DTIC ADA557334: Fiber Optic Coupled Raman Based Detection of Hazardous Liquids Concealed in Commercial Products

1.2 MB·English

by  Defense Technical Information Center

REPORT DOCUMENTATION PAGE
Form Approved OMB NO. 0704-0188
The public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions,
searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments
regarding this burden estimate or any other aspect of this collection of information, including suggesstions for reducing this burden, to Washington
Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington VA, 22202-4302.
Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any oenalty for failing to comply with a collection
of information if it does not display a currently valid OMB control number.
PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS.
1. REPORT DATE (DD-MM-YYYY) 2. REPORT TYPE 3. DATES COVERED (From - To)
New Reprint -
4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER
Fiber Optic Coupled Raman Based Detection of Hazardous W911NF-11-1-0152
Liquids Concealed in Commercial Products
5b. GRANT NUMBER
5c. PROGRAM ELEMENT NUMBER
206022
6. AUTHORS 5d. PROJECT NUMBER
Michael L. Ramírez-Cedeño, Natalie Gaensbauer, Hilsamar Félix-
Rivera, William Ortiz-Rivera, Leonardo Pacheco-Londoño, Samuel P. 5e. TASK NUMBER
Hernández-Rivera
5f. WORK UNIT NUMBER
7. PERFORMING ORGANIZATION NAMES AND ADDRESSES 8. PERFORMING ORGANIZATION REPORT
NUMBER
University of Puerto Rico at Mayaguez
R & D Center
Call Box 9000
Mayaguez, PR 00681 -9000
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS 10. SPONSOR/MONITOR'S ACRONYM(S)
(ES) ARO
U.S. Army Research Office 11. SPONSOR/MONITOR'S REPORT
P.O. Box 12211 NUMBER(S)
Research Triangle Park, NC 27709-2211 58949-PH-REP.28
12. DISTRIBUTION AVAILIBILITY STATEMENT
Approved for public release; distribution is unlimited.
13. SUPPLEMENTARY NOTES
The views, opinions and/or findings contained in this report are those of the author(s) and should not contrued as an official Department
of the Army position, policy or decision, unless so designated by other documentation.
14. ABSTRACT
Raman spectroscopy has been widely proposed as a technique to nondestructively and noninvasively interrogate the
contents of glass and plastic bottles. In this work, Raman spectroscopy is used in a concealed threat scenario where
hazardous liquids have been intentionally mixed with common consumer products to mask its appearance or
spectra. The hazardous liquids under consideration included the chemical warfare agent (CWA) simulant triethyl
phosphate (TEP), hydrogen peroxide, and acetone as representative of toxic industrial compounds (TICs). Fiber
optic coupled Raman spectroscopy (FOCRS) and partial least squares (PLS) algorithm analysis were used to
15. SUBJECT TERMS
Fiber Opics Coupled Raman Spectroscopy, chemical warfare agents simulants, hazardous liquids
16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 15. NUMBER 19a. NAME OF RESPONSIBLE PERSON
a. REPORT b. ABSTRACT c. THIS PAGE ABSTRACT OF PAGES Samuel Hernandez-Rivera
UU UU UU UU 19b. TELEPHONE NUMBER
787-832-4040
Standard Form 298 (Rev 8/98)
Prescribed by ANSI Std. Z39.18
Report Title
Fiber Optic Coupled Raman Based Detection of Hazardous Liquids Concealed in Commercial Products
ABSTRACT
Raman spectroscopy has been widely proposed as a technique to nondestructively and noninvasively interrogate the
contents of glass and plastic bottles. In this work, Raman spectroscopy is used in a concealed threat scenario where
hazardous liquids have been intentionally mixed with common consumer products to mask its appearance or spectra.
The hazardous liquids under consideration included the chemical warfare agent (CWA) simulant triethyl phosphate
(TEP), hydrogen peroxide, and acetone as representative of toxic industrial compounds (TICs). Fiber optic coupled
Raman spectroscopy (FOCRS) and partial least squares (PLS) algorithm analysis were used to quantify hydrogen
peroxide in whiskey, acetone in perfume, and TEP in colored beverages. Spectral data was used to evaluate if the
hazardous liquids can be successfully concealed in consumer products. Results demonstrated that FOC-RS systems
were able to discriminate between nonhazardous consumer products and mixtures with hazardous materials at
concentrations lower than 5%.
REPORT DOCUMENTATION PAGE (SF298)
(Continuation Sheet)
Continuation for Block 13
ARO Report Number 58949.28-PH-REP
Fiber Optic Coupled Raman Based Detection of. .H.azardous Liquids Concealed in Commercial Products
Block 13: Supplementary Note
© 2012 . Published in International Journal of Spectroscopy, Vol. Ed. 0 2012, (0) (2012), (, (0). DoD Components reserve a
royalty-free, nonexclusive and irrevocable right to reproduce, publish, or otherwise use the work for Federal purposes, and to
authroize others to do so (DODGARS §32.36). The views, opinions and/or findings contained in this report are those of the
author(s) and should not be construed as an official Department of the Army position, policy or decision, unless so designated by
other documentation.
Approved for public release; distribution is unlimited.
HindawiPublishingCorporation
InternationalJournalofSpectroscopy
Volume2012,ArticleID463731,7pages
doi:10.1155/2012/463731
Research Article
Fiber Optic Coupled Raman Based Detection of Hazardous
Liquids Concealed in Commercial Products
MichaelL.Ramı´rez-Ceden˜o,1,2NatalieGaensbauer,3HilsamarFe´lix-Rivera,1
WilliamOrtiz-Rivera,1LeonardoPacheco-London˜o,1andSamuelP.Herna´ndez-Rivera1
1ALERT-DHSCenterofExcellence,CenterforChemicalSensorsDevelopment,DepartmentofChemistry,
UniversityofPuertoRico-Mayagu¨ez,P.O.Box9000,Mayagu¨ez,PR00681-9000,USA
2U.S.DepartmentoftheTreasury,BureauofEngravingandPrinting,Washington,DC20228,Boston,USA
3MechanicalandIndustrialEngineeringDepartment,NortheasternUniversity,360HuntingtonAvenue,Boston,
MA02115-5000,USA
CorrespondenceshouldbeaddressedtoSamuelP.Herna´ndez-Rivera,[email protected]
Received15July2011;Revised1October2011;Accepted7October2011
AcademicEditor:MaherS.Amer
Copyright©2012MichaelL.Ram´ırez-Ceden˜oetal. This is an open access article distributed under the Creative Commons
AttributionLicense,whichpermitsunrestricteduse,distribution,andreproductioninanymedium,providedtheoriginalworkis
properlycited.
Raman spectroscopy has been widely proposed as a technique to nondestructively and noninvasively interrogate the contents
of glass and plastic bottles. In this work, Raman spectroscopy is used in a concealed threat scenario where hazardous liquids
havebeenintentionallymixedwithcommonconsumerproductstomaskitsappearanceorspectra.Thehazardousliquidsunder
considerationincludedthechemicalwarfareagent(CWA)simulanttriethylphosphate(TEP),hydrogenperoxide,andacetone
as representative of toxic industrial compounds (TICs). Fiber optic coupled Raman spectroscopy (FOCRS) and partial least
squares(PLS)algorithmanalysiswereusedtoquantifyhydrogenperoxideinwhiskey,acetoneinperfume,andTEPincolored
beverages.Spectraldatawasusedtoevaluateifthehazardousliquidscanbesuccessfullyconcealedinconsumerproducts.Results
demonstrated that FOC-RS systems were able to discriminate between nonhazardous consumer products and mixtures with
hazardousmaterialsatconcentrationslowerthan5%.
1.Introduction The possibility of using liquid explosives for terrorism
purposes is accompanied by the use of chemical warfare
InAugust2006,aterroristplottodestroyaircraftsontrans- agents(CWAs)ortoxicindustrialcompounds(TICs)forthe
atlanticflightswasdiscoveredandtimelystoppedinLondon. same purposes. Such chemicals, which can be accessed in
Theplaninvolvedtheuseofliquidexplosivesstoredinbev- large quantities, can be extremely toxic on contact or in-
erage bottles that would pass checkpoints without being halation even at small quantities [2, 3]. The establishment
detected[1].Theliquidsweregoingtobemixedinflightto of current regulations also recognizes the challenges of de-
create an improvised explosive device (IED) that was going tectingdangerousliquidswhenhiddenormixedwithincom-
to be left in the aircraft and detonated remotely after the mon nondangerous products. For this reason, it is impor-
aircraftlanded.Immediatelyaftertheevent,airportsecurity tant to develop methodologies that would enable to dif-
agencies in the United Kingdom and the United States of ferentiate between common products and compounds that
America established a ban on all liquids except medicines canbecombinedforterrorismintents[4].
andinfantfoodbeyondcheckpoints.Thepositionofairport NormalorspontaneousRamanspectroscopy(RS)isone
security officials has changed several times to allow certain ofthemostpromisingtoolsunderconsideration[5–7].Cur-
amount of liquid or gel-based products. However, when or rentsystemsareportable,sensitive,andhaveawidevariety
howterroristswouldtrytopasshazardousliquidsintoports, of accessories to tailor applications in the field and in the
government buildings, or public areas to generate threats laboratory[8–10].Thetechniqueisbased oninelastic scat-
usingexplosivesorchemicalagentsisunpredictable. tering which brings information on vibrational signatures
2 InternationalJournalofSpectroscopy
of molecules; therefore, it can be used as a specific means
of detection. One of the strengths of RS is that it is able to
analyzesamplesthroughvarioustransparentglassandplastic
containers[11–14].Thisallowsnoninvasive,nondestructive
inspection of the content of a container without opening a.u.)
minimizes exposure to potentially harmful substances and y (
helps to speed the screening process. Since water does nsit
e
not present a strong signature in RS measurements, such nt
I
systems are able to characterize aqueous-based commercial
products.Inaddition,RSdatacanbesuccessfullycombined
withchemometricsalgorithmssuchasprincipalcomponents
analysis(PCA),partialleastsquares(PLS)regressionanaly- 200 400 600 800 1000 1200 1400 1600 1800
sis, and discriminant analysis (DA) for improved detection Raman shift (cm−1)
andquantificationinavarietyofmatrices[15–17].
Tequila gold
Principalcomponentsanalysis(PCA)isveryusefulasa
White rum
datareductiontechniqueforspectroscopicanalysis.InPCA,
Whiskey
spectraareprojectedinspaceasorthogonalvariablesknown
as vectors or principal components (PCs). The goal is to Figure1:Ramanspectrumoftequila,rum,andwhiskeyacquired
captureasmuchvarianceaspossiblewiththeleastamountof with FOCRS. Parameters were 785nm laser, 200mW, and 2s
vectors(PCs)whilereducingunwantednoisebytransferring integration time. All liquors exhibited similar spectra preserving
spectralfeatures.
tolesscontributingPCs.Usuallydataispretreatedtosmooth
random differences and to bring out subtle differences not
perceivablebysimpleevaluationofspectrabyeyesight.Each
componenthasaloadingandascore.Theloadingdescribes rum, Tequila Cuervo Gold, and Ralph perfume. Hydrogen
thespectralfeaturesintheoriginalspectrathatarecaptured peroxide (50wt.% in water) and acetone (99+%) were ob-
by the component, and the score tells how the spectrum of tainedfromFisherScientificInternational,Chicago,IL,USA.
eachPCcompares/contributestotheoriginal(experimental) Cyclohexane (≥99%) and triethyl phosphate (TEP; ≥99%)
spectraintermsofthatparticularcomponentbyassigninga wereobtainedfromSigma-AldrichChemicalCompany(St.
value.Dataisthenvisualizedasaplotofscoreswhichshows Louis,MO,USA).
how different or similar are samples, data groups (classes),
orthewholedataset.InPLS,thevariationobservedduring
2.2.MeasurementsandAnalysis. PortableFOC-basedinstru-
aPCAregressionanalysisiscorrelatedtochangesinconcen-
ment(RamanSystemsmodelR-3000HR)wasusedtoeval-
tration or other measurable property performing a regres-
sionofdatavariance(x)versusaknownvariable(y)[18,19]. uatethecontentofcommercialproducts.TheRSsystemhad
anexcitationwavelengthof785nm(red)with250mWmax-
In this study, FOCRS was used to characterize samples
imumoutputpowermeasuredattheheadoftheprobe.The
in which hazardous liquids were concealed in commercial
spectra of the original liquid in its container were collected
products by mixing them with common liquids in their
though the container walls. The liquid was then replaced
original containers. Common commercial products such
with30mLofahazardousliquidandanalyzed.Liquidswere
as liquors, perfumes, and beverages can exhibit color or
mixedatrandomvolumetovolumeratios(v/v)topreparea
chemical composition that can mask the presence of the
totalof50mL.Ramanspectrawereacquiredfrom200cm−1
contaminanteitherbyspectralcongestionormorecritically
to1800cm−1.ThesystemwascalibratedusingHPLCgrade
by exhibiting endogenous fluorescence. Hydrogen peroxide
cyclohexaneasanexternalstandardbymeasuringtheband
was used as a simulant for peroxide-based materials or
at Raman shift location about 801.3cm−1. The data was
precursors that can be used as homemade explosives. TEP
exportedandmanipulatedusingGramsv.9(ThermoFisher
is a structural analog of a CWA and therefore is commonly
Scientific,Waltham,MA,USA).PretreatmentsandPLSanal-
used as a chemical simulant for it. Acetone was used as
ysesandothercalculationswereperformedusingMATLAB
representativeofindustrialsolventsclassifiedasTICs,butit
v.8 (The Mathworks, Natick, MA, USA.), and PLS Tool-
is also a precursor of the family of cyclic organic peroxides
box Solo v. 6.3 (Eigenvector Technologies, Wenatchee, WA,
collectively referred to as acetone peroxide homemade ex-
USA.),andExcelv.2010(Microsoft,Redmond,WA,USA).
plosives. The results suggested that FOCRS can be used to
discriminateandquantifythehazardousliquidconcealedin
thecommercialproducts. 3.ResultsandDiscussion
2.ExperimentalSection 3.1. Detection of Peroxide in Liquors. The spectra of the
three liquors: whiskey, rum, and tequila were obtained in
2.1.MaterialsandReagents. Thecommercialliquidproducts their original clear glass bottle. The results are presented in
employed in this study were Gatorade Frost Glacier Freeze Figure1. The spectra of the liquors were substantially
(blue color), Fruit Punch (red color) and Orange (orange similaralthoughthereweresomedifferencesinrelativepeak
color), Dewar’s White Label whiskey, Ron Bacardi light intensities. The additives that render the yellowish (gold)
InternationalJournalofSpectroscopy 3
u.)
a.u.) y (a.
ntensity ( Intensit
I
200 400 600 800 1000 1200 1400 1600 1800
200 400 600 800 1000 1200 1400 1600 1800
Raman shift (cm−1)
Raman shift (cm−1)
White rum
Acetone
Hydrogen peroxide
Acetone peroxide
Acetone
Hydrogen peroxide
Figure 3: Spectroscopic signatures of rum versus acetone and
Figure 2: Raman spectra of acetone, hydrogen peroxide, and
peroxide. Parameters were 785nm, 10s, and a laser power of
an acetone peroxide mixture contained in a clear glass bottle. 200mW.Thecharacteristicbandofacetoneispresentat798cm−1.
Parameterswere785nm,10sacquisitiontimeat200mWoflaser The ethanol peak at 881cm−1 makes difficult the differentiation
power. The characteristic peak for acetone is located at 798cm−1 fromtheperoxidepeakat878cm−1.
andforperoxideat878cm−1.Themainacetoneperoxidepeaksare
locatedat588cm−1,780cm−1,890cm−1,and936cm−1.
colorofthetequilaandwhiskeypresentedsomefluorescence
identifiedbytheshiftinthebaselineofthespectra.However, u.)
it was possible to identify the main peaks associated with y (a.
the alcohol content. These spectra as well as the spectrum nsit
for ethanol, the main component of hard liquors, present e
nt
peaks at 881cm−1 (C–C–O symmetric stretch), 1048cm−1 I
(C–C–Oasymmetricstretch)and1086cm−1 (C–Ostretch),
1280cm−1 (C–CH deformation), and 1456cm−1 (C–H
3
stretch in CH and CH ) [20]. Then the contents of the
2 3
clear glass bottle were replaced with acetone and hydrogen 200 400 600 800 1000 1200 1400 1600 1800
peroxidesothatthecontentsappearedtobeclearrum.These Raman shift (cm−1)
consumerproductsareeasilyacquiredatbeautysuppliesand
Whiskey
drugstores,buttheyarealsoprecursorsofacetone-peroxide,
20% peroxide in whiskey
awell-knownhomemadeexplosive(HME)[21,22]. 60% peroxide in whiskey
The Raman spectra of acetone, peroxide, and acetone-
Figure 4: Raman spectra of a mixture of hydrogen peroxide and
peroxidearepresentedinFigure2.Thespectrumofacetone
showed high-intensity peaks at 530cm−1 (C=O deforma- whiskey.Parameterswere785nmlaser3scansat10swith200mW
laserpower.Theadditionofperoxidecausednopeakshiftsbutdid
tion), 789cm−1 (C–C symmetric stretch), 1225cm−1 (C–C
increasetheintensityofthebandat874cm−1.
asymmetric stretch), and 1691cm−1 (C=O stretch band).
ThemostprominentvibrationalRamansignatureislocated
at 789cm−1 [23]. Hydrogen peroxide is characterized by a
strongpeakat878cm−1,characteristicoftheO–Ostretching presents clear characteristic bands at 881cm−1, 1048cm−1,
mode. The main acetone peroxide peaks are located at 1086cm−1, 1280cm−1, and 1456cm−1 that together with
400cm−1 (ring breathing), 594cm−1 (O–C–O bending), knownbandratioscanbeusedtodifferentiatetherumfrom
779cm−1 (O–C–O symmetric stretch and Me–C–Me sym- otherliquids.Inthiscase,acetoneisclearlyidentifiablewith
metricstretch),880cm−1(O–OstretchandMerockingsyn- a unique band at 789cm−1. However the strong band at
chronous and asynchronous), 940cm−1 (C–C stretch), 881cm−1 due to C–C–O stretching that all liquors present
1067cm−1(C–CandringC–Ostretch),and1451cm−1(H– islocatedinthesameregionofthehydrogenperoxideC–O
C–Hbend)[22–24].Theposition,intensity,anduniqueness band at 878cm−1. This may represent a challenge for the
ofthemajorpeaksinacetoneandperoxidemakethemeasily detectionofperoxidemixedwithliquors.
detectedbyRSandcanbeusedasmarkersforthisexplosive Figure4 shows the Raman spectra of 20% v/v and 60%
mixture[24,25]. v/vhydrogenperoxidepreparedfrom50%H O solutionin
2 2
InFigure3,theacetoneandperoxidespectroscopicsig- water(10%and30%,H O resp.)inwhiskey.Aspreviously
2 2
natures are compared to those of rum. The rum spectrum mentioned,thespectroscopicsignatureofperoxideishidden
4 InternationalJournalofSpectroscopy
0.015
0.1
0.01
0.005
%) 0.05
9 0 gs
0 n
2 (13. −0.005 Loadi
C 0
P −−0.00.1051 C oncentration
−0.05
−0.02 200 450 700 950 1200 1450 1700
0.008 0.013 0.018 0.023 0.028 0.033 0.038 Raman shift (cm−1)
PC1 (86.89%) Loadings vector 1 (86.89%)
Loadings vector 2 (13.09%)
Figure5:ScoresofPC1andPC2(99.98%ofthespectralvariance)
are related to the change in concentration of hydrogen peroxide. Figure 6: Loadings vectors for PC1 and PC2. The data confirms
Quantificationofmixturescanbeefficientlycarriedoutusingthese thatspectralchangesintheregionofthehydrogenperoxidepeak
twocomponentsonly. (878cm−1)areresponsibleformostofthespectralvariance.
100
ormaskedbytheethanolicpeakat881cm−1.Samplesfrom
90
0% to 90% peroxide in whiskey were analyzed without 80 y=1.0109x−0.82
observing changes in the location of peaks. However, the 70 R2=0.9945
presenceofperoxidehasaneffectontheintensityoftheband
881cm−1. d (%) 5600
PLS was used to build a model to determine the con- e
ct 40
centrationofperoxideinwhiskey.Thedatawasevaluatedin di
the range of 200 to 1800cm−1, after spectral intensity nor- Pre 30
20
malization. Figure5 shows the scores for principal compo-
10
nents(PC’s)1and2.Thesetwocomponentsareresponsible
0
for 99.98% of the variance in the spectra. The score on
−10
eachspectrumforthesecomponentsconfirmsthatthereare 0 10 20 30 40 50 60 70 80 90 100
spectraldifferencesinthedatasetasaconsequenceofcon- Actual (%)
centrationofperoxide.
Figure 7: Prediction of the concentration of peroxide in whiskey
An evaluation of the loadings provides information on
showingasignificantrelationbetweenconcentrationandspectral
theoriginofthevariancecapturedbyPC1andPC2.Figure6
changes. This relation suggests that it is possible to discriminate
confirmsthatmainlytheoriginofthevariancecapturedby
betweenexpectedalcoholconcentrationinwhiskyandthepresence
PC1 and PC2 is within the area of the 881cm−1 peak. The
ofothermaterialcontributingtothissignal.
PC2suggeststhattherearecontributionsintheincreaseand
the decrease of the intensity of the spectrum as well. This
information is used to create a PLS regression of the
contain multiple ingredients such as alcohols, peroxides,
spectral variance (x) versus concentration of hydrogen
andotherswithchemicalcompositionrelatedtohazardous
peroxide (y). This model was evaluated by a leave-one-out
chemicals. Many personal care products are also colored
cross-validation.Theresultsofthevalidationwereevaluated
and present fluorescence, obscuring the persistent Raman
by the root mean square error of calibration (RMSEC) and
signatures. The spectra of a perfume is shown in Figure8.
the root mean square error of cross-validation (RMSECV).
The spectrum was obtained in its original bottle. The
Inthefirstone,thevaluesinthedatasetarepredictedwith
spectrum of 10% acetone in the perfume is included.
thecalibrationcurveobtainedandinthesecondthevalues
The fluorescence and peak of the perfume dominates the
areestimatedusingacalibrationcurvethatdidnotinclude
signatureoftheliquid.
thisvalue.ForthecaseofperoxideIwhiskey,theprediction
Acetone was mixed with the perfume at concentration
of%ofperoxideversusactualconcentration(Figure7)can
between 10–50% (volume by volume). At higher concen-
beachievedwithaRMSECof2.23andRMSECVof2.69.The
trations, the spectrum of acetone becomes more evident.
linearityofthefitisdescribedbytheR2value(0.9945).
Afterafirstassessmentofprincipalcomponentsandloadings
vectors, the spectral range of 600cm−1 to 1200cm−1 was
3.2.DetectionofToxicIndustrialCompoundsinPersonalCare usedtoestimatetheconcentrationofacetoneintheperfume.
Products. Consumer products such as personal care and The data was first mean centered. Figure9 presents the
cosmetics present a challenge. Commonly these products predicted versus actual concentration obtained with a PLS
InternationalJournalofSpectroscopy 5
nsity (a.u.) ensity (a.u.)
nte Int
I
200 400 600 800 1000 1200 1400 1600 1800 200 400 600 800 1000 1200 1400 1600 1800
Raman shift (cm−1) Raman shift (cm−1)
Perfume Red
10% acetone in perfume Orange
Blue
Figure8:Ramanspectraofaperfumeandamixtureofperfume
and10%acetone.Parameterswere785nmlaser,100mW,and10s Figure10:Ramanspectraofred,blue,andorangeGatoradebev-
integrationtime.Presenceofacetoneisonlysuggestedbythesmall erages.Parameterswere785nmlaser,100mW,and10sintegration
peakat880cm−1. time.
53
48
y=0.9973x+ 0.0676
43
38 R2=0.9973 u.)
%) 33 y (a.
edicted ( 122838 Intensit
Pr
13
8
3
−2 200 400 600 800 1000 1200 1400 1600 1800
0 5 10 15 20 25 30 35 40 45 50
Raman shift (cm−1)
Actual (%)
TEP 18% TEP in red
Figure 9: PLS determination of acetone in perfume. Conditions: 34% TEP in red 6% TEP in red
spectralrange:600to1200cm−1,meancenteringofspectra,three
components(PC’s)used. Figure11:SpectrumofneatTEPandatdifferentconcentrationsin
redGatorade.Thespectralrangeof200to1800cm−1wasusedfor
PLS-basedquantification.
basedmodelinthisspectralregion.Threecomponentswere
usedtoachieveaRMSECof0.887andaRMSECVof1.287.
34%ofTEP.Atconcentrationshigherthan50%ofTEP,the
signaturecouldbeeasilydistinguishedoverthespectrumof
3.3. Detection in Colored Liquids. The detection of a CWA thebeverage.
simulant in heavily colored liquids was studied by FOCRS. AdiscriminationexerciseinwhichPLSwasusedtodis-
TEPiscommonlyusedasasimulantofSoman(GD),anerve criminate between samples with and without TEP was per-
agent. Figure10 presents the spectra of different flavors of formed. For this experiment, all Gatorade samples that did
Gatorade described by colors. The red presents a higher notcontainTEPwereassignedavalueof“−1”,andthosethat
baseline associated to fluorescence. No characteristic peaks containedTEPwerelabeledas“1”.ThesampleswithTEPhad
were observed for the red, blue, or orange liquids at these concentrations of the analyte in Gatorade as follows: 13%
conditions. TEP in blue, 11% TEP in orange, 18% TEP in red, and
TEPRamanspectrum,asshowninFigure11,haschar- 6% in red. A PLS-DA regression was performed, and the
acteristicspeakssuchas733cm−1 (PO symmetricstretch), results are summarized in Figure12. Specificity and sensi-
3
813cm−1 (PO asymmetric stretch), 1032 and 1098cm−1 tivity for each class (y = 1 and y = −1) was one (1).
3
(C–Ostretch),1162cm−1(CH rocking),and1279cm−1(P– However, it is seen that, for samples in red Gatorade, this
3
O symmetric stretch) [26]. TEP was then mixed randomly discriminationwasnotasevidentasforothercolors.Specif-
with the different colors of gatorade at concentration up icallythegroupof6%TEPinredGatoradeobtainedvalues
to 85% v/v of TEP. Figure11 shows the spectrum of TEP close to zero, suggesting that this concentration of TEP is
anditsmixtureswithredGatoradeatconcentration6%and closetothedetectionlimitsoftheapproach.Thethresholds
6 InternationalJournalofSpectroscopy
Table1:LODandLOQforTEPincoloredliquids.
Gatoradevariety PC’s Cumulative%variance R2ofpredictioncurve RMSEC LOD LOQ
Red 2 99.99 0.998 1.52 6.5 21.8
Blue 2 95.72 0.967 2.85 5.2 17.2
Orange 2 99.58 0.961 4.30 10.4 35.2
2 Forthisstudytheregionfrom675to855cm−1wasinte-
grated,anditsrelationtotheconcentrationwasevaluatedby
1.5
regressionanalysistoestimateLODandLOQ.Theobjective
n 1 of these graphs is to compare the limit of detection (LOD)
o
dicti 0.5 and the limit of quantification (LOQ) between colored
pre 0 liquids. This region was selected because of the presence
Y of the characteristic peaks for phosphates. This region will
−0.5 6% TEP simulate the characteristic spectral region of real CWA and
−1 relatedsimulants.
0 10 20 30 40 50 Table1 presents a summary of the PLS-based detection
Sample aswellastheLODandLOQobtainedforTEPinGatorade.
The limits of detection for this analysis were at 10% or less
In class (1) Not in class (1) Threshold
with LOQ between 17 and 35%. However, the low RMSEC
In class (−1) Not in class (−1) Threshold
suggests that quantification can be achieved with PLS or
Figure 12: PLS-based discriminant analysis of TEP in different otheralgorithmsatlevelsclosetotheLOD.
varieties of Gatorade. A positive value (1) implies presence of
TEP, and a negative value (−1) do not contain TEP. Values close 4.Conclusions
to the threshold line (x axis) indicate that samples are close in
spectralcharactermakingmorechallengingthediscrimination.For In this work, concealed liquids scenarios were studied by
y = −1 and 3 PC’s;R2cal = 0.745R2CV = 0.728, RMSEC = FOCRS. A fiber optic coupled Raman probe was used to
0.249, andRMSECV=0.257. differentiate hazardous liquids from common drinks and
consumer products by acquiring spectra through the con-
tainerwalls.PLSalgorithmswereusedtoquantifytheliquids
with RMSEC of less than 5%. These results demonstrated
valuesforpresencewere0.3156andfornotpresence0.6560. that Raman spectroscopy can be used as a tool to quickly
These values represent the critical value of prediction that determine if the content of a bottle is the intended com-
willseparateclasses.Highervalueswillcorrespondtobetter mercial product or has been used to conceal a hazardous
discrimination.Inthiscase,sampleswithoutcontamination material. The fluorescence of a red liquid did not affect the
(pureliquids)wereclearlydiscriminatedfromsampleswith detectionofaCWAsimulantinaclearglasscontainer.The
contaminationatlowpercentageofconcentration. limitsofdetectionforahazardousliquidinaseriesofcolored
PLS-quantitative analysis was performed on samples liquids were estimated in 6–10% using standard conditions
containingTEPatconcentrationlevelslowertan50%since and no data manipulation. This suggests that trace level
thepresenceofTEPismoreobviousoverthatvalue.Quan- detection can be achieved with enhanced experimental
tificationwasperformedbyindividualcolors.Thedatawas setupsandstatisticalanalysisofthedata.
meancenteredandanalyzedbyPLSintherangeof200cm−1
to 1800cm−1. In order to compare the quantification, all Acknowledgments
results were compared at 2 PC’s. However, improvement in
the model with 3 and 4 components was observed. Data is Support from the US Department of Homeland Security
summarized in Table1. Results showed that quantification under Award no. 2008-ST-061-ED0001 is acknowledged.
ofTEPinthesecoloredmatricescanbeachievedwithgood However, the views and conclusions contained in this doc-
linearityandRMSECVoflessthan5%. umentarethoseoftheauthorsandshouldnotbeconsidered
Finally,thelimitsofdetection(LOD)forthisanalytewere a representation of the official policies, either expressed or
calculated according to the IUPAC (International Union of implied, of the US Department of Homeland Security. Part
PureandAppliedChemistry)criteriaof3σ(3timesstandard of the work presented in this study was supported by
deviation of the peak-to-peak noise related to the slope of the US Department of Defense University Research Initia-
thelinearregressionfunction).Whenthesignalis3timesas tive Multidisciplinary University Research Initiative (URI)-
greatasthenoise,itisreadilydetectablebutstilltoosmallfor MURI Program under the Grant no. DAAD19-02-1-0257.
accuratemeasurement.Asignalthatis10timesasgreatasthe TheauthorsalsoacknowledgecontributionsfromMr.Aaron
noiseisdefinedasthelowerlimitofquantification(LOQ),or LaPointe from the Night Vision and Electronic Sensors Di-
the smallest amount that can be measured with reasonable rectorate,FortBelvoir,VA,DepartmentofDefense,Dr.Jen-
accuracy[27]. nifer Becker, MURI Program Manager, Army Research
InternationalJournalofSpectroscopy 7
Office,DoD,andDr.StephenJ.Lee,ChiefScientist,Science [16] J. D. Driskell, Y. Zhu, C. D. Kirkwood, Y. Zhao, R. A. Dlu-
and Technology, Office of the Director, Army Research Of- hy,andR.A.Tripp,“Rapidandsensitivedetectionofrotavirus
fice/Army Research Laboratory, DoD. The authors also molecular signatures using surface enhanced raman spec-
acknowledge contributions of Omar Rivera and Gabriel troscopy,” PLoS ONE, vol. 5, no. 4, Article ID 10.1371/jour-
Nieves and the Chemical Imaging Center at the Chemistry nal.pone.0010222,2010.
[17] V. Sikirzhytski, K. Virkler, and I. K. Lednev, “Discriminant
DepartmentoftheUniversityofPuertoRico-Mayagu¨ez.
analysisofRamanspectraforbodyfluididentificationforfo-
rensicpurposes,”Sensors,vol.10,no.4,pp.2869–2884,2010.
References [18] W.O.GeorgeandandD.Steele,ComputingApplicationsinMo-
lecularSpectroscopy,RoyalSocietyofChemistry,Cambridge,
[1] Police: Plot to blow up aircraft foiled. http://www.cnn.com. UK,1995.
PostedAugust10,2006. [19] O. Svensson, M. Josefson, and F. W. Langkilde, “Reaction
[2] Y.SunandK.Y.Ong,DetectionTechnologiesforChemicalWar- monitoring using Raman spectroscopy and chemometrics,”
fareAgentsandToxicVapors,CRCPress,BocaRaton,Fla,USA, ChemometricsandIntelligentLaboratorySystems,vol.49,no.
2005. 1,pp.49–66,1999.
[3] T. C. Marrs, R. Maynard, and F. Sidell, Chemical Warfare [20] E. K. Plyler, “Infrared spectra of methanol, ethanol, and n-
Agents:ToxicologyandTreatment,JohnWiley&Sons,London, propanol,” Journal of Research of the National Bureau of
UK,1996. Standards,vol.48,no.4,pp.281–286,1952.
[4] NationalResearchCouncil,FinalReportonExistingandPoten- [21] N.A.MilasandA.Golubovic´,“Studiesinorganicperoxides.
tial Standoff Explosives Detection Techniques, National Acad- XXVI.Organicperoxidesderivedfromacetoneandhydrogen
emyofSciences,Washington,DC,USA,2004. peroxide,” Journal of the American Chemical Society, vol. 81,
[5] S.Farquharson,A.Gift,P.Maksymiuk,andF.Inscore,“Sur- no.24,pp.6461–6462,1959.
face-enhancedramanspectraofVXanditshydrolysisprod- [22] L. Jensen, P. M. Mortensen, R. Trane, P. Harris, and R. W.
ucts,”AppliedSpectroscopy,vol.59,no.5,pp.654–660,2005. Berg, “Reaction kinetics of acetone peroxide formation and
[6] K.L.McNesby,J.E.Wolfe,J.B.Morris,andR.A.Pesce-Rod- structureinvestigationsusingRamanspectroscopyandX-ray
riguez,“FouriertransformRamanspectroscopyofsomeen- diffraction,” Applied Spectroscopy, vol. 63, no. 1, pp. 92–97,
ergetic materials and propellant formulations,” Journal of 2009.
RamanSpectroscopy,vol.25,no.2,pp.75–87,1994. [23] G. Dellepiane and J. Overend, “Vibrational spectra and
[7] W.F.PearmanandA.W.Fountain,“Classificationofchemical assignmentofacetone,αααacetone-d3andacetone-d6,”Spec-
and biological warfare agent simulants by surface-enhanced trochimicaActa,vol.22,no.4,pp.593–614,1966.
Ramanspectroscopyandmultivariatestatisticaltechniques,” [24] J. Oxley, J. Smith, J. Brady et al., “Raman and infrared fin-
AppliedSpectroscopy,vol.60,no.4,pp.356–365,2006. gerprint spectroscopy of peroxide-based explosives,” Applied
[8] I. R. Lewis, N. W. Daniel Jr, N. C. Chaffin, P. R. Griffiths, Spectroscopy,vol.62,no.8,pp.906–915,2008.
and M. W. Tungol, “Raman spectroscopic studies of explo- [25] L. Pacheco-London˜o, O. M. Primera, M. Ram´ırez, O. Ruiz,
sive materials: towards a fieldable explosives detector,” Spec- and S. Hernandez-Rivera, “Review of the various analytical
trochimicaActaPartA,vol.51,no.12,pp.1985–2000,1995. techniquesandalgorithmsfordetectionandquantificationof
[9] M. Donahue, H. Huang, C. Brouillette, W. Smith, and S. TATP,”inProceedingsoftheinternationalsocietyforopticsand
Farquharson, “Detecting explosives by portable Raman ana- photonics(SPIE’05),E.M.Carapezza,Ed.,vol.5778,pp.317–
lyzers:acomparisonof785,976,1064,and1550nm(Retina- 326, Sensors, and C3I Technologies for Homeland Security
Safe)laserexcitation,”Spectroscopy,pp.24–28,2011. and Homeland Defense IV, Bellingham, Wash, USA, month
[10] S.Christesen,B.Maciver,L.Procell,D.Sorrick,M.Carrabba, year.
andJ.Bello,“Nonintrusiveanalysisofchemicalagentidentifi- [26] K.Taga,N.Hirabayashi,T.Yoshida,andH.Okabayashi,“Ra-
cationsetsusingaportablefiber-opticRamanspectrometer,” man spectra and rotational isomerism of dimethylphospho-
AppliedSpectroscopy,vol.53,no.7,pp.850–855,1999. rochloridate and trimethyl phosphate,” Journal of Molecular
[11] S. D. Harvey, M. E. Vucelick, R. N. Lee, and B. W. Wright, Structure,vol.212,pp.157–168,1989.
“Blind field test evaluation of Raman spectroscopy as a fo- [27] V. Thomsen, D. Schatzlein, and D. Mercuro, “Limits of de-
rensictool,”ForensicScienceInternational,vol.125,no.1,pp. tectioninspectroscopy,”Spectroscopy,vol.18,no.12,pp.112–
12–21,2002. 114,2003.
[12] C. Eliasson, N. A. Macleod, and P. Matousek, “Noninvasive
detection of concealed liquid explosives using Raman spec-
troscopy,”AnalyticalChemistry,vol.79,no.21,pp.8185–8189,
2007.
[13] C.Eliasson,N.A.Macleod,andP.Matousek,“Non-invasive
detection of powders concealed within diffusely scattering
plasticcontainers,”VibrationalSpectroscopy,vol.48,no.1,pp.
8–11,2008.
[14] M.Alvarez-Rivera,FiberopticcoupledRamanspectroscopyof
liquid explosives mixtures, M.S. Dissertation, University of
PuertoRico,Mayaguez,PuertoRico,2002.
[15] C.Frausto-Reyesa,C.Medina-Gutierrez,R.Sato-Berrub,and
L.R.Sahaguna,“Qualitativestudyofethanolcontentintequi-
lasbyRamanspectroscopyandprincipalcomponentanalysis,”
SpectrochimicaActaPartA,vol.61,no.11-12,pp.2657–2662,
2005.