THz-TDS for Detecting Glycol Contamination in Engine Oil

Abstract

There continues to be a need for an in-situ sensor system to monitor the engine oil of internal combustion engines. Engine oil needs to be monitored for contaminants and depletion of additives. While various sensor systems have been designed and evaluated, there is still a need to develop and evaluate new sensing technologies. This study evaluated Terahertz time-domain spectroscopy (THz-TDS) for the identification and estimation of the glycol contamination of automotive engine oil. Glycol contamination is a result of a gasket or seal leak allowing coolant to enter an engine and mix with the engine oil. An engine oil intended for use in both diesel and gasoline engines was obtained. Fresh engine oil samples were contaminated with four levels of glycol (0 ppm, 150 ppm, 300 ppm, and 500 ppm). The samples were analyzed with THz-TDS and converted to frequency domain parameters of refractive index and absorption coefficient. While both parameters showed potential, the absorption coefficient had the best potential and was able to statistically discriminate among the four contamination levels.

Full text



Appl.Sci.2020,10,3738;doi:10.3390/app10113738www.mdpi.com/journal/applsci
Article
THz‐TDSforDetectingGlycolContamination
inEngineOil
OdayM.Abdulmunem
1,†
,AliMazinAbdul‐Munaim
2,
*
,†
,MarioMendezAller
3
,SaschaPreu
3,
*
andDennisG.Watson
4
1
DepartmentofPhysics,CollegeofScience,MustansiriyahUniversity,Baghdad10071,Iraq;
munem@uomustansiriyah.edu.iq
2
DepartmentofAgriculturalMachinesandEquipment,CollegeofAgriculturalEngineeringSciences,
UniversityofBaghdad,Baghdad10071,Iraq
3
TerahertzDevicesandSystems,TechnicalUniversityofDarmstadt,64283Darmstadt,Germany;
aller@imp.tu‐darmstadt.de
4
AgriculturalSystems,CollegeofAgriculturalSciences,SouthernIllinoisUniversity,Carbondale,
IL62901,USA;dwatson@siu.edu
*Correspondence:alimazin@coagri.uobaghdad.edu.iq(A.M.A.‐M.);sascha.preu@tu‐darmstadt.de(S.P.)
†
Theseauthorshavecontributedequally.
Received:26April2020;Accepted:20May2020;Published:28May2020
FeaturedApplication:TheabilityoftheTHz‐TDStodetectengineoilcontaminationwithglycol
atlowconcentrations.
Abstract:Therecontinuestobeaneedforanin‐situsensorsystemtomonitortheengineoilof
internalcombustionengines.Engineoilneedstobemonitoredforcontaminantsanddepletionof
additives.Whilevarioussensorsystemshavebeendesignedandevaluated,thereisstillaneedto
developandevaluatenewsensingtechnologies.ThisstudyevaluatedTerahertztime‐domain
spectroscopy(THz‐TDS)fortheidentificationandestimationoftheglycolcontaminationof
automotiveengineoil.Glycolcontaminationisaresultofagasketorsealleakallowingcoolantto
enteranengineandmixwiththeengineoil.Anengineoilintendedforuseinbothdieseland
gasolineengineswasobtained.Freshengineoilsampleswerecontaminatedwithfourlevelsof
glycol(0ppm,150ppm,300ppm,and500ppm).ThesampleswereanalyzedwithTHz‐TDSand
convertedtofrequencydomainparametersofrefractiveindexandabsorptioncoefficient.While
bothparametersshowedpotential,theabsorptioncoefficienthadthebestpotentialandwasableto
statisticallydiscriminateamongthefourcontaminationlevels.
Keywords:terahertz;time‐domainspectroscopy;engineoil;glycol
1.Introduction
Combustionengineusersfrequentlyreplaceengineoilthatisstillusableorkeepusingengine
oilthatiscontaminatedandnolongersuitableforuse.Engineoilisexposedtovariouscontaminants
andoxidationthataffectlubricantefficiency,andthustheoilhasalimitedlife.Additivesforengine
oilsreducetheeffectofcontaminants,oxidation,andpreventcorrosion[1].Forpracticalreasons,
designersofinternalcombustionenginesrecommendchangingengineoilsafteraspecifiedperiodof
operation,eventhoughtheyknowworkingconditionsandfailuresofidenticalenginessystemsare
notthesame.Forinstance,twoidenticalenginesmayevenbeoperatingunderthesameloadand
externalconditions,butthesecondenginehasagasketfailureallowingcoolant(containingglycol)
toleakintotheengineandmixwiththeengineoil.Whilethelifeofthefirstenginewillprobablynot

Appl.Sci.2020,10,37382of9
bereducedbywaitingforthedesigner‐specifiedoilchange,thesecondenginemayrequireavery
expensivecompleterenewingprocessduetopoorlubricationandbearingfailurepriortothe
specifiedscheduleofoilexchange.Engineoiliscontaminatedbyglycolduetoleakagefroman
engine’scoolingsystemintothelubricatingsystem,typicallycausedbyadamagedgasketorseal[2].
Glycolmixeswithengineoilandreduceslubricity[3]duetoglycoloxidation[4]causingsludgein
theengineoilandvarnishdepositsonpistonringsandvalves[5],pluggedlubricantscreens[4],and
possibleengineseizure[6].Ageneral,acceptablelimitofglycolinengineoilhasnotbeendetermined
sofar.Booser[7]reportedthat150ppmofglycoldidnotdamageengineoil.Wang[8]foundthat500
ppmofglycolwasnotdetrimentaltoengineoil.However,anyglycolcontaminationofengineoilis
aconcernsinceitgenerallyoriginatesfromaleakandcontaminationgenerallyincreases.
Contaminationandagingofengineoilcanbedeterminedbyusingestablishedmethods,such
asASTMInternational(ASTM)standardsandmethods.Usedengineoilsamplesmaybecollected
andsenttooneofseverallabsthatwillanalyzetheoilusingASTMormodified‐ASTMprocedures
andprovideareportsummarizingthecontaminants.Whilesomeindustrialusersmakeregularuse
ofsuchservices,mostvehicleownersdonot,asitwouldbetoocumbersomeandtime‐consuming.
Engineusersworldwidewouldbenefitfromin‐situtechnologyforreal‐timeengineoilanalysisto
alertusersofcontaminantsoragedengineoil.Sinceglycolhaslowmolecularweightandvolatility
butahighpolarity,itisdifficulttodetectandquantifyinengineoil.Hyperspectral[9],
electromechanical[10],infraredspectroscopy[11],andresonatingsensors[12]wereassessedfor
potentialinin‐situsensorsystems.Thesesensorsaredesignedtomonitoroilconditionsintermsof
viscosity[13],acidity[14],antioxidants[15],andwaterorfreezingcontaminants[16].However,none
ofthesesensorscouldidentifyallthecontaminants,andmultiplesensorswererecommendedto
monitoroilconditions[17].Whilemanyrecentautomobileshavesomeindicatorofwhentochange
engineoil,thesensorsystemsareprimarilybasedonmileageratherthanthedirectsensingofoil
contaminants.Newtechnologiesshouldcontinuetobedesignedandevaluatedforpotentialfuture
in‐situengineoilmonitoring.Scientistscontinuetodevelopsensors,includingspectralanalysisand,
mostnotably,terahertzspectroscopy,whichhasbeenappliedtomanyapplications[18,19].
THz‐time‐domainspectroscopy(THz‐TDS)hasbeenusedforcharacterizingoranalyzing
petroleumproductsandcontaminants,including:severalgradesoflubricatingoils[20,21],several
lubricatinggreasetypes[22],solidifyingpointfordieselfuel[23],dieselandgasolinefuel[24],
differentgradesofgasolineengineoil[25],threedifferentlevelsofwatercontaminationsindiesel
engineoil[26],fourdifferentlevelsoffuelcontaminationsingasolineengineoil[27],fourdifferent
periodsofoxidationforgasolineengineoil[28],sixdifferentperiodsofoxidationfordieselengine
oil[29].PreviousstudieshavenotassessedtheabilityoftheTHz‐TDStodetectengineoil
contaminationwithglycol.
ThisresearchaimedtodeterminetheabilityofTHz‐TDStoidentifyfourdifferentlevelsofglycol
contaminationinpartspermillion(ppm)unitswith(0ppm,150ppm,300ppmand500ppm)in
engineoil(5W‐30)intendedfordieselandgasolineengines.WhileTHz‐TDSisnotenvisionedasan
in‐situsystem,itwasusedtoevaluatetheappropriate/optimumfrequencyrangesforglycol
concentrationdetermination.Afinalin‐situsystemcouldbeaminiaturized,cost‐effectiveelectronic
system,operatinginanarrowfrequencyrange.
2.MaterialsandMethods
2.1.THz‐TDSSpectrometer
TheTHztimedomainspectrometer(MenloSystemsGmbH,Martinsried,Germany)(Figure1)
consistedofanultrafastfiberlasersystemat1550nmwith~90femtosecondpulsedurationand100
MHzrepetitionrate.Thelasersignalwassplitintwo.Adelaystagecausedatemporaldelaybetween
thelasersignaldrivingtheTHzdetectorphotoconductorandthatdrivingtheTHzsource
photoconductor.BothphotoconductorswerefromtheFraunhoferHeinrichHertzInstitute,Berlin,
Germany.Thepeakdynamicrangewas78dBat0.34THzanddecreasedto37dBat2.5THz.An
encapsulatedchamberholdsthesamplebeingsubmittedtoTHz‐TDS.Aftereachsamplecellwas

Appl.Sci.2020,10,37383of9
fittedinthechamber,itwasfilledwithdrynitrogentominimizehumidity,removingwatervapor
featuresfromtheTerahertzsignal.Thetemperatureinthelabwascontrolledtobe22°C.Duetothe
pulsednatureofTHz‐TDS,thetemperaturechangeinthesamplewasconsideredtobenegligible.

Figure1.Illustrationof(a)THztime‐domainspectrometerand(b)THzpaththroughthe
contaminatedoilsamples.Theglycol‐contaminatedsampleswereshakenpriortoTHzmeasurements
inordertoobtainahomogeneousmixture.
2.2.SamplePreparation
A1LbottleoffullsyntheticSAE5W‐30gradeengineoil(TopTec4600,LiquiMolyGmbH,Ulm,
Germany;APISN/CFservicecategory)marketedforbothgasolineanddieselengineswaspurchased
fromalocalmarketinMannheim,Germany.Foursamplesofengineoilwerepreparedbypouring
50mLofthefreshoilintoseparate60mLamberBostonroundglassbottles(QorpakGLC‐01909,
FisherScientific,Waltham,MA,USA).A1.5Lcontainerofantifreeze(KühlerFrostschutzKonzentrat,
Gut&Günstig,Germany)containingethandiol(glycol)wasalsopurchasedfromalocalretailerin
Mannheim,Germany.Foreachoilsample,anappropriateoilvolumewasremovedbypipetteand
replacedwiththesamevolumeofglycolviapipettetocreatecontaminatedsampleswith0ppm,150
ppm,300ppm,and500ppmofglycol.Theremoved/replacedvolumeswere7.5μl,15μl,and25μl
for150ppm,300ppm,and500ppm,respectively.ThesamplebottleswereshippedtoTechnische
UniversitätDarmstadt(TUD),GermanyforTHz‐TDSanalysis.PriortoTHz‐TDSanalysis,all
sampleswereshakenbyhandfor60sandletstandfor24htoallowairbubblestodissipate,after
whichallthesamplesappearedhomogeneoustothehumaneye.
2.3.THzTime‐DomainSpectroscopy
Allmeasurementsweremadeusingacellwithtwo3‐mm‐thickpolyethylene(PE)windows
separatedbymetaljoints,resultinginapermissibleTHzpathlengththroughtheoilsampleof15.25
mmoftheprobevolume,basedonpriorresearch[27].Theopenapertureoftheprobevolumewas
35×35mm,largeenoughtoaccommodatetheTHzbeamwithoutanynoticeabletruncation.Atightly
enclosedexternalmetalframesurroundedthewindowstoensuretheintegrityofthewindowsand
maintainafixedcellsize.Beforemeasuringanysample,areferencespectrumoftheemptycellwas
recorded.Eachglycolconcentrationwasmeasuredfivetimes,andboththerefractiveindexandthe
absorptionwererecorded.Betweeneachmeasurement,thePEcellwascleanedanddriedtoremove
alltracesoftheprioroilsampleinthecell.Thisprocedurewasadoptedforalltreatmentsand
replicationsinthisexperiment.Divisionofthespectraobtainedfromtheoilmeasurementbythe
referencespectrumlargelyremovesanycontributionoftheplasticwindowsandspectralshapeand
responsivityofsourceandreceiver,respectively,resultinginastrongreductioninthemeasurement
error.Anerroranalysisofsuchameasurementsystemhasbeenperformedin[28].

Appl.Sci.2020,10,37384of9
2.4.DataAnalysisofTHz‐TDS
Thestudywascarriedoutusingacompletelyrandomizeddesign,withfourtreatmentsofglycol
contamination(0ppm,150ppm,300ppm,500ppm)into5W‐30engineoil,withfivereplicationsfor
eachtreatment(20experimentalunits).Descriptivestatisticsofmean,variance,standarddeviation,
and95%confidenceintervalwerecalculatedforrefractiveindexandabsorptioncoefficientforeach
THzfrequency.TheTHzfrequenciesreportedwerederivedfromFouriertransformcalculations.
One‐wayanalysisofvariance(ANOVA)wasusedtodetermineifthereweresignificantdifferences
amongthefourtreatmentsforrefractiveindexandabsorptioncoefficientateachfrequency.A
probability(p)of≤0.05wasusedtoindicatesignificantdifferences.Fisher’sleastsignificant
difference(LSD)wasusedtodeterminesignificantdifferencesbetweenmeansateachfrequency.
Linearregressionanalysiswasconductedtopredictthebestmodel,basedonthecoefficientof
determination(R2),fortheglycolcontaminationlevelinengineoilformultipleTHzfrequencies.
3.ResultsandDiscussion
Refractiveindex,aswellasabsorptioncoefficient,wereobtainedusingtheTHztimedomain
spectrometer.Therangeofthespectralcharacteristics—refractiveindexandabsorptioncoefficient—
was0.25to2.5THz,witharesolutionofapproximately7GHzor615frequencies.
3.1.RefractiveIndex
Figure2illustratestherefractiveindicesoftheglycolcontamination—fourlevels—intheglycol
engineoil.From0.25to2.5THz,themeanrefractiveindexrangestartedat1.467(1)(0.25THzfor500
ppm)andendedat1.464(1)(2.5THzfor0ppm).Therefractiveindexdecreasedwithanincreasein
terahertzfrequency,whichwassimilartopreviousresearchwithotheroils[20,22,25].While
refractiveindexwasmeasuredatall615frequencies,Table1detaileddataforrefractiveindicesat
0.25THzintervals.Wenotethattheseerrors(standarddeviations)donotincludeanysystematic
errorscausedbytheTHzsystem,butrepresentrelativeerrorsbetweensuccessivemeasurements.
Thedirectionofchange(increaseinrefractiveindexwithincreaseinglycolcontaminationinthe
engineoil)wassimilartotheeffectofincreasedoxidation[28,29]orwatercontamination[26],butthe
oppositeofincreasedgasolinefuelcontamination[27].Therewasarelativelyconsistentspacing
amongthefourcurvesoftherefractiveindexalongthe0.25to2.5THzrange.

Figure2.Meanrefractiveindexofeachoffourglycolcontaminationlevelsof5W‐30gasoline/diesel
engineoilwith95%confidenceintervals(forclaritywhenhidden,95%confidenceintervalsarethe
sameaboveandbelowthemean).
Thehighestrefractiveindexwasobtainedatthelevelof500ppmglycolcontamination;the
lowestrefractiveindexwasobtainedatthelevelof0ppmglycolcontamination.Figure2

Appl.Sci.2020,10,37385of9
demonstratedtheoverlapin95%confidenceintervalsbetweentwopairsofglycolcontamination
levels,creatingonegroupingof0ppmand150ppmandasecondgroupingof300ppmand500ppm.
Ingeneral,thisconfidenceintervalwasconsistentacrossthefrequencyrange.Thecurveshapes
showedadifferenceamongthefourlevelsofglycolcontamination.Fromthefrequency0.25–2.5THz,
theone‐wayANOVApresentedhighlysignificantdifferences(p<0.01)intherefractiveindexamong
thelevelsofglycolcontamination.BasedonLSD,theglycolcontaminationconcentrationsof300ppm
and500ppmweresignificantlydifferentfromtheconcentrationsof0ppmand150ppmglycol
contaminationforfrequenciesfrom0.25to2.5THz.Theuncontaminatedsample(0ppm)andthe500
ppmcontaminationwereclearlydiscernible.AlthoughWang[8]foundthat500ppmwasnot
detrimental,thisfindingclearlyshowsthatTHz‐TDSiscapableofdeterminingsignificant
contaminationof300ppmandbeyond,atleastunderlaboratoryconditions.
Table1.Partialrefractiveindexandabsorptioncoefficientmeanandstandarddeviationdatafrom
glycolcontaminatedengineoilsamplesat0.25THzintervalsacrossthe0.253–2.5THzrange.
THzRefractiveIndex*AbsorptionCoefficient(cm‐1)*
0ppm150ppm300ppm500ppm0ppm150ppm300ppm500ppm
0.251.4662b
±0.0001
1.4663b
±0.0001
1.4669a
±0.0003
1.4671a
±0.0004
0.148a
±0.009
0.168a
±0.028
0.162a
±0.008
0.171a
±0.017
0.501.4657b
±0.0001
1.4659b
±0.0001
1.4664a
±0.0003
1.4666a
±0.0003
0.286c
±0.012
0.310bc
±0.026
0.319ab
±0.015
0.343a
±0.029
0.751.4654b
±0.0001
1.4656b
±0.0001
1.4661a
±0.0003
1.4662a
±0.0003
0.477c
±0.009
0.521b
±0.029
0.544ab
±0.014
0.579a
±0.040
1.001.4652b
±0.0001
1.4653b
±0.0001
1.4658a
±0.0003
1.4659a
±0.0003
0.690c
±0.011
0.755b
±0.034b
0.796b
±0.010
0.845a
±0.049
1.251.4649b
±0.0001
1.4651b
±0.0001
1.4655a
±0.0002
1.4656a
±0.0003
0.918d
±0.020
1.005c
±0.043
1.068b
±0.008
1.130a
±0.056
1.501.4647b
±0.0001
1.4648b
±0.0001
1.4652a
±0.0002
1.4653a
±0.0002
1.149d
±0.031
1.261c
±0.052
1.345b
±0.010
1.418a
±0.066
1.751.4645b
±0.0001
1.4647b
±0.0001
1.4650a
±0.0002
1.4651a
±0.0002
1.387d
±0.038
1.524c
±0.060
1.625b
±0.011
1.712a
±0.075
2.001.4644b
±0.0001
1.4645b
±0.0001
1.4648a
±0.0002
1.4649a
±0.0002
1.631d
±0.047
1.800c
±0.069
1.918b
±0.021
2.021a
±0.097
2.251.4642b
±0.0001
1.4643b
±0.0001
1.4646a
±0.0002
1.4647a
±0.0002
1.936d
±0.066
2.142c
±0.090
2.266b
±0.041
2.392a
±0.117
2.501.4641b
±0.0001
1.4642b
±0.0001
1.4645a
±0.0003
1.4645a
±0.0002
2.209c
±0.098
2.457b
±0.117
2.584b
±0.058
2.743a
±0.143
*Means,forthesameparameteratthesamefrequency,withthesamesuperscriptletterarenot
significantlydifferent.
3.2.AbsorptionCoefficient
Figure3showstheabsorptioncoefficientsofthefourlevelsofglycolcontaminationintheSAE
5W‐30engineoil.Themeanabsorptioncoefficientforallglycollevelsincreasedwithfrequency,
startingatalowof0.148cm−1for0ppmat0.25THzandreachingahighof2.743cm−1for500ppmat
2.5THz.Ateachfrequency,theabsorptioncoefficientincreasedwithanincreaseinglycol
contamination,withtheexceptionthatbelow0.35THz150ppm,therewasaslightlyhigher
absorptioncoefficientthan300ppm.Differencesamongthecurvesoftheabsorptioncoefficients
increasedwithfrequency,withthedifferencebetween0ppmand500ppmincreasingfrom0.023
cm−1at0.25THzto0.534cm−1at2.5THz.Whiletheabsorptioncoefficientwasmeasuredatall615
frequencies,Table1detaileddataforabsorptioncoefficientsat0.25THzintervals.Theincreasein
absorptioncoefficientwithanincreaseinfrequencywassimilartopreviousresearchwithotheroils
[20,22,25].Theincreaseinabsorptioncoefficientwithgreatercontaminationlevelswassimilartothe
effectofwatercontamination[26]andgasolinefuelcontamination[27],andwasexpected,asallthese
specieswereorcontainedpolarmoleculeswithstrongTHzabsorption.

Appl.Sci.2020,10,37386of9

Figure3.Meanabsorptioncoefficientofeachoffourglycolcontaminationlevelsof5W‐30
gasoline/dieselengineoilwith95%confidenceintervals(forclaritywhenhidden,95%confidence
intervalsarethesameaboveandbelowthemean).
The95%confidenceintervalsincreasedgraduallywithfrequency,butthedifferenceamong
meansincreasedatagreaterrate.Whiletherewasoverlapamongthe95%confidenceintervalsat
0.25THz,theoverlapdecreasedasfrequencyincreased.BasedonANOVA,thereweresignificant
differences(p<0.05)amongcontaminationlevelsofglycolacrossthe0.355–2.5THzrange,with
highlysignificantdifferences(p<0.01)acrossthe0.476–2.5THzrange.BasedonLSDresults,theonly
statisticallysignificantdifferenceamongmeansfrom0.355–0.428THzwasthat500ppmhada
significantlyhigherabsorptioncoefficientthan0ppm.Commencingwith0.432THz,300ppmwas
alsosignificantlydifferentfrom0ppm.Beginningat0.458THz,150ppmwasalsosignificantly
differentfrom500ppm.Startingat0.604THz,150ppmwasalsosignificantlydifferentfrom0ppm.
Beginningwith0.754THz,500ppmwassignificantlydifferentfromeachoftheothercontamination
levels.From1.01–2.471THz,eachcontaminationlevelwassignificantlydifferentfromeachother.
Thus,therangeof1.01–2.471THzwasfoundtobeabletodiscriminatealllevelsofglycol
contamination.AlthoughWangetal.[8]stated500ppmwasnotdetrimentalinengineoil,thisstudy
clearlyshowedthatTHzabsorptioncoefficientwasadecentdiscriminatorforglycolcontamination,
bydiscriminatingbetweeneachofthecontaminationlevels.ThesensitivityofTHzat150ppmwas
betterthanthatBorinandPoppi[16]foundwithamid‐infraredspectroscopymodeldevelopedfrom
glycol‐contaminatedoilsamplesof14–9990ppmandthenvalidatedwithglycolcontamination
differencesaslowas800ppm.ASTMstandardpracticespecifies1000ppmasthesensitivityof
Fourier‐transforminfraredspectroscopy(FTIR)[6].
3.3.RegressionModels
Absorptioncoefficientvaluesintherangeof1.0–2.5THzin0.25THzincrementswereselected
forregressionanalysistobeclosetothe1.014–2.471THzrange,forwhicheachglycolcontamination
levelwassignificantlydifferentfromeachoftheothers.Thelinearregressionprovidedatentative
relationshipofabsorptioncoefficienttoglycolcontaminationbasedonthisstudy.Thecoefficientof
determination(R2)valuesforthesevenmodelsstartedat0.79at1.0THz,increasedtoapeakof0.84
at1.75THz,anddroppedto0.77at2.5THz.Slopevaluesrangedfromahighof2613ppm×cmat1.0
THz,anddecreasedwitheachfrequencyincreasetoalowof746ppm×cmat2.5THz.X‐intercepts
increasedwithfrequencyfrom0.68cm−1at1.0THzto1.92cm−1at2.5THz.Thelinearmodelat1.75
THzwasthebestmodelbasedonR2(Figure4).Althoughtherangeofabsorptioncoefficientvalues
forsomecontaminationlevelswaswiderthantheothers,noneoftheabsorptioncoefficientranges
amongtheglycolcontaminationlevelsoverlapped,andtheabsorptioncoefficientmeansofeach
contaminationlevelweresignificantlydifferentfromallothercontaminationlevels.

Appl.Sci.2020,10,37387of9

Figure4.Absorptioncoefficientat1.75THzpredictingglycolcontamination(ppm)withalinear
model.
4.Conclusions
Basedonthisexperiment,therefractiveindexfromTHz‐TDSidentifiedglycolcontamination
levelsof300+ppmfromuncontaminatedoil.Whilenoglycolfromcoolantshouldbepresentin
engineoil[7],the300ppmcontaminationlevelwasbelowthe500levelthatwasnotfoundtocause
enginedamage[8].Thus,acriticalleakofenginecoolant—includingglycolintoengineoil—couldbe
detectedbyTHz‐TDSbeforeenginedamageandsignaltheneedtorejecttheengineoilandperform
coolingsystemrepairs.Thedifferenceinmeanrefractiveindexwasgreatestatthelowestfrequency,
butstillnearthelimitofdetectionforTHz‐TDS.WhileaTHz‐TDSsystemmayreliablydetectthis
difference,eachTHz‐TDSsystemmayresultinaslightlydifferentrangeofrefractiveindexvalues
andmayrequirecalibrationtodetermineanengineoilrejectlevelforglycolcontaminationbasedon
refractiveindex.
TheabsorptioncoefficientresultsfromTHz‐TDSshowedbetterpotentialfordistinguishing
differencesinglycolcontaminationlevels(0ppm,150ppm,300ppm,and500ppm)of5W‐30
gasoline/dieselengineoil.Acrossthefrequencyrangeof1.01–2.471THz,eachofthefour
contaminationlevelswassignificantlydifferentfromtheotherthreelevels.Themeandifferences
betweencontaminationlevelswereallover0.1cm−1andwellwithinthelimitsofdetectionfor
absorptioncoefficient.Basedontheresultsofthisstudy,THz‐TDSdemonstratedsensitivityto
identifyglycolcontaminationinengineoil.Whilethisindicatespotential,asubsequentstudywould
berequiredtodeterminetheabilityofTHz‐TDStodiscriminatelevelsofcoolant(glycolandwater
mixtures)contaminationinengineoil.Furtherstudieswillrequireinvestigationofthediscrimination
ofdifferentcontaminantsatthesametime.Asnoresonantglycol‐specificfeatureswerefound,afinal
sensorsystemfordetectingmultiplecontaminantsatthesametimemayrequiredifferentmethods—
possiblyamulti‐spectralsystem.ThispaperhasshownthatacompactifiedTHzsensingelement
couldbepartofsuchasensorconcept.
AuthorContributions:Conceptualization,O.M.A.,A.M.A.‐M.andD.G.W.;methodology,O.M.A,A.M.A.‐M.
andM.M.A.;formalanalysis,O.M.A.,A.M.A.‐M.andD.G.W.;resources,O.M.A.andS.P.;writing—original
draftpreparation,O.M.A.andA.M.A.‐M.;writing—reviewingandediting,S.P.andD.G.W.Allauthorshave
readandagreedtothepublishedversionofthemanuscript.
Funding:Thisresearchreceivednoexternalfunding.
Acknowledgments:TheauthorswouldliketoextendtheirthankstothephysicsfacultyattheTechnische
UniversitätDarmstadt(TUD),Germany,wheretheTHz‐TDSworkwasperformedinthelaboratoryofProf.Dr.
SaschaPreu.

Appl.Sci.2020,10,37388of9
ConflictsofInterest:Theauthorsdeclarenoconflictofinterest.
References
1. Mackney,D.;Clague,N.;Brown,G.;Fish,G.;Durham,J.AutomotiveLubricants.InAutomotiveLubricants
andTesting;Tung,S.C.,Totten,G.E.,Eds.;ASTM:WestConshohocken,PA,USA,2012;pp.23‐46.
2. Lansdown,A.R.LubricationandLubricantSelectionAPracticalGuide,3rded.;ProfessionalEngineering
Publishing:London,UK,2004.
3. Liddell,G.J.;Marshall,R.ATestforAntifreezeContaminationinCrankcaseOil;SAETechnicalPaper,(No.
520054);SAE:Warrendale,PA,USA,1952;pp.1‐8.
4. Deere&Company.Fuels,Lubricants,Coolants,andFilters;Deere&Company:Moline,IL,USA,2008.
5. Bonner,D.W.TheProblemofCoolantContamination.Ind.Lubr.Tribol.1974,26,133–135.
6. ASTM.ASTMD2982‐07StandardTestMethodsforDetectingGlycol‐BaseAntifreezeinUsedLubricatingOils;
ASTM:WestConshohocken,PA,USA,2013;doi:10.1520/D2982‐07R13.
7. Booser,E.R.CRCHandbookofLubricationandTribology,VolumeIII:Monitoring,Materials,SyntheticLubricants,
andApplications,VolumeIII;CRCPress:BocaRaton,FL,USA,1993.
8. Wang,S.S.;Lee,H.S.Theapplicationofacimpedancetechniquefordetectingglycolcontaminationin
engineoil.Sens.ActuatorsBChem.1997,40,193–197,doi:10.1016/S0925‐4005(97)80261‐4.
9. Lulu,J.;Xinjie,Y.;Yong,H.Identificationofautomobiletransmissionfluidusinghyperspectralimaging
technology.Int.J.Agric.Biol.Eng.2014,7,81–85,doi:10.3965/j.ijabe.20140704.009.
10. Soleimani,M.;Sophocleous,M.;Glanc,M.;Atkinson,J.;Wang,L.;Wood,R.J.K.;Taylor,R.I.Engineoil
aciditydetectionusingsolidstateionselectiveelectrodes.Tribol.Int.2013,65,48–56,
doi:10.1016/j.triboint.2013.02.030.
11. Kasberger,J.;Fromherz,T.;Saeed,A.;Jakoby,B.MiniaturizedintegratedevanescentfieldIR‐absorption
sensor:Designandexperimentalverificationwithdeterioratedlubricationoil.Vib.Spectrosc.2011,56,129–
135,doi:10.1016/j.vibspec.2011.01.003.
12. Mohammed,A.J.;Hassan,M.A.Piezesensorsformonitoringdegradationofautomotiveengineoil.Sci.
Tech.2013,3,42–45,doi:10.5923/j.scit.20130301.05.
13. Jakoby,B.;Scherer,M.;Buskies,M.;Eisenschmid,H.Anautomotiveengineoilviscositysensor.IEEESens.
J.2003,3,562–568,doi:10.1109/JSEN.2003.817164.
14. Agoston,A.;Ötsch,C.;Jakoby,B.Viscositysensorsforengineoilconditionmonitoring‐Applicationand
interpretationofresults.Sens.ActuatorsAPhys.2005,121,327–332,doi:10.1016/j.sna.2005.02.024.
15. Price,R.J.;Clarke,L.J.Chemicalsensingofamineantioxidantsinturbinelubricants.Analyst1991,116,1121–
1123,doi:10.1039/AN9911601121.
16. Borin,A.;Poppi,R.J.ApplicationofmidinfraredspectroscopyandiPLSforthequantificationof
contaminantsinlubricatingoil.Vib.Spectrosc.2005,37,27–32,doi:10.1016/j.vibspec.2004.05.003.
17. Schmitigal,J.;Moyer,S.EvaluationofSensorsforOn‐BoardDieselOilConditionMonitoringofU.S.Army
GroundEquipment;SAE:Warrendale,PA,USA,2005;SAETechnicalPaper2005‐01‐1810.
18. Jepsen,P.U.;Cooke,D.G.;Koch,M.Terahertzspectroscopyandimaging–Moderntechniquesand
applications.LaserPhotonicsRev.2011,5,124–166,doi10.1002/lpor.201000011.
19. Mathanker,S.K.;Weckler,P.R.;Wang,N.Terahertz(THz)applicationsinfoodandagriculture:Areview.
Trans.ASABE2013,56,1213–1226,doi:10.13031/trans.56.9390.
20. Naftaly,M.;Foulds,A.P.;Miles,R.E.;Davies,A.G.Terahertztransmissionspectroscopyofnonpolar
materialsandrelationshipwithcompositionandproperties.Int.J.InfraredMillim.Waves2005,26,55–64,
doi:10.1007/s10762‐004‐2033‐6.
21. Tian,L.;Zhou,Q.;Jin,B.;Zhao,K.;Zhao,S.;Shi,Y.;Zhang,C.Opticalpropertyandspectroscopystudies
ontheselectedlubricatingoilintheterahertzrange.Sci.ChinaSer.G:Phys.Mech.Astron.2009,52,1938–
1943,doi:10.1007/s11433‐009‐0310‐z.
22. Lu,T.;Qing‐Li,Z.;Kun,Z.;Yu‐Lei,S.;Dong‐Mei,Z.;Song‐Qing,Z.;Cun‐Lin,Z.Consistency‐dependent
opticalpropertiesoflubricatinggreasestudiedbyterahertzspectroscopy.Chin.Phys.B2011,20,010703.
23. Zhao,H.;Zhao,K.;Bao,R.Predictingcoldflowpropertiesofdieselbyterahertztime‐domainspectroscopy.
ISRNSpectrosc.2012,1–4,doi:10.5402/2012/876718.
24. Arik,E.;Altan,H.;Esenturk,O.Dielectricpropertiesofdieselandgasolinebyterahertzspectroscopy.J.
InfraredMillim.TerahertzWaves2014,35,759–769,doi:10.1007/s10762‐014‐0081‐0.

Appl.Sci.2020,10,37389of9
25. Adbul‐Munaim,A.M.;Reuter,M.;Koch,M.;Watson,D.G.Distinguishinggasolineengineoilsofdifferent
viscositiesusingterahertztime‐domainspectroscopy.J.InfraredMillim.TerahertzWaves2015,36,687–696,
doi:10.1007/s10762‐015‐0164‐6.
26. Abdul‐Munaim,A.M.;Reuter,M.;Abdulmunem,O.M.;Balzer,J.C.;Koch,M.;Watson,D.G.Using
terahertztime‐domainspectroscopytodiscriminateamongwatercontaminationlevelsindieselengineoil.
Trans.ASABE2016,59,795–801,doi:10.13031/aim.20152162399.
27. Abdul‐Munaim,A.M.;Aller,M.M.;Preu,S.;Watson,D.G.Discriminatinggasolinefuelcontaminationin
engineoilbyterahertztime‐domainspectroscopy.Tribol.Int.2018,119,123–130,
doi:10.1016/j.triboint.2017.10.026.
28. MéndezAller,M.;Abdul‐Munaim,A.M.;Watson,D.G.;Preu,S.ErrorSourcesandDistinctnessofMaterials
ParametersObtainedbyTHz‐TimeDomainSpectroscopyUsinganExampleofOxidizedEngineOil.
Sensors2018,18,2087,doi:10.3390/s18072087.
29. Abdul‐Munaim,A.M.;Ornik,J.;Koch,M.;Watson,D.G.TerahertzTimeDomainSpectroscopytoDetect
DifferentOxidationLevelsofDieselEngineOil.Lubricants2019,7,18,doi:10.3390/lubricants7020018.


©2020bytheauthors.LicenseeMDPI,Basel,Switzerland.Thisarticleisanopenaccess
articledistributedunderthetermsandconditionsoftheCreativeCommonsAttribution
(CCBY)license(http://creativecommons.org/licenses/by/4.0/).
