cm to 2 m

10 cm

Sta

ge

Sta

ge

~8.5 m

ICC

D trig

ger

ICC

D trigger

monitor

Synchronized delaygenerator (SDG)

LIBRA femtosecond

amplifier

Mechelle

spectrometer

P, T

P, Tθ

Figure 2 Femtosecond standoff (up to 2 m, configuration 1) and remote (∼8.5 m, configuration 2) LIBS setup.In the figure, M, A, HWP, BP, L, D, P, and T stand for mirror, aperture, half wave plate, Brewster plate, lens,collection system, plasma, and target, respectively. Source: Copyright @OSA. Reproduced with permission fromOptics Express, 26(7), 8069–8083, 2018.

utilized fs pulses for remote sensing of DNT(dinitrotoluene) coated on aluminum [60].Chin et al. demonstrated the capability of fsfilamentation for remote sensing of chem-ical and biological pollutants [61]. Brownet al. utilized filament LIBS for detection ofmilitary grade trinitrotoluene (TNT) [62],and Baudelet et al. demonstrated that theself-channeling of fs pulses can be utilizedfor remote interrogation of organic samples[63]. Recently, we have successfully demon-strated classification of metals, bimetallicalloy strips in standoff mode (at a distanceof ∼6.5 m away from the plasma) using fsfilament-induced breakdown spectroscopy(fs-FIBS) [70]. Our group has also reporteddiscrimination methodologies for explo-sives (ANTA, DADNE, HMX, NTO, andRDX) recorded in argon atmosphere usingfs-LIBS technique [64]. fs-LIBS techniqueoffers several advantages compared to otherspectroscopic techniques, and our initialstudies have demonstrated promising results[71, 72] for utilization of fs pulses towarddetection of explosives in remote (∼8.5 m)and standoff (up to ∼2 m) detection. In thesestudies, we have reported remote and stand-off detection of novel explosive molecules(nitroimidazoles and nitropyrazoles) using

the fs filament-induced LIBS technique. Incombination with fs-LIBS data, PCA wasemployed to discriminate/classify the explo-sives, and the obtained results were verypromising. Figure 2 shows the schematic ofthe experimental setup, and Figure 3 showssome of the results obtained toward clas-sification. Our efforts will now be focusedtoward standoff detection of trace explosivemolecules utilizing fs pulses and study thesubstrate effects.

In summary, there are advantages ofboth ns-LIBS and fs-LIBS configurations.However, there are several challenges tobe addressed before LIBS technique canbe routinely used for standoff detection[73]. They are (i) miniaturization of lasers,(ii) stable beam profile at remote distances,(iii) design and optics of detector, (iv) relyingonly on the traces left out during the handlingof explosives (e.g. while burying a landmineor someone carrying it in luggage/vehicle),(iv) scanning area, and (v) development of alibrary or database of all common explosives.The essential requirements of these versatilefs/ns laser sources are to deliver high intensi-ties at the point of interest with a stable beamprofile, should be compact, and portable.Over the past few years, ns and fs lasers have


(a)

(c)

(b)

(d) Wavelength (nm)

Zn

300

−0.07

−0.07

0.07

0.14

−0.05

0.05

−0.04

0.00

0.00

0.00

0.10

−0.1

0.1

0.0

0.1

−0.1

0.1

0.0

0.1

0.1

0.0

0.2

0.1

0.04

PC

3 (

5%

)P

C2 (

21%

)P

C1 (

43%

)

400 500 600 700 800

300

PC

1P

C2

PC

3

400 500 600 700 800

ZnFe

Fe

CN

CN

CN

O

O

Zn

Al

Al

Cu

Cu

Cu

N

NN

HC2

PC3

PC2

PC1

PC3

PC2

PC1

8.5 m

8.5 m

Wavelength (nm)

PC

1 (82%)

PC1 (43%

)

PC1 (43%

)

PC

3 (

5%

)P

C3 (

5%

)

PC2 (21%

)

PC2

(5%

)

PC3

(1%

)

−20

−20

0

0

20

4060

2010

−10

10

0

4NIm

4NPy

1M345TNPy13DNPy

1M24DNIm

8

4

0

−4

−8

−12

−8−4

−4

0

4

8

12

0

4

×10 4

×10

4

×104

SS

Br

Cu

Al

Figure 3 (a) Principal component scores plot; (b) the first three PCs for the processed LIBS spectra of explosives(nitroimidazoles and nitropyrazoles) obtained at 8.5 m; (c) PC scores plot; and (d) the first three PCs for R-LIBS ofdifferent metals obtained at 8.5 m. Source: Copyright @OSA. Reproduced with permission from Optics Express,26(7), 8069–8083, 2018.

been subjected to several developments interms of miniaturization, power, and theirversatility. As mentioned earlier, backpackMP-LIBS systems were already designedusing ns lasers. Though ns lasers are minia-turized, their beam profile is not stable atvery long distances, so alternatively fs lasershave to be investigated. However, fs lasersare slightly bulky, and miniaturization of fsamplifiers is still in progress. As an example,Libra (M/s Coherent Ltd.) is a single box fsamplifier system with both seed and pumplasers being fixed in it. Owing to this, they canbe portable and mounted on a vehicle/trolleyfor on-field applications. However, Pellegrino

et al. [74] have demonstrated near-infrared(IR) low-energy ultrashort laser pulse fiberlasers for portable applications. Undoubt-edly, fs amplifiers have niche applications,and exploiting the capabilities of these laserssuch as delivering energy to long distances(few meters to few kilometers) for stand-off detection of bulk/trace/vapor form ofexplosives is a massive challenge for all thescientists. Appropriate technique (either ns-or fs-LIBS) has to be identified, and in somecases, it may require scanning a small area oreven a larger area. In this case, time shouldnot be a constraint. Further, once the data isobtained, one needs to do statistical analysis


and arrive at an unambiguous decision onthe presence/absence of the explosive. Inves-tigations on the utilization of nanoparticles(NPs) in LIBS technique to achieve low limitof detection (LOD) by enhancing the signalof analyte molecule could also be extendedfor trace detection of high-energy materials(HEMs) [75, 76]. Nevertheless, lot moreefforts are required to understand the LIBSdata of several explosives, mixtures, com-posites, simulants, etc. Additionally, we needto develop robust algorithms for identifica-tion/classification/discrimination of thesematerials. Moreover, improvement of supe-rior optics, fibers, lasers, detectors (ICCDs)is inevitable for development of compact,robust, cost-effective LIBS instrument foron-field applications. Further, the losses inthe fibers should be minimal (coupling andpropagation losses) for exceptional results.

3 Raman Spectroscopyand Coherent Anti-StokesRaman Spectroscopy (CARS)

Several spectrometric techniques were estab-lished to detect explosives, but Raman-baseddetection became a promising potentialtool for simple, quick, nondestructive,and label-free molecular identification ofsamples. The Raman effect was originallyobserved in 1928 by the great physicist SirC.V. Raman. Two years after the report of“a new type of secondary radiation,” i.e. in1930, he received the Nobel Prize in Physicsfor his work on the scattering of light [77].He used an astronomical telescope to focussunlight in benzene liquid and observed theRaman scattering effect due to in-elasticallyscattered photons. Here, the photons of theincident light interact with molecules ofa sample and are then scattered. The fre-quency of scattered photons is frequencyshifted (up or down) in comparison with theoriginal monochromatic frequency, which iscalled the Raman effect. This shift provideskey information about vibrational, rota-tional, and other low-frequency transitions

in molecules. Raman scattering, as such, isa weak process with extremely small crosssections (typically ranging from 10−31 cm2

to 10−26 cm2), thus providing a poor SNR.This situation generally fails to deliver essen-tial, useful information of the importantmolecular fingerprints under investigation.To prevail over the ambiguity caused by thepoor scattering cross sections, many exper-imental techniques, for instance stimulatedRaman scattering, CARS, and SERS, weredeveloped. Using a combination of fs andpicosecond pulses for broadband excitationand narrowband probing allows monitoringof the temporal evolution of a whole setof excited vibrational coherences with highspectral resolution. The Raman spectrometercomponents improved by replacing sunlightand human eyes via mercury arc lamp andphotographic plate, the laser source, andmultichannel detectors. The new-generationportable/handheld Raman microscopes arenow far easier to operate and provide on-sitedetection and analysis in real time. Thedevelopment of optical filter technology hasalso made it possible to manufacture thecompact Raman spectrometers.

3.1 The Basic Components of RamanSpectrometer

1. Light sources (lasers): These are neededto excite the target species to obtainunique vibrational characteristic Ramanfrequencies. Over the years, Raman spec-trometers have been subjected to variousmodifications and improvizations. Mod-ern Raman spectrometers use standardmonochromatic sources at various exci-tation wavelengths ranging from UVto near infrared (NIR) [266 nm (UV),532 nm, 632 nm (visible), 785 nm, and1064 nm (NIR)]. For shorter wavelengthexcitation, the Raman scattering efficiencyincreases, but simultaneously the possibil-ity of fluorescence and burning/damageof sample may also increase. However,this is not always true as the emission


of fluorescence is specific to a molecule.In few cases, the Raman spectrum couldbe found just between the absorptionand emission range because of the gapbetween absorption and fluorescence,where the latter appears at the longerwavelengths. The scattering efficiencywill decrease with increasing excitationwavelength (from green to red to NIR).Therefore, longer integration times orhigher power lasers are required foracquisition of the Raman spectrum. If theexcitation wavelength is in the UV-Visibleregion and matches with the absorptionwavelength of analyte, it can result inthe increase in the scattering intensitiesby factors of 102–106, called resonanceRaman spectroscopy. It is necessary touse the laser that has stable frequency andnarrow bandwidth to avoid the errors inthe Raman shift. The quality of the Ramanpeaks is directly affected by the sharpnessand stability of the excitation light source.Semiconductor lasers have the problemof “mode hopping,” which is caused bylaser temperature, injection current intothe laser, and optical feedback.

2. Sampling optics: To transmit the lightwaves through the system with low powerloss, optics should be extremely sensitiveand suitable for the selection/choice of alllaser wavelengths.

3. Mirrors: To guide light through the spec-trometer and the system.

4. Lenses: To focus the light onto the sampleand also to collect the scattered lightfrom the sample. The solid angle of col-lection of light by wide-angle microscopeobjectives effect on the Raman spectra.The collection efficiency of the objectivedirectly depends on it’s numerical aper-ture (NA) i.e., the objective with high NAcan suspend a large solid angle and thuscollects maximum signal and vice-versa.However, using a higher NA we canobserve a change in the relative intensi-ties, and changing the input wavelengthwill manifest in the Raman peak shifts.

5. Filters: To collect the Raman scatteredlight (Stokes) which contains the samplesignature and to filter out the intenseRayleigh scattering signal. Double or eventriple monochromators were routinelyused to reject the intense diffuse laserscattering and Rayleigh scattering. Theefficient dielectric or holographic filtersare also being developed.

6. Detectors: Detectors form an essentialcomponent of Raman spectrometers.Detectors should be very sensitive asthey have to detect the weak inelasticscattered light from the sample. Variousdetectors have been used in Raman spec-troscopy, such as CCD (charge coupleddevice), EMCCD (electron multiplyingcharge coupled device), InGaAs, andInGaAsP diodes. These array detectorspixel density will affect the resolution ofthe spectra. Each pixel of the detectorarray will collect and hold charge basedon the number of photons that strike. Thesignal-to-noise (S/N) ratio is dependenton the construction and material qualityof the pixel.

7. All the spectrometer components can beintegrated through electronic hardware,and an interfaced computer makes itpossible for automatic operation.

Bremer and Dantus [78] introduced asensitive method for laser-based standoffdetection of chemicals based on stimulatedRaman scattering. Selective excitation of aparticular Raman transition was detectedby measuring the diffusely reflected laserlight from a distant surface. The methodsimultaneously measured stimulated Ramanloss and gain within a single laser shot and isinsensitive to the optical properties (reflec-tivity/absorptivity) of the substrate. Theydemonstrated the specificity and sensitivityof the Raman spectroscopy technique bydetecting and imaging nanogram analytemicrocrystals on paper, fabric, and plasticsubstrates at 1–10 m standoff distance usingonly 10 mW of laser power from a single fslaser. Figure 4 shows typical standoff SRS


Figure 4 Standoff SRS images ofNH4NO3 on cotton (a) and bluetextured plastic (b); and TNT oncotton (c). The sample distribution oneach substrate corresponds to<100 μg cm−2, although the localconcentration is higher. With 20 lasershots per pixel in the 30 × 30 images,the distribution of the analyte isrecorded by observing SRL. Statisticswere used to eliminate points lessthan 0.8 standard deviations of themean above zero. The black lines areguides to the eye. On (off) resonanceis 1043 cm−1 (950 cm−1) for NH4NO3and 1360 cm−1 (1043 cm−1) for TNT.Data for (a) and (c) were collected at10 m and (b) at 7.5 m [78]. Source:Copyright @AIP Publishing.Reproduced from M. T. Bremer et al.,Appl. Phys. Lett. 103, 061119 (2013);doi: 10.1063/1.4817248 with thepermission from AIP Publishing.

Photo

(a)1 mm

1 mm

1 mm

(b)

(c)

On resonance Off resonance

images of the sample ammonium nitrate.Glenn and Dantus [79] have summarized theadvantages of the technique of single ultra-fast pulse excitation for remote-stimulatedRaman scattering (SUPER-SRS) whereinthey were able to simultaneously detect twoexplosives (TNT and RDX), which weredeposited on a barcode.

3.1.1 Coherent Anti-Stokes RamanSpectroscopy (CARS)One of the variants of the Raman spec-troscopy, CARS, has emerged as a powerfultechnique for sensing of low concentrationanalyte molecules. In this technique, insteadof the traditional single laser system, twopowerful laser pulses irradiate a samplecollinearly. The frequency of the first laseris usually constant, while the frequencyof the second one can be tuned in a waythat the frequency difference between thetwo lasers equals exactly the frequency ofsome Raman-active mode of interest. Thisparticular mode will be the only extremelystrong mode in the Raman signal. WithCARS technique we can obtain a strong,desired Raman peak of interest. In this case,

a monochromator is not really required. Awideband interference filter and a detectorbehind the filter would do the job. However,by scanning the Stokes laser pulse or by usinga broadband (supercontinuum) source, it ispossible to obtain a full spectrum. Two laserbeams with frequencies 𝜐1 and 𝜐2 (𝜐1 > 𝜐2)interact coherently, and because of the wavemixing, produce strong scattered light of fre-quency 2𝜐1 − 𝜐2. If the frequency differencebetween the two lasers 𝜐1 − 𝜐2 is equal to thefrequency 𝜐m of a Raman-active rotational,vibrational, or any other mode, then a stronglight of frequency 𝜐1 + 𝜐m is emitted. Inother words, to obtain strong Raman sig-nal, the second laser frequency should betuned in a way that 𝜐2 = 𝜐1 − 𝜐m. Then, thefrequency of strong scattered light will be2𝜐1 − 𝜐2 = 2𝜐1 − (𝜐1 − 𝜐m) = 𝜐1 + 𝜐m, whichis higher than the excitation frequency 𝜈1and, therefore, considered to be anti-Stokesfrequency.

In 1977, Tolles et al. [80] have presentedthe fundamental theory and its applica-tions in surface studies, reaction dynamics,photochemistry, etc. Wallin et al. [18] havereported a section on CARS technique


toward explosive detection in their review(laser-based standoff detection of explosives)back in 2009. A schematic of energy leveldiagram illustrating the CARS principle hasbeen depicted in this review. Imaging withCARS [81] technique in various configu-rations or schemes (such as E-CARS andF-CARS) has wide imaging applications inchemical and biological systems where vibra-tional imaging with high sensitivity, highspectral resolution can be obtained. Natanet al. [82] have recently reported demonstra-tion of a single-beam coherently controlledfs pulse CARS technique for remote detec-tion and identification of minute amountsof solids and liquids at a standoff (>10 m)distance. They had succeeded in rapidlyresolving the vibrational spectrum of traceamounts of contaminants, such as explo-sives and nitrate samples, from the weakbackscattered photons under ambient lightconditions. In their single-beam technique,they carried out the multiplex measure-ment of characteristic molecular vibrationsusing a single broadband phase-shaped fslaser pulse, which supplied the pump, andStokes and probe photons simultaneously.They have demonstrated that fs-CARS spec-troscopy exhibited higher efficiency at lowaverage powers compared to the longer (ns)pulses used in conventional CARS tech-niques – a merit for the nondestructiveprobing of sensitive samples such as explo-sives. Bremer et al. [83] recently presented anondestructive and highly selective methodof standoff detection. The method was foundto be orders of magnitude more sensitivethan previous coherent spectroscopy meth-ods, identifying concentrations as low as2 μg cm−2 of an explosive simulant mixed ina polymer matrix as shown in Figure 5. Theapproach used a single amplified fs laser togenerate high-resolution multiplex CARSspectra encompassing the fingerprint region(400–2500 cm−1) at standoff distance. Addi-tionally, a standoff imaging modality wasintroduced, visually demonstrating similarsensitivity and high selectivity, providingpromising results toward highly selective

trace detection of explosives or warfareagents. Li et al. [84] reported the detection ofcharacteristic Raman lines for several chem-icals using a single-beam CARS techniquefrom a 12-m standoff distance. Single lasershot spectra were obtained with sufficientSNR allowing molecular identification. Back-ground and spectroscopic discriminationwere achieved through binary phase pulseshaping for optimal excitation of a singlevibrational mode. These results provide apromising approach to standoff detectionof chemicals, hazardous contaminants, andexplosives. Katz et al. [85] demonstratedsingle beam standoff CARS (>10 m) andidentified various components KNO3, RDX,sulfur, Teflon, urea, and chloroform underambient conditions as shown in Figure 6 at12 and 5 m standoff distances. Brady et al.[86] demonstrated the detection of sev-eral chemical warfare simulants (dimethylmethylphosphonate and 2-chloroethyl ethylsulfide) using multiplex CARS and at adistance of 1 m. They claim that fs pulsescan be used to obtain a complete Ramanspectrum in milliseconds time with highchemical specificity. Portnov et al. [87, 88]have demonstrated detection of explosives(RDX) via backward coherent anti-StokesRaman spectroscopy (B-CARS) in near aswell as standoff regime. Their results showedthat the signal intensity obtained in B-CARStechnique was comparable to that of sponta-neous Raman scattering, which implies thatB-CARS allows favorable detection as com-pared to the Raman technique. Furthermore,Portnov et al. believe that the detection oftrace amounts of samples from distances of∼10–200 m is possible with minimal pulseenergies of the pump and Stokes laser beams,depending on the species investigated.

3.2 Standoff Raman Spectroscopyand Spatial Offset Raman Spectroscopy(SORS)

Hirschfield first proposed the standoff detec-tion using Raman spectroscopy [89]. Standoff


Figure 5 CARS spectra acquired at 1-mstandoff on <5-μm PS films (a), <2.5-μmPMMA films (b), and 200-nm PMMA filmcontaining 10% DNT (c). Percentagesrefer to the concentration of 2,4 DNT inthe film relative to polymer mass.Unprocessed (a) and processed spectra(b) show detection of the 1350-cm−1 DNTfeature at 2% concentration.Unprocessed spectra in (c) show signalfrom a blank substrate (bottomspectrum) and 200-nm film (middle)integrated for 100 s to clarify features inthe low signal-to-noise 1 s exposure(top). CO2 and the rovibrational featuresof O2 are also visible in (c) [83]. Source:Copyright @AIP Publishing. Reproducedfrom M. T. Bremer et al., Appl. Phys. Lett.99, 101109 (2011); doi:10.1063/1.3636436 with the permissionfrom AIP Publishing.

(a)

9 cm−1

1050

900 1200 1500

900600 1200 1500 1800

PMMAPMMA

Toluene O2

0%

2%

5%

10%

DNT

DNT

PMMA

200 nm film

200 nm film (100s)

Substrate (100s)

Inte

nsity

Inte

nsity

1200 1400

CO2 CO2

O2

1600

Wavenumber (cm−1)

20001800

Inte

nsity

1200 1350 1500

20%

20%

DNT

O2

10%

5%

2%

(b)

(c)

Raman spectroscopy defined as Raman spec-troscopy performed where the spectrometeris at a distance from the sample, i.e., nophysical contact between a threat and theoperator. This physical separation avoidsthe possible severe damage to individualsand the data acquisition device. The basicdifficulty will be to record the weak Ramansignal with high sensitivity and selectiv-ity for identification of explosives due tointerference from the background as wellas the environment. Availability of powerful

lasers and highly sensitive CCD detectorshas overcome this problem. Unfortunately,the use of high-power lasers creates signif-icant practical problems in the field, caninflict damage on target substrates (explo-sive), risk of eye injury, or skin injury tooperators and bystanders. Hobro and Lendl[90] reviewed the standoff Raman instru-ments (laser, optical lay-out, wavelengthdispersion, and detection) and overviewedits application in explosive detection. Wen


0.6

0.4

0.2

01000 10001200 1200

cm−1

1400

KNO3 Sulfur RDX

ChloroformUreaTeflon

1600

0.8

0.6

0.4

0.2

0200

cm−1

400

200

cm−1 cm−1

300 500400

600

600

200

cm−1

400 600

0.4

0.3

0.2

0.1

0

0.3

0.2

0.1

0 0

0.15

0.05

0.1

0.3

0.2

0.1

0800

800

1400

cm−1

1000 1200

(a) (b) (c)

(d) (e) (f)

Figure 6 Resolved femtosecond CARS vibrational spectra of several scattering samples (dashed blue),obtained at standoff distances of (a–c) 12 m and (d–f ) 5 m. (a) <1000 μg crystallized KNO3, (b) <500 μg sulfurpowder, (c) cyclotrimethylenetrinitramine (RDX/T4) explosive particles with a total mass of <4 mg, (d) bulkPTFE, (e) <4 mg of crystallized urea particles, and (f ) 1-cm-long cuvette containing chloroform and scatteringZnTe particles (200 nm diameter). Each spectrum was resolved from a single measurement with an integrationtime of: (a–c) 3 s, (d) 1 s, (e) 300 ms, and (f ) 350 ms. The Raman vibrational lines and their relative strengths areplotted by gray bars for comparison [85]. Source: Copyright @AIP Publishing. Reproduced from O. Katz et al.,Appl. Phys. Lett. 92, 171116 (2008); https://doi.org/10.1063/1.2918014 with the permission from AIP Publishing.

et al. [91] performed deep-UV Raman spec-troscopy for optical standoff trace detectionof explosives. Hadi et al. [92] designed smallstandoff Raman spectroscopy system anddetected ammonium nitrate, TNT, andurea nitrate at a 4-m distance and ammo-nium nitrate and nitromethane (NM), atdistances of 20 and 12.5 m, respectively.Muhammed Shameem et al. [93] designedtime-gated Raman system by assemblingpulsed laser and high-resolution echellespectrograph. He tested standoff distance at5 m by providing high-quality Raman signalsof sulfur with low background noise and low-ered fluorescence. Zachhuber et al. [94, 95]demonstrated the detection of explosives andrelated compounds in standoff mode usingSORS technique, wherein they employed532- or 355-nm laser excitation wavelengths,operating at 10 Hz with a 4.4-ns pulse lengthand variable pulse energy (maximum of180 mJ/pulse at 532 nm and 120 mJ/pulseat 355 nm). The scattered Raman light was

collected by a coaxially aligned 6′′ telescopeand then transferred via a fiber optic cableand spectrograph to a fast gating ICCDcamera capable of gating at 500 ps. Addi-tional research was needed to develop theinevitable systems for high signal-to-noise,low signal acquisition time, and long operatordistances with safe laser sources. Moros et al.[96] highlighted the fundamentals of standoffRaman for the detection of a wide range ofhigh explosives and associated compoundsunder different analysis conditions. Guliaet al. [97] detected RDX (down to 100 ppm)and a mixture of RDX with KBr in standoffmode at a distance of 5 m using time-gatedRaman spectroscopy technique.

Asher’s group [98–102] have thoroughlyinvestigated the capabilities of the UV res-onance Raman spectroscopy for explosivesdetection. They achieved <1 ppm detec-tion limit for common explosives such asNH4NO3, PETN (pentaerythritol tetrani-trate), TNT, HMX, and RDX dissolved in


ACN/water solutions with 229-nm laserexcitation. They did not observe any nonlin-ear spectral response or sample degradationat the input fluences, spectral accumulationtimes used. Further, they have reviewedthe potential of deep UV resonance-Ramanspectroscopy in the standoff mode [82].They highlight the unique advantages of thistechnique: (i) deep UV excitation results inincreased Raman intensities due to the shortwavelength excitation, (ii) occurrence of res-onance enhancement because of the Ramanspectral coupling of vibrational motionwith electronic transitions, (iii) absence offluorescence interference, and (iv) damagethreshold of the eye is higher for deep UVlight when compared to that of visible andnear-IR sources. They have also performedstudies on the UV-excited photochemistryof PETN in acetonitrile and photoproducthydrolysis. Gaft and Nagli [103] establishedthat UV (248 nm)-excited Raman signals ofexplosives are 100–200 times stronger com-pared to Raman with green color excitation(532 nm). Malka and Bar [104] designeda home-built Raman spectrometer withline excitation for fast, compact, and lowcost detection and mapping of explosivemolecules. Malka et al. [105] also developedcompact green laser pointer-based Ramanspectrometer and measured the Raman spec-tra of different explosives (e.g. KNO3, AN,urea, urea nitrate, DNT, TNT, RDX, HMX,and PETN), liquid samples (hydrogen per-oxide, acetone, methanol, and isopropanol),and even performed chemical imaging. Fur-ther, they could also detect explosive residueson latent fingerprints with concentrations aslow as 1 ng in short timescales (∼10 s).

SORS technique is based on the recordingof Raman spectra from regions, which arespatially offset on the sample surface fromthe laser beam interaction point [106]. It hasbeen shown that the Raman spectra fromdifferent spatial offsets contain differentrelative contributions from various depthsdue to the wider spread of photons inter-acting with deeper layers emerging onto the

sample surface. Cletus et al. [107] have suc-cessfully developed a direct-coupled, inverseSORS spectrometer toward the detection ofdifferent chemical hazards even when thebackground light and daylight are present.Their device was efficacious in recognizingthe disguised chemicals in different types ofpackaging (multilayer and nontransparent)systems. Eliasson et al. [108] used SORSfor detection of hydrogen peroxide, whichis a critical constituent of number of liquidexplosives. Recently, Agilent [109] has devel-oped SORS-based devices for use in airports.They have demonstrated a handheld systemalong with a bigger system (Insight200Mweighing 25 kg only) for screening bottles,aerosols, and gels. Some of these systemsapparently are being used in airports in theUnited Kingdom.

3.3 Surface-Enhanced RamanSpectroscopy (SERS)

The critical drawback of Raman spectroscopyis the low intensity of the inelastically scat-tered photons, which carries informationabout the sample. This weakness leads toits inability to detect trace contaminants,especially in real-life applications. SERS wasdiscovered in 1974 by Fleischmann et al.[110], who observed high-intensity Ramanspectra of the molecule pyridine adsorbedonto a specially prepared roughened sil-ver surface than expected. Van Duyne andAlbrecht et al. independently explained theelectric field enhancement, which is respon-sible for Raman scattering [111, 112]. Severalreports and reviews described the majorenhancement mechanisms of (i) electromag-netic (EM) nature (which can be as high as∼1014) and (ii) chemical nature (which canreach values up to ∼106). EM enhancementattributed to the localized surface plasmonresonance (LSPR) in the near field metallicsurface and chemical enhancement is dueto charge transfer (CT) mechanism betweenthe substrate and the analyte. Generally, theenhancement due to the CT is one to threeorders of magnitude smaller than that of EM.


The interaction of the incident EM field withmetal NPs possessing negative real and smallpositive imaginary (absorption) dielectricconstants induces collective and coherentelectron oscillations, called plasmons. Thesuperior local field enhancement was presentat the vicinity of the overlap of the near-fieldregions between adjacent NPs, called “hotspots” [113].

SERS has the potential to provide effec-tive in-field detection of chemical warfareagents (CWAs) due to the combinationof high label-free sensitive molecular fin-gerprinting ability [114]. Effective SERSdetection requires that target substancesare attached to suitably nanostructurednoble metal substrates. The developmentof robust and cost-effective nanofabricationprocesses for the production of SERS sub-strates, in particular structures that containthe so-called EM “hot spots” is, therefore akey requirement toward the widespread useof this technique [17, 115]. The degree ofcontrol and reproducibility of commercialas well as research-based SERS substratesare still disputed, and it is likely that signif-icant improvements regarding performanceand standardization have to be made inorder to transform SERS technique into awidespread technique for routine chemi-cal quantification. Laser ablation in liquids(LAL) is a green technique for producingNPs and NSs (nanostructures) in a sin-gle experiment/exposure. Several recentadvances have demonstrated the capabil-ity of LAL with ultrashort laser pulses invarious applications [116]. Hot-spot gener-ation is a key issue in SERS, which enablesgigantic signal enhancements. Additionally,controllable aggregation of NPs to enablecontrollable hotspots is a critical issue tomeasure reproducible SERS signals of theanalytes. It is important to identify the cor-rect/appropriate substrate(s) for each analytemolecule (in this case different explosivemolecules, e.g. TNT, AN, RDX, and PA). Theenhancements for each molecule should behigh for the same substrate so that a singlesubstrate should enable us to detect multiple

analytes, if not all the analytes. Further, thesubstrate should be recyclable for multipleusages to cut down the cost and make themversatile [117–119]. Several research groupswere testing the SERS technique sensitivity indetecting traces of explosives, nerve agents,and other hazardous chemicals [120–126].Ben-Jaber et al. detected RDX (nM) andvapor phase detection of DNT using silvernanocubes (AgNCs) [127]. Hakonen et al.detected femtograms of picric acid (PA)using Ag pillars as SERS substrates using asimple handheld Raman spectrometer [128].Chen et al. demonstrated portable siliconsensor with AgNPs as a SERS analytical plat-form for signal-on detection of trace TNTexplosive vapors [129].

Several techniques, including chemicaland lithography based, have been employedto fabricate SERS substrates (nanostructuredmetal patterns). Some of them are electronbeam lithography, focused ion beam lithog-raphy, atomic layer deposition, and ultrafastpulsed laser ablation. Over the past decadeour group has been working on SERS todetect the explosives by employing differentSERS substrates using ultrafast laser ablationtechnique. In our earlier works, we studiedindividual metals as SERS substrates includ-ing (a) aluminum by fs ablation [130], (b) cop-per by multiple/single line ps ablation [131],and (c) silver nanoentities through ultrafastdouble ablation [132] techniques. Further-more, we have also investigated (a) the effectof oblique incidence on silver nanomaterials[133] and their applications for SERS, (b)the influence of nondiffracting Bessel beamon the Ag targets [134], and (c) Au NPs/NSby fs ablation for long-term stability [135].Additionally, bimetallic/alloy NPs/NS werestudied to adopt the advantage of the metalssuch as silver–gold NPs/NS (Ag0.65–Au0.35,Ag0.5–Au0.5, and Ag0.35–Au0.65) [136], and adetailed study on the various compositions ofAg–Au (Ag, Au20Ag80, Au30Ag70, Au50Ag50,Au70Ag30, and Au80Ag20, Au) [137], Ag@Au,and Cu@Au NPs was performed using thetechnique of fs laser ablation [138]. Like-wise, Ag NPs decorated ZnO nanostructures


were synthesized by a two-step process (ionbeam sputter deposition of Ag NPs ontothe ZnO nanostructure grown on borosil-icate glass by rapid thermal oxidation) andutilized as SERS substrates. Recently, wehave also demonstrated the fabrication ofeco-friendly filter papers embedded withaggregated Ag/Au NPs, which act as effi-cient SERS substrates toward the detectionof various analytes including three explo-sive molecules [139]. The aggregation wasachieved using NaCl and SERS signal wasoptimized for different concentrations ofNaCl. Furthermore, commercially availableSERS substrates are added advantage forhandheld Raman spectrometers for variousanalyte detection [140]. A summary of theseSERS studies used for the detection of explo-sives (with the limits of detection) is providedin Table 2. The challenge that remains to beconquered is preparation/development ofsimple, robust, recyclable, and versatile(should detect all common explosives withhigh sensitivity) yet cost-effective SERS sub-strates for practical applications. This hasto be combined with compact and powerfulspectrometers, which should be portableor handheld. Furthermore, powerful algo-rithms are needed, which can analyze andidentify the Raman spectra of unknowncompound(s) and/or compare them withcommon explosives/mixtures along with alibrary of the Raman spectra of all commonand new explosives molecules/mixtures.

4 Terahertz (THz)Spectroscopy

Terahertz spectral region is important forexplosive materials since in this region mostof these materials have unique signatures.Several advantages of this technique include(a) capability to detect concealed weaponsbecause many nonmetallic and nonpolarmaterials are transparent in the THz spec-tral range, (b) explosives have characteristicpeaks in the THz spectral region, which canbe used to identify these compounds, and

(c) THz radiation as such does not pose anyhealth risk even while scanning. There areseveral reports on the detection of explosivesusing THz radiation. Teraview group hasdone some pioneering work in this area.For example, see http://www.teraview.com/applications/homeland-security-defense-industry/explosive-detection.html [154],wherein they discuss about detection ofexplosives concealed by clothing and shoes.Further, they have also shown a stand-off system (TPS spectra 3000) from theircompany.

The important components of a THz sys-tem include (a) fs laser source (typically,we are not touching upon other ways ofgenerating THz radiation), (b) THz gen-erator (photo-conducting antenna or thinfilms or nonlinear optical crystals, etc.), and(c) THz detector. For imaging applicationsthere will be additional components involved.Puc et al. [155] utilized THz time-domainspectroscopy combined with imaging tostudy the effects of different paper and textilebarriers while collecting the spectral featuresof drug, explosive simulants. They achievedrapid detection along with identification ofconcealed simulants in the 1.5–4.0 THz fre-quency range using an organic-crystal-basedTHz time-domain system. Davies et al. [156]examined the THz spectra of a wide range ofpure and plastic explosives. Palka et al. [157]demonstrated a THz time-domain techniquefor extracting spectral fingerprints frommaterials, which was based on the frequencyanalysis of the reflected signal from the sam-ple. They have successfully presented resultsobtained for the case of covered samples ofRDX-based materials.

There are several review articles highlight-ing the various aspects of THz technologyfor explosives detection including (a) sourcesused, (b) different explosives data, (c) imag-ing capability of a few explosives, and(d) future prospects for THz technology tobe applied in the field and airports. Federiciet al. [158] have reviewed the techniques usedfor explosives detection using THz radiation.THz spectra of common explosives such as


Table 2 Summary of the identification of explosives by SERS studies using either micro or portable Ramanspectrometers.

SERS substrate AnalyteEnhancementfactor (EF) Raman spectrometer Refs.

Ag NSs ANTA (1 μM) 106 Micro-Ramanspectrometer (532,785 nm)

[133]

Ag–Au NSs FOX-7 (5 μM) 106 Micro-Ramanspectrometer (532 nm)

[136]

Ag NSs CL-20 (5 μM) 106 Micro-Ramanspectrometer (532 nm)

[134]

Ag NSs RDX (50 mM) 104 Micro-Ramanspectrometer (532,785 nm)

[132]

Cu NSs TNT (100 μM) 105 Micro-Ramanspectrometer(WITec-Alpha 300)(532 nm)

[141]

Au NS/NPs PA (5 μM); AN (50 μM) 104; 103 Portable (785 nm) [135]Ag–Au NS and NPs DNT (1 μM) 105 Portable (785 nm) [137]Cu@Au and Ag@Au NPs PA (5 μM); DNT (1 μM) 104; 105 Portable (785 nm) [138]Electrochemical roughgold film

2,4DNT (3 × 10−11 M) — Portable 785 nm [142]

Klarite AN (75 pg) — Renishaw inVia ReflexRaman Microscope(785 nm)

[143]

i) Klariteii) Au NPs (30 nm)

TNT (100 ppm) i) 108

ii) 106Portable 780 nm [144]

GO-Au nanocage TNT (10 f M); RDX(500 fM)

1.6 × 1011 Portable 670 nm [145]

Ag NPs with NaCl TNT (10−10 M) — Confocal (LabRAMAramis, Horiba JobinYvon) 532 nm

[146]

Au NPs on paper TNT (94 pg); DNT (7.8 pg) — Portable (785 nm) Ramanspectrometer

[147]

Ag nanofilm NTO (35 μg L−1 in DI and0.35 mg L−1 in aged tapwater)

— Nicolet Almega XRdispersive (785 nm)

[148]

Au popcorn NPs TNT (10−10 M) — Portable (785 nm) Ramanspectrometer

[149]

Au NPs TNT (10−7 M) — 830 nm [150]Monolayer of 5 nm Au NPs AN (7.7 ppm); RDX

(0.19 ppm)5 × 104 Jobin-Yvon micro-Raman

(532 nm)[121]

Au triangular nanoprisms TNT (100 fM); RDX(10 fM); PETN (10 fM)

6 × 106 Foster + Freman Foram785 HPRaman system (785 nm)

[117]

Au–Ag alloy nanoplates TNT; RDX (10 nM) — Renishaw inVia (633 nm) [118]


Table 2 (Continued)

SERS substrate AnalyteEnhancementfactor (EF) Raman spectrometer Refs.

Ni–Au, Ni–Pd, Ni–Ag,Ni–Ptnanostructures

TNT (10−7 M); DNT(10−7 M)

RDX (10−6 M)

1010 WiTec GmbH,Alpha-SNOM

CRM 300 (633 nm)

[119]

Ag nanocubes DNT (10−15 M);RDX (10−9 M)

DNT vapor

8.7 × 1010 Renishaw 1000spectrometer (633 nm)

[127]

Gold-coated sapphireNS

2,4-DNT; 2,4-DNCB;

p-Nitroaniline (vapor phase)

— Renishaw inViaRaman microscope(785 nm)

[120]

Klarite TNT; RDX; PETN; EGDN(200 pg)

— Portable (785 nm) Ramanspectrometer

[122]

Au NPs TNT (22.7 ng L−1) — Renishaw inVia Raman(785 nm)

[123]

Ag NPs on graphenesheets

TNT (10−11 M) — Renishaw 2000 modelconfocal (514.5 nm)

[124]

Ag nanopillars PA (20 ppt) — Handheld Raman (785 nm) [128]Ag micro/nanopillar PA (10−12 M); NTO

(10−13 M)2.4 × 109 Portable Raman

spectrometer[125]

ZnO–Ag NWs ANTA; Fox-7; CL20 (10 μM) 107 Confocal micro-Raman(532 nm)

[151]

Sea urchin-like nanoZnO-paper

Sulfite in wine — Portable (785 nm) Ramanspectrometer

[152]

AgNPs decorated Siwafer chip

TNT−PABT (10−8 M: fromlake water, soil, envelope,liquor, 10−6 M: from vaporsdiffusing from residues)

— Portable handheld(785 nm) spectrometer

[129]

Si NW Ag NPs AP (50 μM) 104 Confocal micro-Raman(632 nm)

[153]

Filter paper loaded withaggregated Ag/Au NPs

MB (5 nM); PA (5 μM);

DNT (1 μM); NTO (10 μM)

3.4 × 107;

2.4 × 104;

2.0 × 104;

2.1 × 104

Portable (785 nm) Ramanspectrometer

[139]

Ag nanopillar Cyclosarin-(20 μg mL−1);

RDX -(2 ng mL−1);

Amphetamine-(2 ng mL−1);

PA- (0.4 ng mL−1)

40 ng;4 pg;4 pg;0.8 pg

Handheld Ramanspectrometer

[140]

CL-20, hexanitrohexaazaisowurtzitane; EGDN, ethylene glycol dinitrate.


RDX, TNT, PETN, Semtex-H, HMX, C-4,and ammonium nitrate are reported in theliterature. Shen et al. [159] performed reflec-tion THz spectroscopic imaging of RDX.Choi et al. [160] performed reflection THztime-domain spectroscopy of RDX, HMXin the 0.3–3 THz spectral region. Further,they successfully determined the refractiveindex and extinction coefficient accurately.Kemp [161] identified the potential prob-lems associated with THz spectroscopyfor practical applications. He argues that(a) explosives features in this spectral regionare relatively weak and broad and, thereby,tend to be camouflaged by the combinedeffects of atmospheric water vapor absorp-tion and barrier attenuation. Further, thereare issues/challenges such as scattering fromclothing and the target explosives. Chen et al.[162] measured the absorption propertiesof 17 explosive-related compounds usingTHz technique. The absorption spectra wereobtained in the range of 0.1–3 THz. Theirdata clearly demonstrated that most of thosecompounds had THz fingerprints (caused byboth the intramolecular and intermolecularvibrational modes). Leahy-Hoppa et al. [163]reported the THz absorption spectra in the0.5–6 THz spectral range for four explosives(RDX, HMX, TNT, and PETN) and identifiedadditional unique spectral features in thoseexplosives and claim that their data mayaid in their spectral identification. Further,Leahy-Hoppa et al. [164] also reviewed thepros and cons of THz technique for a widerange of explosives detection.

Though there are several advances inthe THz technologies for detection andimaging of explosives, there remain severalchallenges: (a) efficient, robust, compactfs laser sources for generation of the THzradiation, (b) efficient materials for gener-ating broad-band THz radiation design ofbasic components (detectors and optics) inthe frequency range of 0.1–10.0 THz arerelevant and important for spectroscopyand imaging applications, (c) understandingthe THz spectral behavior of explosivesmixed with common interferants (paper,

wood, clothes, plastics, etc.), (d) extensivedatabase of all explosive molecules and theircompositions/mixtures, (e) combining theTHz data with complementary techniquessuch as Raman for unambiguous detec-tion of explosives, (f ) efficient mechanism(including algorithms) for scanning largeareas and for collecting the THz radiation inreflection geometry for imaging applications,and (g) propagation studies of THz radia-tion in atmosphere, reflection from varioussources, and collimation/focusing of thereflected/scattered THz radiation are essen-tial for developing THz-based devices forexplosives detection, which can be used inthe field (e.g. airports).

5 Photoacoustic Spectroscopy(PAS)

Photothermal spectroscopy has been usedrecently to detect trace levels of gases usingoptical absorption and following thermalperturbations of the gases. The fundamentalprinciple involved is the measurement ofphysical changes (e.g. temperature, den-sity, or pressure), which results from thephoto-induced change in the sample’sthermal state. Photothermal methods areclassified as indirect methods because thetransmission of the input light used to excitethe sample is measured indirectly. A fewexamples of the photothermal techniquesinclude (a) photothermal interferometry,(b) photothermal lensing, (c) photothermaldeflection, and (d) PAS. In most photother-mal techniques, one measures the refractiveindex using combinations of probe sourcesand detectors, whereas in PAS, one measuresthe pressure wave produced by heating ofthe sample. PAS is a particularly sensitivetechnique, capable of trace gas detectionat parts-per-trillion (ppt) levels. Holthoffet al. [165] suggested that though the pho-toacoustic sensors are very sensitive, thetotal system size represents a large logisticsproblem in terms of cost, size, and powerconsumption. There are several reports on


Optical

chopper

Reflective

mirrors

Excitation laser

beam from QCLProbe laser beam

from LDV

QCL

Sample and

substrate

(a) (b)

Figure 7 (a) Experimental setup at one end of 100-m distance and (b) long-distance LDV at the other end.Source: Copyright @Elsevier. Reproduced from Fu et al. [169], Optics and Lasers in Engineering, 107, 241-246(2018) with permission.

the PAS of explosives in the near field aswell as in standoff detection. Marcus et al.[166] recently reported on the PAS detectionexplosives such as RDX, PETN, and TNT at adistance of 1 m. They demonstrated and dis-cussed the evaluation of an interferometricPAS sensor or these studies. They argue thatthe strength of PAS allows for the buildingof tailored spectroscopic sensors that aredesigned to achieve specific tasks. Holthoffand Pellegrino [167] recently summarized theresults obtained during the development ofphotoacoustic sensing platforms at the ArmyResearch Laboratory, USA. Chen et al. [168]reported the standoff detection of explo-sives using tunable quantum cascade lasers(QCLs) and the PA technique. They utilizeda mid-infrared QCL (emission wavelengthnear 7.35 μm) as the laser source. Directstandoff detection of TNT was achievedusing an ultrasensitive microphone. With theaid of a sound reflector, narrow band-passfilters, and amplifiers, standoff detectionat a distance of 8 ft was also successfullydemonstrated.

Fu et al. [169] very recently reported detec-tion of explosives at a distance of 100 m inthe standoff mode using a QCL-inducedphoto-vibrational signal on explosivesusing a laser Doppler vibrometer (LDV).

Figure 7 shows the experimental schematicand actual setup. They tested the tracesamples of three high-explosive substances:(i) TNT, (ii) PETN, and (iii) a mixture ofTNT and PETN with a ratio of 50:50% byweight. The results acquired from theirexperiments were compared with the stan-dard absorbance spectra (achieved usinga Perkin–Elmer Frontier (Fourier transferinfrared) FT-IR/NIR spectrometer), and thedata obtained clearly demonstrated that anoptical interferometer (e.g. LDV) is a decentnoncontact sensor for detection of the pho-tovibrational signal in the standoff mode andin an open environment.

There have been several other reports onexplosives detection [170–173], which havesuccessfully demonstrated isolated cases ofdetection. El-Sharkawy et al. [174–176] haverecently performed remarkable experimentswith PAS toward detection of explosiveresidues. They have reported instanta-neous trace detection of TNT, RDX, andHMX using customized PAS technique bysimultaneously validating the optical sen-sor signals with traditional piezoelectrictransducer, where the customized opti-cal sensor has shown fast response, highsensitivity with 10 cm standoff detec-tion capabilities [174]. They have also


recorded PAS signal of explosive-relatedmaterials such as urea nitrate, ammoniumnitrate, and ammonium perchlorate withdistinguished peaks referring to good dis-crimination [175]. Furthermore, the presenceof explosophorous bonds [nitrates (NO3

−)and perchlorates (ClO4

−)] has been inves-tigated using phase-shift domain analysisbetween explosive and similar nonexplosivematerials [176]. They have also demonstratedthe capability of PAS technique to identifythe residues of PETN, TNT, and AN with 1-saccumulations from a 3-m standoff distance.Thus, PAS can offer possibly fingerprintspectral option for each explosive material,with (a) novel discrimination capability and(b) real-time detection possibility along withstandoff capability since there are no twomaterials with the same optical, thermal, andacoustical properties [177]. Patel et al. [178]have earlier demonstrated the capability ofPAS technique, using QCLs, for the detectionof CWAs and explosives with high sensitivity.They demonstrated detection capability ofCWAs at parts-per-billion levels with falsealarm rates of <1:108. However, concreteefforts are still required to arrive at usabledevices for a variety of applications. Themajor applications for THz radiation identi-fied include (a) screening of humans/baggageat airports and screening of moving vehi-cle at various places, (b) detection of landmines, (c) detection from minute quantitiesof samples (postblast scenario or forensics),and (d) screening for explosives in malls andpublic places. Further concerted efforts fromscientists, engineers, and the users togetherwill enable development of robust sensingtechnologies.

6 Concluding Remarks

In conclusion, there are several laser-basedspectroscopic techniques reported in theliterature, with some of them in advancedstages of research, capable of detectingexplosives from a distance. However, none

of these techniques are ready to be useddirectly in the field for applications (thoughthere a few devices based on THz and SORSthat have been employed for detection inselected airports). The main and unresolvedchallenge still remaining today is the detec-tion of trace explosives from a distance (e.g.detection of land mines is extremely diffi-cult even today). Several factors need to beconsidered and researched before each ofthese techniques has full-fledged applica-tions. No one technique can detect all theexplosives and in different scenarios. Onemust identify the niche applications of eachof these techniques and try to develop hybridmethods to fulfil most of the requirements.SORS can identify concealed explosives,while Raman and LIBS have the potential ofstandoff detection of more than 100 m, whileTHz technique has the capability of imaginghidden materials.

1. Novel laser sources (ns and fs) thatare powerful, compact, robust, andcost-effective are need of the hour.

2. Compact/miniature detectors/spectro-meters are again needed. They should notfully compromise on the resolution ordetection capability even when the size isminiaturized.

3. Broad-band optics (which can transmit inthe entire UV, visible, and NIR spectralrange) with antireflection coatings areessential to collect the weak light signalfrom a distance.

4. Efficient telescope designs are warrantedto collect tiny amount of light emanatingfrom the traces of explosives (especiallyfor LIBS and Raman cases).

5. Development of simple, robust, recy-clable, and cost-efficient SERS substratesis necessary along with portable and hand-held spectrometers with high efficacy.

6. Understanding of the interaction of THzradiation with several materials (clothes,paper, and wood) and in different atmo-spheric conditions (rain, humid, dry, snow,etc.) for propagation and collection of THz


radiation, especially for standoff configu-rations.

7. A complete database (Raman, LIBS, SERS,etc.) of all the explosive molecules, mix-tures, interferants, liquids, etc. is desirablealong with appropriate powerful algo-rithms to identify/detect/discriminate/classify these molecules fromothers.

8. Ideally, for standoff applications, onewishes to have drones [179] fitted withlaser source and detector and wirelessnetwork such that the drone can be sentto point of interest and collect the data(wireless connection) continuously.

Abbreviations

CCD charge-coupled deviceCL-20 hexanitrohexaazaisowurtzitaneDNT dinitrotolueneEGDN ethylene glycol dinitrateFIBS filament-induced breakdown

spectroscopyfs femtosecondFT-IR Fourier transfer infraredHMX octahydro-1,3,5,7-

tetranitro-1,3,5,7-tetrazocineICCD intensified charge-coupled deviceIR infraredLIBS laser-induced breakdown

spectroscopyLOD limit of detectionNM nitromethaneNS nanostructureNs/ns nanosecondPAS photoacoustic spectroscopyPCA principal component analysisPETN pentaerythritol tetranitrateQCL quantum cascade laserRDX cyclotrimethylenetrinitramineSERS surface-enhanced Raman

spectroscopySORS spatial offset Raman spectroscopyTHz TerahertzTNT trinitrotolueneUV ultraviolet

Glossary

ablation The process of removing amaterial from the surface of a solid byirradiating it with a laser pulse. Irradiationresults in the material getting heated bythe absorbed laser energy and evaporatesor sublimates.

explosives Materials that store largeamount of energy within them and releaseit within a short period of time whentriggered.

femtosecond A very short timescale that isequal to 10−15 s.

laser ablation in liquids The process ofablating a material/target placed in aliquid media and short laser pulses. Theinput laser pulse could be a nanosecondor a femtosecond pulse.

nanoparticle (nanopowder, nanocluster,or nanocrystal) A microscopic particlewith at least one dimension <100 nm.

Raman spectroscopy A technique thatprovides information about the molecularvibrations and is generally used for sampleidentification. The technique is based oninelastic scattering of monochromaticlight, usually from a laser source.

SEM (scanning electron microscope) Atype of electron microscope that producesvery high-resolution (submicron) imagesof a sample by scanning it with a focusedbeam of energetic electrons.

TEM (transmission electron microscopy)A technique capable of imaging at aconsiderably higher resolution thannormal light microscopes.

Related Articles

Raman Spectroscopy InstrumentationGeometric OpticsRaman Scattering for Speciation and AnalysisMolecular SpectroscopySpectroscopy, Atomic


References

1 Woodfin, R.L. (2006). Trace ChemicalSensing of Explosives. John Wiley &Sons.

2 Yun-Shik, L. (2008). Principles of Ter-ahertz Science and Technology. Berlin:Springer.

3 Schubert, H. and Kuznetsov, A. (2006).Detection and Disposal of ImprovisedExplosives. Springer Science & BusinessMedia.

4 Brown, K.E., Greenfield, M.T., McGrane,S.D., and Moore, D.S. (2016). Anal.Bioanal. Chem. 408 (1): 35–47.

5 Brown, K.E., Greenfield, M.T., McGrane,S.D., and Moore, D.S. (2016). Anal.Bioanal. Chem. 408 (1): 49–65.

6 Singh, S. and Singh, M. (2003). SignalProcess. 83 (1): 31–55.

7 Gudmundson, E., Jakobsson, A., andStoica, P. (2009). NQR-based explosivesdetection—an overview. InternationalSymposium on Signals, Circuits andSystems, IEEE: ISSCS 2009, 1–4.

8 Tourné, M. (2013). J. Forensic Res. 12:1–10.

9 Bielecki, Z., Janucki, J., Kawalec, A. et al.(2012). Metrology Meas. Syst. 19 (1):3–28.

10 Steinfeld, J.I. and Wormhoudt, J. (1998).Annu. Rev. Phys. Chem. 49 (1): 203–232.

11 Reber, E.L., Blackwood, L.G., Edwards,A.J. et al. (2007). Sensing Imaging Int. J.8 (3–4): 121–130.

12 Osán, T.M., Cerioni, L.M., Forguez, J.et al. (2007). Phys. B Condens. Matter389 (1): 45–50.

13 Wynn, C., Palmacci, S., Kunz, R., andRothschild, M. (2010). Opt. Express 18(6): 5399–5406.

14 Toal, S.J. and Trogler, W.C. (2006).J. Mater. Chem. 16 (28): 2871–2883.

15 Singh, S. (2007). J. Hazard. Mater. 144(1): 15–28.

16 Dragoman, D. and Dragoman, M. (2004).Prog. Quantum Electron. 28 (1): 1–66.

17 Mukherjee, A., Von der Porten, S., andPatel, C.K.N. (2010). Appl. Opt. 49:2072–2078.

18 Wallin, S., Pettersson, A., Östmark, H.,and Hobro, A. (2009). Anal. Bioanal.Chem. 395 (2): 259–274.

19 Pellegrino, P.M., Holthoff, E.L., andFarrell, M.E. (2015). Laser-Based OpticalDetection Of Explosives, vol. 40. CRCPress.

20 Mogilevsky, G., Borland, L., Brickhouse,M., and Fountain, A.W. III (2012). Int. J.Spectrosc. 808079.

21 Moore, D.S. (2004). Rev. Sci. Instrum. 75(8): 2499–2512.

22 Cremers, D.A., Yueh, F.Y., Singh, J.P., andZhang, H. (2006). Laser-Induced Break-down Spectroscopy, Elemental Analysis.Wiley Online Library.

23 Labutin, T.A., Lednev, V.N., Ilyin, A.A.,and Popov, A.M. (2016). J. Anal. At.Spectrom. 31 (1): 90–118.

24 Fortes, F. and Laserna, J. (2010). Spec-trochim. Acta B At. Spectrosc. 65 (12):975–990.

25 Qing-Yu, L. and Yi-Xiang, D. (2017).Chin. J. Anal. Chem. 45 (9): 1405–1414.

26 Bauer, A.J.R. and Buckley, S.G. (2017).Appl. Spectrosc. 71 (4): 553–566.

27 Wiens, R.C., Maurice, S., Barraclough, B.et al. (2012). Space Sci. Rev. 170 (1–4):167–227.

28 Laxmiprasad, A., Raja, V.S., Menon,S. et al. (2013). Adv. Space Res. 52 (2):332–341.

29 Portnov, A., Rosenwaks, S., and Bar, I.(2003). J. Lumin. 102–103: 408–413.

30 Portnov, A., Rosenwaks, S., and Bar, I.(2003). Appl. Opt. 42 (15): 2835–2842.

31 Doucet, F.R., Faustino, P.J., Sabsabi,M., and Lyon, R.C. (2008). J. Anal. At.Spectrom. 23 (5): 694–701.

32 Grégoire, S., Boudinet, M., Pelascini, F.et al. (2011). Anal. Bioanal. Chem. 400(10): 3331–3340.

33 Hahn, D.W. and Omenetto, N. (2010).Appl. Spectrosc. 64 (12): 335A–366A.


34 Hahn, D.W. and Omenetto, N. (2012).Appl. Spectrosc. 66 (4): 347–419.

35 Rao, E.N., Mathi, P., Kalam, S.A. et al.(2016). J. Anal. At. Spectrom. 31 (3):737–750.

36 DeLucia, F., Samuels, A.C., Harmon,R.S. et al. (2005). IEEE Sensors J. 5 (4):681–689.

37 Gottfried, J.L., De Lucia, F.C. Jr.,Munson, C.A., and Miziolek, A.W.(2007). Spectrochim. Acta B At. Spec-trosc. 62 (12): 1405–1411.

38 De Lucia, F.C. Jr., Gottfried, J.L.,Munson, C.A., and Miziolek, A.W.(2008). Appl. Opt. 47 (31): G112–G121.

39 Gottfried, J.L., De Lucia, F.C. Jr., andMiziolek, A.W. (2009). J. Anal. At. Spec-trom. 24 (3): 288–296.

40 Gottfried, J.L., De Lucia, F.C., Munson,C.A., and Miziolek, A.W. (2009). Anal.Bioanal. Chem. 395 (2): 283–300.

41 Lopez-Moreno, C., Palanco, S., Laserna,J.J. et al. (2006). J. Anal. At. Spectrom. 21(1): 55–60.

42 Lazic, V., Palucci, A., Jovicevic, S. et al.(2009). Spectrochim. Acta B At. Spec-trosc. 64 (10): 1028–1039.

43 González, R., Lucena, P., Tobaria, L., andLaserna, J. (2009). J. Anal. At. Spectrom.24 (8): 1123–1126.

44 Lucena, P., Dona, A., Tobaria, L., andLaserna, J. (2011). Spectrochim. Acta BAt. Spectrosc. 66 (1): 12–20.

45 Moros, J. and Laserna, J.J. (2011). Anal.Chem. 83 (16): 6275–6285.

46 Lucena, P., Gaona, I., Moros, J., andLaserna, J. (2013). Spectrochim. Acta BAt. Spectrosc. 85: 71–77.

47 Gaona, I., Serrano, J., Moros, J., andLaserna, J.J. (2014). Anal. Chem. 86 (10):5045–5052.

48 Gottfried, J.L., De Lucia, F.C., Munson,C.A., and Miziolek, A.W. (2008). Appl.Spectrosc. 62 (4): 353–363.

49 Moros, J., Lorenzo, J., and Laserna, J.(2011). Anal. Bioanal. Chem. 400 (10):3353–3365.

50 Rigaku-LIBS-Katana. http://www.rigaku.com/en/products/libs/katana (accessed13 August 2018).

51 Sciaps high performance handheldLIBS analyzer. http://sciaps.com/sciaps-introduces-the-z-a-high-performance-handheld-libs-analyzer/ (accessed 13August 2018).

52 NEW Oxford Instruments mPulse lasermetals analyzer. http://chansurveying.com/products/NEW-Oxford-Instruments-mPulse-laser-metals-analyzer.html (accessed 13 August 2018).

53 Chemlite-portable-libs-metal-analyzer.http://tsi.com/chemlite-portable-libs-metal-analyzer (accessed 13 August2018).

54 De Lucia, F.C. Jr., Gottfried, J.L.,Munson, C.A., and Miziolek, A.W.(2007). Spectrochim. Acta B At. Spec-trosc. 62 (12): 1399–1404.

55 De Lucia, F.C., Gottfried, J.L., andMiziolek, A.W. (2009). Opt. Express17 (2): 419–425.

56 Sunku, S., Gundawar, M.K., Myakalwar,A.K. et al. (2013). Spectrochim. Acta BAt. Spectrosc. 79: 31–38.

57 Rao, E.N., Sunku, S., and Rao, S.V.(2015). Appl. Spectrosc. 69 (11):1342–1354.

58 Kalam, S.A., Murthy, N.L., Mathi, P.et al. (2017). J. Anal. At. Spectrom. 32(8): 1535–1546.

59 Myakalwar, A.K., Anubham, S.K., Paidi,S.K. et al. (2016). Analyst 141 (10):3077–3083.

60 Mirell, D., Chalus, O., Peterson, K., andDiels, J.-C. (2008). J. Opt. Soc. Am. B 25(7): B108–B111.

61 Chin, S., Xu, H., Luo, Q. et al. (2009).Appl. Phys. B 95 (1): 1–12.

62 Brown, C., Bernath, R., Fisher, M. et al.(2006). Remote Femtosecond LaserInduced Breakdown Spectroscopy (LIBS)in a Standoff Detection Regime, 62190B.Enabling Technologies and Designof Nonlethal Weapons, InternationalSociety for Optics and Photonics.


63 Baudelet, M., Richardson, M., andSigman, M. (2009) Self-channeling offemtosecond laser pulses for rapid andefficient standoff detection of energeticmaterials. Technologies for HomelandSecurity, 2009. HST’09. IEEE Conferenceon, IEEE. 472–476.

64 Sunku, S., Rao, E.N., Kumar, G.M. et al.(2013). Proc. SPIE 8726: 8726H.

65 Sreedhar, S., Rao, E.N., Kumar, G.M.et al. (2013). Spectrochim. Acta B At.Spectrosc. 87: 121–129.

66 Myakalwar, A.K., Sreedhar, S., Barman,I. et al. (2011). Talanta 87: 53–59.

67 Sreedhar, S., Gundawar, M.K., and Rao,S.V. (2014). Def. Sci. J. 64 (4): 332.

68 Gurevich, E. and Hergenröder, R. (2007).Appl. Spectrosc. 61 (10): 233A–242A.

69 Rodriguez, M., Bourayou, R., Méjean, G.et al. (2004). Phys. Rev. E 69 (3): 036607.

70 Shaik, A.K., Ajmathulla, and Soma, V.R.(2018). Opt. Lett. 43 (15): 3465–3468.

71 Kalam, S.A., Rao, E.N., and Rao, S.V.(2017). Laser Focus World 53: 24–28.

72 Shaik, A.K., Epuru, N.R., Syed, H. et al.(2018). Opt. Express 26 (7): 8069–8083.

73 Johnson, J.B., Allen, S.D., Merten, J. et al.(2014). J. Spectrosc. 2014: 613435.

74 Pellegrino, P.M., Schill, A.W. IV, andStratis-Cullum, D.N. (2008). Proc. SPIE6954: 695408.

75 Dell’Aglio, M., Alrifai, R., andDe Giacomo, A. (2018). Spectrochim.Acta B At. Spectrosc. 148: 105–112.

76 Kalam, A., Srikanth, V., and Soma, V.R.Nanoparticle Enhanced Laser InducedBreakdown Spectroscopy with Femtosec-ond Pulses. International Conferenceon Fibre Optics and Photonics, OpticalSociety of America. Th3A. 89.

77 Raman, C.V. and Krishnan, K.S. (1928).Nature 121 (3048): 501.

78 Bremer, M.T. and Dantus, M. (2013).Appl. Phys. Lett. 103 (6): 061119.

79 Glenn, R. and Dantus, M. (2015). J. Phys.Chem. Lett. 7 (1): 117–125.

80 Tolles, W.M., Nibler, J.W., McDonald,J.R., and Harvey, A.B. (1977). Appl.Spectrosc. 31 (4): 253–271.

81 Djaker, N., Lenne, P.-F., Marguet, D.et al. (2007). Nucl. Instrum. MethodsPhys. Res. Sect. A 571 (1): 177–181.

82 Natan, A., Levitt, J.M., Graham, L. et al.(2012). Appl. Phys. Lett. 100 (5): 051111.

83 Bremer, M.T., Wrzesinski, P.J., Butcher,N. et al. (2011). Appl. Phys. Lett. 99 (10):101109.

84 Li, H., Harris, D.A., Xu, B. et al. (2009).Appl. Opt. 48 (4): B17–B22.

85 Katz, O., Natan, A., Silberberg, Y., andRosenwaks, S. (2008). Appl. Phys. Lett.92 (17): 171116.

86 Brady, J.J., Farrell, M.E., and Pellegrino,P.M. (2013). Opt. Eng. 53 (2): 021105.

87 Portnov, A., Rosenwaks, S., and Bar, I.(2008). Appl. Phys. Lett. 93 (4): 041115.

88 Portnov, A., Bar, I., and Rosenwaks, S.(2010). Appl. Phys. B 98 (2-3): 529–535.

89 Hirschfeld, T. (1974). Appl. Opt. 13 (6):1435–1437.

90 Hobro, A.J. and Lendl, B. (2009). TrACTrends Anal. Chem. 28 (11): 1235–1242.

91 Wen, P., Amin, M., Herzog, W.D., andKunz, R.R. (2018). TrAC Trends Anal.Chem. 100: 136–144.

92 Hadi, N.M., Mohammad, M.R., andAbdulzahraa, H.G. (2017). J. Al-NahrainUniv. Sci. 20 (3): 60–66.

93 Muhammed Shameem, K.M., Tamboli,M., Devangad, P. et al. (2017). J. RamanSpectrosc. 48 (6): 785–788.

94 Zachhuber, B., Gasser, C., Chrysostom,E.T., and Lendl, B. (2011). Anal. Chem.83 (24): 9438–9442.

95 Zachhuber, B., Ramer, G., Hobro, A.J.,and Lendl, B. (2010). Stand-Off RamanSpectroscopy of Explosives, 78380F.Optics and Photonics for Counterterror-ism and Crime Fighting VI and OpticalMaterials in Defence Systems Technol-ogy VII, International Society for Opticsand Photonics.

96 Moros, J., Lorenzo, J.A., Novotný, K.,and Laserna, J.J. (2013). J. Raman Spec-trosc. 44 (1): 121–130.

97 Gulia, S., Gulati, K.K., Gambhir, V. et al.(2016). Vib. Spectrosc. 87: 207–214.


98 Munro, C.H., Pajceni, V., and Asher,S.A. (1997). Appl. Spectrosc. 51 (11):1722–1729.

99 Gares, K.L., Bykov, S.V., and Asher,S.A. (2017). J. Phys. Chem. A 121 (41):7889–7894.

100 Hufziger, K.T., Bykov, S.V., and Asher,S.A. (2017). Appl. Spectrosc. 71 (2):173–185.

101 Gares, K.L., Hufziger, K.T., Bykov,S.V., and Asher, S.A. (2016). J. RamanSpectrosc. 47 (1): 124–141.

102 Gares, K.L., Bykov, S.V., Brinzer, T., andAsher, S.A. (2015). Appl. Spectrosc. 69(5): 545–554.

103 Gaft, M. and Nagli, L. (2008). Opt.Mater. 30 (11): 1739–1746.

104 Malka, I. and Bar, I. (2016). Appl. Phys.B 122 (2): 42.

105 Malka, I., Petrushansky, A., Rosenwaks,S., and Bar, I. (2013). Appl. Phys. B 113(4): 511–518.

106 Macleod, N.A. and Matousek, P. (2008).Appl. Spectrosc. 62 (11): 291A–304A.

107 Cletus, B., Olds, W., Fredericks, P.M.et al. (2013). J. Forensic Sci. 58 (4):1008–1014.

108 Eliasson, C., Macleod, N., and Matousek,P. (2007). Anal. Chem. 79 (21):8185–8189.

109 Explosives Detection Through OpaqueContainers with Agilent Resolve a Hand-held SORS System. www.agilent.com/chem/raman (accessed 13 August 2018).

110 Fleischmann, M., Hendra, P.J., andMcQuillan, A.J. (1974). Chem. Phys. Lett.26 (2): 163–166.

111 Jeanmaire, D.L. and Van Duyne, R.P.(1977). J. Electroanal. Chem. InterfacialElectrochem. 84 (1): 1–20.

112 Albrecht, M.G. and Creighton, J.A.(1977). J. Am. Chem. Soc. 99 (15):5215–5217.

113 Moskovits, M. (1985). Rev. Mod. Phys. 57(3): 783.

114 Hakonen, A., Andersson, P.O.,Stenbæk Schmidt, M. et al. (2015).Anal. Chim. Acta 893: 1–13.

115 Dasary, S.S., Singh, A.K., Senapati, D.et al. (2009). J. Am. Chem. Soc. 131 (38):13806–13812.

116 Rao, S.V., Podagatlapalli, G.K., andHamad, S. (2014). J. Nanosci. Nanotech-nol. 14 (2): 1364–1388.

117 Liyanage, T., Rael, A., Shaffer, S. et al.(2018). Analyst 143 (9): 2012–2022.

118 Sun, M., Qian, H., Liu, J. et al. (2017).RSC Adv. 7 (12): 7073–7078.

119 Sajanlal, P. and Pradeep, T. (2012).Nanoscale 4 (11): 3427–3437.

120 Chou, A., Jaatinen, E., Buividas, R. et al.(2012). Nanoscale 4 (23): 7419–7424.

121 Chen, T., Lu, S., Wang, A. et al. (2014).Appl. Surf. Sci. 317: 940–945.

122 Almaviva, S., Palucci, A., Botti, S. et al.(2016). Challenges 7 (2): 14.

123 Jamil, A.K., Izake, E.L., Sivanesan, A.,and Fredericks, P.M. (2015). Talanta 134:732–738.

124 Liu, M. and Chen, W. (2013). Biosens.Bioelectron. 46: 68–73.

125 He, X., Liu, Y., Xue, X. et al. (2017). J.Mater. Chem. C 5 (47): 12384–12392.

126 Hakonen, A., Rindzevicius, T., Schmidt,M.S. et al. (2016). Nanoscale 8 (3):1305–1308.

127 Ben-Jaber, S., Peveler, W.J.,Quesada-Cabrera, R. et al. (2017).Nanoscale 9 (42): 16459–16466.

128 Hakonen, A., Wang, F., Andersson, P.O.et al. (2017). ACS Sensors 2 (2): 198–202.

129 Chen, N., Ding, P., Shi, Y. et al. (2017).Anal. Chem. 89 (9): 5072–5078.

130 Podagatlapalli, G.K., Hamad, S.,Sreedhar, S. et al. (2012). Chem. Phys.Lett. 530: 93–97.

131 Hamad, S., Podagatlapalli, G.K., Tewari,S.P., and Rao, S.V. (2013). J. Phys. DAppl. Phys. 46 (48): 485501.

132 Krishna Podagatlapalli, G., Hamad, S.,Tewari, S.P. et al. (2013). J. Appl. Phys.113 (7): 073106.

133 Podagatlapalli, G.K., Hamad, S.,Mohiddon, M.A., and Rao, S.V. (2014).Appl. Surf. Sci. 303: 217–232.


134 Podagatlapalli, G.K., Hamad, S.,Mohiddon, M.A., and Rao, S.V. (2015).Laser Phys. Lett. 12 (3): 036003.

135 Byram, C., Moram, S.S.B., Shaik, A.K.,and Soma, V.R. (2017). Chem. Phys. Lett.685: 103–107.

136 Podagatlapalli, G.K., Hamad, S., and Rao,S.V. (2015). J. Phys. Chem. C 119 (29):16972–16983.

137 Byram, C. and Soma, V.R. (2017).Nano-Structures Nano-Objects 12:121–129.

138 Sree Satya Bharati, M., Byram, C., andSoma, V.R. (2018). Front. Phys. 6: 28.

139 Moram, S.S.B., Byram, C., Shibu,S.N. et al. (2018). ACS Omega 3 (7):8190–8201.

140 Hakonen, A., Wu, K., Stenbæk Schmidt,M. et al. (2018). Talanta 189: 649–652.

141 Hamad, S., Podagatlapalli, G.K.,Mohiddon, M.A., and Soma, V.R. (2014).Appl. Phys. Lett. 104 (26): 263104.

142 Sylvia, J.M., Janni, J.A., Klein, J.D., andSpencer, K.M. (2000). Anal. Chem. 72(23): 5834–5840.

143 Farrell, M.E., Holthoff, E.L., andPellegrino, P.M. (2014). Appl. Spectrosc.68 (3): 287–296.

144 Calzzani, F., Sileshi, R., Kassu, A. et al.(2008). Detection of Residual Traces ofExplosives by Surface Enhanced RamanScattering Using Gold Coated Sub-strates Produced by Nanospheres ImprintTechnique, 69451O. Optics and Pho-tonics in Global Homeland Security IV,International Society for Optics andPhotonics.

145 Kanchanapally, R., Sinha, S.S., Fan, Z.et al. (2014). J. Phys. Chem. C 118 (13):7070–7075.

146 Zhang, C., Wang, K., Han, D., and Pang,Q. (2014). Spectrochim. Acta A Mol.Biomol. Spectrosc. 122: 387–391.

147 Fierro-Mercado, P.M. andHernández-Rivera, S.P. (2012). Int. J.Spectrosc. 2012: Article ID 716527, 7pages. doi: 10.1155/2012/716527.

148 Xu, Z. and Meng, X. (2012). Vib. Spec-trosc. 63: 390–395.

149 Demeritte, T., Kanchanapally, R., Fan,Z. et al. (2012). Analyst 137 (21):5041–5045.

150 Kneipp, K., Wang, Y., Dasari, R.R. et al.(1995). Spectrochim. Acta A Mol. Biomol.Spectrosc. 51 (12): 2171–2175.

151 Shaik, U.P., Hamad, S.,Ahamad Mohiddon, M. et al. (2016).J. Appl. Phys. 119 (9): 093103.

152 Chen, M., Yang, H., Rong, L., and Chen,X. (2016). Analyst 141 (19): 5511–5519.

153 Vendamani, V., Nageswara Rao, S.,Venugopal Rao, S. et al. (2018). J. Appl.Phys. 123 (1): 014301.

154 Teraview. http://www.teraview.com/applications/homeland-security-defense-industry/explosive-detection.html(accessed 13 August 2018).

155 Puc, U., Abina, A., Rutar, M. et al.(2015). Appl. Opt. 54 (14): 4495–4502.

156 Davies, A.G., Burnett, A.D., Fan, W.et al. (2008). Mater. Today 11 (3): 18–26.

157 Palka, N. (2013). Opt. Eng. 53 (3):031202.

158 Federici, J.F., Schulkin, B., Huang, F.et al. (2005). Semicond. Sci. Technol. 20(7): S266.

159 Shen, Y., Lo, a.T., Taday, P. et al. (2005).Appl. Phys. Lett. 86 (24): 241116.

160 Choi, K., Hong, T., Ik Sim, K. et al.(2014). J. Appl. Phys. 115 (2): 023105.

161 Kemp, M.C. (2011). IEEE Trans. Tera-hertz Sci. Technol. 1 (1): 282–292.

162 Chen, J., Chen, Y., Zhao, H. et al. (2007).Opt. Express 15 (19): 12060–12067.

163 Leahy-Hoppa, M., Fitch, M., Zheng, X.et al. (2007). Chem. Phys. Lett. 434 (4–6):227–230.

164 Leahy-Hoppa, M.R., Fitch, M.J., andOsiander, R. (2009). Anal. Bioanal.Chem. 395 (2): 247–257.

165 Holthoff, E., Bender, J., Pellegrino, P.,and Fisher, A. (2010). Sensors 10 (3):1986–2002.

166 Marcus, L.S., Holthoff, E.L., andPellegrino, P.M. (2017). Appl. Spectrosc.71 (5): 833–838.

167 Holthoff, E.L. and Pellegrino, P.M.(2017). Appl. Opt. 56 (3): B74–B84.


168 Chen, X., Guo, D., Choa, F.-S. et al.(2013). Appl. Opt. 52 (12): 2626–2632.

169 Fu, Y., Liu, H., and Xie, J. (2018). Opt.Lasers Eng. 107: 241–246.

170 Witinski, M.F., Blanchard, R., Pfluegl,C. et al. (2018). Opt. Express 26 (9):12159–12168.

171 Pettersson, A., Johansson, I., Wallin,S. et al. (2009). Propellants ExplosivesPyrotechnics 34 (4): 297–306.

172 Carter, J.C., Angel, S.M.,Lawrence-Snyder, M. et al. (2005). Appl.Spectrosc. 59 (6): 769–775.

173 Skvortsov, L.A. and Maksimov, E.(2010). Quantum Electron. 40 (7): 565.

174 El-Sharkawy, Y.H. and Elbasuney, S.(2017). Forensic Sci. Int. 277: 215–222.

175 El-Sharkawy, Y.H., Elbasuney, S.,El-sherif, A. et al. (2018). TrAC TrendsAnal. Chem. 106: 151–158.

176 El-Sharkawy, Y.H. and Elbasuney, S.(2018). Spectrochim. Acta A Mol. Biomol.Spectrosc. .

177 Zrimsek, A.B., Bykov, S.V., and Asher,S.A. (2018). Appl. Spectrosc. doi:10.1177/0003702818792289.

178 Patel, C.K.N. (2008). Eur. Phys. J. SpecialTop. 153: 1–18.

179 SpectroDrone-Detecting Explosives froma Safe Distance-Airobotics. https://www.airoboticsdrones.com/press-releases/spectrodrone-detecting-explosives-safe-distance (accessed 13 August 2018).

photonics for explosives detection' in: encyclopedia of

Documents