<%BANNER%>

Optimization of La2CuO4 Sensing Electrodes for a NOx Potentiometric Sensor

Permanent Link: http://ufdc.ufl.edu/UFE0042139/00001

Material Information

Title: Optimization of La2CuO4 Sensing Electrodes for a NOx Potentiometric Sensor
Physical Description: 1 online resource (131 p.)
Language: english
Creator: MACAM,ERIC R
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: CERAMICS -- ELECTROCHEMISTRY -- GEOMETRY -- MATERIALS -- MICROSTRUCTURE -- NOX -- PROCESSING -- SENSOR
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Gas sensors show great potential in the effort to reduce the amount of emissions of harmful by-products of combustion and to maintain efficient operation of these systems. There is an obvious need for these devices with the current social and political climate that points towards a cleaner and less consumptive culture. Although sensors for carbon monoxide, nitrogen oxides, hydrogen, etc. have been studied, they have yet to reach the level of commercial success and widespread use as the oxygen lambda sensor. The metric of success in gas sensor development must include developing devices that are sensitive, selective, and stable. Sensitivity refers to the sensor?s ability to detect a given amount of gas (often in the parts per million range) and produce a large enough signal in response. A selective sensor is one that is able to detect a given gas without being affected by other gases. Stability is a parameter that should include the properties of accuracy, precision, and repeatability. In addition, an understanding of the sensor mechanism is of utmost importance in terms of fundamental and practical application and can help with transitioning this technology to commercialization. This work examined the geometric properties of La2CuO4 sensing electrodes and their influence on the performance of a ceramic based NOx potentiometric sensor. These properties, which include electrode thickness, area, and configuration also helped illustrate and explain the different contributions of the proposed sensing mechanism ?Differential Electrode Equilibria?. This comprehensive mechanism includes contributions due to Mixed Potential, heterogeneous catalysis, and changes in semiconductive properties due to adsorption of gas species on the electrode surface. Actual sensor devices (with the varying sensing electrode geometric properties) were successfully produced and tested in simulated gas environments to assess sensor performance and understand sensing mechanism. Impedance spectroscopy and Kelvin probe microscopy were also used to further relate the semiconductive properties of La2CuO4 to actual sensor performance. The results of this work help further explain "Differential Electrode Equilibria? as well as provide design parameters in order to make sensors that meet the requirements of sensitivity, selectivity, and stability.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by ERIC R MACAM.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Wachsman, Eric D.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0042139:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042139/00001

Material Information

Title: Optimization of La2CuO4 Sensing Electrodes for a NOx Potentiometric Sensor
Physical Description: 1 online resource (131 p.)
Language: english
Creator: MACAM,ERIC R
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: CERAMICS -- ELECTROCHEMISTRY -- GEOMETRY -- MATERIALS -- MICROSTRUCTURE -- NOX -- PROCESSING -- SENSOR
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Gas sensors show great potential in the effort to reduce the amount of emissions of harmful by-products of combustion and to maintain efficient operation of these systems. There is an obvious need for these devices with the current social and political climate that points towards a cleaner and less consumptive culture. Although sensors for carbon monoxide, nitrogen oxides, hydrogen, etc. have been studied, they have yet to reach the level of commercial success and widespread use as the oxygen lambda sensor. The metric of success in gas sensor development must include developing devices that are sensitive, selective, and stable. Sensitivity refers to the sensor?s ability to detect a given amount of gas (often in the parts per million range) and produce a large enough signal in response. A selective sensor is one that is able to detect a given gas without being affected by other gases. Stability is a parameter that should include the properties of accuracy, precision, and repeatability. In addition, an understanding of the sensor mechanism is of utmost importance in terms of fundamental and practical application and can help with transitioning this technology to commercialization. This work examined the geometric properties of La2CuO4 sensing electrodes and their influence on the performance of a ceramic based NOx potentiometric sensor. These properties, which include electrode thickness, area, and configuration also helped illustrate and explain the different contributions of the proposed sensing mechanism ?Differential Electrode Equilibria?. This comprehensive mechanism includes contributions due to Mixed Potential, heterogeneous catalysis, and changes in semiconductive properties due to adsorption of gas species on the electrode surface. Actual sensor devices (with the varying sensing electrode geometric properties) were successfully produced and tested in simulated gas environments to assess sensor performance and understand sensing mechanism. Impedance spectroscopy and Kelvin probe microscopy were also used to further relate the semiconductive properties of La2CuO4 to actual sensor performance. The results of this work help further explain "Differential Electrode Equilibria? as well as provide design parameters in order to make sensors that meet the requirements of sensitivity, selectivity, and stability.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by ERIC R MACAM.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Wachsman, Eric D.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0042139:00001


This item has the following downloads:


Full Text

PAGE 1

OPTIMIZATIONOFLA 2 CUO 4 SENSINGELECTRODESFORANO X POTENTIOMETRICSENSOR By ERICR.MACAM ADISSERTATIONPRESENTEDTOTHEGRADUATESCHOOL OFTHEUNIVERSITYOFFLORIDAINPARTIALFULFILLMENT OFTHEREQUIREMENTSFORTHEDEGREEOF DOCTOROFPHILOSOPHY UNIVERSITYOFFLORIDA 2011

PAGE 2

c r 2011EricR.Macam 2

PAGE 3

Tomyparents,RenatoandVioletaMacam Withouttheirsacrices,Iwouldnotevenhavetheopportuni tytobewhereIamtoday.I thankthemfortheirinspiration,support,andcomfortwhen everIneededit 3

PAGE 4

ACKNOWLEDGMENTS Iwouldrstliketothankmyparents,RenatoandVioleta,and mysisterMichelle forbeingsupportiveandlovingthroughoutgraduateschool .ThankstoalltheteachersI havehadateverylevelschooling,whoallplayedaverylarge partoffosteringlearning inallaspectsoflife.IwouldalsoliketothankProfessorEr icWachsmanfornotonly beingagreatadvisor,butforalsobeingagreatmentor.Iwou ldhaveneverdecided togotograduateschoolwithouthisencouragementandhisbe liefinmewhenIwas stillanundergraduate.ThankstoProfessorEnricoTravers a,whoalongwithProfessor Wachsmangavemetheopportunitytospendayearofmygraduat estudiesatthe Universit adiRoma“TorVergata”.Thanksalsotoallmycommitteemembe rs.Iwouldbe remisstonotthankeveryoneatmyresearchgroupsattheUniv ersityofFloridaandthe Universit adiRoma,especiallythoseinRome,Italywhohelpedmetrans itiontolivingin anewcountry.Nobodyelsecouldfullyunderstandthisjourn eycalledgraduateschool. ThankstoallmyfriendsandcommunityinGainesville,FLand Rome,Italy,especiallyall ofmyroommates,whogavemethepeaceofmindtolovelifeouts ideofschool.Lastly, andmostimportantly,IwouldliketothankJesusChrist,mys avior,fornothingwouldbe possiblewithoutHim. 4

PAGE 5

TABLEOFCONTENTS page ACKNOWLEDGMENTS .................................. 4 LISTOFTABLES ...................................... 8 LISTOFFIGURES ..................................... 9 LISTOFABBREVIATIONS ................................ 13 ABSTRACT ......................................... 14 CHAPTER 1BACKGROUNDINFORMATION .......................... 16 1.1Introduction ................................... 16 1.2Signicance ................................... 16 1.2.1EnergyEfciency ............................ 16 1.2.2AirPollution ............................... 17 1.2.2.1Gaschemistry ........................ 17 1.2.2.2Pollutionformation ...................... 17 1.2.2.3Environmentalaspects ................... 18 1.2.2.4Healthaspects ........................ 18 1.3LiteratureBackground ............................. 19 1.3.1SensorTechnologyandDevelopment ................ 19 1.3.1.1Solidelectrolytesensors .................. 19 1.3.1.2Catalyticsensors ....................... 20 1.3.1.3Semiconductingsensors .................. 21 1.3.2SensorMechanisms .......................... 22 1.3.2.1Mixedpotentialtheory .................... 23 1.3.2.2DifferentialElectrodeEquilibriaTheory .......... 24 1.4Goals ...................................... 25 2EXPERIMENTALMETHODS ............................ 31 2.1MaterialsUsed ................................. 31 2.1.1LanthanumCopperOxide(La 2 CuO 4 ) ................. 31 2.1.2Yttria-StabilizedZirconia(YSZ) .................... 32 2.1.3Platinum ................................. 33 2.2TechniquesUsed ................................ 33 2.2.1PowderSynthesis ............................ 33 2.2.2TapeCasting .............................. 34 2.2.3ScreenPrinting ............................. 35 2.2.4EMFMeasurement ........................... 36 2.2.5ImpedanceSpectroscopy ....................... 36 5

PAGE 6

3ELECTRODETHICKNESS ............................. 40 3.1Introduction ................................... 40 3.2Experimental .................................. 40 3.2.1SensorFabrication ........................... 40 3.2.2TestingParameters ........................... 41 3.3Results ..................................... 42 3.3.1FE-SEMThicknessMeasurements .................. 42 3.3.2SensorResponse ............................ 42 3.4Analysis ..................................... 44 3.4.1SensorMechanism ........................... 44 3.4.1.1HeterogeneousCatalysis .................. 44 3.4.1.2Electrocatalysis ....................... 45 3.4.1.3SemiconductingEffect .................... 46 3.4.2Discussion ................................ 47 3.5Conclusions ................................... 49 4ELECTRODEAREA ................................. 65 4.1Introduction ................................... 65 4.2Experimental .................................. 65 4.2.1PowderSynthesis ............................ 65 4.2.2SensorFabrication ........................... 66 4.2.3TestingParameters ........................... 66 4.2.4SensorCharacterization ........................ 67 4.3Results ..................................... 67 4.3.1SensorCharacterization ........................ 67 4.3.2PotentiometricTests .......................... 68 4.3.3ImpedanceTesting ........................... 70 4.3.3.1Highfrequencyimpedance ................. 71 4.3.3.2Totalelectrodeimpedance ................. 72 4.4Analysis ..................................... 73 4.5Conclusions ................................... 75 5ELECTRODECONFIGURATION .......................... 87 5.1Introduction ................................... 87 5.2Experimental .................................. 87 5.2.1PowderSynthesis ............................ 87 5.2.2SensorFabrication ........................... 87 5.2.2.1Conguration1 ........................ 88 5.2.2.2Conguration2 ........................ 88 5.2.2.3Conguration3 ........................ 88 5.2.2.4Conguration4 ........................ 88 5.2.2.5Conguration5 ........................ 89 5.2.3TestingParameters ........................... 89 6

PAGE 7

5.3ResultsandDiscussion ............................ 90 5.3.1SensorResponse ............................ 90 5.3.1.1Sensitivity ........................... 90 5.3.1.2Selectivity/crosssensitivity ................. 91 5.3.1.3Repeatability ......................... 92 5.3.2SensorMechanism ........................... 93 5.4Conclusions ................................... 96 6CONCLUSION .................................... 104 APPENDIX AADDITIONALDATAFORCHAPTER4 ....................... 108 BADDITIONALDATAFORCHAPTER5 ....................... 111 CKELVINPROBEMETHOD ............................. 121 C.1Introduction ................................... 121 C.2ExperimentalProcedure ............................ 122 C.3ResultsandDiscussion ............................ 122 REFERENCES ....................................... 124 BIOGRAPHICALSKETCH ................................ 130 7

PAGE 8

LISTOFTABLES Table page 3-1Electrodethicknessmeasurements ......................... 50 4-1Electrodethicknessmeasurementsforpotentiometrics ensorelectrodes .... 76 4-2Electrodethicknessmeasurementsforimpedancesample s ........... 76 C-1 CPD NOx ....................................... 123 8

PAGE 9

LISTOFFIGURES Figure page 1-1Bondstructureofnitricoxide. ............................ 26 1-2Molecularstructureofnitricoxide. ......................... 26 1-3Bondstructureofnitrogendioxide. ......................... 27 1-4Molecularstructureofnitrogendioxide. ...................... 27 1-5Bondstructureofcarbonmonoxide. ........................ 27 1-6Molecularstructureofcarbonmonoxide. ...................... 27 1-7Schematicofa)anoxygensensormountedinanexhaustpip e[1]andb)cross sectionofanautomotivesensorwithametallicshroud[2]. ............ 28 1-8DefectEquilibriumDiagram ............................. 29 1-9FermiLevel ...................................... 29 1-10SensitivityofSnO 2 sensorsversusconcentrationofH 2 andCO[3]. ....... 30 1-11Current-voltageplotforNOandNO 2 reactionsonaWO 3 sensor[4]. ...... 30 2-1ThreedimensionalstructuralarrangementofLa 2 CuO 4 modiedperovskite crystalstructure.Onetetragonalunitcellisrepresented [5]. ........... 38 2-2YSZcrystalstructure ................................. 38 2-3XRDplotofLa 2 CuO 4 ................................ 39 2-4Nyquistplot ...................................... 39 3-1Schematicofsensorwithx=10to30 m ..................... 50 3-2FESEMcrosssectionsofLa 2 CuO 4 sensingelectrodes .............. 51 3-3NO 2 voltageresponseversustimeat550 C .................... 51 3-4NO 2 Voltageresponseversustimeat600 C .................... 52 3-5NO 2 voltageresponseversustimeat650 C .................... 52 3-6NOvoltageresponseversustimeat400 C .................... 53 3-7NOvoltageresponseversustimeat450 C .................... 53 3-8NOvoltageresponseversustimeat500 C .................... 54 3-9NOvoltageresponseversustimeat550 C .................... 54 9

PAGE 10

3-10NOvoltageresponseversustimeat600 C .................... 55 3-11Voltageversuslog(NO 2 Concentration)forallthicknessesat550 C ....... 56 3-12Voltageversuslog(NO 2 Concentration)forallthicknessesat600 C ....... 57 3-13Voltageversuslog(NO 2 Concentration)forallthicknessesat650 C ....... 57 3-14Voltageversuslog(NOConcentration)forallthicknes sesat400 C ....... 58 3-15Voltageversuslog(NOConcentration)forallthicknes sesat450 C ....... 58 3-16Voltageversuslog(NOConcentration)forallthicknes sesat500 C ....... 59 3-17Voltageversuslog(NOConcentration)forallthicknes sesat550 C ....... 59 3-18Voltageversuslog(NOConcentration)forallthicknes sesat600 C ....... 60 3-19NO 2 sensitivityversuselectrodethickness ..................... 60 3-20NOsensitivityversuselectrodethickness ..................... 61 3-21NO 2 TPRoverLa 2 CuO 4 [6] ............................. 62 3-22NO 2 TPDoverLa 2 CuO 4 [6] ............................. 63 3-23NOTPDoverLa 2 CuO 4 [6] .............................. 64 4-1Schematicofsensorwithx=8,6,4,2mm .................... 76 4-2FE-SEMcrosssectionsofLa 2 CuO 4 sensingelectrodesforpotentiometricsensors: a)4mm 2 x5.41 mb)16mm 2 x5.62 mc)36mm 2 x5.08 mandd)64mm 2 x11.40 m ...................................... 77 4-3FE-SEMcrosssectionsofLa 2 CuO 4 electrodesforimpedancetesting:a)4 mm 2 x5.52 mb)16mm 2 x7.91 mc)36mm 2 x4.22 mandd)64mm 2 x 6.45 m ........................................ 78 4-4NOvoltageresponseversusTimeforallareasat550 C ............. 78 4-5Voltageversuslog(NOConcentration)forallareasat55 0 C ........... 79 4-6NOvoltageresponseversusTimeforallareasat500 C ............. 79 4-7Voltageversuslog(NOConcentration)forallareasat50 0 C ........... 80 4-8NO 2 sensitivitiesversusLa 2 CuO 4 electrodeareaat500,550,and600 C ... 80 4-9NOsensitivitiesversusLa 2 CuO 4 electrodeareaat400,450,and500 C .... 81 4-10NO 2 sensitivitiesversusLa 2 CuO 4 electrodevolumeat500,550,and600 C .. 81 4-11NOsensitivitiesversusLa 2 CuO 4 electrodevolumeat400,450and500 C ... 82 10

PAGE 11

4-12Impedancein3%O 2 ,200ppmNO 2 ,andNOat550 C .............. 83 4-13HighfrequencyImpedancein3%O 2 ,200ppmNO 2 ,andNOat550 C. .... 84 4-14NormalizedhighfrequencyNO 2 impedanceversusLa 2 CuO 4 versus500,550, and600 C ...................................... 85 4-15NormalizedhighfrequencyNOimpedanceversusLa 2 CuO 4 at400,450,and 500 C ......................................... 85 4-16NormalizedtotalNO 2 impedanceversusLa 2 CuO 4 at500,550,and600 C ... 86 4-17NormalizedtotalNOimpedanceversusLa 2 CuO 4 at450and500 C ...... 86 5-1SensorCrossSection ................................ 96 5-2SchematicofthedifferentLa 2 CuO 4 electrodecongurations ........... 96 5-3Temperatureandtestgasproleofexperiments ................. 97 5-4Voltageversustimeplot ............................... 97 5-5Voltageversuslog(GasConcentration) ...................... 98 5-6NO 2 Sensitivityof“Sensor1”foreachcongurationat550,600, and650 C .. 98 5-7NOSensitivityof“Sensor1”foreachcongurationat550 ,600,and650 C .. 99 5-8COSensitivityof“Sensor1”foreachcongurationat550 ,600,and650 C .. 99 5-9NO 2 selectivitywithrespecttoNOat550,600,650 C .............. 100 5-10NO 2 selectivitywithrespecttoCOat550,600,650 C .............. 100 5-11NOselectivitywithrespecttoCOat550,600,650 C ............... 101 5-12NO 2 sensitivitycomparisonsbetweenRun1andRun2at550,600,a nd650 C 101 5-13NOsensitivitycomparisonsbetweenRun1andRun2at550 ,600,and650 C 102 5-14COsensitivitycomparisonsbetweenRun1andRun2at550 ,600,and650 C 102 5-15ComparisonofCOsensitivityfor3samplesofCongurat ion3at600 C .... 103 A-1NO 2 voltageresponseversustimeata)600 Cb)550 Cc)500 C ....... 108 A-2NOvoltageresponseversustimeata)500 Candb)450 C ............ 109 A-3NO 2 voltageresponseversusNO 2 concentrationata)600 Cb)550 Cc)500 C 109 A-4NOvoltageresponseversusNOconcentrationata)500 Candb)450 C .... 110 11

PAGE 12

B-1Conguration1:NO 2 ,NO,andCOvoltageresponseversustimeata)650 C b)600 Candc)550 C ................................ 111 B-2Conguration1:voltageversusgasconcentrationata)6 50 Cb)600 Cand c)550 C ........................................ 112 B-3Conguration2:NO 2 ,NO,andCOvoltageresponseversustimeata)650 C b)600 Candc)550 C ................................ 113 B-4Conguration2:voltageversusgasconcentrationata)6 50 Cb)600 Cand c)550 C ........................................ 114 B-5Conguration3:NO 2 ,NO,andCOvoltageresponseversustimeata)650 C b)600 Candc)550 C ................................ 115 B-6Conguration3:voltageversusgasconcentrationata)6 50 Cb)600 Cand c)550 C ........................................ 116 B-7Conguration4:NO 2 ,NO,andCOvoltageresponseversustimeata)650 C b)600 Candc)550 C ................................ 117 B-8Conguration4:voltageversusgasconcentrationata)6 50 Cb)600 Cand c)550 C ........................................ 118 B-9Conguration5:NO 2 ,NO,andCOvoltageresponseversustimeata)650 C b)600 Candc)550 C ................................ 119 B-10Conguration5:voltageversusgasconcentrationata) 650 Cb)600 Cand c)550 C ........................................ 120 12

PAGE 13

LISTOFABBREVIATIONS [x]Gasconcentrationof“x”EISElectrochemicalimpedancespectroscopyEMFElectromotiveforceeVElectronvoltsFFaraday'sconstantFE-SEMField-emissionscanningelectronmicroscopeHCsHydrocarbonsmVMillivoltNO 2 Nitrate NO 3 NItrite NO x Nitrogenoxides pO 2 Partialpressureofoxygen ppmPartspermillionRUniversalgasconstantssecondssccmStandardcubiccentimetersTTemperatureTPRTemperatureprogrammedreactionTPDTemperatureprogrammeddesorptionXPSX-RayphotoemissionspectroscopyXRDX-RaydiffractionYSZYttriastabilizedzirconia 13

PAGE 14

AbstractofDissertationPresentedtotheGraduateSchool oftheUniversityofFloridainPartialFulllmentofthe RequirementsfortheDegreeofDoctorofPhilosophy OPTIMIZATIONOFLA 2 CUO 4 SENSINGELECTRODESFORANO X POTENTIOMETRICSENSOR By EricR.Macam May2011 Chair:EricD.WachsmanMajor:MaterialsScience&Engineering Gassensorsshowgreatpotentialintheefforttoreducethea mountofemissionsof harmfulby-productsofcombustionandtomaintainefcient operationofthesesystems. Thereisanobviousneedforthesedeviceswiththecurrentso cialandpoliticalclimate thatpointstowardacleanerandlessconsumptiveculture.A lthoughsensorsforcarbon monoxide,nitrogenoxides,hydrogen,etc.havebeenstudie d,theyhaveyettoreach thelevelofcommercialsuccessandwidespreaduseastheoxy genlambdasensor. Themetricofsuccessingassensordevelopmentmustinclude developingdevicesthat aresensitive,selective,andstable.Sensitivityreferst othesensor'sabilitytodetecta givenamountofgas(ofteninthepartspermillionrange)and producealargeenough signalinresponse.Aselectivesensorisonethatisabletod etectagivengaswithout beingaffectedbyothergases.Stabilityisaparameterthat shouldincludetheproperties ofaccuracy,precision,andrepeatability.Inaddition,an understandingofthesensor mechanismisofutmostimportanceintermsoffundamentalan dpracticalapplication andcanhelpwithtransitioningthistechnologytocommerci alization. ThisworkexaminedthegeometricpropertiesofLa 2 CuO 4 sensingelectrodesand theirinuenceontheperformanceofaceramicbasedNO x potentiometricsensor. Theseproperties,whichincludeelectrodethickness,area ,andcongurationalsohelped illustrateandexplainthedifferentcontributionsofthep roposedsensingmechanism “DifferentialElectrodeEquilibria”.Thiscomprehensive mechanismincludescontributions 14

PAGE 15

duetoMixedPotential,heterogeneouscatalysis,andchang esinsemiconductive propertiesduetoadsorptionofgasspeciesontheelectrode surface.Actualsensor devices(withthevaryingsensingelectrodegeometricprop erties)weresuccessfully producedandtestedinsimulatedgasenvironmentstoassess sensorperformanceand understandsensingmechanism.Impedancespectroscopyand Kelvinprobemicroscopy werealsousedtofurtherrelatethesemiconductivepropert iesofLa 2 CuO 4 toactual sensorperformance.Theresultsofthisworkhelpfurtherex plain“DifferentialElectrode Equilibria”aswellasprovidedesignparametersinorderto makesensorsthatmeetthe requirementsofsensitivity,selectivity,andstability. 15

PAGE 16

CHAPTER1 BACKGROUNDINFORMATION 1.1Introduction Thequalityoftheairthatpeoplebreathehasbeenaconcernd atingbacktothe 14thcenturywhencoalheaterswererstusedinhomesinEuro pe.Thisconcern remainstodayalthoughperhapsinadifferentcapacitythan inthe14thcentury.Today, muchattentionisfocusedongaspollutantsthataregenerat edbycombustionsources, particularlyon-roadvehicles.Thesegasesincludecarbon monoxide(CO),nitrogen oxides(NO x ),andhydrocarbons(HC).Accordingtoa2007reportwritten bythe EnvironmentalProtectionAgency,about75%oftotalCOemmi ssionsand50%oftotal NO x emissionsintheUnitedStateswereattributedtoon-roadve hicles[ 7 ]. AnothermajorconcernintheUnitedStatesisenergyconsump tion.Energyusage continuestobeontherisedespitethefactthattypicalfoss ilfuelbasedenergysources arebecominglimited.Anobviousimpactcommonlyobservedb ytheconsumeris theincreaseingasolineprices.Onewaythisproblemcanbes olvedisthroughthe developmentofsystemsthatuseenergymoreefciently. Thedevelopmentofsensorsthatmeasurethegaseousby-prod uctsofcombustion processesisimportanttosolvingthementionedproblems.S ensorshavebeen developedformanyyears;however,theyhavenotbeenreliab lenorfullyunderstood.It istheauthor'sgoaltooptimizesensordesignandmaterials processinginordertobetter understandtheworkingmechanismofsensorssotheycanbeus edtocreateabetter solutionforairpollutionandenergyefciency. 1.2Signicance 1.2.1EnergyEfciency Inrecentyears,thesubjectoffueleconomyhasgarneredmuc hattention. Fluctuatingfuelpricesandincreasingpublicknowledgeof alimitedfuelsupplyare factorsthathavecontributedtothisconcern.Onepotentia lsolutiontothisproblemisthe 16

PAGE 17

useoflean-burncombustionengines.Lean-burnenginesope rateatanairtofuelratio greaterthanstochiometry(14.7partsofairtooneoffuel). Thisoperationcanleadtoan improvementinfueleconomyupto10%[ 8 ].Anissuewiththeuseofthistypeofengine isthatthethree-waycatalystthatreducesHC,CO,andNO x emissionsatstochiometry hasproblemsreducingNO x whenrunatlean-burnconditions,i.e.excessoxygen. AnothermethodofreducingNO x emissionssuchasselectivecatalyticreduction[ 9 10 ] wouldberequired,andasaresultaNO x sensorisessentialtomonitorthisprocess. 1.2.2AirPollution1.2.2.1Gaschemistry Thegasesthatareofprimaryconcerninthisworkarenitrico xide(NO),nitrogen dioxide(NO 2 ),andCO.Themolecularstructureofeachofthesegasesplay asignicant roleonhowtheyaredetectedbyasensor.Thebondstructurea ndmolecularstructure ofNOareshowninFigs. 1-1 and 1-2 .Itisalinearmoleculethatconsistsofaresonant doublebondwithalengthof115pm.Incontrast,NO 2 (Figs. 1-3 and 1-4 )isabent moleculewithapairofresonanthybriddoublebondswithale ngthof119.7pm.NOcan oxidizeNO 2 accordingtoEquation 1–1 2 NO + O 2 2 NO 2 (1–1) CO(Figs. 1-5 and 1-6 )isalinearmoleculewithatriplebondoflengthof112.8pm. TheCOmoleculeisknowntohaveadipolemoment.Theimbalanc eoftheelectron cloudsurroundingtheCatomenhancesitspolarizability,w hichcanaffectprocesses suchasgasadsorption.1.2.2.2Pollutionformation Theproductionofthemostsignicantpollutants(CO,NO x ,andHC)inaspark-ignition combustionengineoccurswithintheenginecylinder.COfor msasaby-productwhen fuelisincompletelyoxidized,i.e.,notenoughoxygenisav ailabletoproducecarbon 17

PAGE 18

dioxide(CO 2 ).[ 11 12 ].Undernormaloperatingconditions,COaccountsfor0.5vo l.%of thetypicalexhaustgascomposition[ 13 ]. NO x formswhennitrogenmoleculescollidewithoxygenmolecule sathigh temperatureswithenoughenergytobreaktheirbonds.Onceb roken,aNOmolecule canform.InuentialparametersintheproductionofNO x includegastemperatureand oxygenconcentration. NO x canalsoformwhenfuelsthatcontainnitrogenareusedinthe formof ammoniaorboundnitrogeninhydrocarboncompounds.Thepro cessislessdependent ontemperaturesincecarbon-nitrogenbondscanbebrokenmo reeasilythandiatomic nitrogenbonds.Thisisparticularlytrueforcoalcombusto rs[ 14 ]. 1.2.2.3Environmentalaspects Alackofcontrolofpollutantgasemissionshasnegativeeff ectsontheenvironment. Whenultra-violet(UV)radiationreactswithanatmosphere thatispollutedwith hydrocarbonsandoxidesofnitrogenitproducesphotochemi calsmog,abrownhaze intheair.Nitrogendioxide(NO 2 )initiatestheprocesswhenitabsorbsUVlightto formNO.Oxygenthatisfreedupbythisprocesscanalsoreact toformozone(O 3 ). Photochemicalsmogdamagesplantsandisdetrimentaltohum anhealth[ 15 ]. Acidrainisanotherenvironmentaleffectofpollutantgase s.Rainfallcanbeacidic inthepHrangeof5to7duetonaturallyoccurringCO 2 ,sulfurdioxide(SO 2 ),andNO x AcidrainisdenedasrainwithapHlessthan5.Itisproduced whenSO 2 andNO x converttosulfuricacidandnitricacid(asinEquation 1–2 ),respectively.Theacidrain canaffecttheacidityofsoilandwatersourcestherebyaffe ctingplantandanimallife. 3 NO 2 + H 2 O NO +2 HNO 3 (1–2) 1.2.2.4Healthaspects Airpollutantsarefactorsinanumberofhealthproblems.Pr oblemscanrangefrom minorirritationstofatalconditions.COisanodorlessgas thatattachesitselftothe 18

PAGE 19

hemoglobininredbloodcellstoformcarboxyhemoglobin.He moglobinhas200timesan afnityforCOthanO 2 ,resultinginthereductionofO 2 deliverytothebody.Thiscauses cellularanoxia.Cellularanoxiaistheconditionwherethe reisnotenoughO 2 inthe cellsformetabolicprocesses.Thisconditionaffectspart softhebodythatrequirehigh levelsofO 2 suchasthebrainandotherpartsofthenervoussystem.Lowle velsofCO exposuremayalsoaffectpeoplewhosufferfromheartdiseas e[ 11 ]. NO x causesavarietyofeffectsaswell.Forexample,NO 2 appearsasareddish browngaswithanirritatingodor.Itisadeeplungirritantt hatwiththepotentialtoreach thebronchiolesandalveoli.Furthermore,itcanreachthea queousliningmediumofthe nasalandpharyngealmucosawhereitcanbeconvertedtonitr ousandnitricacid.At highenoughconcentrations,NO 2 exposureresultsinpulmonaryedema,evendeath.In addition,NO x canirritatetheskinandeyesandcausevisualimpairment.[ 15 ] 1.3LiteratureBackground 1.3.1SensorTechnologyandDevelopment Gassensorshavebeenstudiedforalmost40years.Thesesens orsweredeveloped primarilytomonitorautomotivecombustionstochiometry. Gassensorsaregenerally placedintothreecategories:solidelectrolyte,catalyti c,andsemiconducting[ 16 ]. 1.3.1.1Solidelectrolytesensors Solidelectrolytesensorsarebasedonelectrolyteswhichh avehighionicconductivity. Themostsuccessfulsensordevelopedthusfaristheoxygenl ambdasensor[ 17 ].These sensorsarepresentineveryautomobilethathasbeenproduc edsincethe1970's.A schematicofthissensorisshowninFigure 1-7 Thesensorconsistsofaclosedtubeelectrolytemadeofyttr ia-stabilizedzirconia (YSZ)andtwoPtelectrodes.OnePtelectrodeislocatedinth einsideofthetubewhile theotheroneisontheoutsideofthetube.Theelectrolytese paratesareferenceair atmosphereinsidethetubeandtheengineexhaustatmospher eoutsideofthetube. EachPtelectrodeprovidessitesforthefollowinghalfcell reactions: 19

PAGE 20

4 e + O 2 2 O 2 ( ExhaustPtElectrode )(1–3) 2 O 2 4 e + O 2 ( ReferenceAirPtElectrode )(1–4) Oxygenionsareconductedthroughtheelectrolyteduetothe concentration gradientbetweenthetwoatmospheres.Thevoltagegenerate dbythiselectrochemical cellcanbedeterminedbytheNernstEquationshowninEquati on( 1–5 )whereRis theuniversalgasconstant,FisFaraday'sconstant,andTis thetemperatureinK. Thesensoroperatesattemperaturesabove600 C.Belowthistemperature,theionic conductivityistoolowtogenerateameasurablevoltage.Us ingthisvoltage,onecan determinethepartialpressureofoxygen( pO 2 )intheexhauststream. EMF = RT 4 F ln pO 2( reference ) pO 2( exhaust ) (1–5) Ingeneralsolidelectrolytesensorshavethefollowingadv antages:(i)quickand continuousmonitoring,(ii)becausesensoroutputisanele ctricalsignal,itcanbe modulatedforelectroniccontroldevices,(iii)accuracy, (iv)simpledesign,and(v)weak temperaturedependence.Itshouldbenotedthesetypesofse nsorsmeasurethe equilibriumpartialpressureofO 2 ,notthetrueO 2 concentration.Thesesensorsalso losesensitivitywhenthedifferencein pO 2 betweentheexhaustandreferenceissmall [ 16 ]. 1.3.1.2Catalyticsensors Catalyticcombustionsensorsconsistofanaluminabeaddis persedwithacatalyst (suchasPt,Pd,Rh,etc.)andmountedonaPtcoil.Thetempera tureofthesensoris heatedtoabout673-723KbyapplyingacurrenttothePtcoil. Whenacombustible gascomesincontactwiththesensor,itreactswiththecatal yst.Thereactioncauses anincreaseinthesensortemperatureandconsequentlyanin creaseintheresistance ofthesensormaterial.Theincreaseinresistanceisthusre latedtotheincreasein 20

PAGE 21

theamountofcombustiblegas.Thesesensorsareprimarilyu sedforgasessuchas C 2 H 5 OH,CO,orH 2 Anothertypeofcatalyticsensorconsistsoftworesistivet emperaturedetectors,one whichiscoatedwithacatalyticmaterial.Thecatalystmaya ctivateareactionforthe detectedgas.Thesereactionscauseadifferenceinresista ncebetweenthedetectors, whichrelatestothemolefractionofthedetectedgas.Altho ughsensorsofthistype havebeendevelopedforCOandNO x ,thesegenerallyhavepoorselectivityandare subjecttopoisoning[ 18 ]. 1.3.1.3Semiconductingsensors Semiconductingsensorsworkontheprinciplethatproperti esofsemiconducting materialscanchangebasedontheadsorptionand/orreactio nofforeignspecies.This behaviorcanrstbeillustratedwiththeuseofadefectequi libriumdiagramforan acceptor(A')dopedn-typesemiconductingoxide,suchasSn O 2 (Figure 1-8 )[ 19 ].The majoritychargecarriersforthismaterialinthelow pO 2 regionareelectrons.Alarge enoughchangein pO 2 willchangetheconcentrationofdefects,andconsequently the conductivityofthematerial.However,changesin pO 2 inthescopeofthisresearchare notlargeenoughtochangeconductivitybythismechanism. Theconductivityofthesematerialsisoftendescribedinte rmsofitsFermiEnergy. TheFermiEnergyisdenedastheenergyatwhichtheFermidis tributionisequalto 1 2 i.e.theenergyinwhichthereisa50%thatanelectroncanbef oundinthatstate[ 20 ]. ThechangeinFermilevelduetotheadsorptionofspeciesiss eeninFigure 1-9 .The gureshowsthatanaccumulationofchargeonthesurfaceoft hesemiconductorcauses abendingoftheconductionandvalencebands.Thisoccursbe causeoftherepelling actionoftheaccumulationofcharge.Theregioniscalledth edepletionlayer.Thus, theFermiEnergyrelativetothevalencebandandconduction haschangedwithinthis depletionlayerandresultsinachangeintheconductivityo fthematerial. 21

PAGE 22

Semiconductingoxideswereusedinordertoenhancebothsen sitivityandstability. Thepropertiesofoxidescanbeaffectedbyreactionwithoxy gen.Inaddition,the surfaceandgrainboundaryresistancearecontrolledbythe adsorptionofvarious gaseousspecies.Varyingdegreesofchemicaladsorptionof differentgasesonthese oxidesallowsforselectivityingassensing.Asimpliedpr oposedmechanismforgas sensingbyasemiconductingoxidecanbeexplainedfromares istanceperspectiveofa n-typesemiconductor.Adsorptionofoxygenonthesurfaceo fann-typesemiconductor causesanincreaseintheresistanceofthematerial.Ifared ucinggasweretoreactwith thesurfaceoxygen,theresistancewouldbelowered. Anumberofsemiconductingoxides(bothn-andp-type)haveb eenstudied assensorsforgasessuchasH 2 ,CO,andNO x .N-typesemiconductorsareoften preferredoverp-typesemiconductorssothattheintroduct ionofareducinggascauses adecreaseinresistance.SnO 2 isoneofthemoststudiedmaterialsforsensors.Liet al.[ 3 ]studiedSnO 2 forH 2 andCOsensitivity.ThechangeinresistanceofSnO 2 pellets underdifferentH 2 andCOconcentrationsat300 Cweremeasuredunderanapplied2 Vbias.Thesensitivityofthesensorwasdenedas: S = R air R gas (1–6) where R air istheresistanceofthesensorinairand R gas istheresistanceinreducing gases.Thesensitivityplottedversustheconcentrationof H 2 andCOisshowninFigure 1-10 1.3.2SensorMechanisms Sensorsstudiedmorerecentlyuseelementsfromeachofthep reviouslymentioned sensortypes.Forexample,alargenumberofthesesensorsco nsistofazirconia electrolyteandmetaloxidesensingelectrodes.Thesesens orscanoperatepotentiometrically (voltageoutput)oramperometrically(currentoutput).Us ingacombinationofdifferent elementsinasensorandaccountingforthecomplexityofthe behaviorofgaseslike 22

PAGE 23

COandNO x makesensingbehaviordifculttounderstand.Twotheories havebeen proposedtoexplaintheworkingmechanismofthesesensors: MixedPotentialTheory andDifferentialElectrodeEquilibriaTheory.1.3.2.1Mixedpotentialtheory MixedPotentialTheoryoriginatesfromcorrosiontheory[ 21 ].Thisconceptas relatedtosensorswasrstinvestigatedinordertoaddress potentiometricresponses thatdeviatedfromtheNernstEquation[ 4 22 25 54 – 59 ].Duringthecorrosionprocess apairofreductionandoxidation(redox)electrochemicalr eactionsoccursimultaneously. Eachreactionhasanassociatedvoltagewhichisideallyrel atedtocurrentbyTafel behavior.Asboththeoxidationandreductionreactionssim ultaneouslycometo equilibrium,theyestablishacorrosionpotential.Thisbe haviorcanbeplottedona current-potentialdiagram.Onthisdiagram,thecurrentat thecorrosionpotential correspondstoacurrentdensitywhichisproportionaltoth ecorrosionrate. Miuraetal.usesthisargumenttodescribethebehaviorofth eirsensors[ 22 ].These sensorshaveasimilardesigntotheoxygenlambdasensor;ho wever,thereference andsensingelectrodesareametalandmetaloxide,respecti vely.AccordingtoMixed PotentialTheory,multiplegasreactionssimultaneouslyc ometoequilibriumoverthe electrodes.Eachelectrodeestablishesamixedpotentialb asedontheequilibrium andratesofthesereactions.ForaNOsensor,Equations 1–7 and 1–8 representthe correspondingredoxreactions: 2 NO +2 O 2 2 NO 2 +4 e (1–7) O 2 +4 e 2 O 2 (1–8) Equations 1–9 and 1–10 representthecorrespondingredoxreactionsforNO 2 sensing: NO 2 +2 e NO + O 2 (1–9) 23

PAGE 24

2 O 2 O 2 +4 e (1–10) Anexampleofacurrent-potentialdiagramforthesesetofre actionstakingplaceon aWO 3 sensorisshowninFigure 1-11 .Consequently,thesensingelectrodeand referenceelectrodeobtaindifferentmixedpotentialsdue tothedifferenceinreaction ratesoccurringovereachelectrode.Themetalreferenceel ectrodeisexposedtoanair referenceatmosphereandthemetaloxidesensingelectrode isexposedtothetestgas atmosphere.OnFigure 1-11 ,E M 1designatesthemixedpotentialforthereactionsin Equations 1–7 and 1–8 overtheWO 3 electrodeuponexposuretoNO.E M 2designates themixedpotentialforthereactionsinEquations 1–9 and 1–10 overtheWO 3 electrode uponNO 2 exposurewhilethemixedpotentialoverthereferenceelect rodeis0mV.The voltageoutputofthesensoristhedifferencebetweeneache lectrode'smixedpotential. ThistheoryhasbeenappliedtoNO x ,CO,andHCsensors[ 4 16 22 – 25 ]. 1.3.2.2DifferentialElectrodeEquilibriaTheory Wachsmanetal.havefoundlimitationstotheMixedPotentia lTheoryandinstead formulatedtheideaofDifferentialElectrodeEquilibria[ 26 ].OnemainproblemisthatNO doesnotcometothermodynamicequilibriumatsensoroperat ingconditionsbecause ofthekineticlimitations.SinceMixedPotentialTheoryme asuresequilibriumlevelsof NO,itcannotaccuratelydescribeNOsensing.Theotherlimi tationisthatgenerally atleastoneoftheelectrodesinasensorisasemiconducting material.Upongas adsorption,semiconductorsexhibitachangeinFermiEnerg y.Forasemiconductor, theFermiEnergyliesbetweenthevalenceandconductionban ds.Thedevelopmentof chargeonthesurfaceduetotheadsorptionofgascausesthev alenceandconduction bandstobend.ThischangestherelativeleveloftheFermiEn ergy.Theoutputofa potentiometricsensor,whichmeasuresthepotentialbetwe entwoelectrodes,isaffected bythisphenomenonsinceFermiLevelisthelocalizedelectr ochemicalpotentialof thematerial'selectrons( e = E F ).Mixedpotentialalsodoesnottakeaccountforgas phasecatalysisthatoccursoverelectrodeswhichexhibitc atalyticbehavior.Therefore, 24

PAGE 25

DifferentialElectrodeEquilibriaisamoreinclusivetheo rywhichtakesintoaccount multiplecontributionsasacauseforthedifferenceinpote ntialbetweentwoelectrodes. Theseadditionalcontributions,namelytheadsorptiveand catalyticproperties ofsemiconductingoxides,includingLa 2 CuO 4 ,havebeenthoroughlystudiedaswell asrelatedtopotentiometricsensitivityinNO x sensors.Forexample,Temperature ProgrammedDesorption(TPD)experimentsexhibitedwhatad sorbedspecies (suchaschemisorbednitratesandnitrites)remainonthema terialsurfaceatwhich temperature[ 6 31 ].Correspondingsensorsusingthesematerialsconsistent lyexhibit maximumsensitivityatthetailendofthedesorptionpeakso ftheseTPD'sand sensitivityisoftenlimitedbysaturationoftheadsorbeds pecies[ 6 31 37 ].Athigher temperatureswheredesorptionofspeciesnolongeroccurs, thesensingmechanism cannolongerbeattributedtoasemiconductingresponse. TemperatureProgrammedReaction(TPR)experimentsexhibi tthecatalytic behaviorofsomesemiconductingmaterialsusedinsensors. Forexample,NO 2 begins toreducetoNOoverLa 2 CuO 4 at300 Candfurtherreducesastemperaturesincrease untilfullreductionoccursabove600 C[ 6 ].Thisissignicantinthatthisbehavior changestheconcentrationofspeciesthatreachtheelectro de/electrolyteinterface andthusaffectstheelectrocatalyticcontributionasdesc ribedbyMixedPotential.This phenomenaalsooftenoccursattemperatureshigherthanwhi chasemiconducting responsenolongeroccursasstatedaboveandthuscanbeasso ciatedtothesensing behaviorinthistemperaturerange. 1.4Goals Theoverallscopeofthisworkistoprovideafundamentalund erstandingofthe mechanismofNO X sensingforpotentiometrictypegassensorswhiledevelopi ng sensorstofurthermakeprogresstocommercialization.Nom atterthetypeofsensor, themetricsofagoodsensorshouldincludesensitivity,sel ectivity,andreliability[ 23 ]. Inthisregardsensitivityreferstotheabilitytodetectas mallamountofadesiredgas 25

PAGE 26

(oftenintheppmrange)whileproducingalargeenoughsigna lresponse.Selectivity referstheabilityofthesensortodetectaspecicgaswitho utthatdetectionbeing affectedbythepresenceofotherspecies.Finally,reliabi lityreferstothesensorsability torespondfastenoughandtobeusedwithcondenceinrepeat ability. Theapproachtoaddressingbothofthesegoalsistoanalyzet heeffectsof geometryoftheLa 2 CuO 4 sensingelectrodeincludingparametersofelectrode conguration,area,andthickness.Bycomparingsensorper formancewithadditional electrochemicalanalysistechniques,conrmationofcont ributionstosensormechanism asoutlinedbyDifferentialElectrodeEquilibriacanbeach ieved.Inaddition,thiswork canserveasaguidetodesignelectrodesappropriateforthe metricsofsensitivity, selectivity,andreliabilitysetforperformance. Figure1-1.Bondstructureofnitricoxide. Figure1-2.Molecularstructureofnitricoxide. 26

PAGE 27

Figure1-3.Bondstructureofnitrogendioxide. Figure1-4.Molecularstructureofnitrogendioxide. Figure1-5.Bondstructureofcarbonmonoxide. Figure1-6.Molecularstructureofcarbonmonoxide. 27

PAGE 28

Figure1-7.Schematicofa)anoxygensensormountedinanexh austpipe[ 1 ]andb) crosssectionofanautomotivesensorwithametallicshroud [ 2 ]. 28

PAGE 29

Figure1-8.GeneralDefectEquilibriumDiagramforanaccep tor(A')dopedn-type semiconductingoxide(Dashedlineanddottedlinerepresen toxygen vacancyconcentrationandelectronconcentration,respec tively)[ 19 ]. An-typesemiconductor Bp-typesemiconductor Figure1-9.FermiLevelofa)an-typesemiconductorandb)ap -typesemiconductorwith asurfacecharge[ 20 ]. 29

PAGE 30

Figure1-10.SensitivityofSnO 2 sensorsversusconcentrationofH 2 andCO[ 3 ]. Figure1-11.Current-voltageplotforNOandNO 2 reactionsonaWO 3 sensor[ 4 ]. 30

PAGE 31

CHAPTER2 EXPERIMENTALMETHODS 2.1MaterialsUsed 2.1.1LanthanumCopperOxide(La 2 CuO 4 ) La 2 CuO 4 hasbeenselectedasthecandidateforaNO x sensingelectrodeinthis researchbecauseofitsp-typesemiconductingresponseupo nNOadsorptionandits lackofcatalyticactivityforNOreductionoroxidation[ 10 ].Ithasamodiedperovskite structure,whichisanorthorhombicdistortionoftheI4/mm mspacegroup[ 5 27 ].The crystalstructureisdisplayedinFigure 2-1 .ManystudieshavebeendoneonLa 2 CuO 4 sinceitisaparentcompoundforthehigh-T c superconductingoxide,La 2 x M x CuO 4 (M=Ba,Sr).Itisamaterialthatcanbesynthesizedusingman yprocessingroutes includingco-precipitation,auto-ignitioncombustion,f reezedrying,sol-gel,andinorganic reactionusinganamorphousheteronuclearcomplexprecurs or[ 28 – 30 ].Theperovskite crystallinephaseofLa 2 CuO 4 isformedatacalcinationtemperatureof650 C[ 28 ]. ThesensormechanismforLa 2 CuO 4 basedNO x andCOpotentiometricsensors haspreviouslybeeninvestigatedviatemperatureprogramm edreactions(TPR) andtemperatureprogrammeddesorption(TPD)experiments[ 31 ].Inaddition,the compositionoftheNO x adsorbatesonLa 2 CuO 4 andthemechanismofoxygen exchangebetweenNO x andLa 2 CuO 4 hasbeenstudiedextensivelyusingInfrared/X-ray PhotoemissionSpectroscopy[ 32 ]andTPR/TPDexperimentswithisotopicallylabeled O 2 [ 33 ],respectively.NOsensitivitywasattributedtoNOadsorp tionandtheassociated relativechangeinFermiEnergyatthesurface.Thischangei scausedbytheremovalof electronsfromtheLa 2 CuO 4 bulkwhentheadsorbedNOinteractswithlatticeoxygento formthenitritecomplex(NO 2 )onthesurface. ThenatureofNO 2 ismorecomplicated.NO 2 beginstoreducetoNOoverLa 2 CuO 4 at350 Candcompletelydecomposesby600 C.However,NO 2 behavessimilarlyover Pt.Thus,thesensingmechanismcannotbeattributedtoacat alyticdifferencebetween 31

PAGE 32

electrodesofPtandLa 2 CuO 4 .Rather,NO 2 isadsorbedontothesurfaceofLa 2 CuO 4 similarlyasNO,alsoresultinginarelativechangeinFermi Energyatthesurface. However,whenanitriteisformedwithNO 2 adsorptionelectronholesareproducedwith noLa 2 CuO 4 latticeoxygenexchange.Thusasexpected,theresponseofN O 2 ispositive andoppositetothatoftheNOresponse.NO 2 adsorbatescanalsoformanitrate complex(NO 3 )viatwodifferentreactionpathswhicheitherproduceelec tronholesor consumeelectrons.Theoverallvoltageisthusdependenton therelativeamountof nitriteandnitrateformation.2.1.2Yttria-StabilizedZirconia(YSZ) Yttria-stabilizedzirconia(YSZ)hashistoricallybeenon eofthemostimportant electrolytestosolidstateionicdevices.WaltherNernstu sedzirconiaaspartofhis NernstLightin1899.YSZwasrstusedinasolidoxidefuelce ll(SOFC)in1937by BauerandPries.TheoxygensensordevelopedbyFlemingin19 77usedYSZasits electrolyte.Zirconia(ZrO 2 )exhibitsthreepolymorphs:monoclinicat1100 C,tetragonal at2370 C,andtheuoritecubicstructureabove2370 C[ 34 ].Itistheopennessof theuoritestructurethatmakesZrO 2 idealforoxygenionconduction.Intheuorite structure,theZr 4+ ionsareinfacecenteredcubicpositionsandtheO 2 ionsareinthe tetrahedralpositions.ThustheZr 4+ ionsaresurroundedby8O 2 ionswhiletheO 2 ionsaresurroundedby4Zr 4+ ions. Y 2 O 3 2 ZrO 2 2 Y 0 Zr + V O +3 O x O (2–1) Whiletheuoritestructureisdesired,thestructureisonl ystableabove2370 C.In ordertostabilizethestructure,elementssuchasYttrium( Y)areaddedasadopant.Y isaddedaccordingtoequation 2–1 .Astheequationshows,oxygenvacancies V O are producedwhichalsoaidinoxygenconduction.HowY 2 O 3 structuresubstitutesintothe ZrO 2 latticeisshowninFigure 2-2 32

PAGE 33

2.1.3Platinum Inthiswork,Platinum(Pt)isbeingusedasthecounterelect rodematerial.Ptis anoblemetalanditsatomscoordinateinaface-centeredcub icstructure[ 35 ].As mentionedearlierinthischapter,Ptelectrodesareusedin theautomobileoxygen lambdasensor.Itisacommonlyusedmaterialusedforotherc eramicbasedsensors thatdetectothergasessuchascarbonmonoxideandnitrogen oxides[ 24 ],[ 36 ].Ptis alsousedforasbondingwiresforsensors,aswellasotherel ectrochemicaldevices.Pt isanidealcandidatefortheseapplicationsbecauseofitsh ightemperaturecapabilities (meltingpointof1780 C)anditscatalyticproperties. 2.2TechniquesUsed 2.2.1PowderSynthesis Inthiswork,theLa 2 CuO 4 electrodematerialwassynthesizedusinganamorphous citrategelcombustionmethod.Thismethodwaschosenafter studyingvariousmethods ofLa 2 CuO 4 powdersynthesisandcorrelatingtheresultingpowdermicr ostructuresand morphologiestosensorperformance[ 37 ]. Theprecursorstothenalpowderarenitratesolutionsofla nthanum(La) andcopper(Cu).Thesesolutionswerepreparedbyaddingdeionizedwaterto hydrouslanthanumnitrateLa(NO 3 ) 3 6H 2 OandcoppernitrateCu(NO 3 ) 2 3H 2 O.The concentrationsofionsinthesesolutionswereconrmedusi nginductivelycoupled spectroscopy.Themeasuredconcentrationswere[La 3+ ]=1.75mol/Land[Cu 2+ ]=1.47 mol/L.Intheamorphouscitrategelcombustionmethod,stoi chiometricamountsofthe solutionsoflanthanumandcoppernitrateswithagivenamou ntofcitricacid,which gaveacitrate-nitrateratioof0.22.Thisratiowascritica lbecausethenitrateandcitrate ionsundergoaredoxcombustionreactionwherethenitrates actastheoxidantand thecitratesactasthereductant.Thiscombustionprocessp rovidesenoughenergyto synthesizeLa 2 CuO 4 atalowertemperaturecomparedtostandardasolidstaterea ction andhelpsreducetheparticlesize. 33

PAGE 34

Fora10gbatchofLa 2 CuO 4 ,16.8mLofCusolutionwasmixedwith28.4mLofthe Lasolution.8.40gofcitricacidwasaddedinordertogetthe 0.22citrate-nitratemolar ratio.Thesolutionwasheatedtoabout95 Cuntilmostofthewaterevaporatedanda gelwasleft.Thisgelwassubsequentlyredinatubefurnace at650 Cfor10hours. X-raydiffraction(XRD)analysiswasusedforstructuralch aracterizationoftheresulting powder. Figure 2-3 displaystheXRDplotobtainedfromtheLa 2 CuO 4 powder.Thegure alsodisplaystheanglesforthefourstrongestintensities forCuOandLa 2 O 3 which couldnotbeidentiedintheXRDplot.ThisdisplaysthatLa 2 CuO 4 isobtainedfromthe amorphouscitrategelcombustionsynthesisroute.2.2.2TapeCasting TheYSZelectrolytesthatwereusedwerepreparedbytape-ca sting.Tapecasting isaprocesstypicallyusedforproducingthinsheetsofcera micsinlargequantities.Itis acastingmethodinwhichapreparedslipslurryisspreadacr ossaatsurfacerather thanamold,likeinslipcasting.Itisusedmostoftenforthe fabricationofdielectricsfor multilayercapacitorsorAl 2 O 3 substratesforintegratedcircuits.Adoctorbladeprocess isthemostcommonlyusedfortapecasting.Inthisprocess,a bladeisusedtospread aslurryacrossacarriersurface,suchascelluloseacetate ,Teon,Mylar,etc.ata controlledthickness.Eitherthebladeorthecarriersurfa cecanbemoving. Aswithmostceramicdepositiontechniques,thepreparatio noftheceramicslip slurrysystemisverycritical.Theslurrysystemmustconta inabalancedmixtureof additivesincludingbinders,plasticizers,solvents,dis persants,etc.Abinderwillallowthe greentape(unsinteredbody)tohaveenoughstructuralstre ngthforhandling,etc,while aplasticizerwillgivethegreentapepliability.Fortapec asting8mol%YSZ,thesolvent isusuallyamixtureoftolueneandethanol.Fishoil,polyvi nylbutyral,andpolyethylene glycolareusedasthedispersant,binder,andplasticizer, respectively[ 38 ]. 34

PAGE 35

InitialsensorfabricationutilizedYSZelectrolytespurc hasedfromMarkatech. Thesehadthedimensionsof13x20x0.1mm.Asresourcesallow edforlargerscale production,theelectrolytesweretapecastinhouse.Tapec astingslurriescantypically containharshsolventssuchastoluene,asstatedabove.How ever,PolymerInnovations suppliedawaterbasedsystemwhichincludesaproprietaryw aterbasedsolvent (WB4101),dispersant(DS001),deocculant(DF002),andpl asticizer(PL005). 2.2.3ScreenPrinting Screenprintingisanimportantdepositionmethodthatisus edinmanyindustries fromtextilestoelectronics.Withthistechniquebothsimp leandcomplexdesignscan beprintedrepetitivelywithease.Screenprintingndsits originsinstenciling,oneof theoldestformsofartwithexamplesfoundintheancientcav esofGargasandTibiran. Screeningevolvedinordertoprovideamoreadvancedandeco nomicalalternative tostenciling.Photographicadvanceswouldalsomakemorec omplexscreendesigns possible[ 39 ]. Inelectronics,discretecomponents,suchasdiodes,trans istors,andintegrated circuitsareoftendepositedusingscreenprinting[ 40 ].Justlikeintapecasting,a ceramicslurryformulatedwithproperadditivesisnecessa ryforscreenprinting.A parameterthatisveryimportantforscreenprintingisslur ryviscosity.Properslurry formulationsshouldmakeitthixotropicinnature,viscous atrestandowingwiththe applicationofashearforce.Oncedeposited,thelmusuall yrangesfrom12to25 min thickness. Screenmeshesareusuallymadeofstainlesssteelwires.The numberofopenings perunitareaareclassiedjustasmeshesusedforsieves.Th emeshisstretchedwith agiventensiontoanaluminumframe.Thetensionisanimport antparameterandwill determinehowmuchthescreenwillexwhenpressureisappli ed.Thepatternonthe screenisappliedasanemulsionusingaphotolithographypr ocesswiththeuseofa photoresistmaterial. 35

PAGE 36

Inthecourseofthiswork,twoscreenprinterswereused.Int heearlystages,a manualtabletopscreenprinterwasused,butasproductiona ndprecisionrequirements increased,asemi-commercialautomaticscreenprinter(DE K)wasused.Theautomated screenprinterallowedforcontrolofcriticalparameterss uchassqueegeepressure, squeegeespeed,printalignment,etc.2.2.4EMFMeasurement Apotentiometricsensor,asitsnameimplies,generatesasi gnalintheformofa potentialdifference.Inthecaseofthesensorsinthiswork ,thispotentialisanelectrical potentialdifferenceknownasavoltageandismeasuredbetw eenthesensingelectrode andthecounterelectrode.Thepotentialoneachelectrodei sanelectrochemical potential.Inthecaseofthesemiconductorsusedassensing electrodesinthiswork, theFermiLevelisthelocalizedelectrochemicalpotential ofthematerial'selectrons ( e = E F ). ThevoltageismeasuredbyaKeithleymultimetercontinuous lyovertimewhilethe sensorisexposedtovariousenvironments.Amajordetailth atshouldnotbeoverlooked isthatinmeasuringthepotentialthepositiveterminaloft hemultimeterisconnectedto thesensingelectrodeandthenegativeterminaltothecount erelectrode. 2.2.5ImpedanceSpectroscopy Impedancespectroscopyhasproventobeavaluabletoolfort heanalysisof electrochemicalandelectricalmaterials.Itcanbeusedto investigatedifferentsystem processessuchasionictransport,grainboundaryphenomen a,andgas-electrode interactions.Impedancespectroscopyhasthebenetofbei ngabletoidentifytime constantsthatareassociatedwiththeseprocesses.Theset imeconstantsusuallyrange ontheorderof10 7 to10 3 s 1 ,withsomeoftheprocessesbeingseparatedby1to 2ordersofmagnitude.Ithasbeenutilizedinstudyingmanys olidstateionicdevices includingbothsolidoxidefuelcellsandsensors. 36

PAGE 37

Theimpedanceofadevicecanbedescribedasthecomplexresi stance.Anideal resistorfollowsOhm'sLawatalcurrentandvoltage,andhas aresistancevaluethatis independentoffrequency.WhileanidealresistancehasanA Ccurrentthatisinphase withitsvoltage,thisnotnecessarilythecaseforrealworl dcircuits.Theimpedanceof asystemmeasuredbyapplyingavoltagetothedevicewithagi venperturbationand measuringtheresultingcurrent.Thevoltageandresulting currentarerepresentedby equations 2–2 and 2–3 .Theimpedance,isthusrepresentedbyequation 2–4 .The impedance Z hasarealcomponentandanimaginarycomponent, j E ( t )= E 0 cos( t )= E 0 exp ( j t )(2–2) I ( t )= I 0 cos( t )= I 0 exp ( j t j )(2–3) Z = E ( t ) I ( t ) = Z 0 exp ( j )= Z 0 (cos + j sin )(2–4) Impedancespectroscopywasperformedusingafrequencyres ponseanalyzer suchasaSolartron1260.“Z-plot”softwarewasusedtosetth eamplitudeofthevoltage andfrequencyrangeoftheperturbation.“Z-view”software wassubsequentlyused toplottheresultingdata.Onewayofinterpretingtheimped ance,isbyplottingthe imaginarycomponentversustherealcomponentinaNyquistp lot.Anexampleis seeninFigure 2-4 .Thistypeofplotcanprovidesinformationrelevanttothis work.For example,thevalueattheRealZintercept( =0)representstheresistance.Inthe caseofsemiconductingoxidesusedforsensors,thehighfre quencyimpedancehas beenshowntoberelatedtochangesinitssurfaceconductanc e[ 41 ].Thus,inorderthis methodcouldbeusedtoisolatethecontributionofchangesi nsurfaceconductivitydue togasadsorption. 37

PAGE 38

Figure2-1.ThreedimensionalstructuralarrangementofLa 2 CuO 4 modiedperovskite crystalstructure.Onetetragonalunitcellisrepresented [ 5 ]. Figure2-2.YSZcrystalstructure 38

PAGE 39

Figure2-3.XRDplotofLa 2 CuO 4 Figure2-4.Nyquistplot 39

PAGE 40

CHAPTER3 ELECTRODETHICKNESS 3.1Introduction Sensorsarenecessarydevicesforminimizingharmfullevel sofgasessuchasNO x andCO.Abetterunderstandingofthesensormechanismisnec essaryforsensor commercialization.Thisworkaddressesthisbystudyingth eeffectofelectrode thickness.DifferentialElectrodeEquilibria[ 26 ]explainsthatapotentiometricsensor generatesavoltagebecauseofsomeformofasymmetrybetwee ntwoelectrodesin thesameatmosphereincludingheterogeneouscatalysis,el ectrochemicalreactionsat triple-phaseboundary(gas/electrode/electrolyte)site s,andchangesinsemiconducting propertiesduetogasadsorption.Theeffectsofsensorelec trodemicrostructureon NO x sensitivityhavebeenpreviouslyexamined[ 37 ].Byvaryingelectrodethickness, wearechangingpropertiesthatcanfurtherberelatedtosen singmechanism.Athicker electrodehasmoretotalsurfacesitesandprovidesamoreto rtuouspathforgasto reachtheelectrode/electrolyteinterface.However,thet hicknessoftheelectrodedoes notaffecttheamountoftriplephasereactionsitesattheel ectrode/electrolyteinterface. Inaddition,variationsinelectrodethickness,ifnotprop erlycontrolledduringprocessing, maycauseinconsistenciesinsensorresponse. 3.2Experimental 3.2.1SensorFabrication Thebasicschematicofthesensorspreparedandtestedissee ninthecross-section diagraminFigure 3-1 .The8-mol%yttria-stabilizedzirconia(YSZ)electrolyte was commerciallypurchasedfromMarketechwithdimensionsof1 3x20x0.1mm.La 2 CuO 4 powdersynthesizedviaanamorphouscitrategelcombustion method[ 30 ]wasused forthesensingelectrodes.AscreenprintableLa 2 CuO 4 ink(20wt%)waspreparedby mixingLa 2 CuO 4 powderwithHeraeusV-015screenprintingvehicle.Theinkw asmixed 40

PAGE 41

usingacentrifugalmixer(Thinky)at2000rpmfor5minfollo wedbyde-airingat500rpm for5min. Anindustrialgradescreenprinter(DekInt'l)wasusedtode posittheLa 2 CuO 4 ink asan8x8cmsquareelectrodeontotheYSZelectrolytes.Theu seoftheindustrial screenprinterhelpedreducevariabilitybycontrollingim portantparameterssuchas squeegeepressureandspeed.Afterasinglelayerwasdeposi ted,aninfraredheatlamp wasusedtodrytheelectrode.Thickerelectrodeswereprepa redbyscreenprinting additionallayers(upto4total)ontothedriedelectrode.A heatlampwasusedinstead ofadryingoventopreservethepositionoftheelectrolytes onthescreenprintingcarrier forthesubsequentscreenprints.TheLa 2 CuO 4 sensingelectrodeswereredat800 C for10hoursaftera1hourbinderburnoutat400 C. Pt(Heraeus)wasscreenprintedasthecounterelectrode(8x 8mm)usingthe Dekscreenprinter.PtinkwasalsousedtoattachthePtwires tobothelectrodes.The sampleswereredat750 Cfor4hoursafter1hourbinderburnoutat400 C.The thicknessoftheelectrodesweremeasuredusingaFieldEmis sionScanningElectron Microscope(FE-SEM).3.2.2TestingParameters Sensorperformancewastestedinagas-owapparatusunderv ariousgas environmentscontrolledusingmassowcontrollers(MKS). AcustomLabView programwasusedtoautomaticallycontroltemperaturesand thegasowrates.The gasenvironmentconsistedofachangingstepconcentration ofNOorNO 2 in3%O 2 balancedbyN 2 atatotalowrateof300sccm.ThestepconcentrationsofNOa ndNO 2 were0,50,100,200,and400ppm.Theseexperimentswereperf ormedateach50 C incrementbetween400-700 C.Thevoltagedifference(opencircuitpotential)between thesensingandcounterelectrodewasmonitoredandrecorde dusingaKeithley Multimeter.Forthevoltagemeasurements,thesensingelec trodewasconnectedto 41

PAGE 42

thepositiveterminalofthemultimeter,andthecounterele ctrodewasconnectedtothe negativeterminal. 3.3Results 3.3.1FE-SEMThicknessMeasurements CrosssectionsoftheLa 2 CuO 4 sensingelectrodethicknessesareshownin Figure 3-2 .Theseimageswereusedtomeasureastatisticalaveragethi cknessof eachelectrode.Atleast10measurementsweretakenacrosst hecrosssectionof theelectrode.Thethicknessmeasurementsalongwiththeco rrespondingstandard deviationsarelistedinTable 3-1 .ThePtcounterelectrodesweremeasuredas13.05 m 1.76 m. 3.3.2SensorResponse Withinthemeasuredrangeof400-700 C,NO 2 andNOsensitivitywasobserved at550-650 Cand400-600 C,respectively.Thisisdisplayedinthevoltageresponse versuschangesingasconcentrationforNO 2 andNOinFigures,respectivelyinthe aforementionedtemperatureranges.Eachsensorisdesigna tedbyitsaverageLa 2 CuO 4 sensorelectrodethickness. WiththeexceptionofthesensorwiththethickestLa 2 CuO 4 electrode,theNO 2 voltageresponseispositiveandincreasesinmagnitudeast emperaturedecreases. Themagnitudeofthesensorresponsedecreaseswithincreas ingsensingelectrode thicknessintherangeof6.86-16.55 mateachtemperature.Thesensorwiththe thickestelectrode(32.88 m)hasapositiveresponseat550 C(Figure 3-3 ),transitions tobeinginsensitiveat600 C(Figure 3-4 ),andhasanegativeresponseat650 C (Figure 3-5 ). TheNOresponseforeachsensorisnegativeinvoltageateach temperature.At thelowesttemperature(400 C,Figure 3-6 ),themagnitudeofthesensorresponseis largeforthelowestNOconcentration,butsaturatesattheh igherconcentrations.As temperatureincreasesthemagnitudeofsensorresponsedec reases,butthevoltage 42

PAGE 43

stepsreachmoreofasteadystateandaremorewelldened.At 600 C,themagnitude ofresponseforthesensorswiththethreethickestsensinge lectrodesisalmostequal andsignicantlylessthanthesensorwiththethinnestelec trode(Figure 3-10 ). WhenthevoltageresponsesareplottedversusNO x concentrationonalogarithmic scale(Figs. 3-11 3-18 ),therelationshipsarelinear.Thevoltagesintheseplots arean averageoftheresponseateachconcentration.Theaveragev oltagewascalculated from4voltagesmeasuredateachconcentrationof50,100,an d200ppmNO x .Twoof thesevoltagesweremeasuredduringtheupwardstepincreas esofNO x whiletheother 2weremeasuredduringthedownwardstepsofthegascycle.Th eaveragevoltagefor 400ppmwascalculatedfrom2measurements.Theerror(stand arddeviationofthe calculatedaverages)isshownasverticalerrorbarsinFigu res 3-11 3-18 .Thelineart wascalculatedbyaminimizationoferrortechniquethrough thestandarddeviationof eachpoint.Theslopeofthistprovidesthemetricofsensit ivitythatthesensorscanbe comparedby.Ingeneralagreementwithourprevioussensorw orkthatusedLa 2 CuO 4 andPtasthesensingandcounterelectrodes,respectively, NO 2 andNOsensitivities wereofoppositepolarity.Anexceptionwasobservedforthe sensorwiththethickest La 2 CuO 4 sensingelectrode(32.88 m)whichhadanegativeNO 2 sensitivityat650 C. Thesesensitivitieswereplottedasafunctionofthesensin gelectrodethickness (Figs. 3-19 and 3-20 )ateachtemperature.Thehorizontalerrorbarsrepresentt he standarddeviationinLa 2 CuO 4 thicknesswhiletheverticalerrorbarsrepresentthe standarddeviationinsensitivity.NO 2 sensitivitydecreasesforeachsample(ofeach thickness)astemperatureincreases(Figure 3-19 ).At550 C,sensitivitydecreaseswith increasingthickness.Astemperatureincreasesthedepend enceonthicknessbecomes lessprominentforsensingelectrodethicknessesgreatert han10 m.At650 C,the sensorswithsensingelectrodesgreaterthan10 mbecomeessentiallyinsensitiveto NO 2 (sensitivity 0). 43

PAGE 44

TheeffectofLa 2 CuO 4 electrodethicknessonNOsensitivityisalsostrongly temperaturedependent(Figure 3-20 ).Whilethesensorsaresensitiveat400 C( -17 mV/decade),thesensitivityisindependentofelectrodeth icknessabove 10 m.At 600 Cthesensorswithelectrodethicknessesgreaterthan 10 mareessentially insensitivetoNO(sensitivity 0),andthusalsoindependentofsensingelectrode thickness.However,intheintermediatetemperaturerange (450-550 C)thereisastrong dependenceonthickness.Therelationshipbetweensensiti vityandelectrodethickness becomesnearlyparabolicwithaninectionpointat12.70 mat450 Cand500 C.The parabolicbehaviordecreasesat550 Candsensitivitybecomesnearlyindependent ofelectrodethicknessabove 10 m.Thisbehaviorintheintermediatetemperature rangeisindicativeofatransitionandoverlapinsensingme chanisms,whichwillbe discussedinthefollowingsection. 3.4Analysis 3.4.1SensorMechanism Inordertounderstandthesensorbehaviorinrelationtothi ckness,onemustrst trytounderstandallthepossiblecontributionstosensorm echanism.Oneoftherst acceptedexplanationsforNO x sensinginnon-Nernstiansolidstatepotentiometric sensorswastheMixedPotentialTheory.Thistheory,whichw asdevelopedinthe eldofcorrosion,wasrstrelatedtosolidstatepotentiom etricsensorsin1982[ 42 ]. However,thistheoryhaslimitationswhensemiconductingo xidesareusedaselectrodes inordertoimprovestabilityandgasselectivity[ 43 – 45 ].Inordertoaccountforthese contributions,amorecomprehensivetheory,Differential ElectrodeEquilibria,was developed[ 26 ]. 3.4.1.1HeterogeneousCatalysis Onereasonthatsemiconductingoxidesaresoattractiveass ensingelectrode materialsisthatmanyofthemexhibitcatalyticbehaviorth atcanbeusedtoimprove selectivity[ 6 31 ].Inthisregard,La 2 CuO 4 isagoodcandidateforthisstudysinceit 44

PAGE 45

isnon-catalyticforthereductionofNO[ 10 ]andhighlycatalyticforthereductionof NO 2 withinthetemperaturerangeoursensorsaretested[ 6 33 ].InYooetal.'swork, temperatureprogrammedreaction(TPR,Figure 3-21 )showedthatNO 2 beginsto reducetoNOandO 2 (asinequation 3–1 )at300 Candessentiallyreducescompletely above600 C. NO 2 n NO + 1 2 O 2 (3–1) ThisreactioncaninuencetherelativeamountofNO 2 toNOandthusaffectthe surfacepropertiesofthesensingelectrodeand/orthereac tionstakingplaceatthe electrode/electrolyteinterface.3.4.1.2Electrocatalysis MixedPotentialTheorybasesitsexplanationonthefacttha tthegasspecies beingsensedtakepartinelectrochemicalreactionsatboth electrodes.Inthecaseof NO 2 sensingthefollowingpairofreactions( 3–2 and 3–3 )occursatboththesensing electrode/electrolyteandcounterelectrode/electrolyt einterfaces: NO 2 +2 e NO + O 2 (3–2) 2 O 2 O 2 +4 e (3–3) Ananalogouspairofreactions( 3–4 and 3–5 )isassociatedwithNOsensingas well: 2 NO +2 O 2 2 NO 2 +4 e (3–4) O 2 +4 e 2 O 2 (3–5) Thedifferenceinreactionratesthatoccurovereachelectr odeestablishesa differenceinpotentialthatisultimatelymeasuredasavol tage.Alimitationtothis 45

PAGE 46

mechanismisthatitassumesthattheNOreaction( 3–4 )comestoequilibriumwhenin factNOshouldthermodynamicallydecomposetoN 2 andO 2 3.4.1.3SemiconductingEffect Semiconductingoxidesalsoexhibitachangeinconductivit yupongasadsorption [ 20 ].Asgasmoleculesadsorbontothematerialsurface,thesem iconductorcan exhibitbandbendinginboththeconductionandvalenceband sduetotheinjection ofelectronsorholesdependingonthegas.Thisbendingcaus esarelativechange intheFermiEnergyofthematerial'selectronswhichresult sinachangeinthe electrochemicalpotential.NO(areducinggas)andNO 2 (anoxidizinggas)shouldcause anoppositechangeinelectrochemicalpotentialforLa 2 CuO 4 ,ap-typesemiconductor. Thisphenomenahasbeenseeninourpreviousworkandwasfurt herunderstoodon La 2 CuO 4 [ 37 ]throughtemperatureprogrammeddesorption(TPD)[ 6 ]andimpedance spectroscopy[ 41 ].Thissemiconductingcontributiononsensorresponseisa lso temperaturedependent. TPDexperimentsforNO 2 andNO(Figures 3-22 and 3-23 )showwhatspecies desorbfromtheLa 2 CuO 4 surfaceastemperatureisincreasedafterNO 2 orNOisinitially adsorbedat300 C[ 6 ].InthecaseoftheNO 2 TPD,thereisamaximumdesorption peakforNO 2 at 300 C.However,NO 2 continuestodesorbfromthesurfaceuntil 600 C,whichindicatesNO x speciesarestilladsorbedatthesetemperatures.The TPDalsoshowsdesorptionpeaksforNOat 300 Cand 425 CandO 2 at 425 C, thelatterindicatingdecompositionofasurfacenitratesp ecies.Thechemistryofthe NO x adsorbatesonmetaloxideshavebeenpreviouslyanalyzed[ 46 ]andwasfurther examinedthroughX-RayPhotoemissionSpectroscopy(XPS)b yVanAsscheet.al. [ 32 ].WhenNO 2 adsorbstothesurfaceoftheLa 2 CuO 4 ,asurfacenitritecomplex(NO 2 ) formswhileproducingelectronholeswithnoLa 2 CuO 4 latticeoxygenexchangeresulting inarelativechangeinFermiEnergy.NO 2 adsorbatescanalsofurtherformanitrate complex(NO 3 ).Dependingonthereactionpathwayofnitrateformation(E quations 46

PAGE 47

3–6 and 3–7 ),theconductivityofLa 2 CuO 4 caneitherincreaseordecrease.The overallvoltageisthusdependentontherelativeamountofn itriteandnitrateformation associatedwithNO x adsorption. NO 2( g ) + O ( ad ) NO 3( ad ) + h (3–6) O x o + NO 2( g ) NO 3( ad ) + V o + e (3–7) TheNOTPD(Fig. 3-23 )indicatesNOdesorptionpeaksat 150 Cand 280 C. VanAsscheshowedthatwhenNOadsorbstothesurfaceofLa 2 CuO 4 thattheNO interactswiththelatticeoxygentoformanitritecomplexo nthesurface[ 32 ].This processremoveselectronsfromtheLa 2 CuO 4 bulkresultinginarelativechangeofFermi Energythatisoppositetotherelativechangecausedbynitr iteformationassociatedwith NO 2 adsorption. 3.4.2Discussion Eachofthecontributionsdiscussedmustbetakenintoconsi derationwhenrelating sensorbehaviortothethicknessofthesensingelectrode.B ecausecontributionsfrom electrocatalysis(mixedpotential)aredependentonreact ionsoccurringatthetriple phaseboundariesattheelectrode/electrolyteinterface, thesecontributionsshould becomepredominantforthethinnerelectrodes.Thickelect rodesresultinpathways (pores)thatgashastodiffusethroughtogettotheelectrod e/electrolyteinterface.Since themeasuredporesizeis 0.4-0.9 m,gasmustdiffusethroughtheelectrodepores beforereachingtheinterfaceevenforthesensorwiththeth innest(6.86 m)sensing electrode.Therefore,heterogeneouscatalysisofNO 2 reductionandNO x surface adsorptionmustalsobeconsidered. Asdiscussedearlier,NO 2 beginsreductiontoNOandO 2 at300 C.However, TPRdata(Figure 3-21 )showsthatNO 2 doesn'tcompletelydecomposetoNOuntil above600 C.ThisissignicantinthatNO 2 sensitivityinthisworkisobservedat 47

PAGE 48

550-650 C.AsNO 2 diffusesthroughthickerelectrodestotheelectrode/elec trolyte interface,itisexposedtoagreateramountofLa 2 CuO 4 tocatalyzeNO 2 reduction. Thus,heterogeneouscatalysisresultsinNObeingtheprima ryspeciesreachingthe electrode/electrolyteinterfaceattemperaturesgreater than600 C.Sincethesegases resultinoppositepolarity,theNOresponsewillcancelthe NO 2 response.Infact,as seeninFigure 3-5 ,forthe32.88 mLa 2 CuO 4 electrode,thepolarityisoppositeat 650 CandisindicativeofNOsensing. At550 C,NO 2 reductionisincompleteandwouldbedependentontheamount ofLa 2 CuO 4 theNO 2 isexposedto.Thisexplainswhythereisagradualdecrease insensitivitywithincreasingLa 2 CuO 4 thickness.However,sincethereductionis incompletethesensorwiththethickestelectrodestillsho wsasensitivityof5mV/decade [NO 2 ]atthistemperature.Astemperatureincreasesto600and65 0 C,NO 2 reduction proceedsmoretocompletionandisreectedinadecreasedse nsitivity.However,with anelectrodethicknessof6.86 m,thesensorstillshowsNO 2 sensitivitiesof18and 9mV/decadeat600and650 C,respectively.ThisindicatesthatNO 2 isnotexposed toenoughLa 2 CuO 4 tocompletelyreduceittoNO.AsobservedintheNO 2 TPD,NO andNO 2 specieswerebeingdesorbedabove600 C.AlthoughthisindicatesNO 2 isstill beingadsorbedontheLa 2 CuO 4 surfaceandcontributestothesensorresponse,the contributionisonlyminor. Asstatedearlier,La 2 CuO 4 isnon-catalytictowardsNOreduction.Therefore, heterogeneouscatalysisdoesnotcontributetotheNOsensi ngmechanism.Sensitivity increasesatlowertemperatures.However,alargeresponse isonlyseenforthe lowestconcentrationwhiletheresponsesaturatesathighe rconcentrations.Atthese temperatures,moreNOisadsorbedandeventuallysaturates thesurface.At400 C, thesensitivity,whilelarge( -17mV/decade)isindependentofelectrodethickness above 10 m.Inaddition,thethermodynamicstabilityofNOmustalsob eaccounted for.Thermodynamiccalculationsshowthatbelow500 C,NO 2 ismorestablethan 48

PAGE 49

NO[ 47 ].Therefore,NOoxidationtoNO 2 mayalsoaffectwhatspeciesreachesthe electrode/electrolyteinterface.Intheintermediatetem peraturerange(450-550 C)that NOsensitivitywasobserved,therelationshipbetweenNOse nsitivityandelectrode thicknessisconvolutedbecauseoftheNO/NO 2 equilibriumandthecompeting mechanismsofsurfaceadsorptionandelectrocatalysisatt heelectrode/electrolyte interface.Thetransitiontoanelectrocatalyticdominate dmechanismisapparentat 550 Catwhichsensitivitydecreasesandbecomesnearlyindepen dentofthickness above 10 m.At600 CallthesensorsareessentiallyinsensitivetoNOexceptth eone withthethinnestelectrode.Atthistemperatureadsorptio nofNOisnolongerobserved makingelectrocatalysisthedominantcontributiontothes ensormechanism.Thus,NO sensitivityshouldbegreatest,asobserved,forthethesen sorwiththethinnestelectrode sincethegaspathwaytotheinterfaceisshortest. 3.5Conclusions NO x potentiometricsensorswerestudiedinordertounderstand theeffect sensorelectrodethicknesshasonNO x sensitivity.NO 2 sensitivitywasattributedto electrocatalysisoccurringattheelectrode/electrolyte interfaceatboththesensing andcounterelectrodeat550-650 C.CatalyticreductionofNO 2 toNOhadtoalso beaccountedforattheLa 2 CuO 4 electrode.WhenNO 2 isconvertedtoNO,NO becomestheprimaryspeciestoreactattheelectrode/elect rolyteinterfaceand thusreducesNO 2 sensitivity.Ateachtemperature,NO 2 sensitivitydecreasedwith increasingthicknessduetoNO 2 beingincontactwithmoreLa 2 CuO 4 beforereaching theelectrode/electrolyteinterface.Inaddition,astemp eratureincreasedNO 2 sensitivity decreasedandcanbeattributedtoincreasedreductionofNO 2 asdescribedbyprevious NO 2 TPRexperiments. TherelationshipsbetweenNOsensitivityandthicknesswas indicativeofcompeting mechanisms.At400 C,thedominantmechanismistherelativechangeinFermi EnergyuponNOadsorptionandsurfacenitrate/nitriteform ation.Atthistemperature, 49

PAGE 50

sensitivitywaslarge( -17mV/decade)andindependentofthickness.Theresponse waslargeforthelowestNOconcentration,buttheresponsew assaturatedathigher NOconcentrations,indicativeofsaturationofNOadsorpti on.Astemperatureis increased,therelationshipbetweenNOsensitivityandthi cknessismoreconvoluted duetochangesintheNO 2 /NOequilibriumandthecompetingmechanismsofsurface adsorptionandelectrocatalysisattheeletcrode/electro lyteinterface.At550 C,the transitiontoanelectrocatalyticdominatedmechanismwas evidentasthesensitivity decreasedandbecamenearlyindependentofthicknessabove 10 m.At600 C,the dominantmechanismiselectrocatalysisattheelectrode/e lectrolyteinterface.However, allthesensorsareessentiallyinsensitiveexcepttheonew iththethinnestelectrode. Thus,therelationshipsbetweenNO x sensitivityandthicknessdemonstratemultiple sensingmechanisms,likeheterogeneouscatalysisandchan gesinsemiconducting properties,thataren'tincludedinMixedPotentialTheory Table3-1.Electrodethicknessmeasurements LayersAvg.Thickness( m)StandardDeviation 16.862.98212.707.66316.554.78432.884.10 Figure3-1.Schematicofsensorwithx=10to30 m 50

PAGE 51

Figure3-2.FESEMcrosssectionsofLa 2 CuO 4 sensingelectrode:a)1Layer(6.86 m) b)2Layers(12.70 m)c)3Layers(16.55 m)andd)4Layers(32.88 m) Figure3-3.NO 2 voltageresponseversustimeat550 C 51

PAGE 52

Figure3-4.NO 2 Voltageresponseversustimeat600 C Figure3-5.NO 2 voltageresponseversustimeat650 C 52

PAGE 53

Figure3-6.NOvoltageresponseversustimeat400 C Figure3-7.NOvoltageresponseversustimeat450 C 53

PAGE 54

Figure3-8.NOvoltageresponseversustimeat500 C Figure3-9.NOvoltageresponseversustimeat550 C 54

PAGE 55

Figure3-10.NOvoltageresponseversustimeat600 C 55

PAGE 56

Figure3-11.Voltageversuslog(NO 2 Concentration)forallthicknessesat550 C 56

PAGE 57

Figure3-12.Voltageversuslog(NO 2 Concentration)forallthicknessesat600 C Figure3-13.Voltageversuslog(NO 2 Concentration)forallthicknessesat650 C 57

PAGE 58

Figure3-14.Voltageversuslog(NOConcentration)forallt hicknessesat400 C Figure3-15.Voltageversuslog(NOConcentration)forallt hicknessesat450 C 58

PAGE 59

Figure3-16.Voltageversuslog(NOConcentration)forallt hicknessesat500 C Figure3-17.Voltageversuslog(NOConcentration)forallt hicknessesat550 C 59

PAGE 60

Figure3-18.Voltageversuslog(NOConcentration)forallt hicknessesat600 C Figure3-19.NO 2 sensitivityversuselectrodethickness 60

PAGE 61

Figure3-20.NOsensitivityversuselectrodethickness 61

PAGE 62

Figure3-21.NO 2 TPRoverLa 2 CuO 4 [ 6 ] 62

PAGE 63

Figure3-22.NO 2 TPDoverLa 2 CuO 4 [ 6 ] 63

PAGE 64

Figure3-23.NOTPDoverLa 2 CuO 4 [ 6 ] 64

PAGE 65

CHAPTER4 ELECTRODEAREA 4.1Introduction Theintentofthisworkistolookattheeffectsofvaryingthe electrodeareaand sensingandcounterelectrodeasymmetrywithrespecttoNO x sensitivityinorderto furtherunderstandsensingmechanism.Varyingelectrodea reachangeselectrode propertiesincludingtotalelectrodesurfaceareaandnumb erofreactionsitesatthe electrode/electrolyteinterfacewhichcanaffectNO x sensitivity.ThechangeinLa 2 CuO 4 areaalsoaffectstheasymmetrybetweenthesensingandPtco unterelectrodeswhich couldaffectsensitivityasshowninourworkwithelectrode conguration.Withrespectto engineeringdesign,sensorgeometryconsiderationslikee lectrodeareaareimportant, particularlyifminiaturizationisagoal(asinsensorappl icationsfortheautomobile industry). Electrochemicalimpedancespectroscopy(EIS)hasbeenpre viouslyusedasatool toinvestigateLa 2 CuO 4 basedNO x sensors[ 41 ].EISisamethodtohelpunderstand anddistinguishelectrochemicalprocessesthatoccuronso lidstateelectrochemical devicessuchasfuelcellsandsensors.Forexample,thehigh frequencyimpedance ofaLa 2 CuO 4 electrodeisrelatedtoitsp-typeconductivity.Thus,EISw asusedto investigatechangesoftheconductivityinthesensingelec trodeasaresultofsurface gasadsorption. 4.2Experimental 4.2.1PowderSynthesis La 2 CuO 4 wasselectedasthecandidateforthesensingelectrodeinth isresearch becauseofitsp-typesemiconductingresponseuponNOadsor ptionanditslackof catalyticactivityforNOreductionandoxidation[ 10 26 ].La 2 CuO 4 powderswere synthesizedviaanamorphouscitrategelcombustionmethod usingnitratesofLaand 65

PAGE 66

Cu[ 30 ].StructuralcharacterizationwasperformedusingX-rayD iffractiontoconrmthat noadditionalphases,suchasLa 2 O 3 orCuO,werepresentinthenalpowder. 4.2.2SensorFabrication Eachsensorwascomposedofacommerciallypurchased8-mol% yttria-stabilized zirconia(YSZ)electrolytewithdimensionsof13x20x0.1mm (Markatech).Sensorsfor testinginthepotentiometricmodewerepreparedwithasens ingelectrodeofLa 2 CuO 4 andacounterelectrodeofplatinum(Pt).Thesensingelectr odewasscreenprinted withaslurryconsistingofLa 2 CuO 4 powdermixedwithpolyethyleneglycolandethanol. Forpotentiometricsensingtests,sensingelectrodesofva riousareas(8x8mm,6x6 mm,4x4mm,and2x2mm)wereused.Theseelectrodesweredried for15minutes at120 Cafterprintingandthensubsequentlysinteredat800 Cfor10hoursaftera1 hourburnoutat400 C.Forthecounterelectrode,an8x8mmPtelectrodewasscree n printedusingacommercialPtpaste(Heraeus)ontheopposit esideoftheelectrolyte thanthesensingelectrode.ThePtelectrodewassinteredat 750 Cfor4hours.AfterPt inkwasusedtoattachPtwiretobothelectrodes,thesensors wereagainredat750 C foronehour.AschematicofthesensorisshowninFigure 4-1 SampleswerealsopreparedtoconductEIStests.Thesesampl esweresimilarto thesensorspreparedwiththeexceptionthattheelectrodes wereLa 2 CuO 4 onbothsides oftheYSZandsymmetricalinarea.Foursamplesweremadewit hthefollowingareas: 8x8mm,6x6mm,4x4mm,and2x2mm.Bothelectrodeswereco-re dat800 C for10hours.PtwireswereattachedwithPtpastewiththesam eprocedureusedabove. 4.2.3TestingParameters Sensorexperimentswereconductedinagas-owapparatus.G asenvironments wereestablishedusingmassowcontrollers.Forthepotent iometrictests,acustom LabViewprogramwasusedwhichallowedtemperaturesandgas owratestobe setautomaticallybyacomputer.Thegasenvironmentconsis tedofachangingstep concentrationofNOorNO 2 in3%O 2 ,balancedbyN 2 atatotalowrateof300sccm. 66

PAGE 67

ThestepconcentrationsofNOandNO 2 were0,50,100,200,and400ppm.Each concentrationwasheldfor200sandwerevariedintwohyster esisloops.Experiments weredoneateachtemperaturebetween400-700 Cat50 Cincrements.Thevoltage difference(opencircuitpotential)betweenthesensingan dcounterelectrodewas monitoredandrecordedusingaKeithleyMultimeter. EIStestingwasusedinordertoisolateadditionalelectroc hemicalphenomena notclearlyseeninsensortestsinthepotentiometricmode. Theimpedanceforthe symmetricalLa 2 CuO 4 sampleswasmeasuredusingaSolartron1260frequency responseanalyzer.Measurementsweretakenunderthecondi tionsof3%O 2 andwith theadditionof200ppmNOorNO 2 inabackgroundofN 2 .A20mV(rms)perturbation wasusedinthefrequencyrangeof100mHzto0.1MHz.Measurem entswerealso takenat50 Cincrementsbetween400-700 C. 4.2.4SensorCharacterization Asstatedearlier,changesinconductivityduetogasadsorp tionontheelectrode surfacecontributetothesensingmechanism.Thus,themicr ostructureandtotalvolume ofsurfacesitesarecriticalparameterstotakeintoconsid eration.Aeld-emission scanningelectronmicroscopewasusedtoobservethemicros tructureandtodetermine thethicknessofthesensorelectrodesafterpotentiometri candEIStestswere completed. 4.3Results 4.3.1SensorCharacterization FE-SEMmicrographsareshownforthepotentiometricsensor sandEISsamplesin Figures 4-2 and 4-3 ,respectively.Multiplethicknessmeasurements(atleast 10)were takenandtheaverageandstandarddeviationsareshowninTa bles 4-1 and 4-2 .As expected,themicrostructuresofalltheLa 2 CuO 4 electrodeswerequalitativelythesame. ThePtelectrodesweredeterminedtobe9.1 0.66 mthick.Whiletheintentwas uniformthickness,theLa 2 CuO 4 electrodesrangedfrom5-12 mwiththelargestarea 67

PAGE 68

samplehavingagreaterthicknessthantherestofthesample s.Theelectrodethickness oftheEISsamplesrangedfrom4-8 m.Thusinordertoverifytheintegrityofthedata, thepotentiometricsensitivityresultswerecomparedvers ustheoverallvolumeofthe La 2 CuO 4 electrodeinadditiontotheareaoftheelectrode. 4.3.2PotentiometricTests Withinthemeasurementrangeof400-700 C,signicantNO 2 andNOsensitivity wasobservedat500-600 Cand400-500 C,respectively.Thevaluesofthevoltage responseareverysmallabovethisrange,andwhilethevolta geislargetheresponseis unstableandsaturatesinthehigherNO x concentrationsabove50ppmattemperatures belowthisrange.Thistemperaturedependenceisinagreeme ntwithpreviouswork donewithNO x sensorswithLa 2 CuO 4 asthesensingelectrode[ 37 ].Anexampleofthe responseisshowninFigure 4-4 whichdisplaysthevoltagechangeswithcorresponding changesinconcentrationsofNO 2 forsensorswitheachofthevaryingelectrodeareas at550 C. Ifthevoltagesareplottedversusthecorrespondingconcen trationsonalogarithmic scaleasinFigure 4-5 ,alinearrelationshipresults.Thevoltagesintheseplots arean averageoftheresponseateachconcentration.Theaveragev oltagewascalculated from4voltagesmeasuredateachconcentrationof50,100,an d200ppmNO x .Twoof thesevoltagesweremeasuredduringtheupwardstepincreas esofNO x whiletheother 2weremeasuredduringthedownwardstepsofthegascycle.Th eaveragevoltagefor 400ppmwascalculatedfrom2measurements.Theerror(stand arddeviationofthe calculatedaverages)isshownasverticalerrorbars.Theli neartwascalculatedbya minimizationoferrortechniquethroughthestandarddevia tionofeachpoint.Theslope ofthistprovidesthemetricofsensitivitythatthesensor scanbecomparedby.Similar graphsfortheresponsetowardsNOat500 CareshowninFigures 4-6 and 4-7 TheNO 2 andNOsensitivitieswererecordedforeachsensorforthete mperatures 500-600 Cand450and500 C,respectively.Thesensitivitieswereplottedversus 68

PAGE 69

areaforeachtemperatureandareseeninFigures 4-8 and 4-9 .Theverticalerror barsrepresentthestandarddeviationingeneratingthelin eartinwhichtheslope (sensitivity)wastaken.TheNO 2 andNOhavevoltageresponsesthatareoppositein sign,positiveforNO 2 andnegativeforNO.Thesensitivityinbothcases,decrease swith anincreaseintemperature.Thesearebothtrendsthatarety picallyseeninmostsolid statesensorsthatuseoxidesaselectrodes[ 48 49 ]. InthecaseofNO 2 ,adifferentrelationshipbetweensensitivityandareaiso bserved foreachtemperature(Fig. 4-8 ).At500 C,therelationshipisparabolicwhileat600 C,it islinear.Anintermediatetypeofrelationshipisobserved at550 C. ForNO,anon-linearrelationshipwithaminima(maximumsen sitivity)isobserved at400and450 C,andanearlylinearrelationshipat500 CforNO(Fig. 4-9 ).At400 C theresponseisactuallylargefor50ppm[NO].However,ther esponsesaturatesand doesnotincreasemuchatlargerconcentrations,resulting inanoverallsmallerNO sensitivitycomparedto400 C. Themaximuminsensitivityispresentat 30mm 2 ( 22mV/decade[NO 2 ])at600 C andat16mm 2 ( -22mV/decade[NO])at450 CforNO 2 andNO,respectively.Atthe highesttemperatureofoperationsensitivitydecreaseswi thareaforbothNO 2 andNO. Thechangesinpotentialcanbeattributedtoelectrocataly sis,heterogeneouscatalysis, andchangesinthesemiconductingpropertiesofLa 2 CuO 4 duetotheinteraction betweenNO x gasspeciesandsensingelectrode.Thesecontributionswil lbediscussed furtherintheanalysissection. Eachsensorelectrodewasdepositedbyscreenprintingasin glelayerofLa 2 CuO 4 withtheintentofcontrollingthickness.Inordertocompen sateforslightvariationsin electrodethicknessseeninposttestingFESEMcrosssectio nanalysis,NO x sensitivity wasalsoplottedversuselectrodevolume(Figures 4-10 and 4-11 ).Thesimilaritiesto Figures 4-8 and 4-9 conrmthatthethicknessvariationshadminimaleffectont he sensitivityversusareabehavior. 69

PAGE 70

4.3.3ImpedanceTesting Figure 4-12 isanexampleoftheentireimpedancespectraforasymmetric al La 2 CuO 4 samplewithanareaof64mm 2 whenexposedto3%O 2 andwiththeaddition of200ppmNO 2 and200ppmNOat550 C.Figure 4-13 focusesonthehighfrequency endofthatspectra.AlthoughitappearsinFigure 4-12 thatthespectraintersectsthe x-axis(realimpedance)athighfrequency,Figure 4-13 showsthattherecouldbean additionalelectrochemicalprocessnotclearlyresolveda tthisfrequencyrange.Higher frequenciesabove0.1MHzwerenotusedbecauseofthedecrea seinaccuracyat higherfrequenciesforthesesamples.Therefore,theimped anceat0.1MHzwasusedto denethehighfrequencyimpedance. Atthehighfrequencyendthereisadecreaseinimpedancewit hNO 2 exposureand anincreasewithNO(Fig. 4-13 ).Inourpreviousimpedancework,asimilarsymmetrical cellwith64mm 2 La 2 CuO 4 electrodeshadasignicantlylargerhighfrequencyinterc ept in3%O 2 ( 250 n /cm 2 )thanitsPtsymmetricalcellcounterpartin3%O 2 ( 25 n /cm 2 ) at500 C.Inaddition,thehighfrequencyimpedanceforthePtsymme tricalcelldid notchangeuponintroductionto200ppmNO 2 andNO,whereasthehighfrequency interceptchangedfortheLa 2 CuO 4 cellwithNO 2 andNO.Thisindicatesthatthe La 2 CuO 4 electrodeisthemaincontributortothehighfrequencyimpe danceandthe changeswithNO x exposurecanberelatedtochangesinLa 2 CuO 4 'sconductanceupon gasadsorption[ 41 ]. Incontrast,thereisadecreaseinthelowfrequencyimpedan ceuponexposure tobothNO 2 andNO(Fig. 4-12 ).Thetotalimpedanceincludescontributionsatlow frequencywhichtypicallyarerelatedtomasstransportpro cessessuchasgasdiffusion andadsorptionprocessesthatdonotnecessarilychangethe electronicstateofthe materialviachargetransfer. 70

PAGE 71

Inordertomakecomparisonsoftheimpedancevalues,theres ultswerenormalized accordingtoEquation 4–1 ,where Z o istheimpedanceunder3%O 2 and Z NO x isthe impedancewiththeadditionof200ppmNO 2 orNO.Thenormalizedimpedance( Z n ). Z n = Z NO x Z o (4–1) Thenormalizedhighfrequencyimpedance( Z n hf )isdenedastheimpedance measuredat0.1MHzwithexposuretoNO x ( Z NO x hf )dividedbythecorresponding impedancemeasuredin3%O 2 ( Z o hf )(Equation 4–2 ). Z n hf = Z NO x hf Z o hf (4–2) Thenormalizedtotalelectrodeimpedance, Z n tot ,is Z NO x tot dividedby Z o tot ,where totalimpedanceisdenedasthedifferencebetweenthelowf requencyandthehigh frequencyrealimpedanceintercepts( Z lf Z hf )in200ppmNO x ( Z NO x tot )and3%O 2 ( Z o tot )(Equation 4–3 ). Z n hf and Z n tot wereplottedversustheareaoftheLa 2 CuO 4 electrodeforNO 2 andNOexposure.Avalueofunityfor Z n wouldindicatenochange intheimpedancewithexposuretoNO x .Avaluegreaterthanunitywouldindicatean increaseinimpedancewhileavaluelessthanunityadecreas einimpedance. Z n tot = Z NO x tot Z o tot = Z NO x lf Z NO x hf Z o lf Z o hf (4–3) 4.3.3.1Highfrequencyimpedance The Z n hf resultsareshownforNO 2 andNOinFigures 4-14 and 4-15 ,respectively. Asnotedbefore,NO 2 causesadecreaseinimpedancewhileNOcausesanincrease. Thesechangesreectachangeintheconductivityfromthead sorptionofNO x on thesurfaceresultinginarelativechangeinFermiEnergy.A tthelowesttemperature (500 C),valuesof Z n hf forNO 2 deviatethegreatestfromunityandshowthegreatest dependenceonelectrodearea.At600 C, Z n hf isclosetounityandnearlyindependent 71

PAGE 72

ofarea.Aninectionpointispresentat 30mm 2 foreachtemperature.Thisphenomenon ismostpronouncedat500 C,whereNO 2 potentiometricsensitivityisamaximumat 30mm 2 .InthecaseofNO(Fig. 4-15 ),therelationshipbetween Z n hf andareaat 400and450 Cisnon-linearwithamaximaatanelectrodeareaof36mm 2 .At500 C, Z n hf decreaseswithareainarelationshipthatisnearlylinear. Thelinearbehavior at500 Candnon-linearresponseatlowertemperaturesarealsocon sistentwith thepotentiometricresponse.However,themaximainpotent iometricsensitivityand impedanceoccuratdifferentelectrodeareas.4.3.3.2Totalelectrodeimpedance The Z n tot resultsareshownforNO 2 andNOinFigures 4-16 and 4-17 ,respectively. Incontrastto Z n hf ,allvaluesof Z n tot arelessthan1forNO 2 andNO,duetoadecrease inimpedancewithNO x exposure..Therelativechangeintotalelectrodeimpedanc eis greaterthanthehigherfrequencycounterparts;thevalues ofZ n tot approach0.2and0.1 forNO 2 andNO,respectivelyatthelowertemperatures.Thisbehavi orisinagreement withpreviouswork[ 41 ]. Ateachtemperature(Fig. 4-16 ),Z n tot withNO 2 exposureapproaches1as electrodeareaincreases.Foreachelectrodearea,Z n tot withalsoapproaches1as temperatureincreases.Thegreatestdecreaseinimpedance (smallestvalueofZ n tot ) isobservedat500 C,andZ n tot isleastdependentonareaatthehighesttemperature, similartothe Z n hf behaviorforNO 2 (Fig. 4-14 ).Incontrastto Z n hf ,thereisnoinection pointat30mm 2 TherelationshipforNO(Fig. 4-17 )isnon-linearwithaminimaatanelectrodearea of16mm 2 for450and500 C.ItshouldbenotedthatZ n tot forNOat400 Cwasnot includedastheimpedanceatthelowestfrequencydidnotcro sstherealimpedance axisandanassociatedcircuitcouldnotbetwithoutsigni canterror.Thenon-linear behaviorforZ n tot andNOpotentiometricsensitivityareverysimilarat450 Cwith 72

PAGE 73

identicalminima(16mm 2 )correspondingtobothmaximumdeviationinimpedancewith NOexposureandmaximumpotentiometricsensitivity. 4.4Analysis “MixedPotential”iscommonlyproposedinordertoexplaint henon-Nernstian potentialgeneratedbysolidstatepotentiometricsensors .Thistheoryclaimsthe potentialisgeneratedbyadifferenceinthereactionrates ofcorrespondingredox reactionpairsestablishedoverthetwoelectrodes,asseen incorrosion[ 50 – 53 ].The potentialofeachelectrodecanberelatedtoeitherthecurr entorcurrentdensity associatedwiththeredoxreactionpairasextrapolatedfro mTafelplots. Theoretically,mixedpotentialisoftendescribedinterms ofcurrentdensity[ 54 55 ]. However,currentisusuallyplottedversuspotentialinexp erimentalTafelplotsinorderto extractthemixedpotential[ 4 25 54 – 59 ].Intermsofcurrentdensitythemixedpotential, andthusthesensorresponse,shouldbeindependentofelect rodearea.Whereasin termsofcurrent,themixedpotentialshouldincreaselogar ithmicallyinmagnitudewith electrodearea.However,weobservethatNO 2 andNOdecreaseinsensitivitylinearly withelectrodeareaathightemperatures,whilesensitivit yisnon-linearlydependentwith areaatlowertemperatures. Therefore,theremustbeothercontributionstothesensing mechanism,as proposedinDifferentialElectrodeEquilibria[ 26 ].Thistheoryexplainsthatapotentiometric sensorgeneratesavoltagebecauseofsomeformofasymmetry betweentwo electrodesinthesameatmosphereincludingotherprocesse ssuchasheterogeneous catalysisandgasadsorption.Forexample,resistivesenso rsworkontheprincipleofa changeinsemiconductingpropertiesduetogasadsorption[ 60 – 63 ].Theheterogeneous catalyticactivityandadsorptionbehaviorofLa 2 CuO 4 havebeenexploredbyTemperature ProgrammedReaction(TPR),TemperatureProgrammedDesorp tion(TPD),andX-ray PhotoemissionSpectroscopy(XPS)[ 6 32 33 ]. 73

PAGE 74

Instudyingtheadsorptiveandcatalyticpropertiesofsemi conductingoxidesin general,andLa 2 CuO 4 specically,forpotentiometricNO x sensorapplicationswehave foundconsistentlythatmaximumsensitivityisobservedat temperaturescorresponding tothetailendofthedesorptionpeakinTPDs[ 6 31 – 33 ].Atlowertemperature sensitivityislimitedbysaturationoftheoxidesurfacewi thchemisorbednitrateand nitritespecies.Whileathighertemperaturethekineticso fdesorptionaregreaterthan adsorption,andthesensingmechanismtransitionsfromase miconductingresponse duetotheadsorbedspeciestoamechanismthatismorecataly ticorelectrocatalyticin nature. Thatsametemperaturedependenceisobservedhereinbothth epotentiometric andimpedancedata.Atthehighesttemperatureofpotentiom etricsensitivity(600 C forNO 2 and500 CforNO)boththehighfrequency(bothNO 2 andNO)andtotal impedance(NO 2 )haveminimalsensitivityindicatingsemiconductorbehav iorhas minimalimpactonpotentiometricsensorresponse.Moreove r,weobservealinearly decreasingpotentiometricsensitivitywithelectrodeare a,whileboththehighfrequency (bothNO 2 andNO)andtotalimpedance(NO 2 )areessentiallyareaindependent. Incontrast,atthelowesttemperaturesofpotentiometrics ensitivity(500 Cfor NO 2 and400-450 CforNO)boththehighfrequencyandtotalimpedanceexhibit high sensitivitytobothNO 2 andNOindicatingsemiconductorbehaviorisamajorcontrib utor topotentiometricsensorresponse.Moreover,atthesetemp eraturesweobservehighly non-linearelectrodeareadependenceforbothpotentiomet ricsensitivityandhigh frequencyandtotalimpedance.Notonlythat,buttheelectr odeareathatexhibited maximumpotentiometricNO 2 sensitivity(30mm 2 )wastheinectionpointintheNO 2 highfrequencyimpedancedataindicatingacommonalityina mechanisticchangeat thisarea.Similarly,forNOpotentiometricresponseamaxi mumsensitivitywasobserved at16mm 2 whichdirectlycorrespondedtotheelectrodeareaformaxim umNOtotal impedancesensitivity.Clearlytheeffectofareaonpotent iometricresponseinthelower 74

PAGE 75

temperatureregionisrelatedtotheimpactofsemiconducti ngbehavioronthesensing mechanism. Theelectrodesinthisstudyweresimilarinthickness( 5 m)tothethinnest electrodeinthethicknesschapterinordertominimizethis effect.However,at600 C, NO 2 sensitivitydecreasedlinearlywithincreasingareawhich mayindicatethatthe greaterareastillcontributestogreaterheterogenouscat alyticreductionofNO 2 toNO. Asnotedearlier,whileMixedPotentialcouldpredictanare aeffectonsensitivity,itwould bealogarithmicincreaseasopposedtoalinearincreaseass hownhere. 4.5Conclusions Theworkpresentedherefurtherstheunderstandingoftheme chanismof non-NernstianNO x sensinginsolidstatepotentiometricsensorsbyexhibitin g mechanisticcontributionsnotincludedinMixedPotential Theory.Thiswasachieved bytestingtheeffectofthesensorelectrodearea,animport antengineeringdesign parameter,onNO x sensitivity.SensorsstudiedinthisworkshowedNO 2 sensitivity inthetemperaturerangeof500-600 C.TherelationshipbetweenNO 2 andelectrode areawasparabolicat500 Candlinearat600 C.Atthelowesttemperature500 C boththehighfrequencyandtotalimpedanceexhibithighsen sitivitytoNO 2 indicating semiconductingbehaviorisamajorcontributortothepoten tiometricsensorresponse, whileat600 Cthehighfrequencyandtotalelectrodeimpedancehavemini mal sensitivity.Instead,NO 2 sensitivityat600 Cwaslinkedtoamechanismmorecatalytic orelectrocatalyticinnature.Thelineardecreasewascaus edbythecatalyticreduction ofNO 2 toNObyLa 2 CuO 4 .At550 C,bothofmultiplemechanisticphenomenaare likelyoccurringandaccordinglythebehaviorofsensitivi tywithelectrodeareashowsa resultingmixtureofparabolicandlinearbehavior. NOsensitivityshowedauniquedependenceofareadepending ontemperature aswell.At400and450 Ctherelationshipisnon-linearwithaminimaofmaximum sensitivity( -19mV/decadeand -22mV/decade,respectively)atanelectrodearea 75

PAGE 76

of16mm 2 andcorrespondswelltothehighNOsensitivityinhighfrequ encyandtotal electrodeimpedance.Thus,thebehavioratthesetemperatu rescanalsobeattributedto changesinsemiconductingpropertiesofLa 2 CuO 4 .At500 C,NOsensitivitydecreases linearlywithareawhilehighfrequencyimpedanceshowsmin imalNOsensitivity indicatingminimalcontributionbychangesinsemiconduct ingproperties. Table4-1.Electrodethicknessmeasurementsforpotentiom etricsensorelectrodes ElectrodeArea(mm 2 )Avg.Thickness( m)StandardDeviation 45.411.23 165.621.13365.080.986411.404.49 Table4-2.Electrodethicknessmeasurementsforimpedance samples ElectrodeArea(mm 2 )Avg.Thickness( m)StandardDeviation 45.520.21 167.911.85364.220.71646.450.78 Figure4-1.Schematicofsensorwithx=8,6,4,2mm 76

PAGE 77

Figure4-2.FE-SEMcrosssectionsofLa 2 CuO 4 sensingelectrodesforpotentiometric sensors:a)4mm 2 x5.41 mb)16mm 2 x5.62 mc)36mm 2 x5.08 mand d)64mm 2 x11.40 m 77

PAGE 78

Figure4-3.FE-SEMcrosssectionsofLa 2 CuO 4 electrodesforimpedancetesting:a)4 mm 2 x5.52 mb)16mm 2 x7.91 mc)36mm 2 x4.22 mandd)64mm 2 x 6.45 m Figure4-4.NOvoltageresponseversusTimeforallareasat5 50 C 78

PAGE 79

Figure4-5.Voltageversuslog(NOConcentration)forallar easat550 C Figure4-6.NOvoltageresponseversusTimeforallareasat5 00 C 79

PAGE 80

Figure4-7.Voltageversuslog(NOConcentration)forallar easat500 C Figure4-8.NO 2 sensitivitiesversusLa 2 CuO 4 electrodeareaat500,550,and600 C 80

PAGE 81

Figure4-9.NOsensitivitiesversusLa 2 CuO 4 electrodeareaat400,450,and500 C Figure4-10.NO 2 sensitivitiesversusLa 2 CuO 4 electrodevolumeat500,550,and600 C 81

PAGE 82

Figure4-11.NOsensitivitiesversusLa 2 CuO 4 electrodevolumeat400,450and500 C 82

PAGE 83

Figure4-12.Impedancein3%O 2 ,200ppmNO 2 ,andNOat550 C 83

PAGE 84

Figure4-13.HighfrequencyImpedancein3%O 2 ,200ppmNO 2 ,andNOat550 C. 84

PAGE 85

Figure4-14.NormalizedhighfrequencyNO 2 impedanceversusLa 2 CuO 4 versus500, 550,and600 C Figure4-15.NormalizedhighfrequencyNOimpedanceversus La 2 CuO 4 at400,450, and500 C 85

PAGE 86

Figure4-16.NormalizedtotalNO 2 impedanceversusLa 2 CuO 4 at500,550,and600 C Figure4-17.NormalizedtotalNOimpedanceversusLa 2 CuO 4 at450and500 C 86

PAGE 87

CHAPTER5 ELECTRODECONFIGURATION 5.1Introduction Inthischapter,theeffectofelectricalcontactcongurat ionfortheLa 2 CuO 4 sensing electrodesonthesensorresponseisinvestigated.Althoug hthisdesignaspectisoften overlooked,itprovedtohaveaneffectonsensorsensitivit y,selectivity/crosssensitivity, andrepeatability. 5.2Experimental 5.2.1PowderSynthesis La 2 CuO 4 wasselectedasthecandidateforthesensingelectrodeinth isresearch becauseofitsp-typesemiconductingresponseuponNO x adsorption.Inaddition,itis non-catalyticforthereductionofNO[ 26 ]andhighlycatalyticforthereductionofNO 2 andoxidationofCO[ 6 ].La 2 CuO 4 powdersweresynthesizedviaanamorphouscitrate gelcombustionmethod[ 30 ].SolutionsofnitratesofLaandCuwereprepared.The concentrationsofthesesolutionsweredeterminedusingIn ductivelyCoupledPlasma Spectroscopy(ICPS,Perkin-ElmerPlasma3200). Appropriateamountsofthesolutionweremixedwithcitrica cidinordertoobtain thestochiometricratioofLaionstoCuionsandamolarratio ofcitricacidtonitrateions equalto0.22.Themixturewasheatedto100 Cuntilmostofthewaterinthesolution evaporated,leavingbehindagel.Thisgelwascalcinedfor1 0hoursat650 C. 5.2.2SensorFabrication Eachsensorwascomposedofacommerciallypurchased8-mol% yttria-stablized zirconia(YSZ)electrolyte(Marketech).Theseelectrolyt eshaddimensionsof13 x20x0.1mm.Tofabricatethesensingelectrode,ascreen-pr intableslurrywas preparedbymixingLa 2 CuO 4 powder(17.5%solidsloading)withpolyethyleneglycol andethanol.Thisslurrywasusedtoscreenprinta8x8mmelec trodeononesideof aYSZelectrolyte.Theelectrodewasdriedat120 C,followedbya1hourburnoutat 87

PAGE 88

400 C,andthensinteredat800 Cfor10hours.Thisprocedurewasrepeatedon14 additionalYSZelectrolytes. A8x8mmPtcounterelectrodewasscreenprintedontheopposi tesideofeach oftheYSZelectrolytesusingcommercialaPtPaste(Heraeus )indirectalignmentwith theLa 2 CuO 4 electrode.ThePtelectrodeswereburnedoutfor1hourat400 Candthen redat750 Cfor4hours.Figure 5-1 showsaschematicofthesensorswithaSEM micrographofthesensorcrosssection. The15sampleswerethenseparatedinto5groupsof3toprepar ethedifferent congurationsofthesensingelectrode.Thedifferencesin the5differentcongurations aredescribedbelowandshowninFigure 5-2 .AfterPtorLa 2 CuO 4 inkwasusedto attachPtwiretotherespectiveelectrodes,thesensorswer eagainredat750 Cfor onehour.5.2.2.1Conguration1 PtinkwasusedtoattachaPtwiretothePtcounterelectrodew hiletheLa 2 CuO 4 slurrywasusedtoattachaPtwiretotheLa 2 CuO 4 electrode. 5.2.2.2Conguration2 PtinkwasusedtoattachPtwirestoboththePtcounterelectr odeandLa 2 CuO 4 sensingelectrode.5.2.2.3Conguration3 Conguration3isidenticaltoConguration1exceptthatit hasanadditional4x4 mmPtelectrodescreenprintedontheYSZelectrolyteadjace nttotheLa 2 CuO 4 sensing electrode.5.2.2.4Conguration4 Conguration4isidenticaltoConguration1exceptthatit hasanadditional4x4 mmPtelectrodescreenprintedontopoftheLa 2 CuO 4 sensingelectrodewithoutmaking contactwiththePtwire. 88

PAGE 89

5.2.2.5Conguration5 PtinkwasusedtoattachaPtwiretothePtcounterelectrode. A4x4mmPt electrodewasscreenprintedontopoftheLa 2 CuO 4 electrode.APtwirewasattached tothe4x4mmPtelectrodeusingPtink.5.2.3TestingParameters Sensorexperimentswereconductedinatypicalgas-owappa ratus.Gas environmentswerecontrolledusingmassowcontrollers.A customLabViewprogram wasusedwhichallowedtemperaturesandthemassowcontrol lerstobeautomatically controlledbyacomputer.Sensorswereexposedto3%O 2 balancedbyN 2 .Thetotal owratewassetataconstant300sccm.NO 2 ,NO,andCOeachwereexposedtothe sensorsinthefollowingconcentrations:0,50,100,200,40 0,and650ppm,holding eachconcentrationfor200s.Thegasconcentrationswerest eppedforeachgasintwo hysteresistypeloopsateachtemperaturebetween400-700 Cat50 Cincrements.The orderofthegasestestedateachtemperatureisdisplayedin Figure 5-3 Thevoltagebetweenthesensingandcounterelectrodewasme asuredduringthe stepchangesusingaKeithleyMultimeter.Foreachsensor,t hesensingelectrodewas connectedtothepositiveterminalofthemultimeterwhilet hecounterelectrodewas connectedtothenegativeterminal.Thedesignofthetestin gapparatusallowedfortwo sensorstobetestedsimultaneously.Thus,thersttwosamp lesofeachconguration weretestedtogether.Foreachconguration,thethirdsamp lewasthentestedalong withtherstsensor.Testingonesensortwicefromeachcon gurationallowedoneto lookatthesensor'sbehaviorbasedongashistory.Furtherm ore,itshouldbenotedthat allelectrodesfrombothsensorswereinthesamegasatmosph ere,withnoreference gasused. 89

PAGE 90

5.3ResultsandDiscussion 5.3.1SensorResponse Thetypicalsensorresponse(Conguration2at600 C)forallthreedifferentgases (NO 2 ,NO,andCO)canbeseeninFigure 5-4 .Theguresshowthechangeinvoltage (measuredbetweenthesensingelectrodeandcounterelectr ode)withtimeasthegas concentrationvaried. Ifthevoltagesareplottedversusalogplotofgasconcentra tionalinearrelationship isvisibleasseeninFigure 5-5 .Fromthisgraph,thesensitivityofthesensorforeach particulargasiscalculatedfromtheslopeofeachline. Thesensitivitiesofthesensorswerecalculatedandanalyz edforeachconguration intherangeof550-650 C.Thisisthetemperaturerangethateachsensorofeach congurationhadastableresponseforeachgastested.5.3.1.1Sensitivity Foreachtemperatureonecanmakesensitivitycomparisonsf orthedifferent congurations.ThisisseenintheNO 2 ,NO,andCOsensitivityvs.temperatureplots (Figures 5-6 5-8 )fortherstsensoroutofthethreemadeforeachcongurati on(i.e., Sensor1).Ineachofthesegures,theslopedescribesthech angeofsensitivitywith changeoftemperature.Forexample,ifthesensitivityispo sitive,apositiveslope indicatesthatthesensitivityincreaseswithincreasingt emperaturewhileanegative slopecorrespondstoadecreaseinsensitivitywithincreas ingtemperature.Theopposite isobservedforsensorswithanegativesensitivity. Dependingontheconguration,thechangeinNO 2 sensitivitywithtemperature, asseeninFigure 5-6 ,displaysbothpositiveandnegativechangesinslope.The mostsimplecase,Conguration1(C1),showsaanegativeslo pewiththeincreasein temperature.Thisdecreaseinsensitivityisusuallyseeni nmostsolidstatesensorswith metaloxidesensingelectrodes.C2,C4,andC5displayaposi tiveslopebetween550 and600 Candanegativeslopebetween600and650 C.Eachofthesecongurations 90

PAGE 91

havesometypeofPt/La 2 CuO 4 interfaceinadditiontothePtwire/sensingelectrode interfacepresentineachofthecongurations.C3,themost distinctcongurationwith anadditionalareaofPtscreenprintedadjacenttothesensi ngelectrode,deviatedthe mostfromtherestofthecongurationsinthatitproducedan egativesensitivity(with apositiveslope).ThedifferencesinsensitivityforNO 2 ismostlikelyduetothemore complexnatureofNO 2 interactionwithLa 2 CuO 4 .Thiswillbediscussedfurtherinthe analysissection. TheslopesforNOandCOsensitivity,Figures 5-7 and 5-8 ,indicateadecreasein sensitivitywithincreasingtemperature.Thissuggeststh atthesensorcongurationplays alesssignicantroleinaffectingsensitivityathigherte mperatures.Asinthecasefor NO 2 ,C3deviatesfromtherestofthecongurationshavingaNOse nsitivityofopposite signtotheothercongurations.Accordingtosemiconducto rtheory[ 16 20 44 ],one wouldexpectNO 2 tohaveasensitivityoftheoppositepolaritythanthatofNO ,aswas thecaseforallcongurations.InthecaseofCO,thecongur ationthatdeviatedthe mostfromtherestwasC2.5.3.1.2Selectivity/crosssensitivity Agassensorisselectiveifitrespondstoaparticulargasbu tnottoothers.One mayneedtodetectNO 2 withoutitssensitivitybeingaffectedbychangesinanothe rgas, suchasCO.Theissueofselectivityisparticularlyimporta ntforNO x sinceNOisfound inmuchlargerconcentrationsthanNO 2 incombustionengines. Theselectivityisquantitativelydescribedastheratioof thesensorsensitivityfor onegastothesensitivityofanother.Figures 5-9 5-11 showhowelectrodeconguration alterssensorselectivityforNO 2 ,NO,andCO.Anegativeratiosimplyindicatesthatthe sensitivityforeachgaswasoppositeinpolarity.Figure 5-9 showsthatallcongurations showgoodselectivitytowardsNO 2 withrespecttoNO,particularlyat650 Cwhereall congurationsshowedlittlesensitivitytowardsNO.Theda tafollowsasimilartrendfor allcongurationsexceptat650 CwheretheratioispositiveforC1.Atthistemperature, 91

PAGE 92

theNOsensitivityswitchestoaverysmallpositivevalue.T heseresultssuggestthat whenusingLa 2 CuO 4 asasensingelectrode,operationtemperaturecanbeusedas a parametertocontrolsensorselectivitybetweenNO 2 andNO.Thisprinciplehasbeen demonstratedinamultifunctionalsensorarray[ 64 ]. TheresultsinFigures 5-10 and 5-11 implythatthepresenceofaPtelectrode ontopoftheLa 2 CuO 4 electrode(C4andC5)improvestheselectivityforNO x inthe presenceofCO.WhilethePtlayerdoesnotdrasticallyimpro veNO x sensitivity,it effectivelymakesthesensorinsensitivetoCOresultingin thehighNO x selectivity. AlthoughtheCOsensitivityislowforC4,theresponseswitc hesdirectionbetween550 and600 CcausingtheswitchinthesignoftheNO x selectivityatthesetemperatures. 5.3.1.3Repeatability Anotherimportantparameter,particularlyinregardstoco mmercializingsensor technology,isrepeatability.Repeatabilitycanrefertot woaspects:abilityforone sensortoproducesimilarresultsafterrepeatedtesting(r untorun)andtheabilityto manufacturemultiplesensorsthatgivesimilarresultstoe achother(sampletosample). Ifsensitivitycanberelatedtotheaccuracyofthesensor,r epeatabilityshouldthenbe relatedtotheprecision. Figures 5-12 5-14 showhowtestingeachsensorasecondtimeaffectedthe sensitivity.Onegeneraltrendisthatthesensitivityfore achgasusuallydecreasesupon therepeatedtests.Inseveralinstances,thedirectionoft heresponseevenswitched. ThisisespeciallytrueforC4whichchangeddirectionateac htemperatureforNO 2 andNO.Nevertheless,C2andC3havethebestrepeatabilityperformanceintermsof havingtheleastdegradationinsensitivity,withoutchang ingresponsedirectionfora particulargas.Ontheotherhand,C4andC5showtheworstrep eatableperformance. Intheirwork,Penzaetal.[ 65 ]notedthatlongtermstabilitywasanissuefortheWO 3 electrodeswhichhadaPtlayerontopofthem. 92

PAGE 93

Thetwoadditionalsamplesfromeachcongurationwerealso testedinthesame conditionsastherstsensorinordertotestsampletosampl erepeatability.Results showedthattwoadditionalsensorswithinagivencongurat ionproducedsimilar,butnot identical,resultstotherstsensorasshowninFigure 5-15 .Thus,furtherimprovement andcontrolinsensorfabricationisneededtoensuresample tosamplerepeatability beforethesesensorscanbemadeforlargescalecommerciali zation. 5.3.2SensorMechanism Someinsightintothesensorbehaviorinthisworkmaybedraw nfromunderstanding theconceptof“DifferentialElectrodeEquilibria”[ 26 ].Thisconceptexplainsthatthe signalgeneratedbythesensoristhedifferenceinpotentia lbetweenthesensing electrodeandcounterelectrode.Thepotentialoneachelec trodeisaffectedby gas/electrodeinteractionprocesseslikeelectrochemica lreactions,electrocatalytic activities,and/orchangesinsemiconductingpropertiesd uetogasadsorption[ 66 ]. SinceLa 2 CuO 4 isap-typesemiconductorwithsomecatalyticactivitytowa rdsCO oxidationandNO 2 reduction,alltheseprocessesmustbetakenintoconsidera tion. Ingeneral,eachoftheseprocessesoccuratspecictempera turesforeachsensing material;however,atagiventemperaturemorethanonecanc ontributetoeach electrodepotential.Thepotentialdifferencearisesfrom havingtwoelectrodeswith someformofasymmetry,inmostcasesadifferenceinmateria l. Thedifferenceinsensorelectrodecongurationcouldfurt hercontributeto asymmetry,andthusthesensorsignal.Inordertorationali zetheeffectsofconguration, onemustremembertheinherentpropertiesofLa 2 CuO 4 mentionedaboveandexceptfor C1thecontributionsduetotheadditionofPt,whichalsosho wscatalyticactivity. ThesensormechanismforLa 2 CuO 4 basedNO x andCOpotentiometricsensors haspreviouslybeeninvestigatedviatemperatureprogramm edreactions(TPR) andtemperatureprogrammeddesorption(TPD)experiments[ 31 ].Inaddition,the compositionoftheNO x adsorbatesonLa 2 CuO 4 andthemechanismofoxygen 93

PAGE 94

exchangebetweenNO x andLa 2 CuO 4 hasbeenstudiedextensivelyusingInfrared/X-ray PhotoemissionSpectroscopy[ 32 ]andTPR/TPDexperimentswithisotopicallylabeled O 2 [ 33 ],respectively.NOsensitivitywasattributedtoNOadsorp tionandtheassociated changeofpotentialwithrespecttotheFermiEnergylevelat thesurface.Thischange iscausedbytheremovalofelectronsfromtheLa 2 CuO 4 bulkwhentheadsorbedNO interactswithlatticeoxygentoformthenitritecomplex(N O 2 )onthesurface.Thesame canbesaidaboutNOsensitivityinthiswork.ThenatureofNO 2 ismorecomplicated. NO 2 beginstoreducetoNOoverLa 2 CuO 4 at350 Candcompletelydecomposes by600 C.However,NO 2 behavessimilarlyoverPt.Thus,thesensingmechanism cannotbeattributedtoacatalyticdifferencebetweenelec trodesofPtandLa 2 CuO 4 Rather,NO 2 isadsorbedontothesurfaceofLa 2 CuO 4 similarlyasNO,alsoresultingin achangeofpotentialwithrespecttotheinFermiEnergyleve l.However,whenanitrite isformedwithNO 2 adsorptionelectronholesareproducedwithnoLa 2 CuO 4 lattice oxygenexchange.Thusasexpected,theresponseofNO 2 ispositiveandopposite tothatoftheNOresponse.NO 2 adsorbatescanalsofurtherformanitratecomplex (NO 3 ).Dependingonthereactionpathwayofnitrateformation,t hesurfaceconductivity ofLa 2 CuO 4 caneitherincreaseordecrease.Theoverallvoltageisthus dependenton therelativeamountofnitriteandnitrateformation.Thisc omplexityhelpsexplainthe behaviorseeninFigure 5-6 COshowsbothadsorption/desorptionbehavioraswellasoxi dationtoCO 2 over La 2 CuO 4 above300 C.COalsooxidizesoverPthowevertoalesserdegreethanove r La 2 CuO 4 .Thus,theresultingsensitivitycouldbeattributedtobot hadifferenceinlocal pO 2 (andacorrespondingNernstianVoltage)inadditiontoacha ngeinenergylevels duetosurfaceadsorption.ThismakesCOsensitivitymoredi fculttoexplainandcould giveanexplanationastowhyinsomecasesasensorrespondsi noppositedirections forNOandCO,althoughbotharereducinggases. 94

PAGE 95

Anadditionalcontributionthatmustbeconsideredisthein teractionbetweena catalystandsemiconductorpresentinC2,C4,andC5.Itwasp reviouslyshownthatthe presenceofPt(aswellasotheradditivecatalystslikePd)o nasemiconductingoxide (e.g.,InO x andSnO 2 )canaffecttheelectrochemicalstateoftheoxide[ 67 ],[ 68 ].There aretwoproposedmechanismsofhowtheelectrochemicalstat eisaltered.Oneisa “spillover”mechanisminwhichspeciesadsorbontotheaddi tivesurface,whereitthen reacts.Thisreactedspeciessubsequentlyspillsovertoth esemiconductor,causinga changeinitsresistance.Inthesecondmechanism,adsorpti onofaspeciesontothe additiveresultsinanelectronexchange.Thisexchangecon tinuesinthesemiconductor, alsoresultinginachangeinsemiconductorresistance. ThePt/La 2 CuO 4 interactionplaysaroleforeachgastested.ForNO 2 ,thesensors withaPtinterfaceonLa 2 CuO 4 (C2,C4,andC5)havethelowestsensitivity(interms ofabsolutemagnitude)at550 C,andthehighestsensitivityat650 C.Whilethe amountofPtaffectsthesensitivityatlowertemperatures, theeffectonNO 2 sensitivity isindependentoftheamountofPtat650 C.Theinteractionalsocausesanincreasein NOsensitivitybetween550and650 C,withtheeffectlessprominentabove600 C.The catalyst/semicondcutorinteractionplaysthegreatestro lewithCOsensitivity.Asmall amountofPt(asinC2)increasedthesensitivitysignicant ly,whilesensorswithlarger amountsofPt(asinC4andC5)hadverylittleCOsensitivitya t550to650 C. TheresultsforC3areperhapsthemostdifculttounderstan d.Thereareno Pt/La 2 CuO 4 interfacespresent,butanadditionalscreenprintedPtare anexttothe La 2 CuO 4 electrode.Althoughitmightbeassumedthatthiscongurat ionwouldbehave similarlytoC1,itsbehaviortowardsNO 2 andNOdeviatesthemostfromtheother congurations.Whileproducinglargesensitivities,Con guration3hasresponsesof oppositepolaritythantheothers:anegativeresponseforN O 2 andpositiveforNO.As farasCO,C3behavessimilarlytoC1. 95

PAGE 96

5.4Conclusions Thesensorfabricationprocessanddesignisoftenoverlook edintermsof understandinggassensorbehavior.Thisworkhasattempted toshowhowsensor design,inparticularsensingelectrodecongurationcana ffectsensitivity,selectivity,and repeatabilityforgasessuchasNO x andCO.Sensorswithvedifferentcongurations ofthesensingelectrodeweretested.Thesensorsresponded bestinthetemperature rangeof550-650 C.Thedifferencesincongurationprovedtochange,andsom etimes improve,sensitivityandselectivityandalsoaffectedthe runtorunrepeatability.In particular,sensorselectivitytowardsNO x withrespecttoCOwasshowntobeenhanced bydepositingPtontopoftheLa 2 CuO 4 sensingelectrode.However,thesensorswith thePtelectrodeontopofthesensingelectrodeprovedtobet heleaststableintermsof runtorunrepeatability.Thus,longtermstabilitymuststi llbeaddressed. ASensor BSensorSem Figure5-1.SchematicofsensorandSEMcrosssection Figure5-2.SchematicofthedifferentLa 2 CuO 4 electrodecongurations 96

PAGE 97

Figure5-3.Temperatureandtestgasproleofexperiments Figure5-4.Voltageversustimeplot 97

PAGE 98

Figure5-5.Voltageversuslog(GasConcentration) Figure5-6.NO 2 Sensitivityof“Sensor1”foreachcongurationat550,600, and650 C 98

PAGE 99

Figure5-7.NOSensitivityof“Sensor1”foreachcongurati onat550,600,and650 C Figure5-8.COSensitivityof“Sensor1”foreachcongurati onat550,600,and650 C 99

PAGE 100

Figure5-9.NO 2 selectivitywithrespecttoNOat550,600,650 C Figure5-10.NO 2 selectivitywithrespecttoCOat550,600,650 C 100

PAGE 101

Figure5-11.NOselectivitywithrespecttoCOat550,600,65 0 C Figure5-12.NO 2 sensitivitycomparisonsbetweenRun1andRun2at550,600,a nd 650 C 101

PAGE 102

Figure5-13.NOsensitivitycomparisonsbetweenRun1andRu n2at550,600,and 650 C Figure5-14.COsensitivitycomparisonsbetweenRun1andRu n2at550,600,and 650 C 102

PAGE 103

Figure5-15.ComparisonofCOsensitivityfor3samplesofCo nguration3at600 C 103

PAGE 104

CHAPTER6 CONCLUSION Thedevelopmentofsensorshasencompassedmanyyearsofres earchand development.Theeffortstounderstandandcommercializet hesedevicesstemfroman immediateneedtobothreduceemissionsofharmfulgasesint otheenvironmentand tousethelimitedresourceoffossilfuelsasefcientlyasp ossible.Oxygensensors, forexampleinautomobiles,haveprovedovertheyearstoadd ressthoseneedswitha highdegreeofsuccess.Thefurtherdevelopmentofsensorsf orothergasessuchas NO x ,CO,etc.canonlyhelpfurtheraddressthoseneeds.Anumber ofissues,suchas alimitedunderstandingofboththefundamentalsensingmec hanismandtheeffects ofprocessingconditions,hasstalledthecommercializati onandpracticaluseofthese devices.Inordertomakethesedevicesmorepracticaltheym ustreachthebenchmark ofsensitivity,selectivity,andstability.Thescopeofth isworkhashelpedcontinuethe understandingofbothsensingmechanismandtheimportance ofengineeringdesign andprocesscontrolintermsofmakingasuitableNO x sensor. La 2 CuO 4 wasexhibitedtobeagoodchoiceinmaterialasasensingelec trode forasolidstateNO x potentiometricsensor.Itwasshowntobebothsensitiveand selectiveforevenppmlevelsofNO 2 andNO.Becauseofitstemperaturedependent catalyticandadsorptivebehavior,La 2 CuO 4 couldbeselectivetowardsNO 2 orNO simplybycontrollingthetemperatureofthedevice/sensin gelectrode.Inthiswork, geometricparameterswerecontrolledinordertoexhibitth eeffectstheyhadonsensor performance.Theeffectsshowedthepromiseofcontrolling theseparametersinorderto improvesensitivityandselectivity. Inthepreviousworkoftheadsorptiveandcatalyticpropert iesofsemiconducting oxidesingeneral,andLa 2 CuO 4 specically,forpotentiometricNO x sensorapplications ithasbeenfoundconsistentlythatmaximumsensitivityiso bservedattemperatures correspondingtothetailendofthedesorptionpeakinTPDs. Atlowertemperature 104

PAGE 105

sensitivityislimitedbysaturationoftheoxidesurfacewi thchemisorbednitrateand nitritespecies.Whileathighertemperaturethekineticso fdesorptionaregreaterthan adsorptionattheexpenseofsenstiivty,andthesensingmec hanismtransitionsfrom asemiconductingresponseduetotheadsorbedspeciestoame chanismthatismore catalyticorelectrocatalyticinnature. Thistemperaturedependencewasobservedintheresultsoft hisworkforeachof thesensorgeometricparametersexamined.Therstparamet erexamined,thickness, displayedalargeeffectonbothsensitivityandselectivit y.Forexample,NO 2 sensitivity decreasedwithincreasingthicknessat550-650 C,behaviorexplainedbythecatalytic reductionofNO 2 toNObyLa 2 CuO 4 beforereachingtheelectrode/electrolyteinterface. Atthehighesttemperatureofsensitivity(650 CforNO 2 and600 CforNO),NO x sensitivitywasindependentofthickness.Howeverthesens orwiththethinnestelectrode hadthehighestNO 2 andNOsensitivity. NO 2 adsorptionandtheresultingrelativechangeinconductivi tyalsocontributes totheNO 2 sensingmechanismasindicatedbytheincreaseinsensitivi tyatlower temperatures.ForNO,largebutsaturatedsensitivitiesin dependentofthickness indicativeofsemiconductingbehaviorwereobservedat400 C,whilethetransitionfrom asemiconductingresponsetoacatalyticallydrivenmechan ismcausedanon-linear thicknessdependenceat450-550 C. Theareaofthesensingelectrodewasalsoconsidered.While thicknessdoesnot affecttheamountofTPBreactionsites,increasingthetota lareaincreasestheamount ofTPB's.Inaddition,increasingareaalsoincreasestheto talsurfaceofelectrode materialthatisexposedtothegasspeciesthatarebeingdet ected.Thisresultedin observingmultipleaffectsonNO x sensitivitythatwasalsotemperaturedependent.At thehighesttemperatureofpotentiometricsensitivity(60 0 CforNO 2 and500 CforNO) boththehighfrequency(bothNO 2 andNO)andtotalimpedance(NO 2 )haveminimal sensitivityindicatingsemiconductorbehaviorhasminima limpactonpotentiometric 105

PAGE 106

sensorresponse.Moreover,weobservealinearlydecreasin gpotentiometricsensitivity withelectrodearea,whileboththehighfrequency(bothNO 2 andNO)andtotal impedance(NO 2 )areessentiallyareaindependent. Incontrast,atthelowesttemperaturesofpotentiometrics ensitivity(500 Cfor NO 2 and400-450 CforNO)boththehighfrequencyandtotalimpedanceexhibit high sensitivitytobothNO 2 andNOindicatingsemiconductorbehaviorisamajorcontrib utor topotentiometricsensorresponse.Moreover,atthesetemp eraturesweobservehighly non-linearelectrodeareadependenceforbothpotentiomet ricsensitivityandhigh frequencyandtotalimpedance.Notonlythat,buttheelectr odeareathatexhibited maximumpotentiometricNO 2 sensitivity(30mm 2 )wastheinectionpointintheNO 2 highfrequencyimpedancedataindicatingacommonalityina mechanisticchangeat thisarea.Similarly,forNOpotentiometricresponseamaxi mumsensitivitywasobserved at16mm 2 whichdirectlycorrespondedtotheelectrodeareaformaxim umNOtotal impedancesensitivity.Clearlytheeffectofareaonpotent iometricresponseinthelower temperatureregionisrelatedtotheimpactofsemiconducti ngbehavioronthesensing mechanism. Finally,sensorelectrodeconguration(e.g.electrodepl acement,presenceofan additionalcatalyst,etc)wasstudiedinordertoanticipat eformorecomplexsensor designs.Thedifferencesincongurationprovedtochange, andsometimesimprove, sensitivity,selectivity,andrepeatability.Thepresenc eofPt(alsoacatalystforNO 2 reduction)onthesurfaceoftheLa 2 CuO 4 sensitivityincreasedbothNO 2 andNO sensitivityat600and650 C(temperaturesatwhichsensitivityismorecatalytically driven)anddecreasedsensitivityat550 C(wheresemiconductivepropertiescould becomemoreofaninuence).ThepresenceofaPtelectrodene arthesensing electrodeevencausedareverseinpolarityforNO 2 andNOsensitivityat550-650 C. ThiseffectalsocouldplayaroleinaffectingNO x sensitivityintheelectrodeareastudy sincechangingthesensingelectrodeareaalsochangesthea symmetryofLa 2 CuO 4 106

PAGE 107

andPt.Inaddition,sensorselectivityforNO x withrespecttoCOwasenhancedbythe presenceofanadditionPtelectrodeontopoftheLa 2 CuO 4 electrode.However,asmall amountofPtusedtoattachaPtwireleadhelpedincreaseCOse nsitivity. Inconclusion,thisworkservedthedualpurposeoffurthere xplainingsensor mechanismandemphasizingtheimportanceforprocessing/d esigncontrol.Whilethese resultsdemonstratethatelectrodegeometriesandcongur ationshavesignicanteffects onpotentiometricresponse,furtherworkisnecessaryatel ucidatingtheindividual mechanisticcontributionsandhowtheyintegrateintoacon volutedmuti-mechanism sensorresponse. 107

PAGE 108

APPENDIXA ADDITIONALDATAFORCHAPTER4 FigureA-1.NO 2 voltageresponseversustimeata)600 Cb)550 Cc)500 C 108

PAGE 109

FigureA-2.NOvoltageresponseversustimeata)500 Candb)450 C FigureA-3.NO 2 voltageresponseversusNO 2 concentrationata)600 Cb)550 Cc) 500 C 109

PAGE 110

FigureA-4.NOvoltageresponseversusNOconcentrationata )500 Candb)450 C 110

PAGE 111

APPENDIXB ADDITIONALDATAFORCHAPTER5 FigureB-1.Conguration1:NO 2 ,NO,andCOvoltageresponseversustimeata)650 C b)600 Candc)550 C 111

PAGE 112

FigureB-2.Conguration1:voltageversusgasconcentrati onata)650 Cb)600 Cand c)550 C 112

PAGE 113

FigureB-3.Conguration2:NO 2 ,NO,andCOvoltageresponseversustimeata)650 C b)600 Candc)550 C 113

PAGE 114

FigureB-4.Conguration2:voltageversusgasconcentrati onata)650 Cb)600 Cand c)550 C 114

PAGE 115

FigureB-5.Conguration3:NO 2 ,NO,andCOvoltageresponseversustimeata)650 C b)600 Candc)550 C 115

PAGE 116

FigureB-6.Conguration3:voltageversusgasconcentrati onata)650 Cb)600 Cand c)550 C 116

PAGE 117

FigureB-7.Conguration4:NO 2 ,NO,andCOvoltageresponseversustimeata)650 C b)600 Candc)550 C 117

PAGE 118

FigureB-8.Conguration4:voltageversusgasconcentrati onata)650 Cb)600 Cand c)550 C 118

PAGE 119

FigureB-9.Conguration5:NO 2 ,NO,andCOvoltageresponseversustimeata)650 C b)600 Candc)550 C 119

PAGE 120

FigureB-10.Conguration5:voltageversusgasconcentrat ionata)650 Cb)600 Cand c)550 C 120

PAGE 121

APPENDIXC KELVINPROBEMETHOD C.1Introduction Thesensorresponsesinthisworkareattributedtomultiple contributionsthatare describedintheDifferentialElectrodeEquilibriaTheory [ 26 ].Oneofthesecontributions isachangeinFermiEnergyofthesensingelectrodematerial uponadsorptionof gaseousspecies.Gasadsorptiononsensormaterialscanbes hownviaTemperature ProgrammedAdsorption[ 31 ]experiments,howevertheeffectofgasadsorptionis somewhatlimitedtosensorperformancedata[ 37 ]. OnewaythechangeinFermiEnergycanbedirectlyexaminedis throughtheKelvin ProbeMethod[ 69 ].TheKelvinProbeMethoddeterminesachangeinworkfuncti on betweentwomaterials.Sincetheworkfunctionisdenedast heamountofenergy requiredtomoveanelectronfromtheFermilevelintovacuum ,thechangeinwork functionalsocangiveameasureofthechangeinFermiEnergy .TheKelvinmethod measuresthecontactpotentialdifference(CPD)betweentw osurfaces(theprobetip andsample).Thismethodhasbeenusedtodeterminetheelect ricalpropertiesof semiconductorswithsuccess[ 70 ]. TheKelvinprobesetupconsistsoftheprobetipandsampleel ectrodethatare inaplane-parallelorientationsuchthatacapacitoriscre ated.Theprobetipissetto oscillatephysicallyoscillate,creatinganoscillatingc apacitance.Asthetwomaterials comeinelectricalcontact,atransferofchargeallowstheF ermilevelstoequalizeand giverisetoasurfacecharge.Atthispoint,acontactpotent ialexiststhatisrelatedtothe differenceinworkfunction.Anapplied“backingvoltage”( V b )whichnullsthissurface chargeisequaltotheCPD.CPDisrelatedtotheworkfunction ( )byEquation C–1 CPD = e ( )(C–1) 121

PAGE 122

C.2ExperimentalProcedure AMcAllisterKP6500KelvinProbewasusedtoperformKelvinP robemeasurements. UnlikeotherKelvinProbesystemsthattypicallyusealockinampliertomeasurethe nullcondition,theMcAllistersystemusesadigitaloscill atoranddigitaltrigger. SamplesconsistedofLa 2 CuO 4 electrodesthatwerescreenprintedontoYSZ similartotheelectrodesusedinthesensorwork.Threesamp leswerepreparedthat mimickedthreeoftheelectrodecongurationsusedinChapt er5:onewithjustLa 2 CuO 4 (LCO),anotherwithaPtelectrodeadjacenttotheLa 2 CuO 4 (LCOnPt),andonewitha PtelectrodescreenprintedwithaslightoverlapontheLa 2 CuO 4 electrode(PtonLCO). KelvinProbemeasurementsweretakenonthesesamplesinana mbientcondition withnogasesadsorbed,andwithNOandNO 2 adsorbedontotheelectrodesurface. NO x wasadsorbedbyexposingthesamplestoaconditionof400ppm NO x ,3%O 2 andabalanceofN 2 atatotalowrateof300sccmovernightat400 Cinagasow apparatus. C.3ResultsandDiscussion CPDmeasurementsweretakenforallthesampleswithexposur etoNO 2 andNO. CPD NOx isdenedas CPD NO x = CPD NO X CPD o (C–2) TheresultsareshowninTable C-1 .Whileitishardtoconjectureanythingwith thisdataitshouldbenotedthatthe CPDvaluesindicateanoppositechangein workfunctionuponadsorptionofNO 2 andNOasexpectedduetosemiconducting behavior.The CPD NO 2 forsample(LCOnPt)isnegativeunliketheothersamples.Th is isrelatedtothesensorperformanceofConguration3inCha pter4,whereitwasthe onlyCongurationthatproducedanegativeNO 2 sensitivity.Thisworkshowspromise butshouldbecontinuedwithbothmorecarefulcalibrationo ftheKelvinProbesystemas 122

PAGE 123

wellasinconditionswithinsitugaschangesandhightemper aturesindicativeofactual sensoroperation. TableC-1. CPD NOx Sample CPD NO 2 (V) CPD NO (V) LCO-0.5410.330 LCOnPt0.3000.427 PtonLCO-0.7230.038 123

PAGE 124

REFERENCES [1] C.Young,StaticanddynamicpropertiesofZrO 2 oxygensensor,in:T.Wheat, A.Ahmad,A.Kuriakose(Eds.),ProgressinSolidElectrolyt es,Energy,Mines,and Resources,1983. [2] E.Logothetis,Automotiveoxygensensors,in:N.Yamazoe(E d.),ChemicalSensor Technology,Vol.3,KodanshaLtd.,1991. [3] G.Li,X.Zhang,S.Kawi,Relationhipsbetweensensitivity, catalyticactivity,and surfaceareasofSnO 2 gassensors,SensorsandActuators:B60(1999)64–70. [4] N.Miura,G.Lu,N.Yamazoe,Progressinmixed-potentialdev icesbasedonsolid electrolyteforsensingredoxgases,SolidStateIonics136 -137(2000)533–542. [5] C.Chaillout,S.Cheong,Z.Fisk,M.Lehmann,M.Marezio,B.M orosin,J.Schirber, ThecrystalstructureofsuperconductingLa 2 CuO 4.032 byneutrondiffraction,Physics C158(1989)183–191. [6] J.Yoo,S.Chatterjee,F.M.V.Assche,E.D.Wachsman,Inuen ceofadsorption andcatalyticreactiononsensingpropertiesofapotentiom etricLa 2 CuO 4 /YSZ/Pt sensor,JournaloftheElectrochemicalSociety154(7)(200 7)J190–J195. [7] O.ofAirQualityPlanning,Standards,Theplainenglishgui detothecleanairact, Tech.Rep.EPA-456/K-07-001,EnvironmentalProtectionAg ency(2007). [8] M.Westbrook,J.Turner,AutomotiveSensors,InstituteofP hysicsPublishing, Bristol,1994. [9] O.ofAir,Radiation,Performanceofselectivecatalyticre ductiononcoal-redsteam generatingunits,Tech.rep.,EnvironmentalProtectionAg ency(1997). [10] E.Wachsman,P.Jaiyaweera,G.Krishnan,A.Sanjuro,Electr ocatalyticreductionof NO x onLa 1 x A x B 1 y B 0y O 3 :evidenceofelectricallyenhancedactivity,SolidState Ionics136(2000)775–782. [11] O.ofMobileSources,Automobilesandcarbonmonoxide,Tech .Rep.EPA 400-F-92-005,U.S.EnvironmentalProtectionAgency(1993 ). [12] O.ofMobileSources,Automobileemissions:Anoverview,Te ch.Rep.400-F-2-007, U.S.EnvironmentalProtectionAgency(1994). [13] R.Heck,R.Farrauto,Automobileexhaustcatalysts,Applie dCatatlysisA:General 221(2001)443–457. [14] K.B.Schnelle,C.A.Brown,AirPollutionControlTechnolog yHandbook,CRC Press,BocaRaton,2002. 124

PAGE 125

[15] S.Bernard,J.Samet,A.Grambsch,K.Ebi,I.Romieu,Thepote ntialimpactsof climatevariabilityandchangeonairpollution-relatedhe altheffectsintheunited states,EnvironmentalHealthPerspectives109. [16] A.Azad,S.Akbar,S.Mhaisalkbar,L.Birkefed,K.Goto,Soli dstategassensors:A review,JournaloftheElectrochemicalSociety129(12)(19 92)3690–3704. [17] W.J.Fleming,Physicalprinciplesgoverningnonidealbeha viorofthezirconia oxygensensor,JournaloftheElectrochemicalSociety124( 1)(1977)21–28. [18] S.C.NicolasDocquier,Combustioncontrolandsensors:Are view,Progressin EnergyandCombusionScience28(2002)107–150. [19] J.Maier,Kroger-vinkdiagramsforboundaryregions,Solid StateIonics32-33Part 2(1989)727–733. [20] R.E.Hummel,ElectronicPropertiesofMaterials,3rdEditi on,Springer,NewYork, 2001. [21] D.Jones,PrinciplesandPreventionofCorrosion,2ndEditi on,PrenticeHall,New Jersey,1996. [22] N.Miura,G.Lu,N.Yamazoe,H.Kurosawa,M.Hasei,Mixedpote ntialtype NO x sensorbasedonstabilizedzirconiaandoxideelectrode,Jo urnalofthe ElectrochemicalSociety143(2)(1996)L33–l35. [23] S.Akbar,P.Dutta,C.Lee,High-temperatureceramicgassen sors:Areview, InternationalJourunalofAppliedCeramicTechnology3. [24] R.Mukundan,E.Brosha,F.Garzon,Alowtemperaturesensorf orthedetectionof carbonmonoxideinhydrogen,SolidStateIonics175(2004)4 97–501. [25] N.F.Szabo,P.K.Dutta,Correlationofsensingbehaviorofm ixedpotentialsensors withchemicalandelectrochemicalpropertiesofelectrode s,SolidStateIonics171 (2004)183–190. [26] E.Wachsman,P.Jaiyaweera,SelectivedetectionofNO x bydifferentialelectrode equilibria,in:W.Weppner,E.Traversa,M.Liu,P.Vanysek, N.Yamazoe (Eds.),SolidStateIonicDevicesII-CeramicSensors,Vol. PV2000-32,The ElectrochemicalProceedingsSeries,Pennington,NJ,2000 ,pp.298–304. [27] P.Radaelli,J.Jorgensen,A.Schultz,B.Hunter,J.Wagner, F.Chou,D.Johnston, StructureofthesuperconductingLa 2 CuO 4+ phases( 0.08,0.12)preparedby electrochemicaloxidation,PhysicalReviewB48(1)(1993) 499–510. [28] Y.Yhu,R.Tan,T.Yi,S.Gao,C.Y.L.Cao,Preparationofnanos izedLa 2 CuO 4 perovskiteoxideusinganamorphousheteronuclearcomplex asaprecursorat low-temperature,JournalofAlloysandCompounds311(2000 )16–21. 125

PAGE 126

[29] J.Bednorz,M.Takashige,K.Muller,Preparationandcharac terizationof alkaline-earthsubstitutedsuperconductingLa 2 CuO 4 ,MaterialsResearchBulletin 22(1987)819–827. [30] S.Roy,W.Sigmund,F.Aldinger,Nanostructuredyttriapowd ersviagelcombustion, JournalofMaterialsResearch14(1999)1524–1531. [31] J.Yoo,F.M.V.Assche,E.D.Wachsman,Temperature-program medreactionand desorptionofthesensorelementsofaWO 3 /YSZ/Ptpotentiometricsensor,J. Electrochem.Soc.153(6)(2006)H115–H121. [32] F.M.V.Assche,J.C.Nino,E.D.Wachsman,Infraredandx-ray photoemission spectroscopyofadsorbatesonLa 2 CuO 4 todeterminepotentiometricNO x sensor responsemechanism,J.Electrochem.Soc.155(7)(2008)J19 8–J204. [33] F.M.V.Assche,E.D.Wachsman,Isoptopicallylabeledoxyge nstudiesofthe NO x exchangebehaviorofLa 2 CuO 4 todeterminepotentiometricsensorresponse mechanism,SolidStateIonics179(2008)2225–2233. [34] A.Moulson,J.Hebert,Electroceramics,2ndEdition,Wiley Publishing,2003. [35] W.D.Callister,MaterialsScienceandEngineering:AnIntr oduction,6thEdition, WileyPublishing,2002. [36] K.-C.Ho,W.-T.Hung,Anamperometricno 2 gassensorbasedonpt/naon electrode,SensorsandActuators:B79(2001)11–16. [37] B.White,S.Chatterjee,E.Macam,E.Wachsman,Effectofele ctrode microstructureonthesensitivityandresponsetimeofpote ntiometricNO x ,Journal oftheAmericanCeramicSociety91(6)(2008)2024–2031. [38] D.Richerson,ModernCeramicEngineering,2ndEdition,Mar celDekker,Inc., 1992. [39] T.R.JohnRoss,ClareRomano,TheCompletePrintmaker:Tech niques,Traditions, Innovations,Vol.2,RoundtablePress,Inc.,1990. [40] R.Ulrich,W.Brown(Eds.),AdvancedElectronicPackaging, secondeditionEdition, JohnWileyandSons,2006. [41] B.White,E.Traversa,E.Wachsman,InvestigationofLa 2 CuO 4 /YSZ/Pt potentiometricNO x sensorswithelectrochemicalimpedancespectroscopy,Jou rnal oftheElectrochemicalSociety155(1)(2008)J11–J16. [42] D.Williams,P.McGeehin,B.Tolled,Solidelectrolytemix edpotentialphenomena, in:R.Metselaar,H.Heijligers,J.Schoonman(Eds.),Proce edingsofthe2nd EuropeanConferenceonSolidStateChemistry,1982. 126

PAGE 127

[43] E.Traversa,Y.Sadaoaka,M.Carotta,G.Martinelli,Enviro nmentalmonitoring eldtestsusingscreen-printedthick-lmsensorsbasedon semicondutingoxides, SensorsandActuators:B65(2000)181–185. [44] E.D.Bartolomea,M.L.Grilli,E.Traversa,Sensingmechani smofpotentiometric gassensorsbasedonstabilizedzirconiawithoxideelectro des:Isitalwaysmixed potential?,JournaloftheElectrochemicalSociety151(5) (2004)H133–H139. [45] J.W.Yoon,M.Grilli,E.D.Bartolomea,R.Polini,E.Travers a,TheNO 2 responseof solidelectrolytesensorsmadeusingnano-sizedLaFeO 3 electrodes,Sensorsand Actuators:B76(2001)483–488. [46] S.-J.Huang,A.Walters,M.Vannice,Adsorptionanddecompo sitionofnoon lanthanumoxide,JournalofCatalysis192(2000)29–47. [47] B.White,Solid-statepotentiometricsensorsfornitrogen monoxide,nitrogen dioxide,andcarbonmonoxide,Ph.D.thesis,UniversityofF lorida(2007). [48] M.L.Grilli,E.D.Bartolomeo,E.Traversa,Electrochemica lNO x sensorsbasedon interfacingnanosizedLaFeO 3 perovskite-typeoxideandionicconductors,Journal oftheElectrochemicalSociety148(2001)H98–H102. [49] V.Plashnitsa,P.Elumalai,Y.Fujio,N.Miura,Zircnonia-b asedelectrochemicalgas sensorsusingnano-structuredsensingmaterialsaimingat detectionofautomotive exhausts,ElectrochimicaActa54(2009)6099–6106. [50] Z.Can,H.Narita,J.Mizusaki,H.Tagawa,Detectionofcarbo nomonoxidebyusing zirconiaoxygensensor,SolidStateIonics79(1995)344–34 8. [51] H.Kurosawa,Y.Yan,N.Miura,N.Yamazoe,Stabilizedzircon ia-basedNO x sensor operativeathightemperature,SolidStateIonics79(1995) 338–343. [52] N.Miura,H.Kato,N.Yamazoe,T.Seiyama,Protonconductors ensorsforH 2 and COoperativeatroomtemperature,in:ProceedingsoftheInt ernationalMeetingof ChemicalSensors,Elsevier,Amsterdam,1983,p.233. [53] F.Shimizu,N.Yamazoe,T.Seiyama,Detectionofcombustibl egaseswithstabilized zirconiasensor,ChemistryLetters(1978)299. [54] F.Garzon,R.Mukundan,E.Brosha,Solid-statepotentialga ssensors:theory, experimentsandchallenges,SolidStateIonics136-137(63 3-638). [55] G.Lu,N.Miura,N.Yamazoe,High-temperaturehydrogensens orbasedon stabilizedzirconiaandametaloxideelectrode,Sens.Actu atorsBChem.35-36 (1996)130–135. [56] N.Miura,G.Lu,N.Yamazoe,High-temperaturepotentiometr ic/amperometricNO x sensorscombiningstabilizedzirconiawithmixed-metalox ideelectrode,Sens. ActuatorsBChem.52(1998)169–178. 127

PAGE 128

[57] N.Miura,S.Zhuiykov,T.Ono,M.Hasei,N.Yamazoe,Mixedpot entialtypesensor usingstabilizedzirconiaandZnFe 2 O 4 sensingelectrodeforno x detection,Sens. ActuatorsBChem.83(2002)222–229. [58] J.Park,B.Yoon,C.Park,W.-J.Lee,C.LEE,Sensingbehviora ndmechanismof mixedpotentialNO x sensorsusingNiO(+YSZ)andCuOoxideelectrodes,Sens. ActuatorsBChem.135(2009)516–523. [59] V.Plashnitsa,T.Ueda,P.Elumalai,N.Miura,NO 2 sensingperformancesofplanar sensorusingstabilizedzirconiaandthin-NiOsensingelec trode,Sens.ActuatorsB Chem.130(2008)231–239. [60] N.Barsan,M.Schweizer-Berberich,W.Gopel,Fundamentala ndpracticalaspects inthedesignofnanoscaledSnO 2 gassensors:Astatusreport,Fresenius'Journal ofAnalyticalChemistry365(1999)287. [61] N.Barsan,J.Stetter,J.M.Findlay,W.Gopel,High-perfoma ncegassensingofCO: Comparativetestsforsemiconducting(SnO 2 -based)andforamperometricgas sensors,AnalyticalChemistry71(1999)2512. [62] M.Carotta,G.Martinelli,L.Crema,C.Malagu,G.Ghiotti,E .Traversa, Nanostructuredthick-lmgassensorsforatmosphericpoll utantmonitoring: Quantitativeanalysisoneldtests,SensorsandActuators B:Chemical76(2001) 336–342. [63] B.White,F.M.V.Assche,E.Traversa,E.Wachsman,Electric alcharacterizationof semiconductingLa 2 CuO 4 forpotentiometricgassensorapplications,ECSTrans. 1(7)(2006)109. [64] B.Blackburn,E.Wachsman,Multifunctionalgassensorarra ywithimproved selectivitythroughlocalthermalmodication,ECSTransa ctions11(33). [65] M.Penza,C.Martucci,G.Cassano,NO x gassensingcharacteristicsofWO 3 thin lmsactivatedbynoblemetals(Pd,Pt,Au)layers,Sens.Act uatorsBChem.50 (1998)52–59. [66] L.Chevallier,E.D.Bartolomeo,M.L.Grilli,M.Mainas,B.W hite,E.Wachsman, E.Traversa,Non-Nernstianplanarsensorsbasedonyszwith aNb 2 O 5 electrode, SensorsandActuatorsB:Chemical129(2008)591–598. [67] S.Morrison,Selectivityinsemiconductorgassensors,Sen s.Actuators12(1987) 425–440. [68] D.Vlachos,C.Papadopoulos,J.Avaritsiotis,Characteris ationofthe catalyst-semiconductorinteractionmechanisminmetal-o xidegassensors,Sens. ActuatorsBChem.44(1997)458–461. [69] D.Woodruff,T.Delchar,ModernTechniquesofSurfaceScien ce,2ndEdition, CambridgeUniversityPress,1994. 128

PAGE 129

[70] J.Nowotny,M.Sioma,SurfaceelectricalpropertiesofBaTi O 3 atelevated temperatures,SolidStateIonics49(1991)129–133. 129

PAGE 130

BIOGRAPHICALSKETCH ItallbeganwhenEricRyanMacamwasbornin1982atthehospit alonNaval AirStationinJacksonville,FloridatoparentsRenatoandV ioletaandsisterMichelle. AsaNavykid,Erichadthepotentialtomoveconstantly,buth adthefortunetospend hisentireyouthinJacksonville.Whilenotnecessarilycon sideredthe“truesouth”,the FirstCoastofFloridaprovidedhimaperfectenvironmentfo rlearningthevaluesofthe SouthandthePhilippines,wherehisparentsgrewup.Fromth emomenthewasborn, Ericwasalwayssurroundedbyacoregroupoffamilyandatigh tknitgroupoffriends, especiallywiththelargepopulationoftheFilipinocommun ityinJacksonville. EricreceivedhisearlyschoolingfromLittleLambsPre-Sch ool,GrovePark Elementary,OrangeParkJuniorHighSchool,andOrangePark HighSchool.Somehow inthattimespan,hefoundthathisfavoritesubjectsinscho olincludedmathand science.Fromaveryyoungage,hebecameafanoftheUniversi tyofFlorida.Itshould benotedthathisfavoritecolorgrowingupwasorange,soheb ecamefondoftheFlorida Gator'sorangehelmetsthersttimehesawthemplayingfoot ballonTV.Hisolder cousinNatalieandsisterwouldeventuallyattendUFanditm adesensethathewould followintheirfootsteps.Ontopofthat,hishighschoolche mistryteacherandimportant mentor,MargaretHansensuggestedthathelookintomateria lsscience&engineering. UFboastsoneofthetopMSEprogramsintheUnitedStates.His responsetotheever importantchildhoodquestionof“Whatdoyouwantwanttobew henyougrowup?” includedthingssuchasbaseballplayer,stockbroker,doct or,andnaturallybothofhis parents'professions.However,EricwouldleaveJacksonvi lleforGainesvilletogrowup andbecomeaMaterialsEngineer. Eric'sinterestintheaerospaceandtheautomotiveindustr iesinuencedhis decisioninspecializinginceramicmaterials.Insearchin gforaadvisortomentorhim intheUniversityScholar'sUndergraduateprogram,hemetP rofessorEricWachsman. Overthecourseofhisundergraduateresearchcareer,someh owDr.Wachsman 130

PAGE 131

convincedhimtogoforaPh.D,somethingEricneverreallyco nsideredbefore.Eric wouldcontinueonwithhisschooling,andwouldalsohaveagr eatlifechanging experiencebyhavingtheopportunitytospendayearofhisgr aduateprogramatthe UniversityofRome“TorVergata”inItaly. WhereEricgoesfromhereissomewhatuncertain.However,he didgenuinelyenjoy histimeattheUniversityofFlorida.Whileheconsidersrec eivingaPh.Dagreathonor andaccomplishment,itissomethingthatwillnotdenehim. Hewillalwaysbedened byhisrelationshipwithhissaviorJesusChristandtheChri stlikelovehetriestoextend toothers. 131