1 HETEROGENEOUS CATALYTIC EVALUATION AN D ADSORPTION CHARACTERIZATION OF THE LA2CUO 4/YSZ/PT POTENTIOMETRIC COUPLE By FREDERICK MARTIN VAN ASSCHE IV A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008
2 by Frederick Martin Van Assche IV
3 To my whole family, for teaching me: Vita non est vivere se d valere vita est
4 ACKNOWLEDGMENTS A strange thing that learning does is to te ach you how much you do not know. Odder still is the realization that you de sire to know exactly how much more you do not know! Whatever I learned (and learned what I want to learn) in my time here has been due to the help of a great many people. I would like to take the opportunity to thank: Dr. Jiho Yoo and Dr. Briggs White. Without th eir tireless assistance with equipment, and limitless patience for discussion, I could not have progressed this far; My 3rd years, Bryan Blackburn, Cynthia Kan, and Eric Armstrong, for your enthusiasm and demands; Everyone in the Wachsman resear ch group, for color and support; Dr. Eric Wachsman, for the resources and advi ce to make my tenure at the University a fruitful and enlightening one; Drs. Scott Perry, Wolfgang Sigmund, and Juan Nino, for serving as my supervisory committee. Their guidance has been helpful, and it is greatly appreciated; Dr. Jason Weaver, for serving as the extern al member of my committee, and for his continued support and direction th roughout my many years at UF; Dr. Merle Battiste, for helping me acquire a diligent scientific method and a formidable research skill set; Eric Lambers and Dr. Susan Si nnott, for useful discussion; The Major Analytical Instrumentation Center and the Engineering Research Center, for use of the facilities. Research is an activity that requires great cooperation to accomplish. Behind every achievement stand ten people who made it happen. Thus, I would be remiss in my
5 acknowledgement without mentioning the individu als whose work on my behalf made this dissertation possible. Dr. Suman Chattergee and Dr. Jiho Y oo developed the early synthesis procedures for autoignition La2CuO4 powder used in Chapters 3 and 4. Sean Bishop, Jin Soo Ahn, Tak-keun Oh, and Dong-Jo Oh provided fa brication techniques for pellet synthesis performed in Chapter 6. In early 2007, during sensor experiments utilizi ng a potentiometric gas sensor with heating elements, Bryan Blackburn noticed large sensor re sponses to small applied heater voltages after a crack formed in one of the heaters. The voltage s caused changes in the po tential of the sensor, and after testing the effect of the voltage gap, ap proached me to help him determine the causes of this behavior. Our discussions resulted in the poss ibility that the electric field had an effect on adsorption/desorption and or chemical/electrochem ical reactions taking place at the electrodes. Thus began our collaborative rese arch involving these external elec tric-field modifications to the sensor and measurements with the mass spectrome ter that is included in Chapter 6 of my dissertation. Bryan and I worked closely on the testing procedures, utilizing my experience with the mass spectrometer and as a chemical engineer /materials scientist and his experience with sensors and as an electrical engi neer/materials scientist. We constructed two new reactors for testing the sensors and this initial work ha s opened new avenues for investigating this phenomenon. My best wishes go to him as he continues it. Special thanks go out to Dr. Heesung Yoon, fo r his contributions to keeping the labs running, and his assistance with equipment repair. Jennifer Tucker deserves my thanks for her diligent management of orders, reservations, supplies, and everything else keeping the group chugging along. My roommates Alan, Eric, and Henry, though they will never read these words, deserve my thanks and gratitude for the use of their 50 ft2.
6 Success on any scale requires a curious comb ination of love, apprehension, support, intimidation, persistence, tenacity, larceny, and a bit of abject terror. Fo r all those effects and more, I have a few people to thank. My fiance, Christine has provided the continual impetus to persevere in my quest for the doctorate, and is ju st as happy as I to see it completed. My mother, Jeri, has been (and continues to be) a s ource of confidence, reason, temperance, and encouragement, and for that I thank her with all my heart. Finally, I thank my father, Fred, for impressing upon me a motto that has always se rved me well, and my tribute to him; Aut viam inveniam aut faciam.
7 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ........10 LIST OF FIGURES................................................................................................................ .......12 ABSTRACT....................................................................................................................... ............19 CHAPTER 1: INTRODUCTION................................................................................................................... .21 1.1 NOX?.............................................................................................................................. ...21 1.2 Advantages of the Potentiometric Sensor.........................................................................22 1.3 Barriers to Commercialization..........................................................................................22 2: BACKGROUND..................................................................................................................... .24 2.1 The Nernstian Sensor....................................................................................................... .24 2.2 The Non-Nernstian Sensor...............................................................................................25 2.3 Initial Theories for Non-Nerstian Behavior......................................................................26 2.4 Mixed Potential............................................................................................................ .....27 2.5 Differential Electrode Equilibria......................................................................................28 2.6 The Semiconducting Response.........................................................................................30 2.7 Materials of Sensing....................................................................................................... ..32 2.8 Gas Phase NOx and CO Chemistry...................................................................................33 2.9 La2CuO4 Background.......................................................................................................34 3: INITIAL TPD/TPR STUDIES OF LA2CUO4 AND PT.........................................................39 3.1 Introduction............................................................................................................... ........39 3.2 Experimental............................................................................................................... ......41 3.2.1 Synthesis of La2CuO4.............................................................................................41 3.2.2 Powder Characterization........................................................................................41 3.2.3 TPD/TPR Setup......................................................................................................42 3.2.4 TPD/TPR Procedure...............................................................................................42 3.3 Results and Discussion.....................................................................................................44 3.3.1 Powder Characterization Results............................................................................44 3.3.2 TPD/TPR Results of NO Gas Exposure.................................................................44 3.3.3 TPD/TPR Results of NO2 Gas Exposure................................................................50 3.3.4 TPD/TPR Results of COx /NOx Gas Mixture Exposure.........................................51 3.4 Summary.................................................................................................................... .......53
8 4: ADSORBATE STUDIES ON LA2CUO4 USING INFRARED SPECTROSCOPY AND X-RAY PHOTOEMISSI ON SPECTROSCOPY...................................................................63 4.1 Introduction............................................................................................................... ........63 4.2 Experimental............................................................................................................... ......64 4.2.1 Sample Preparation.................................................................................................64 4.2.2 Infrared Characterization........................................................................................65 4.2.3 XPS Characterization.............................................................................................66 4.3 Results and Discussion.....................................................................................................67 4.3.1 IR Results for NOx Adsorption...............................................................................67 4.2.2 XPS Results for NOx Adsorption...........................................................................70 4.4 Conclusions................................................................................................................ .......77 5: ISOTOPICALLY LABELED OXYGEN STUD IES OF THE EXCHANGE BEHAVIOR OF NOx OVER LA2CUO4....................................................................................................85 5.1 Introduction............................................................................................................... ........85 5.2 Experimental............................................................................................................... ......93 5.2.1 Powder Preparation................................................................................................93 5.2.2 Isotopically Labeled TPR/ TPD.............................................................................93 5.3 Results and Discussion.....................................................................................................95 5.3.1 18O2 Temperature Programmed Reaction...............................................................95 5.3.2 Determination of NOx Adsorption via TPD/TPR...................................................96 5.4 Summary.................................................................................................................... .....108 6: MASS SPECTROMETRY AND ELECTRICAL AN ALYSIS OF A LA2CUO4/YSZ/PT FIELD MODIFIED DESORPTION SENSOR....................................................................122 6.1 Introduction............................................................................................................... ......122 6.2 Reactor Construction......................................................................................................124 6.3 Experimental............................................................................................................... ....125 6.3.1 Sample Preparation...............................................................................................125 6.3.2 Testing Setup........................................................................................................127 6.3.3 Electric-field Effect on Desorption......................................................................127 6.3.4 Electric-Field Effect on Sensing Behavior...........................................................128 6.4 Results and Discussion...................................................................................................128 6.4.1 Testing of La2CuO4 Capacitor-Type Sample.......................................................128 6.4.2 Testing of External Electric-Field Sensor............................................................132 6.5 Summary.................................................................................................................... .....136 7: CONCLUSIONS...................................................................................................................148 APPENDIX A: CALIBRATION AND CORRECTION OF MASS SPECTROMETRY DATA.................152 B: TEMPERATURE PROGRAMMED DESO RPTION DATA FOR RELEVANT GAS COMBINATIONS................................................................................................................155
9 C: SILICA CONTAMINATION OF THE SENSOR ELECTRODE INTERFACE.................163 D: ADSORPTION XPS OF NO/NO2/CO ON LA2CUO4 POWDER......................................169 E: DECONVOLUTED MASS SPECTRA OF ISOTOPICALLY LABELED NOX/LA2CUO4 TPD EXPERIMENTS..............................................................................182 LIST OF REFERENCES.............................................................................................................202 BIOGRAPHICAL SKETCH.......................................................................................................230
10 LIST OF TABLES Table page 2-1. Metal oxides commonly used as resistan ce-type and potentiometric gas sensors and relevant publications..........................................................................................................37 3-1. TPD experiments........................................................................................................... .........55 3-2. TPR experiments........................................................................................................... .........55 3-3. Desorption energies of products resulting from NO adsorption on La2CuO4........................57 3-4. Desorption energies of products resulting from NO2 adsorption on La2CuO4.......................59 4-1. IR Peak assignments of the NO/NO2 adsorbed samples of La2CuO4.....................................78 5-1. List of Experiments Performed.............................................................................................112 6.1 NO adsorption TPD of device................................................................................................140 6.2. NO2 adsorption TPD of device.............................................................................................141 A-1. Primary and secondary ions found in Chapter 3 experiments.............................................152 B-1. Temperature programmed desorpti on characteristics of 1% NO on La2CuO4 for the 30 oC per minute desorption.................................................................................................155 B-2. Temperature programmed desorp tion characteristics of 1% NO2 on La2CuO4...................156 B-3. Temperature programmed desorption ch aracteristics of 0.67% NO and 0.33% CO on La2CuO4...........................................................................................................................157 B-4. Temperature programmed desorption char acteristics of 500 ppm NO, 500 ppm CO, and 1% O2 on La2CuO4...........................................................................................................159 B-5. Temperature programmed desorp tion characteristics of 500 ppm NO2, 250 ppm CO and 1000 ppm O2 on La2CuO4................................................................................................161 B-6. Temperature programmed desorption ch aracteristics of 500 ppm NO and 1000 ppm CO2 on La2CuO4..............................................................................................................162 E-1: Desorption energy of peaks for 1% NO and 1% 16O2 adsorption on La2Cu16O4.................182 E-2: Desorption energy and adsorp tion capacity for 1% NO and 1% 18O2 adsorption on La2Cu16O4 partially substituted with 18O.........................................................................184 E-3: Desorption energy and adsorption capacity of La2Cu18O4 for adsorption of 1% NO and 1% 16O2............................................................................................................................186
11 E-4: Desorption energy and adsorption capacity for 1% NO2 and 1% 18O2 adsorption on La2Cu16O4........................................................................................................................189 E-5: Desorption energy of peaks for 1% NO2 and 1% 18O2 adsorption on La2Cu16 O4 partially substituted with 18O..........................................................................................................192 E-11: Desorption energy and adsorption capacity for 1% NO2 and 1% 16O2 adsorption on La2Cu18O4........................................................................................................................197
12 LIST OF FIGURES Figure page 2-1. Diagram of the space charge layer at the surface of the electrode. Opposing charges build up inside the material to compensate for the charge established on the surface. The depth of this charged layer is dependa nt on the material (taken from Vayenas)147....36 2-2. Surface band model of a p-type semic onductor involving positive surface charge. The total work function ( ) consists of the external work function ( ) combined with the Fermi energy ( v) and the contribution of the surface charge from adsorption ( S)(taken from Vayenas)147..............................................................................................37 2-3. Plot of equilibrium mole fractions of exhaust gas consisting of 3% O2, 15% CO2, 3% H2O, 650 ppm NO, and balance N2 versus temperature in oC...........................................38 3-1. Pipeline diagram of the Mass Spectrometer gas system.........................................................54 3-2. X-ray diffraction spectrum of La2CuO4 powder synthesized vi a the auto-ignition technique. Circles marked on the spectrum correspond to copper oxide impurities.........56 3-3. SEM image of the La2CuO4 powder generated via auto-ignition..........................................56 3-4. Temperature Programmed Desorption of 1% NO on La2CuO4.............................................57 3-5. Temperature Programmed Desorp tion of 500 ppm NO + 0.95% O2 on La2CuO4.................57 3-6. Temperature Programmed Reaction of 500 ppm NO (a.)/ Temperature Programmed Reaction of 500 ppm NO and 1000 ppm O2 (b.) on La2CuO4...........................................58 3-7. Temperature Programmed Reac tion of 1000 ppm NO / 1000 ppm O2 over (a.) Pt on YSZ and (b.) 500 ppm / 1000 ppm O2 in the gas phase....................................................58 3-8. Temperature Programmed Desorption of 500 ppm NO2 on La2CuO4. (a.)/ Temperature Programmed Reaction (b.) of 400 ppm NO2 on La2CuO4.................................................59 3-9. Temperature Programmed Reaction of (a) 1200 ppm NO2+ 1800 ppm O2 over Pt/YSZ plate (b) 500 ppm NO2 Temperature Programmed Reaction in the gas phase..................59 3-10. Temperature Programmed Desorp tion of 0.67% NO and 0.33% CO on La2CuO4. (a.)/ Temperature Programmed Reaction ( b.) of 250 ppm NO and 1000 ppm CO on La2CuO4.............................................................................................................................60 3-11. Temperature Programmed Desorption of 500 ppm NO, 500 ppm CO, and 1% O2 on La2CuO4. (a.)/ Temperature Programmed Reac tion (b.) of 500 ppm NO, 500 ppm CO, and 1000 ppm O2 on La2CuO4...................................................................................60
13 3-12. Temperature Programmed Reaction of 1000 ppm NO, 1000 ppm CO, and 2000 ppm O2 on Pt/YSZ and 500 ppm NO, 500 ppm CO, and 1000 ppm O2 in blank reactor..........61 3-13. Temperature Programmed Deso rption of 250 ppm CO, 500 ppm NO2 and 1000 ppm O2 on La2CuO4. (a.)/ Temperature Programmed Reaction (b.) of 500 ppm CO, 500 ppm NO2 and 1000 ppm O2 on La2CuO4...........................................................................61 3-14. Temperature Programmed Reaction of 1000 ppm NO2, 1000 ppm CO, and 2000 ppm O2 on Pt/YSZ and 500 ppm NO2, 500 ppm CO, and 1000 ppm O2 in blank reactor........62 3-15. Temperature Programmed Desorp tion of 500 ppm NO and 1000 ppm CO2 on La2CuO4. (a.)/Temperature Programmed R eaction (b.) of 500 ppm NO and 1000 ppm CO2 on La2CuO4........................................................................................................62 4-1a. IR spectra of La2CuO4/KBr mulled mixture after NO adsorption.......................................79 4-1b. IR spectra of La2CuO4/KBr mulled mixture after NO2 adsorption......................................79 4-2a. Fourier Deconvolution of 1550-1200 cm-1 spectrum of NO-adsorbed La2CuO4 (Labeled line corresponds to the Fourier deconvolution)..................................................80 4-2b. Fourier Deconvolution of 1550-1200 cm-1 spectrum of NO2 adsorbed La2CuO4 (Labeled line corresponds to the Fourier deconvolution)..................................................80 4-3. Full range XPS spectra of solid La2CuO4 samples in the (a) freshly sintered case and (b) after 3 weeks of exposure to sensor testing conditions......................................................81 4-4. O1s XPS spectra of NO adsorbed La2CuO4 versus the unadsorbed case...............................82 4-5. O1s XPS spectra of NO2 adsorbed La2CuO4 versus the unadsorbed case..............................82 4-6. Flowchart of proposed NO adsorption mechanism................................................................83 4-7. Flowchart of proposed NO2 adsorption mechanism...............................................................84 5.1 Flowchart of possible isotopic oxygen ex change mechanisms during NO adsorption on La2CuO4...........................................................................................................................110 5.2 Flowchart of possible oxygen exchange mechanisms during NO2 adsorption on La2CuO4...........................................................................................................................111 5-3. Temperature programmed reaction of 1000 ppm 18O2 over La2Cu16O4...113 5-4 (a). 1% NO+ 1% 18O2 TPD over La2Cu16O4 Lattice, (b) Low Concentration Detail, (c).TPD of aged La2CuO4 sample..113 5-5. (a) 1% NO+ 1% 16O2 TPD over heavily enriched La2Cu18O4 Lattice; (b) 1% NO+ 1% 18O2 TPD over lightly enriched La2Cu18O4 Lattice.114
14 5-6. Gaussian Analysis of (a) Desorption Peak of N16O; (b) Desorption Peak of N18O; (c) Desorption Peak of 16O2; (d) Desorption Peak of 16O18O; (e) Desorption Peak of 18O2...114 5-7. (a) 1% NO2 + 1% 18O2 TPD over La2Cu16O4 Lattice; (b) 1% NO2 + 1% 16O2 TPD over La2Cu16O4 Lattice; (c) 1% NO2 + 1% 16O2 TPD over heavily enriched La2Cu18O4 Lattice; (d) 1% NO2 + 1% TPD over lightly enriched La2Cu18O4 Lattice...116 5-8. Gaussian Analysis of: (a) Desorption Peak of N16O2; (b) Desorption Peak of N16O18O; (c) Desorption Peak of N18O; (d) Desorption Peak of 16O2; (e) Desorption Peak of 18O2; (f) Desorption Peak of 16O18O; (g) Desorption Peak of N16O 5-9. (a). 700 ppm N16O+600 ppm 16O2 TPR over La2Cu18O4 Lattice.........................................119 (b). 700 ppm NO+ 300 ppm 16O2 TPR over partially substituted La2Cu18O4 Lattice.....119 5-11. Formation of nitrito via nitrosyl ion/ chelated nitrite intermediate....................................120 5-12. Nitrogen-down bonding to 16O surface to form one-step nitrito charged complex............120 5-13. Simplified Mechanism for NO Adsorption on La2CuO4....................................................120 5-14. Reversible rearrang ement of the nitrite to nitrate formed by NOx adsorption....................121 5-15 .Simplified Mechanism for NO2 Adsorption on La2CuO4..................................................121 6.1 Tubular reactor for sensor reference studies..........................................................................137 6.2 La2CuO4 capacitor device for determining E-field effects on desorption.............................138 6.3 NO desorption from the La2CuO4 capacitor following NO adsorption.................................138 6.4 NO desorption from La2CuO4 powder following NO adsorption.........................................139 6.5 NO2 desorption from the La2CuO4 capacitor following NO adsorption................................139 6.6. NO desorption from the La2CuO4 capacitor following NO2 adsorption...............................140 6.7. NO2 desorption from the La2CuO4 capacitor following NO2 adsorption.............................141 6-8. Sensitivity plot of NO for the sensor....................................................................................142 6-9. R2 values of data fitting to linear plots for NO.....................................................................142 6-10. Sensitivity plot of NO2 for the sensor.................................................................................143 6-11. R2 values of data fitting to linear plots for NO2.................................................................143 6-12. Mass spectrometry comparison of NO in react or effluent from forward biased sensor testing........................................................................................................................ .......144
15 6-13. Mass spectrometry comparison of NO2 in reactor effluent from forward biased sensor testing........................................................................................................................ .......145 6-14. Mass spectrometry comparison of NO in react or effluent from reverse biased sensor testing........................................................................................................................ .......146 6-15. Mass spectrometry comparison of NO2 in reactor effluent from reverse biased sensor testing........................................................................................................................ .......147 B-1. Temperature Programmed Desorption of 1% NO on La2CuO4 at 30 oC per minute (a.) and 5 oC per minute (b.)...................................................................................................155 B-2. Temperature Programmed De sorption of 500 ppm NO2 on La2CuO4.................................156 B-3. Temperature Programmed Desorp tion of 0.67% NO and 0.33% CO on La2CuO4.............157 B-4. Temperature Programmed Desorption of 500 ppm NO, 500 ppm CO, and 1% on La2CuO4...........................................................................................................................158 B-5. Temperature Programmed Desorption of 500 ppm NO2, 250 ppm CO and 1000 ppm O2 on La2CuO4......................................................................................................................160 B-6. Temperature Programmed Desorpti on of 500 ppm NO and 1000 ppm CO2 on La2CuO4..162 C-1. La2CuO4 electrode on 8% YSZ substrate used for heat and gas cycling tests.....................164 C-2. XPS of La2CuO4 electrode surface following 3 weeks of cycling under sensor testing conditions. After 3 weeks in a thermally cycled environment and exposure to NO, NO2, CO, and CO2 gas mixtures, the structur e and composition of the La2CuO4 electrode remains constant according to XPS, indicating no silica deposition from siloxane occurs in the short term.....................................................................................165 C-3. XPS spectra of YSZ electrolyte on the ju st-sintered sample. Some trace peaks of the La2CuO4 electrode can be seen due to slu rry wetting during the screen printing process........................................................................................................................ ......166 C-4. XPS spectra of the YSZ elect rolyte after thermal cycling in sensor testing conditions. The peak for the Si 2p3 orbital was seen to increase due to diffusion of Si to the grain boundaries of the YSZ............................................................................................166 C-5. XPS spectra of the YSZ electrolyte below the La2CuO4 electrode after thermal cycling. The Si 2p3 peak was observed to increase over the uncycled sample, but remained lower than the uncovered and cycled YSZ sample..........................................................167 C-6. SEM photograph of the electrode/elect rolyte interface after sintering at 800 oC for 10 hours. The La2CuO4/YSZ interface was well-adhe red, and showed no sign of auxiliary phases............................................................................................................... .167
16 C-7. The surface of the La2CuO4 electrode shows no visible de posits of silica products after 3 weeks of thermal cycling under sensor testing conditions............................................168 D-1. C1s XPS spectra for unadsorbed (blue) and NO adsorbed (green) La2CuO4 powder.........169 D-2. La3d5 XPS spectra of unadsorbed (blue) and NO adsorbed (green) La2CuO4 powder......170 D-3. C1s XPS spectra of unadsorbed (blue) and NO2 adsorbed (red) La2CuO4 powder.............170 D-4. La3d5 XPS spectra of unadsorbed (blue) and NO2 adsorbed (red) La2CuO4 powder.........171 D-5. O1s XPS spectra of unadsorbed (blu e) and NO+CO adsorbed (purple) La2CuO4 powder......................................................................................................................... .....172 D-6. C1s XPS spectra of unadsorbed (blu e) and NO+CO adsorbed (purple) La2CuO4 powder......................................................................................................................... .....173 D-7. La3d5 XPS spectra of unadsorbed (blue) and NO+CO adsorbed (purple) La2CuO4 powder......................................................................................................................... .....174 D-8. O1s XPS spectra of unadsorbed (blue) and NO2+NO adsorbed (dark blue) La2CuO4 powder......................................................................................................................... .....175 D-9. C1s XPS spectra of unadsorbed (blue) and NO2+NO adsorbed (dark blue) La2CuO4 powder......................................................................................................................... .....176 D-10. La3d5 XPS spectra of unadsorbed (blue) and NO2+NO adsorbed (dark blue) La2CuO4 powder......................................................................................................................... .....177 D-11. O1s XPS spectra of unadsorbed (b lue) and CO adsorbed (orange) La2CuO4 powder......178 D-12. C1s XPS spectra of unadsorbed (b lue) and CO adsorbed (orange) La2CuO4 powder......178 D-13. La3d5 XPS spectra of unadsorbed (blue) and CO adsorbed (orange) La2CuO4 powder..179 D-14. Full range XPS comparison spectra of unadsorbed and adsorbed La2CuO4 powder samples........................................................................................................................ .....180 D-15. O1s XPS comparison spectra of unadsorbed and adsorbed La2CuO4 powder samples....180 D-16. C1s XPS comparison spectra of unadsorbed and adsorbed La2CuO4 powder samples....181 D-17. La3d5 XPS comparison spectra of unadsorbed and adsorbed La2CuO4 powder samples........................................................................................................................ .....181 E-1. Desorption Peak of N16O after NO adsorption on La2CuO4 16 with Gaussian Analysis.......182 E-2. Desorption Peak of 16O2 after NO adsorption on La2Cu16O4 with Gaussian Analysis.........183
17 E-3. Desorption Peak of 18O16O after NO adsorption on La2Cu16O4 with Gaussian Analysis.....183 E-4. Desorption Peak of N16O after NO adsorption on partially substituted La2Cu18O4 with Gaussian Analysis............................................................................................................184 E-5. Desorption Peak of 16O2 after NO adsorption on partially substituted La2Cu18O4 with Gaussian Analysis............................................................................................................185 E-6. Desorption Peak of 16O 18O after NO adsorption on partially substituted La2Cu18O4 with Gaussian Analysis............................................................................................................185 E-7. Desorption Peak of 18O2 after NO adsorption on partially substituted La2Cu18O4 with Gaussian Analysis............................................................................................................186 E-8. Desorption Peak of N16O after NO adsorption on La2Cu18O4 with Gaussian Analysis.......187 E-9. Desorption Peak of N18O after NO adsorption on La2Cu18O4 with Gaussian Analysis.......187 E-10. Desorption Peak of 16O2 after NO adsorption on La2Cu18O4 with Gaussian Analysis.......188 E-11. Desorption Peak of 16O 18O after NO adsorption on La2Cu18O4 with Gaussian Analysis..188 E-12. Desorption Peak of 18O2 after NO adsorption on La2Cu18O4 with Gaussian Analysis.......189 E-13. Desorption Peak of N16O after NO2 adsorption on La2Cu16O4 with Gaussian Analysis....190 E-14. Desorption Peak of N16O2 after NO2 adsorption on La2Cu16O4 with Gaussian Analysis...190 E-15. Desorption Peak of 16O2 after NO2 adsorption on La2Cu16O4 with Gaussian Analysis.....191 E-16. Desorption Peak of 16O 18O after NO2 adsorption on La2Cu16O4 with Gaussian Analysis....................................................................................................................... .....191 E-17. Desorption Peak of 18O2 after NO2 adsorption on La2Cu16O4 with Gaussian Analysis.....192 E-18. Desorption Peak of N16O after NO2 adsorption on partially substituted La2Cu18O4 with Gaussian Analysis............................................................................................................193 E-19. Desorption Peak of N16O2 after NO2 adsorption on partially substituted La2Cu18O4 with Gaussian Analysis....................................................................................................193 E-20. Desorption Peak of N16O18O after NO2 adsorption on partially substituted La2Cu18O4 with Gaussian Analysis....................................................................................................194 E-21. Desorption Peak of N18O after NO2 adsorption on partially substituted La2Cu18O4 with Gaussian Analysis............................................................................................................194 E-22. Desorption Peak of 16O2 after NO2 adsorption on partially substituted La2Cu18O4 with Gaussian Analysis............................................................................................................195
18 E-23. Desorption Peak of 16O18O after NO2 adsorption on partially substituted La2Cu18O4 with Gaussian Analysis....................................................................................................195 E-24. Desorption Peak of 18O2 after NO2 adsorption on partially substituted La2Cu18O4 with Gaussian Analysis............................................................................................................196 E-25. Desorption Peak of N16O after NO2 adsorption on La2Cu18O4 with Gaussian Analysis....198 E-26. Desorption Peak of N16O2 after NO2 adsorption on La2Cu18O4 with Gaussian Analysis...198 E-27. Desorption Peak of N16O18O after NO2 adsorption on La2Cu18O4 with Gaussian Analysis....................................................................................................................... .....199 E-28. Desorption Peak of N18O after NO2 adsorption on La2Cu18O4 with Gaussian Analysis....199 E-29. Desorption Peak of 16O2 after NO2 adsorption on La2Cu18O4 with Gaussian Analysis.....200 E-30. Desorption Peak of 16O18O after NO2 adsorption on La2Cu18O4 with Gaussian Analysis.200 E-31. Desorption Peak of 18O2 after NO2 adsorption on La2Cu18O4 with Gaussian Analysis.....201
19 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy HETEROGENEOUS CATALYTIC EVALUATION AN D ADSORPTION CHARACTERIZATION OF THE LA2CUO 4/YSZ/PT POTENTIOMETRIC COUPLE By Frederick Martin Van Assche IV May 2008 Chair: Eric D. Wachsman Major: Materials Science and Engineering The problem of scarce resources has advanced the development of more efficient lean-burn engine technology for power generation a nd transportation. Thus the issue of NOx production and the need to monitor that production become s increasingly important To better understand NOx sensing, the potentiometric couple of La2CuO4 and platinum across a yttrium-stabilized zirconia (YSZ) substrate was studied. Temperature programmed reaction and desorp tion studies were performed on powder samples to determine desorption strength and catalysis of NOx on La2CuO4. A mass spectrometer with one part per billion reso lution collected spectra for the experiments. NO gas proved nonreactive in both oxygen-rich and oxygen-lean conditions. NO2 gas desorbed in multiple gasphase products, and its reaction wa s catalytically activated by La2CuO4. Adsorption of NOx was studied by X-ray photoelect ron spectroscopy and infrared spectroscopy in diffuse reflection configuration. Results indicated the presence of ionic nitrate and nitrite complexes providing substantial elec tronic charge separation in large quantities. Photoelectron peak splitting of the O1s oxyge n orbital verified NO caused oxygen deprivation and increased resistivity, while NO2 adsorption resulted in crea tion of electron holes and an overall decrease in resistivity.
20 Oxygen isotope exchange was performed to determine the adsorption mechanism of NOx. Temperature programmed experiments using 18O2 gas over La2Cu16O4 and La2Cu18O4 identified bonding behavior and adsorptio n/desorption sequence of NOx. Formation surface complexes were found to occur exclusively with solid pha se oxygen, but substituti on of the primary oxygen in the complex was dependent on available ther mal energy. It could be concluded that the oxygen exchange is a very important component of the overall sensor response, and that the adsorptive component plays a large role in th e establishment of the potential between the electrodes. To adjust the adsorption equilibrium and m odify sensor response, desorption experiments under external electric field bias were performed on a La2CuO4 pellet. Initial bias, aging effects and increases in field strength were shown to change the desorption profile of NO and NO2 gases from the surface of the La2CuO4 pellet. A sensor utilizing a La2CuO4 pellet with similar field generating structures was also constructe d. It displayed sensitivity changes to NOx gases with changes in field voltage and bias.
21 CHAPTER 1 INTRODUCTION 1.1 NOX? The last 25 years have seen increased technol ogical modernization of the world, and with it, a massive demand for energy. This demand coin cides with economic and societal pressures for cleaner and more efficient power generation for transportati on and electricity. While these influences provide a significant impetus to devel op alternate sources of energy for future needs, current methods of power generation (such as combus tion of oil, coal, or na tural gas) will remain the most widely utilized processes for many years. As such, it is in the best interests of the world at large to optimize these processes. The simplest and most effective method to in crease the yield and efficiency of these combustion processes is to combust fuel in an excess of oxygen. This allows for complete conversion of the fuel and maximum heat liber ation. Unfortunately, the excess oxygen has the undesired effect of produci ng oxides of nitrogen (NOx).1 NOx is a problematic by-product for a number of reasons. First, NOx gases are biological irritants, and can react with airborne organics to produce smog and ozone (O3). In concentrated urban areas, NOx and smog cause respiratory problems after extended exposure.2 NOx can react with atmospheric water to form acids, which prec ipitate out as acid rain. Additionally, since this rain can migrate away from the cities, even rural areas are impacted by urban NOx generation.3 In stationary (e.g. power gene ration, industrial heating) a pplications, treatments are available to eliminate the NOx, but require bulky and delicate equipment in order to monitor the treatment process. In mobile applications, this equipment is unavailable to monitor NOx. This
22 fact, combined with the multitude of demands placed on engines, results in transportation applications producing the majority of NOx emissions.4 NOx production from transportation can only gr ow, as more cars fill the roads and consumer pressure for greater fuel efficiency pushes the limits of lean-burn combustion. Thus, treatment of these mobile sour ces will become essential in order to decrease pollution. A compact NOx sensor package would enable the applic ation of pollution removal technology to mobile sources of NOx generation. 1.2 Advantages of the Potentiometric Sensor Solid state potentiometric sensors have key advantages for NOx detection under lean-burn combustion conditions. These devi ces, based on the classic O2 sensor design, have excellent resistance to high temperature, vibration, and corrosion. They re quire no external power source or complicated electronic s to obtain a signal. Potentiometric sensors for NOx detection have some flexibility in design, and so can be constructed as either planar or tubular devices, depending on whic h electrodes are exposed to the sensed gas. Planar construction allows the se nsor fabrication using simple ceramic processing techniques such as tape casting and screen printing. This allows for excellent control of the electrode microstructure, as well as construction of a multi-layered structure that can integrate heaters, temperature sensors, and other elements. 1.3 Barriers to Commercialization At present, there are no commercial potenti ometric sensors available for specific NOx monitoring, instead estimating NOx concentrations indirectly vi a multiple measurement of total pollutant concentration. This is due to the comp licated response of curre nt sensor technology to NOx and other combustion products such as CO, water, and unburned hydrocarbons. Sensors of
23 this type tend to exhibit cross-sensitivity to mixtures of exhaust products, and so the signal generated must be deconvoluted in order to prov ide a useful value. This presents a substantial challenge, as the mechanism for sensing has not been deciphered for individual gases, and the interaction between the gases is not well established. Thus, the contribution from each component of the mixture cannot be accurately assigned and the total signal cannot be reliably determined under all conditions. Some resistive NOx sensors have been commercialized, though these devices suffer the same cross-sensitivity of NOx to other exhaust products. This limits their use to applications where the analyte mixture can be treated to remove products that may interfere with NOx sensing, such as stationary power generation.
24 CHAPTER 2 BACKGROUND 2.1 The Nernstian Sensor It is impossible to discuss se nsors in any depth without firs t discussing the qualities that compose a good example. The key attributes of a good sensor are sensitivity, selectivity, stability, and quick response. Sensitivity (S) can be expressed in a potentiometric sensor as the change in voltage across the elect rolyte with a given change in gas concentration, represented below in Equation 2-1. () ()  Vgas Sgas gas [2-1] Selectivity to a component gas is the ratio of the sensitivity of a desired sensate over the sensitivity of the device to an undesired sensate. Quick response from a sensor means that the time required to achieve a stable value is short, and a sensor w ith good stability will repeatedly deliver that value over time. The most widely used and best known exampl e of a solid electrolyte sensor is the -sensor, lambda sensor, named for its use in measuring the air/fuel ratio (denoted ) during combustion. This sensor consists of two metal electrodes se parated by a solid electro lyte layer capable of conducting O2ions, usually a doped ZrO2. One electrode remains exposed to a reference concentration of oxygen, and the other subm erged in an unknown concentration of O2. The difference in potential between these electrodes is sufficient to calculate the concentration of oxygen at the el ectrode in the unknown atmosphere referred to as the sensing electrode. The voltage between the electrodes can be related to the difference in concentration between the unknown and reference atmospheres by the use of the Nernst equation, assuming
25 that the oxygen atmosphere is in equilibrium w ith the individual electrodes, shown in Equation 2-2. 2 2[()] ln 4[()]O OPUnknown RT V FPReference [2-2] Since the behavior of these sensors is govern ed by the Nernst equation, they are known as Nernstian sensors. 2.2 The Non-Nernstian Sensor Nernstian behavior does not a pply to all sensors or conditi ons. When the gas mixture does not establish oxygen equilibrium with the sensing electrodes, it is possible to see a shift from these Nernstian values, making the potential gene rated across the electrolyte a non-Nernstian potential. Fleming observed in 1977 that the pr esence of several thousand parts per million CO in the exhaust atmosphere region of an oxygen se nsor caused a 100-200 mV shift in the sensor response.5 6 If the Nernstian response to an order of magnitude ch ange in oxygen concentration from 1% to 10% at 800 oC is examined, the result can be seen in Equation 2-3. 22 22[(.1)][(.01)] (lnln)50 4[(.21)][(.21)]OO OOPatmPatm RT VmV FPatmPatm [2-3] The sensitivity can be calc ulated using Equation 2-4. 2 2 2()50 ()555 100,00010,000 VOmV V SO p pm Oppm [2-4] This value can be compared to the sensitivity of CO on the oxygen sensor,7 which responds to a change from 100 ppm CO to 500 pp m CO with a 30 mV non-Nernstian response, seen in Equation 2-5. 3()30 ()7510 500100 VCOmV V SCO p pm COppm [2-5]
26 This was a much larger shift than a co rresponding shift from a change in oxygen concentration. If the selectivity of the two gase s is compared, the sensitivity can be evaluated. (Equation 2-6) 3 27510 () 135 () 555 V SCO ppm V SO ppm [2-6] This means that the sensitivity is 135 times higher for CO than for O2, and so the oxygen sensor is quite selective for CO over oxygen. This non-Nernstian behavior occurs in the presence of NOx and other exhaust products as well, and so the need to unde rstand the factors behind this behavior becomes readily apparent. 2.3 Initial Theories for Non-Nerstian Behavior The non-Nernstian effect has been st udied for many types of sensors. NOx has been studied the most intensely,7 79 but other exhaust products such as hydrocarbons,13 15 80 87 hydrogen sulfide,88 SOx, 89 91 hydrogen,37 38 50 92 and N2O93 95 have also been examined. The overall mechanism for potentiometric sensing has not yet b een totally explained. Since Flemings initial work in 1977, the deviation of potentiometric co uples from the Nernst equation has been an extensively debated topic. Initial ideas to explain the deviation from th eory centered on the modification of the local O2 atmosphere as a way of changing the potential at an electrode without complication of the phenomena associated with the O2 reaction on the surface. It was noted that catalyst powders on resistive semiconductor electrodes had mo re rapid responses to carbon monoxide.96 It was assumed that the catalyst was converting COCO2 by removing the oxygen above the electrode, and this behavior shifted the response.
27 2.4 Mixed Potential In 1982, Williams97et al. applied a concept from aque ous corrosion theory, Mixed Potential, to the gas phase in order to model th e potentiometric response. This theory supposed that multiple electrochemical reactions simultaneously occur over a conducting surface, and the potential of that surface is inte rmediate between the half-cell r eactions of the electrochemistry occurring there. In sensors, the electrodes f unction independently, and the potential difference across the cell would be the addi tive sum of the potentials ge nerated at each electrode. In the case of the potentiometric sensor c oupled around yttrium-stabilized zirconia (YSZ), the basic reactions with the environment occur as follows. YSZ, being an oxygen ion conductor, will equilibrate with the oxygen in the environment according to Equation 2-7. 2()1 ()2'()() 2x gOOOVYSZeexteriorOYSZ [2-7] Pollutant gases, such as CO or NOx, will equilibrate according to the general expression in Equation 2-8. ()()Re()'()x gOgOOYSZOxVneexterior [2-8] Re(g) and Ox(g) refer to the reduced and oxidi zed pollutant gases (NO/NO2, CO/CO2 ) in the mixture. Depending on the kinetics of each reac tion at each electrode, a potential will evolve to drive the charges generated by the reactions around the exterior circuit. This simplified approach constitutes the Mixed Potential theory for sensors, and is more extensively explained in other articles.31 45 81 98 100 It is assumed by this mechanism that the gas concentrations at the interface of the electrodes and electrolytes reach a full equilibrium according to the reactions stated above. The potential between the electrodes comes about from the difference in electrochemical rate of reaction at the electrodes. At th e time, this theory was widely adopted as the explanation for
28 almost all non-Nernstian sensors, though the conclusions of the m echanism had not been verified by gas analysis. 2.5 Differential Electrode Equilibria In 2000, an alternate explanation, ca lled Differential Electrode Equilibria,63 was proposed. This approach postulated that the mixed potenti al theory adequately describes some of the behavior of sensors at high temper atures, where kinetics allows for fast equilibrium, but at lower temperatures the effect of the electrochemical r eaction is only one of the factors affecting the surface. In an environment where an electrochemical reaction can easily occur, such as methane and trace oxygen, the largest source of electrons will come from the establishment of steady state equilibrium by these electrochemical reactions.101 By contrast, the thermodynamic equilibrium for NO gas at high temperature shou ld result in decomposition to N2 and O2, but this outcome does not occur;102 thus, the electrochemical conversion r eaction cannot play a major role. If the reaction does not reach its thermodynamically defi ned equilibrium state, it is unlikely that the electrochemical reaction of the sensed gas is the sole factor in determining the voltage generated between the two electr odes of the sensor. To accurately model the output of the sensor at lower temperatures, the Differential Electrode Equilibria theory suggest ed that the other contributions to the electrode potential must be combined with potentials from electrochemical reactions. Gas concentrati on at the electrode is affected due to non-electrochemical catalytic conversion, and this change must be accounted for in order to predict the change of voltage response to a change in concentration. In addition to this effect, adsorption contributes to the electrode potential by inducing changes in the Fermi level. These behaviors change with temperature and reactant concentration, which can affect the voltage response of the sensor.
29 For sensors utilizing a metal re ference electrode in a relativel y inert reference gas, it is possible for the electrochemical reactions over the sensing electr ode to account for the majority of the potential across the electr olyte. Non-electrochemical catalys is at the sensing electrode will not affect the potential at th e reference electrode directly, though it can modify the local concentration of the desired sensat e gas over the electrode and so a ffect sensor response in that manner. If the sensing electrode catalyzes a reaction away from the three phase boundary (TPB), no electrons will be present in the exterior circuit as a result of the reaction, since no compensating ionic current can flow through the electrolyte from the TPB. Thus, no potential will be generated as a result of the reaction. Since the reference electrode will operate under constant atmosphere, any change in electrochemical reaction rate at the sensing electrode will be very important.103 By contrast, when both electrode s are exposed to the sensing e nvironment, the presence of both electrochemical and non-electr ochemical catalytic reactions at both electrodes magnifies the importance of adsorption. Reactions causing the shifts in the potenti al of the electrode now occur over both the sensing and counter electrodes, and so the contri butions from these reactions cancel each other. This does not necessarily mean the contributions can cel out entirely. If a highly catalytic counter-electrode such as platin um is used, the sensed gas may react before reaching the three phase boundary of the counter-ele ctrode. As a result, the partial pressure of both the sensing gas and oxygen will be differe nt at the three phase boundary (TPB) for each electrode. Because of this, the potential genera ted becomes a function of both the concentration in the gas stream and the differing concentrations of gases at the electrode /electrolyte interface of each electrode.
30 In addition to the effects caused by catalysis, adsorption of gases onto the surface of the electrode may induce a shift in the electrode pot ential. Metals predispose to adsorb gases somewhat independently of the gas atmosphere and the charge concentration generated as a result of this adsorption tends to be small. Voltage s created as a result of the adsorption behavior are assumed confined to the non-metal electrodes. If the sensing electrode is a semiconductor, adsorption can play a large role in potential generation. The Differential Electrode Equilibria model includes semiconductor sensing effects, and it is important to note th em due to the effect of surface morphology on the experimental sensi ng behavior. The grain size of the electrode powder is closely related to th e roughness, and hence, the surf ace area of the electrode. Small grains of powder contribute to the length of the three pha se boundary by providing a more effective interface between the electrode and el ectrolyte. Since the powders used for these experiments often contain substantial amounts of sub-micron sized powders, the surface area of the sensing electrode can be quite large.104 2.6 The Semiconducting Response The semiconducting response of a sensor is ba sed on the change of the work function or surface charge by the influence of adsorbed radi cals at the surface and interface regions. The work function of a compound is the sum of the ener gies required to remove an electron from the compound to a distance outside the su rface. It includes the internal work function, known as the Fermi energy, the external work function, a nd the sum of adsorbate components corresponding to the surface charge. The idea of semiconducting response has its basis in the resistive sens or, where the change in resistance indicates the sensor response. During adsorption bonds form from the adsorbate to the surface, and as a result either reduce or e nhance the concentration of electronic charge carriers at the surface of the semiconductor. The conductive process of the system is described
31 relative to the Debye length, expressed in Equation 2-10, where k, T, q, and n are the Boltzmann constant, absolute temperature, elec tron charge, dielectric constant, and chargecarrier concentration.105 1 2 2 DkT L qn [2-10] This characteristic length can sp an into the micrometer range, and become larger than the grains of the semiconductor, depending mos tly on the temperature and charge-carrier concentration. This can range from 3-5 nm for SnO2 106 107 to roughly 7 nm for TiO2, 108 to 15 nm for WO3 109 for an operating temperature of 400-500 oC. The Debye length controls the depth of the Randschicht (space charge) layer,110 which compensates for the charged species formed at the adsorbate/ substrat e interface surface by the adsorption of gaseous species. (Figure 2-1) Materi al at the interface maintains equivalent charge density, which decreases with distance from the interface. The depth of this conduction pathway is highly dependent on the work func tion of the semiconductor compound. For semiconductors with large ch arge carrier concentrations ( n ) the Debye length will be small and the Randschicht layer will have limite d influence. As a result, adsorption will not greatly affect the resis tivity of the material. In the case wh ere the semiconductor has a very low charge carrier concentration (i.e. an insulator), th e Randschicht layer will be very large in order to balance the charge, and the effect w ill be easily discernable through conductivity measurements. For resistive sensors, the indivi dual grains interconnect to provide an electronic pathway, and the size and connectivity can be influenced by ceramic processing. When LD is small with respect to the grain size, only th e outside of the grains are influenced by the space charge layer.
32 The conduction volume affected by gas interaction is limited to the actual interface between gas and grain, and at the grain/ grain boundary. When grain size is small with respect to LD, the entire grain is affected by the charge layer and the entir e bulk conductivity is modi fied as a result. This effect has been seen on many semiconductors, for many different gases.106 109 111 This interconnection of the space charge layer due to grain size observed for resistive sensors also has implications on the potentiometric sensor. The ch ange in bulk conductivity will naturally have an effect on establishing a charge balance, and small-grained electrodes also will have a greater surface area to establish the pot ential. This can shorten the response time to establish the potential, but may also impose an upper limit to detection capabi lity if a sufficiently large Randschicht layer cannot form to compensate for the charge on the surface. The semiconducting effect can influence th e sensor behavior even without a gas environment. In the full sensor, the initial voltage between the elec trodes may be a non-zero value. The Fermi level difference between a se miconducting sensing electrode and a metallic reference electrode generates a potential difference even in the absence of adsorbant gases. Surface adsorptions modify the surface with charge changing the work function and altering the potential difference between the electrodes112 113 as shown in Figure 2-2. The total work function ( ) consists of the external work function ( ) combined with the Fermi energy ( v) and the contribution of the surface charge from adsorption ( S). 2.7 Materials of Sensing A key point in further deci phering the sensing mechanism for both potentiometric and conductivity sensors will be materi al selection. As mentioned pr eviously, metals would be poor choices for a conductometric sensor, as the ove rall change in conductivity due to gas-phase variations is very small relative to the total c onductivity of the metal. This problem becomes less important with respect to potentiometric sensor s, as the metal can function as a known catalyst
33 and interface. Indeed, Pt6 7 18 23 25 28 30 33 41 43 63 68 70 80 81 88 92 114 127 and Au13 25 50 81 83 85 92 115 126 128 have found extensive use as counter el ectrodes in potentiometric sensors. One type of material that has found a us e in both potentiometric and conductometric sensors are metal oxides. Since the oxides of metals do not retain the full metallic bonding character of their parent, they have a far greater resistivity, and are thus classified as semiconductors. Additionally, metal oxides can extensively interact with gases due to modification of their oxygen chemistry. These ch anges allow catalytic conversion with charge transfer and cause significant ch anges in the concentr ation of charge carriers. Some examples and references of metal oxides used in conducto metric and potentiometric sensors are given in Table 2-1. As different oxides demonstrate di fferent catalytic properties with changes in composition, it is necessary to choose the correct oxide for th e gas environment. 2.8 Gas Phase NOx and CO Chemistry NOx composition in the gas phase can be co mplex, and the thermodynamically favored composition of the gas phase can be quite diffe rent than the actual composition. At low temperatures, a metastable equilibrium occurs. This equilibrium unbalances only in cases of extraordinary driving forces, such as large Keq values seen at temperatures where sufficient thermal energy exists to overcome any barriers of activation energy. To calculate the equilibrium composition of a system, several components are required. First, possible occurring species must be identified, and the Gformation values for each independent formation reaction must be found. The composition can then be optimized to mini mize the free energy of the system at a given temperature. This process is best performed by a numerical solver so ftware package coupled with a database of Keq values. Calculations were performed using TEST software129 130 to determine the equilibrium composition of a simulated exhaust mixture (3% O2, 15% CO2, 3% H2O, and 650 ppm NO) used
34 for potentiometric sensing experiments. The plot of the calculations can be seen in Figure 2-3. At all temperatures shown, thermodynamic equilibri um heavily favors the decomposition of NO to N2 and O2. NO2 is more stable than NO at temperatures below 425 oC, but the overwhelming conversion is for NOx species to decompose. Temperatures above 600 oC are more favorable to the formation of stable nitrogen compounds with oxygen, such as N2O, NO and NO2, with a significant majority of the balance resulting from N2 and O2 formed by decomposition, but these remain minor products until temperatures far above the values shown. Low temperature thermodynamics support the decomposition of NO and NO2 to the component molecules of N2 and O2. However, the large activation energy to achieve that result is difficult to muster at temperatures below 600 oC, so NO and NO2 remain intact at those temperatures in the gas phase. At temperatures far above 600 oC, thermal energy is plentiful enough to provide for fast kinetics, so ther modynamic calculations provide a more accurate estimation of the true composition. 2.9 La2CuO4 Background La2CuO4, the simplest member of the layered per ovskites, is the parent compound of the La2-x(Sr,Ba)xCuO4+y family of high temperature (Tc< 40 K) superconducting oxides. A superconductor itself at temperatures below 30 K, La2CuO4 functions as an intrinsic semiconductor from 30 to 50 K, and above these ra nges, as an extrinsic acceptor-doped (P-type) semiconductor.131 The acceptor dopant in this case is O2 gas, which induces interstitial Oi ions into the rock salt la yer of the Ruddleson-Popper stru cture as the primary defect.132 135 The concentration of inters titial oxygen is dependent on both th e oxygen partial pr essure and the extent to which the gas can equilibrat e with the oxide. Equilibrium of La2CuO4 can be affected by the presence of ionically blocking ad sorbates or very low temperatures.
35 Kanai proposed a valence change for copper fr om plus two to plus three to accommodate the excess oxygen, thus defect equilibrium c ondition can be expressed in Krger-Vink notation in Equations 2-11 and 2-12. 1 2 2()2giOOh [2-11] 2 12 2 ()iOp K PO [2-12] In this case, K represents th e equilibrium constant with regards to oxygen. From these equations, the neutrality condition can be expressed by Equation 2-13. 2iOp [2-13] By substitution, the stoichiometric modifier can be defined as Equation 2-14. 16 2 hpO [2-14] This is in agreement with the p-type behavior of La2CuO4+ La2CuO4 maintains a slight excess of oxygen with regard to stochiometry throughout the oxygen partial pressure range, but decomposes if exposed to a pO2 of less than ~10-6.8 atmospheres at 800 oC. At that partial pressure, the oxide is in stochiometric balance.134 At higher temperatures, the oxide can become sli ghtly hypostoichiometric before decomposition; however, for significant hypostoichiometry to manifest the A site of the material must be doped. This doping greatly complicates the analysis of defects. Due to its use in superconductivity research, the structure and electr onic properties of the material have been studied extensively. The unit ce ll structure consists of a perovskite unit cell of CuO2 sandwiched between layers of rock salt La2O2 lattice.136 137 Below 200 oC, the unit cell organizes to an I4/mmm orthogonal structure; above that temperat ure, it retains tetragonal Fmmm structure.138 The material can be constructed a number of ways, including solid-state
36 synthesis,139 140 co-precipitation,141 mixed-oxalate reaction,132 142 sol-gel synthesis,143 freezedrying,63 64 inverse microemulsions,144 metal nitrate reaction in supercritical water,145 and the Pechini method.146 Most of these synthesis methods produce fine-grained, semi-crystalline powder that fully crystallizes during the calcination step. The orthorhombic perovskite phase is form ed by calcination at temperatures above 650 oC. Calcination temperatures below this threshold re sult in partially or completely amorphous solid oxides that can be crystallized by heating above the threshold.141 Figure 2-1. Diagram of the space charge layer at the surface of the electrode. Opposing charges build up inside the material to compensate for the charge established on the surface. The depth of this charged layer is dependa nt on the material (taken from Vayenas)147
37 Figure 2-2: Surface band model of a p-type semiconductor involving positive surface charge. The total work function ( ) consists of the exte rnal work function ( ) combined with the Fermi energy ( v) and the contribution of the surface charge from adsorption ( S)(taken from Vayenas)147 Table 2-1. Metal oxides commonly used as resi stance-type and potentiom etric gas sensors and relevant publications. Metal Oxide Selected References for Resistance-type Sensors Selected References for Potentiometric Sensors WO3 9109 121 148 154 919 21 23 25 33 34 68 155 TiO2 13108 126 151 156 168 214 57 SnO2 2116 17 62 106 107 156 157 165 169 181 367 70 182 LaFeO3 6156 157 165 183 185 618 19 23 32 34 71 SmFeO3 5165 183 186 188 118 LaCoO3 1189 118 ZnO 937 180 190 196 337 38 196 In2O3 7156 165 176 180 197 199 424 86 200 201 Nb2O5 1202 315 201 203 La2CuO4 3181 193 194 67 64 66 125 204 205
38 10-2110-1910-1710-1510-1310-1110-910-710-50.001 0.1 0100200300400500600700Mole Fraction Temperature (oC) N2O2NO2N2O CO CO2NO H2O Figure 2-3. Plot of equilibr ium mole fractions of exha ust gas consisting of 3% O2, 15% CO2, 3% H2O, 650 ppm NO, and balance N2 versus temperature in oC
39 CHAPTER 3 INITIAL TPD/TPR STUDIES OF LA2CUO4 AND PT 3.1 Introduction A cornerstone tool for discerning the contributio ns of Mixed Potential versus contributions of the semiconducting effect in both potentiomet ric and conductometric sensors will be the evaluation of the effluent gas from the exposure of sensor electrode materials to gas mixtures commonly seen in combustion environments. The relative contribution of each mechanism will largely depend on the composition of the electrode and the temperature at which it operates. For metal electrodes, electrochemical conversio n is the most likely mechanism for changes in the potential, since adsorptiv e effects only contribute a small am ount of charge relative to the large volume of charge carriers already present in the material. Adsorption on metals can cause changes in the local electronic environment through mixing of the adsorbate/metal orbitals206 207 and can also form reaction barri ers to electrochemical reaction208 even without permanently affecting the surface. However, since pure metals cannot supply anything but electrons to form charge-generating complexes at the interface wit hout passivation of the surface, the extent to which potential can generate at the interface is limited. Additionally, the intrinsic conductivity does not change much over the investigated te mperature range, which, though not so important for the potentiometric sensor, lowers the possibilit y of error associated with conductivity. As a result, major changes in response are most likely due to the increased kine tic activity for reaction found at higher temperatures. Conversely, a semiconducting compound involve s different properties that complicate analysis, but also allow for more selective us e. In semiconductors, adsorption causes a much larger relative change in the charge carrier concen tration, and this change can be very important for determining the conductometric and potentiometric mechanisms. Thermal stimulation causes
40 a much greater, and positive, change in conductiv ity due to the excitati on of electronic carriers within the material. This stimulation has the added effect of speeding up the electrochemical reaction kinetics and adsorption / desorption beha vior at the surface, each of which can have substantial effects on the electrode potential. In general, the adsorption phenomena are the determining mechanisms at temperatures below 500 oC, and electrocatalytic reactions dominate above that temperature. This behavior has been examined previously on a TiO2 sensor.57 La2CuO4 demonstrates promise as a sensing electrode material to explore the semiconducting mechanism due to its lack of catalytic activity for the NON2 reaction in the presence of oxygen.102 Therefore, contributions from catal ytic and electrochemical conversion reaction of this type are minimal during NOx sensing. This allows the contribution from gas adsorption to be the primary influence of charge creation for a sensor under NO. Measurement of the adsorption behavior was compared to the se nsor behavior, which better allowed for the separation of the various contributions to the electric potential generation and conductance. Platinum was selected as the counter electrode for the potentio metric couple due to its known catalytic ability and extensive documentation in th e literature as an elect rode in potentiometric sensors.6 7 18 23 25 28 30 33 41 43 63 68 70 80 81 88 92 114 127 Mass spectrometry (MS) was used to examin e the heterogeneous catalytic activity and adsorption activity of the sensing and reference electrodes, specifically La2CuO4 and Pt, for NOx and COx exposure. The use of MS to monito r reaction kinetics has some precedent.209 220 By recording instantaneous changes in gas concen tration with time and/or temperature, the desorption energy, quantity of NOx gas desorption, and the decomposition products formed by reactions with NOx at high temperatures were evaluated.
41 Profiling the heterogeneous catalyt ic activity of a sensing material was very important for determining the contribution of that catalytic reac tion to the potential generated at that electrode for the complete device. NO and NO2 mixtures are kinetically limited in the gas phase from reaching equilibrium, but in the presence of a ca talyst can react in an attempt to establish equilibrium. By altering the react ion conditions and monitoring the result, the initi ation point of catalytic activity can be observed. Additionally, since the gas composition can be recorded after the reaction has taken place, the extent of catalysis can be determined by comparison to the thermodynamic equilibrium. 3.2 Experimental 3.2.1 Synthesis of La2CuO4 La2CuO4 powder was synthesized by the autoignition technique.221 A stoichiometric mixture of La(NO3)3H2O (Kanto Chemical Company, 99.99%) and Cu(NO3)210H2O (Alfa Aesar, 99.999%) was prepared in deionized water. After dissolving the metal nitrates, citric acid (Alfa Aesar, 99.5%) was added to the mixture until 0.25 mole percent was achieved, making this a fuel-lean solution. The solution was heated to ~70 C with constant stirring until a gel-like precipitate dropped from solution. The liquid wa s boiled down and the prec ipitate autoignited by further heating. The powder generated by the auto ignition process was calcined at 600 C for 10 hours. 3.2.2 Powder Characterization X-Ray diffraction (XRD) was performed on a Philips APD 3720 automated diffractometer (Cu K radiation, = 1.54178 ) to verify the phase composition of La2CuO4 powder. Step scans were performed over a 2 interval of 0.02 degrees and a retention time of 5 sec for each step in the 20 2 60 range. Surface area was recorded using a N2-adsorption B.E.T.
42 measurement (NOVA, Quantachrome). Calcined powders were examined using an JEOL JSM6330F Field Emission SEM. 3.2.3 TPD/TPR Setup Temperature Programmed Desorption and React ion (TPD/TPR), as well as isothermal experiments, were performed in a quartz tube flow reactor. The reactor consisted of a 4 mm inner diameter quartz tube joined to a 1 mm inner diameter quartz tube by a glass frit. The tube ends linked to stainless steel tubing th rough Ultra-Torr fittings. A con centric glass tube containing a K type thermocouple was inserted through the upper fitting and suspended directly above the sample to monitor the temperature. Gas flow in to the reactor was controlled using a bank of MKS mass flow controllers (MFC) through a stainless steel tube intake manifold into the top of the reactor. Effluent reactor gases flowed out the reactor exit into an Extrel Merlin 5220 quadrapole mass spectrometer. A full schematic of the gas setup can be seen in Figure 3-1. 3.2.4 TPD/TPR Procedure The procedure of the temperature-progra mmed desorption (TPD) and reaction (TPR) experiments is described below.66 68 69 205 Samples of La2CuO4 containing 43 milligrams of powder (total BET area 0.15 m2) were initially purged of at mospheric adsorbed species by cycling the sample to 800 oC and back to room temperature under an atmosphere of 100 % oxygen, followed by heating under helium to show desorption spectra. The process was repeated until the spectra did not show any desorption pe aks. Once clean, the sample was cooled under oxygen and then maintained under a helium at mosphere until further experiments were performed. For standard TPD experiments, the sample was subjected to 30 cubic centimeters per minute (ccm) flow of a considered gas or gas combination at 300 oC for 30 minutes, followed by cooling to 30 oC at 5 oC / minute. At 50 oC, the sample was purged with 30 ccm helium until a
43 stable baseline for mass analysis was achieved. The sample was heated to 800 oC at 30 oC /minute under 30 ccm flowing helium atmosphere. Du ring heating, the effluent gas was analyzed to determine the quantity and na ture of desorbed products. Parallel experiments utili zing adsorptions from 700 oC to room temperature yielded similar results to adsorptions cooled from 300 oC to room temperature. The lower adsorption temperature was standardized to prevent reaction of gases at high temperature and retain powder microstructure over time. Additional TP D experiments were performed with NOx gases and CO at heating rates of 1, 5, 10, and 15 oC / minute in order to determine the activation energy of desorption of these gases on La2CuO4. TPD experiments are listed in Table 3-1. For TPR experiments, the sample was expos ed to 30 ccm of a reaction mixture at 30 oC, and held until gas flow and mass spectrometer signa l remained steady. The sample was heated to 800 oC at 30 oC / minute, and the effluent spectra record ed to determine the extent of reaction and reaction products. These experiments were performed on La2CuO4 powder, a Pt electrode on YSZ, and in the blank reactor. TPR experiments were performed using Pt and the blank reactor to determine the temperature points of uncatalyzed gas-phase reaction, and the catalytic effect of Pt versus the gas phase reactions. Table 3-2 shows the list of experiments. For all mass spectrometry experiments, specific mass to charge ratios were examined: m/z = 12 (C), m/z = 14 (N), m/z = 28 (CO and N2), m/z = 30 (NO), m/z = 32 (O2), m/z = 44 (CO2 and N2O), and m/z = 46 (NO2). The powder sample was evaluated by TPD and TPR for each combination of gas, and then cycled to 700 oC under helium before beginning the next gas combination. Raw data was collected in arbitrary intensity units for each ratio of mass to charge, which was converted to concentration in parts per million by comparison to previously collected calibration data.
44 Calibration of the system was achieved by m easuring the intensity output of the mass spectrometer for m/z ratio of atomic weight for gas and characteristic fragment gases versus input concentration of sensed gas minus the background spectrum. Experiments were repeated from 6-10 times with variations of <10%. Spectrums shown in later figures are discrete, but representative of the aver age of all experiments. 3.3 Results and Discussion 3.3.1 Powder Characterization Results X-ray diffraction of the calcined powd er revealed formation of the La2CuO4 phase, showing good phase homogeneity with tr aces of copper oxide impurity. The La2CuO4 phase was able to form below the threshold calcination temperature due to the high-temperature powder formation inherent in th e auto-ignition technique.222 (Figure 3-2) Minor st ructural inhomogenaity left from processing was eliminat ed by cycling the sample to 700 oC during sample pretreatment. BET measurements using N2 gas revealed a specific surface area of 3.48 m2/g. SEM images of the powder show fine-microstru ctured particles of irregular shape. The particles have average sizes of 100 to 300 nm and appear to contain networ ks of small pores in each particle.223 (Figure 3-3) 3.3.2 TPD/TPR Results of NO Gas Exposure Initial studies of the La2CuO4 powder focused on comparing the adsorption of NO gas in the absence of oxygen and contrast ing that spectrum to NO adsorp tion in the presence of oxygen. Mass spectrometry spectra of the 1% NO in He adsorbed case showed clear desorption peaks at low temperature. (Figure 3-4) The low-te mperature desorption begins just below 100 oC, peaking just above 150 oC. As heating continued, a larger deso rption peak appeared, with its maximum coming around 275 oC. This higher temperature desorption continued as temperature increased, with a shoulder in the peak extending to 450 oC. Above 450 oC, the surface has been exhausted
45 of NO adsorbates, and shows no further NO desorption. NO2 desorption was noticeably absent during the NO TPD experiment, so no conversion from NONO2 takes place solely from interaction with the La2CuO4 powder. When La2CuO4 adsorbed NO in the presence of oxygen, no change occurred in the desorption spectrum, seen in Figure 3-5. The quantity of NO desorbed from the surface for the 500 ppm NO / 0.95% O2 adsorption was almost identical to the quantity desorbed from the 1% NO adsorption. No NO2 desorbed during the experiment, though this was omitted from the plot for purposes of clarity. This means that O2 has no effect on the adsorption of NO, nor does the presence of O2 contribute to the NONO2 reaction, even in 0.95% O2. Further, it was demonstrated that the La2CuO4 powder can saturate to maximum surface coverage in 500 ppm NO. Thus, the NO adsorption was quite favorable on La2CuO4. To examine the energies of adsorption, the NO adsorption and desorption cycle was repeated at multiple heating rates, as seen in the literature. First, following Redheads flashdesorption work,224 the rate of desorption was defined (Equation 3-1). ()exp()n naNtddtERT [3-1] In this equation, n is the desorption reaction order, is the surface coverage (in molecules/cm2), n is the rate constant, and Ea is the activation energy of desorption in Joules/mole. By identifying the temperature fo r which gas emission from the surface peaks (Tp), the following relations can be expa nded (Equations 3-2, 3-3, 3-4). 2 1(/)exp()apapERTERT : if n =1 [3-2] 2 2(2/)exp()appap E RTERT : if n =2 [3-3] 2 02()exp()apapERTERT : if n =2 [3-4]
46 In Equations 3-2, 3-3, and 3-4, is the heating rate in oK / second, 0 is initial surface coverage, p is the coverage at T=Tp (oK), 1 and 2 are reaction rate constants. 1 and 2 are assumed such that 1013> n/ >108 (oK-1). For these equations, p and 0 can be calculated from the individual TPD plots by measuring the total desorption from the surface. Using the known flow rate and time per scan, the mole value of gas desorbed can be calculated on a per second basis. Using z ccm flow of carrier gas as a basis, the amount of gas stored on the powder can be ca lculated (Equations 3-5 and 3-6). 3 83 6()1min11sec 1.6710() min60sec10 zcm zcmppmscan ppmscan [3-5] 6 837 3110 1.6710()s7.4410() 100022.4 Lmol zcmppmcanzmolppmscan cmL [3-6] The total storage capacity of the powder for NO gas at each adsorption site type is represented by Equation 3-7. 7 0 2 17.4410() ()n scan scan x ppmzmolppmscan ycm [3-7] The NO remaining on the surface is given by Equation 3-8. 7 0 2 17.4410() ()p scan p scan x ppmzmolppmscan ycm [3-8] In Equations 3-7 and 3-8, y is given as an arbitrary basis for the surface area of the sample. From these equations, the adsorption energy of NO for 1st and 2nd order desorption can be calculated. Second-order desorption can occu r as a result of incomplete adsorption,209 and while based on previous TPD that is unlikely, the equati ons needed to calculate that value have been included.
47 Using the assumed value for 1, the Ea value can be calculated at each ramp rate by substituting into the empirical equati on developed by Redhead. (Equation 3-9) 1ln3.64apERT [3-9] The value of Ea can also be found without assuming 1 by using a second empirical equation and plotting log vs. log Tp. (Equation 3-10) (log) 2 (log)ap pd ERT dT [3-10] To be thorough, the activation energy cal culations were repeated using a method previously applied to Temperatur e Programmed Reduction by Wimmers et al.216 217 Adapting their conversion reaction to a gas-phase indepe ndent term transforms the expression into Equation 3-11. 1 ()() d kTf dT [3-11] In this equation, is the degree of desorp tion for that type and is the heating rate. Further, the Arrhenius relationshi p can be substituted to remove k and produce Equation 3-12. ()aERTdA ef dT [3-12] Following Kissingers method for determining activation energies,212 these equations can be related: At the peak temperature Tp, Equation 3-13 holds. 0pTTdd dTdT [3-13] Combining Equation 3-13 and 3-12 forms Equation 3-14.
48 exp()()0pa TTdA ERTf dT [3-14] This transforms to Equation 3-15. 2() exp()0ppa ap TTTT pE ddfA ERT dtRTd [3-15] Since p TTd dT does not equal zero, this expression reduces to Equation 3-16. 2() exp()pa a TT pE df ERT ARTd [3-16] This expression further simplifies to Equation 3-17. 2() lnlnln p aa TT ppEE df TARRTd [3-17] Making the assumption that f( ) and at Tp are independent of and that () p TTdf d does not equal zero, this equation fu rther simplifies to Equation 3-18. 2lnlna ppaE AR C TRTE [3-18] In Equation 3-18, C is assumed to be a constant. Thus, a plot of 2lnpT vs. 1pT should give a straight line with slope aE R The results of these calculatio ns can be seen in Table 3-3. In the case of NO desorption, the difference between the three methods of calculation is readily apparent. While these activat ion energies are rather different in the first case versus the second and third case, it must be noted the Wi mmers and Redhead 7 methods are particularly
49 susceptible to problems from a single outlier, as the desorptions take place at relatively low temperatures. Surveys of other metal oxides indi cated the possibility of charged complexes on the surface of the powder due to adsorption.225 226 This could cause lateral interactions, which might affect the peak position.227 Each of the values obtained fr om Redheads empirical equation relating only ramp rate and temperature for the desorption system were within 2 kJ/mol of one another for each ramp rate, and for that reason the calculations in the first column are assumed to be the correct ones within 5% and the othe r values provided as a consistency check. Desorption peaks present in the TPD app ear during the NO TPR, but no changes in concentration consistent with NO conversion occur with or without the presence of oxygen in the experiments. La2CuO4 was exposed to 500 ppm NO in He and no decomposition of NO to N2 and O2 was observed. (Figure 3-6a) Similarly, in the NO and O2 TPR, when exposed to 500 ppm NO in 1000 ppm O2 (Figure 3-6b), no NO2 formation was observed. This experiment was repeated in a background of 21% O2, and no NO2 formation was observed. Therefore, La2CuO4 was catalytically inactive for NO decomposition or oxidation. Thermodynamic calculations suggested a near-complete conversion to NO2 at low temperatures. However, gas phase and Pt electrode measurement of the NO / O2 system showed no activity towards NO conversion. (Figure 3-7) This deviation was unsurprising in the gas phase case, as per the discussion in Chapter 2. The experiments ove r Pt reinforced that oxidation from NONO2 was not occurring over either the La2CuO4 or Pt electrodes. From this, it can be stated that the response coming from the sensing electrode is not a result of electrochemical conversion reactions of NO. Further, the poten tial from the semiconducting effect was not influenced by changes in concentra tion as a result of reacting species.
50 3.3.3 TPD/TPR Results of NO2 Gas Exposure The desorption profiles obtained from the NO2 on La2CuO4 powder show extensive surface adsorption. (Figure 3-8a) NO2 exhibited strong chemisorpti on, but the desorption profile contained multiple unique adsorbed NOx surface species. NO2 begins desorption at 230 C with a peak maximum near 280 C. This NO2 desorption peak coincided with a NO desorption peak near 300 oC, larger than the desorption peak from NO alone. A second NO desorption peak near 400 C superimposed itself over an O2 desorption peak indicating decomposition of a surface NOx species. From these patterns, it can be obs erved that a series of complex reaction steps occurred on the surface of La2CuO4 at the lower end of the temper ature range. At temperatures above 600 C, lattice oxygen discharged from th e sample. This oxygen emission resulted due to favorable energy conditions that allowed rearrangement of the de gree of superstoichiometry of the La2CuO4. Temperature programmed reaction of NO2 showed that La2CuO4 was catalytically active for the decomposition of NO2. (Figure 3-8b) The NO2 began to decompose at 250 oC. The TPR for NO2 was compared to previous experiments usi ng a blank reactor. In the blank, which simulates gas phase reaction, NO2 began to decompose at about 600 oC, and the reaction rate increased greatly above that temperature. NO2 reacted over the Pt/YSZ sample in the presence of O2. (Figure 3-9a) The initial decomposition temperature was slightly higher than in the La2CuO4 case, beginning at 450 oC, and coming to steady-state equilibrium values by 500 oC. The surface area for reaction was much smaller in the Pt/YSZ case, explaining why the temperature must be higher for it to react, even though Pt is the more catalytic material. In th e gas phase, the kinetics of reaction were so unfavorable that NO2 did not begin to decompose until 600 oC, even in the absence of oxygen. (Figure 3-9b)
51 Calculations performed for the NO desorption were repeated for each peak in the NO2 desorption spectrum, and the results can be seen in Table 3-4. 3.3.4 TPD/TPR Results of COx /NOx Gas Mixture Exposure While pure NO bonds weakly to the La2CuO4 powder during adsorption, the addition of CO causes a radical shift in the bonding behavior of NO, reflected in the spectra. In the presence of CO, the NO bonds much more strongly, and large desorption peaks appear at 500 and 600 oC in addition to the small peaks at 150 and 200 oC seen in the NO TPD. (Figure 3-10a). CO was observed to desorb sligh tly, increasing quasi-linear ly from zero ppm at 50 oC to 8 ppm at 800 oC in the O2 deficient case. In the TPR of CO and NO, if oxygen is absent the two gases will reac t with one another to form N2O and CO2 (Figure 3-10b). The boost in NO adsorption stre ngth seen in the TPD is mirrored in the TPR experiment, visible in th e slight rebound in NO gas-phase concentration near 600 oC. This occurred once the majority of gas-phase CO was consumed and the system heated past the NO desorption threshold. This phenomenon revealed some information about the cross-sensitivity of La2CuO4 to NO and CO, and revealed part of the problem of eliminating that cross-sensitivity. Since CO reacted with NO, then this means the concentration the sensor sees in an anaerobic environment is not the true concen tration, regardless if the actual reaction is electrochemical. The further problem is that the cross-adsorption essentia lly poisoned the surface of La2CuO4 with NO. Since the desorption temperatur e (and thus, desorption energy) of NO rose dramatically, it will now respond much less quic kly to downward changes in NO concentration at the sensor operation temperature. The important consequence of the addition of O2 to an environment containing NO and CO is the similar increased amount of NO adsorption seen without oxygen, and the increase in the strength of that adsorption versus the adsorption without CO. This effect was clearly seen in the
52 TPD. CO2 and N2O emission cannot be separated due to their identical mass (44). However, concurrent peaks of NO and mass 44 near 300 oC, as well as overlapping peaks of CO and 44 at 400 oC strongly suggested conversion of both NO a nd CO on the surface of the powder. In the presence of oxygen, the la rge peaks at 500 and 600 oC shrink to ~10-15 ppm, and the low T peaks become slightly larger, gr owing to ~20 ppm (Figure 3-11a). In the presence of oxygen, CO will selectively react with O2 to produce CO2, leaving the NO to behave as if neither O2 nor CO were present (Fi gure 3-11b). The dip in oxygen concentration near 450 oC likely corresponds to the consumpti on of carbon-containing species on the surface before settling to a steady-state value above 550 oC. The combination of NO, CO, and O2 over Pt and YSZ showed ex tensive conversion of CO beginning near 200 oC (Figure 3-12) The reaction with CO is entirely selective to O2, and the speed of the reaction increased as higher temper atures made the reaction kinetics more favorable. The reaction rate did not s cale upward as quickly as La2CuO4, on account of the smaller available surface area. In the gas phase, CO reacted with O2 beginning near 300 oC, but without catalytic promotion, the overall exte nt of reaction was rather small. The simultaneous adsorption of NO2, CO and O2 affected an interesting change on the desorption pattern. (Figure 3-13a) The spectrum resembles NO2 desorption, except that peaks corresponding to NO desorption have been reduced in size. Additionally, the 44 m/z peak corresponding to N2O and CO2 increased, but remained in th e same temperature range of desorption as in the NO2 desorption experiment. CO clearly ad sorbed, but did not desorb without first oxidizing to CO2. Heating the La2CuO4 system containing NO2, O2, and CO flow caused simultaneous reduction of NO2 and oxidation of CO during the TP R (Figure 3-13b). Predictably, NO2 began to
53 react near 200 oC, and completely reacted above 300 oC, in the presence of strongly reducing CO. The oxygen dip seen previously for CO-c omplex consumption again appeared near 450 oC. Exposure of the Pt/YSZ system to CO, NO2 and O2 showed similar light-off temperatures to the La2CuO4 system for the NO2+CONO+CO2 reaction, but the extent of conversion was limited by the surface area of the electrode versus an intended platinum cat alyst (Figure 3-14). In the blank reactor, the re action begins near 350 oC, but does not approach a near-complete conversion until nearly 800 oC, compared to 500 oC over Pt. CO2, by contrast, inhibited NO adsorption, and it adsorbs to a significant extent in the presence of NO (Figure 3-15a). CO2 and NO do not react with each other upon heating. While CO2 on its own cannot be detected by the La2CuO4 sensing electrode, it appears that its presence may inhibit the adsorption of NO on the powder surface, which might explain the response reversal of the sensor in greater CO2 concentration. These occu rrences are all indicative of complex adsorptive behavior on the sensing electrode surface. 3.4 Summary TPD and TPR experiments on La2CuO4 show quantifiable adso rption and reaction of NOx and COx mixtures on the surface. Examination of the desorption peaks for Temperature Programmed Desorption allows for the calculation of desorption en ergy of gas species adsorbed on the powder, and gas phase conversion with respect to temperatur e could be tracked by examining concentration profiles during Temp erature Programmed Reaction. NO was found to be mostly unreactive to other ga s species while in the presence of oxygen, indicating that its response in the sensor is more strongly influenced by adsorpti on rather than mixed potential. NO2 decomposed catalytically at sensor operating temperatures, a nd was found to fragment into several products at low temperat ure, indicating mixed potential ma y play a significant role for the sensing behavior of NO2 on the potentiometric sensor.
54 Figure 3-1. Pipeline diagram of the Mass Spectrometer gas system
55 Table 3-1. TPD experiments Gas Mixture Concentration (ppm) Rate (oC/minute) NO 1% NO 30 NO+O2 500 ppm NO /0.95% O2 30, 15, 10, 5 NO2 500 ppm NO2 30, 15, 10, 5 NO+CO 0.67% NO/0.33% CO 30 NO+CO+O2 500 ppm NO/500 ppm O2/1% O2 30 NO2+CO+O2 500 ppm NO2 /250 ppm CO /1000 ppm O2 30 NO+CO2 500 ppm NO2 /1000 ppm CO2 30 Table 3-2. TPR experiments Gas Mixture Concentration (ppm) Sample NO 500 ppm NO La2CuO4 NO+O2 500 ppm NO / 1000 ppm O2 La2CuO4, Blank, Pt on YSZ 500 ppm NO / 21% O2 La2CuO4, Blank, Pt on YSZ NO2 500 ppm NO2 La2CuO4, Blank, Pt on YSZ NO+CO 250 ppm CO / 1000 ppm NO La2CuO4, Blank NO+CO+O2 500 ppm NO / 500 ppm CO /1000 ppm O2 La2CuO4, Blank, Pt on YSZ NO2+CO+O2500 ppm NO2 /500 ppm CO / 1000 ppm O2 La2CuO4, Blank, Pt on YSZ NO+CO2 500 ppm NO /1000 ppm CO2 La2CuO4
56 Figure 3-2. X-ray diffr action spectrum of La2CuO4 powder synthesized via the auto-ignition technique. Circles marked on the spectrum correspond to copper oxide impurities. Figure 3-3. SEM image of the La2CuO4 powder generated via auto-ignition.
57 0 5 10 15 0100200300400500600700Concentration (ppm)Temperature (oC)NO NO2 Figure 3-4. Temperature Programmed Desorption of 1% NO on La2CuO4. 0 50 100 150 0100200300400500600700800Concentration (ppm)Temperature (oC) O2NO Figure 3-5. Temperature Programmed Desorption of 500 ppm NO + 0.95% O2 on La2CuO4. Table 3-3. Desorption energies of pro ducts resulting from NO adsorption on La2CuO4. Peak Number Assignment Desorption Energy Equation 9 Average Desorption Energy Equation 10 Desorption Energy Wimmers Derivation 0 Covalent/Ionic NO 73.78 kJ/mol N/A N/A 1 NO Nitrites 110.23 kJ/mol 82.3 kJ/mol 86.3 kJ/mol 2 NO Nitrates 127.56 kJ/mol 172.9 kJ/mol 168.85 kJ/mol
58 0 100 200 300 400 500 0100200300400500600700800Concentration (ppm)Temperature (oC) NOa 400 500 600 700 800 900 1000 1100 0100200300400500600700800Concentration (ppm)Temperature (oC) O2NOb Figure 3-6. Temperature Program med Reaction of 500 ppm NO (a .)/ Temperature Programmed Reaction of 500 ppm NO and 1000 ppm O2 (b.) on La2CuO4. 0 200 400 600 800 1000 1200 0100200300400500600700800Concentration (ppm)Temperature (oC) O2O2 Calculated NO Calculated NO2 Calculated NO NO2 a 0 200 400 600 800 1000 1200 0100200300400500600700800Concentration (ppm)Temperature (oC) O2NO NO2 b Figure 3-7. Temperature Programme d Reaction of 1000 ppm NO / 1000 ppm O2 over (a.) Pt on YSZ and (b.) 500 ppm / 1000 ppm O2 in the gas phase.
59 0 10 20 30 40 50 60 0100200300400500600700800Concentration (ppm)Temperature (oC) NO2NO O2 N2O/ CO2a 0 100 200 300 400 500 0100200300400500600700800Concentration (ppm)Temperature (oC) NO2NO O2b Figure 3-8. Temperature Progr ammed Desorption of 500 ppm NO2 on La2CuO4. (a.)/ Temperature Programmed R eaction (b.) of 400 ppm NO2 on La2CuO4. 0 500 1000 1500 2000 2500 3000 0100200300400500600700800 O2O2 (Calculated) NO2NO2 (Calculated) NO (Calculated) NO Temperature (oC)Concentration (ppm) a 0 100 200 300 400 500 0100200300400500600700800NO2O2 NOConcentration (ppm)Temperature (oC)b Figure 3-9. Temperature Program med Reaction of (a) 1200 ppm NO2+ 1800 ppm O2 over Pt/YSZ plate (b) 500 ppm NO2 Temperature Programmed Reaction in the gas phase. Table 3-4. Desorption energies of products resulting from NO2 adsorption on La2CuO4. Peak Number Assignment Desorption Energy Equation 9 (Redhead A) Average Desorption Energy Equation 10 (Redhead B) Desorption Energy Wimmers Derivation 1 NO, NO2, N2O 115.2 kJ/mol 114.9 kJ/mol 118.8 kJ/mol 2 NO, O2 131.575 kJ/mol 148.46 kJ/mol 153.72 kJ/mol 3 N2O 152.79 kJ/mol 154.45 kJ/mol 158.5 kJ/mol
60 0 10 20 30 40 50 60 0100200300400500600700800Concentration (ppm)Temperature (oC) NO CO2/ N2O NO2CO O2a 0 200 400 600 800 1000 1200 0100200300400500600700800Concentration (ppm)Temperature (oC) NO CO CO2/ N2Ob Figure 3-10. Temperature Programmed Deso rption of 0.67% NO and 0.33% CO on La2CuO4. (a.)/ Temperature Programmed Reaction (b.) of 250 ppm NO and 1000 ppm CO on La2CuO4. 0 5 10 15 20 25 0100200300400500600700800Concentration (ppm)Temperature (oC) CO2/ N2O NO NO2O2COa 0 200 400 600 800 1000 1200 0100200300400500600700800Concentration (ppm)Temperature (oC) O2CO2CO NOb Figure 3-11. Temperature Programmed Desorption of 500 ppm NO, 500 pp m CO, and 1% O2 on La2CuO4. (a.)/ Temperature Programmed Reactio n (b.) of 500 ppm NO, 500 ppm CO, and 1000 ppm O2 on La2CuO4.
61 0 500 1000 1500 2000 0100200300400500600700800Concentration (ppm)Temperature (oC) O2 -Pt/YSZ NO Pt/YSZ CO Pt/YSZ CO2-Pt/YSZ O2 -Blank CO2-Blank CO-Blank NO-Blank Figure 3-12. Temperature Programmed Reaction of 1000 ppm NO, 1000 ppm CO, and 2000 ppm O2 on Pt/YSZ and 500 ppm NO, 500 ppm CO, and 1000 ppm O2 in blank reactor 0 10 20 30 40 50 60 70 100200300400500600700Concentration (ppm)Temperature (oC) NO2O2 NO CO2/ N2O CO a 0 200 400 600 800 1000 1200 0100200300400500600700800Concentration (ppm)Temperature (oC) O2NO CO2/ N2O CO NO2 b Figure 3-13. Temperature Programmed Desorption of 250 ppm CO, 500 ppm NO2 and 1000 ppm O2 on La2CuO4. (a.)/ Temperature Programmed Reaction (b.) of 500 ppm CO, 500 ppm NO2 and 1000 ppm O2 on La2CuO4.
62 0 200 400 600 800 1000 1200 0100200300400500600700800 NO Pt/YSZ CO Pt/YSZ CO-Blank NO-Blank NO2 Pt/YSZ NO2-Blank CO2-Blank CO2-Pt/YSZConcentration (ppm)Temperature (oC) Figure 3-14. Temperature Progr ammed Reaction of 1000 ppm NO2, 1000 ppm CO, and 2000 ppm O2 on Pt/YSZ and 500 ppm NO2, 500 ppm CO, and 1000 ppm O2 in blank reactor. 0 5 10 15 20 25 30 35 0100200300400500600700800Concentration (ppm)Temperature (oC) CO2NOa 400 500 600 700 800 900 1000 1100 02004006008001000Concentration (ppm)Temperature (oC) CO2NOb Figure 3-15. Temperature Programmed De sorption of 500 ppm NO and 1000 ppm CO2 on La2CuO4. (a.)/Temperature Programmed Reac tion (b.) of 500 ppm NO and 1000 ppm CO2 on La2CuO4.
63 CHAPTER 4 ADSORBATE STUDIES ON LA2CUO4 USING INFRARED SPECTROSCOPY AND XRAY PHOTOEMISSION SPECTROSCOPY 4.1 Introduction The conventional explanation for non-Nernstian behavior for potentiom etric sensors is the Mixed Potential theory. By this method, the voltage genera ted between the electrodes of the sensors arises purely as a result of the electr ochemical reactions taking place at the electrodes. The difference between the non-equ ilibrium kinetics at the electr odes will cause a net voltage to arise.189 228 231 However, this could account for the entire ty of the deviation from Nernst potential only when the electrode pair is a working electrode coupled to a true reference electrode232 and adsorption can be neglected. This mechan ism operates on the premise that multiple electrochemical conversion processes simultaneously occur on the surface of the electrodes, and that only the kinetics of these reactions determine the potential generated between the electrodes. As a result, the mixed potential theory may adequa tely describe the respon se behavior of sensors to a gas mixture at high te mperature (i.e. above 600 oC), where fast kinetics allow for equilibrium. However, at lower temperatures th e effect of these electrochemical reactions are only one of the factors affecting the surface. To address the shortcomings in Mixed Po tential theory, a new approach known as Differential Electrode Equilibria was suggested.63 This approach adds the contributions of adsorption/desorption, semiconducting behavior, a nd non-electrochemical catalytic reactions to the Mixed Potential contribution of electrochemical reactions, to explain the voltage response to a concentration change. The adsorption of gase s forms charged species at the surface of the electrode, leading to an accumula tion or depletion of electrons, which in turn can create an electric field across the elect ronically blocking electrolyte.233 234
64 La2CuO4 demonstrates promise as a material fo r a sensing electrode to explore this mechanism. The compound is catalytically unresponsive to the NON2 and NONO2 reactions in the presence of oxygen,102 and so the contributions from th ese catalytic or electrochemical reactions during NOx sensing are minimal. This allows th e contribution from gas adsorption to be the primary potential generator for a sensor under NO. Th e material functions as a p-type semiconductor, and so the adsorption behavior wi ll have an effect on th e conductivity as well as the potential. The adsorption behavior can be measured and compared to conductometric and potentiometric sensor behavior, which better allo ws for the separation of contributions to the electric potential generation. In this work, the techniques of diffuse re flectance infrared spec troscopy (DRIFT) and Xray photoelectron spectroscopy (XPS) were used to investigate the ad sorptive behaviors of La2CuO4 for NOx gases. 4.2 Experimental 4.2.1 Sample Preparation La2CuO4 powder was synthesized via an am orphous citrate autoignition route.235 To prepare solid electrode samples, La2CuO4 powder was used to form a screen-printable slurry. This slurry was prepared by mixing La2CuO4 powder with polyvinyl buty ral, alpha terpineol, and di-n-butyl phthalate dissolved in ethyl alcoho l (8:47:6 ratios) until the desired viscosity was reached. The suspension was then screen printed onto 8% mole Y2O3doped ZrO2 substrates (Marketech International Inc., YSZ-8Y, 20 x 10 x 0.1 mm) in an identi cal configuration to sensors previously tested. Samples were dried at 400 oC for 1 hour to remove volatile organics before firing. The screen-printed sample was sintered in air at 800 oC for 10 hours, with a heating and cooling rate of 5 oC per minute. The La2CuO4 electrodes had a thickness of ~15 m.
65 4.2.2 Infrared Characterization The adsorbed sample was analyzed using di ffuse reflective infrared (DRIFT) spectroscopy on a Nicolet MAGNA 760 IR. Spectra were obtain ed at room temperature under dry nitrogen. DRIFT was chosen to examine surface complexes af ter experiments with transmission infrared (IR) failed to generate sufficient signal for identification of surface complexes. Adsorbed samples were analyzed using identically mulled weight KBr/La2CuO4 mixture as a baseline in order to clearly observe changes on the surface as a result of adsorption. To use this subtraction technique, it was necessary to ensure the uniform ity of the mulled mixture between the sample of La2CuO4 used as a background and the adsorbed samp le, ruling out the possibility of mixing the pure powder with KBr following the adsorption pro cedure. This technique has some precedent in separating substrate signal from the signal due to material at interfaces236 238 and was valid for these experiments due to the homogeneity of the La2CuO4/KBr mixture for both samples. For infrared spectroscopy measurements a 5% / 95% weight mixture of La2CuO4 / KBr powders was prepared using a Wig-LBug muller and dried in air at 120 oC for one hour. The mulled mixture was loaded into the reactor used for mass spectrometry measurements.239 The mixture was heated under 20 cubic centimet ers per minute (ccm) flowing He at 2 oC / min to 225 oC. At that temperature the sample was exposed to 30 ccm of 1% NO or NO2 / balance He and cooled to room temperature at 2 oC/min. Once cooled to room temperature, the sample chamber remained under helium flow until transferred to an IR spectrometer. In the absence of water and acidic conditions, KBr remains stable under the conditions and temperatures listed above.240 Use of dry gases assured that ther e would be no interac tion between the KBr and gas phase during adsorption, and exposure temperatures remained well below the 730 oC melting point.
66 4.2.3 XPS Characterization X-ray photoelectron spectroscopy (X PS) was performed on both the La2CuO4 electrode and on La2CuO4 powder. The solid sample was examined at the gas/electrode interface before and after exposure to a testing cycle in order to see the effect of aging on the electrode surface. Powder samples were tested in the unadsorbed condition, as well as af ter adsorptions of NO, NO2, CO, NO+CO, and NO+NO2. Sensor treatment was conducted in a typical gas-flow reactor. Gas environments were controlled using MKS mass flow controllers. The sensors were exposed to step changes of the gas in 3% O2 with N2 balance. The total flow rate was se t at a constant 300 ccm. NO steps were set at 0, 50, 100, 200, 400 and 650 ppm while NO2 steps were set at 0, 50, 100, 200, 300 and 333 ppm. Each step was held for 200 seconds a nd was delivered in an upward and downward hysteresis loop. After the aging treatment was completed, th e sample was re-surveyed to examine the effects of gas exposure. Initial XPS measurements were obtained for the solid sample directly after firing. A XPS/ESCA Perkin-Elmer PHI 5100 ESCA System acquired spectra using an Mg anode at a 45 degree incidence angle. The sample was surveyed from 1-1000 eV in 0.5 eV steps over an area covered by the La2CuO4 electrode. Following the XPS measurement of the freshly sintered sample, it was placed in a sensor te sting chamber and cycled over multiple testing conditions for a period of 3 weeks to examine th e effects of thermal hysteresis and gas exposure on the composition of the electrode. The results of these studies were covered in more detail in Appendix B. Powder samples of La2CuO4 were heated to 550 oC and held at temperature under flowing helium and oxygen for 1 hour to clear the surface of impurities. The adsorption sample chamber was identical to those used previously for mass spectrometry measurements.239 The samples
67 cooled under various gas atmospheres to 50 oC. Adsorption samples of the following types were obtained: Helium, 1% NO, 1% NO2, 0.5% CO+5% O2, 0.5% NO/0.5% CO, and 0.5% NO/0.5% NO2. All samples were heated and cooled under flow of 10 cubic centimeters gas mixture per minute, at a rate of 4 oC per minute. The adsorbed powder samples were evaluated using a Kratos Analytical Surface Analyzer XSAM 800. Freshly adsorbed samples were dus ted on double sided carbon tape and mounted on a revolving sample holder. The testi ng chamber was then evacuated to 10-7 Torr and held overnight. Each sample was scanned from 1 eV to 1000 eV in 0.5 eV steps to establish a survey spectrum. After obtaining a full range spectrum, the samples were scanned to measure the spectra of the C 1s, La 3p5, and O 1s electron orbitals. These measurements were obtained over the ranges of 270-295 eV, 830-870 eV, and 500550 eV binding energy, respectively, in 0.1 eV steps. NO and NO2 adsorptions are examined here. Other experimental results ar e described briefly in Appendix C. 4.3 Results and Discussion 4.3.1 IR Results for NOx Adsorption IR spectra of the species adsorbed on th e powders were obtained by subtracting the spectrum of the unadsorbed powder from the spectra of the adsorbed samples. By removing all the characteristic IR traces from the La2CuO4 powder as well as any unintentional carbonate traces from atmosphere exposure, the NOx adsorbates can be more cl early seen on the surface of the powder.241 Subtraction of the unadsorbed spectrum was necessary in orde r to accentuate the adsorption peaks, as La2CuO4 adsorbs the near-infrared spectrum quite strongly.242 Peaks were fit using Fourier deconvolution of a Kubelka-Mu nk spectrum transformation in ThemoNicolet OMNIC 6.1a. On the NO adsorbed sample (Figure 4-1a, 4-2a), relevant peaks can be seen at 2400, 1766, 1510, 1468, 1435, 1385, 1355, 1338, 1310, 1046, and 833 cm-1. In the NO2 sample
68 (Figure 4-1b, 4-2b), the peak s are located at 2396, 1763, 1500, 1465, 1436, 1414, 1390, 1351, 1325, 1305, 1039, and 825 cm-1. Peaks seen near 3000-3500 cm-1 correspond to vibration of the simple ionic complexes, and were superfluous for assignment of species. However, in the NO2 adsorption case, the peak near 3400 is much larg er than in the NO adso rption case, indicating a greater amount of ionic NO formation on the surface due to gas adsorption. Numerations of the peak assignments are ta bulated in Table 1. The NO sample showed greatest peaks corresponding to ch elated nitrite and nitrito form ation, but minor peaks supporting the presence of room-temperature unidentate nitrate were also observed. Species present on the surface were identified by isolating specifi c vibrational signatures and comparing these signatures to patterns f ound in the literature. Unidentate nitrate can be seen in the asymmetrical NO2 stretch at 1510 cm-1, symmetrical NO2 stretching vibration at 1310 cm-1, and 1046 cm-1 NO vibrations.243 246 Nitrito complexes were observed from stretching vibrations of (N=O) at 1466 cm-1, NO vibrations at 1046 cm-1, and 833 cm-1 ONO bending vibrations.244 245 247 248 Chelated nitrite groups were evident from asymmetric NO2 vibrations at 1355 cm-1, symmetric NO2 vibrations near 1338 cm-1, and the 833 cm-1 ONO bending vibrations.248 254 Additionally, ionic NO+ was revealed by the NO vibrations at 2400 cm-1, and covalent bonding of the NO to meta l sites was shown by the peak at 1766 cm-1 wavenumbers.255 Free ionic NO3 is seen at 1435 cm-1, assigned for the general case.256 This NO3 variant represents the transition stat e between structur es of nitrates. Numerations of the peak assignments for NO2 are also tabulated in Table 1. In the NO2 adsorbed case, nitrito complexes were observed from stretching vibrations of (N=O) at 1463 cm1, NO vibrations at 1039 cm-1, and ONO bending vibrations at 825 cm-1. Chelated nitrite groups were evident from asymmetric NO2 vibrations at 1351 cm-1, symmetric NO2 vibrations near 1325
69 cm-1, and the ONO bending vibrations at 825 cm-1. Nitro groups can be seen by asymmetric NO2 vibration at 1436 cm-1, symmetric NO2 vibrations at 1351 cm-1, ONO bending at 825 cm-1, and evidence of conjugation w ith the surface at 1394 cm-1 was observed.256 Ionic NO+ was revealed by the NO vibrations at 2396 cm-1, and covalent bonding of the NO to metal sites was shown by the peak at 1763 cm-1. Unidentate nitrate can be seen in the asymmetrical NO2 stretch at 1500 cm-1, symmetrical NO2 stretching vibration at 1305 cm-1, and NO vibrations at 1039 cm-1. In discussing the composition of surface adsorbat es, the limitations of IR must be stressed. The presence of nitrate (NO3 -) and nitrite (NO2 -) groups was well documented in the literature for metal oxides.225 244 250 254 257 The composition of these adsorbate groups is somewhat malleable, depending on temperature. Isolating a specific nitrate or nitr ite compound structure in the IR spectrum is mostly sy mbolic, as the different structures of the surface groups only represent the transien t properties of the oxygen connection to the lattice involved in the formation of that group.225 The individual structures within the nitrate and nitrite designations are not functionally different, and so in speaking of one st ructure in the group, similar characteristics can be assigned to other structures for that same compound. Further, the nitrate, nitrite, and bonded NO groups observed via IR lik ely possess the ability to shift from one configuration to another as pa rt of a larger mechanism, suggested by the presence of the indefinable free NO3 on the surface after adsorption. Previous results from TPD a nd TPR experiments show manife stly different desorption and reaction products resulting from the NO vs. NO2 adsorptions, with total adsorbed NOx a factor of two larger in the NO2 case.66 The IR signal in the spectrum following NO2 adsorption was also observed to be twice as large as in the NO case, reinforcing the previ ous results for TPD/TPR studies. The IR results for NO adsorption versus NO2 adsorption show very similar patterns, with
70 the exception of nitro groups visibly present on the NO2 adsorbed sample spectrum. The formation of very similar complexes on the surfa ces is important because it shows that while the mechanisms of their formation are dis tinct, the end products are related. 4.2.2 XPS Results for NOx Adsorption The solid La2CuO4 electrode on YSZ substrate was ev aluated by XPS immediately after firing. XPS measurements were repeated after thermal cycling and gas cycling in order to determine the extent of change on the surface of the electrode. This comparison can be seen in Figure 4-3. The peaks from the XPS measurement in the as received sample are consistent with those in the literature for La2CuO4. 258 261 The La2CuO4 was well visualized by the La 3d5 peak at 835 eV, O 1s peak at ~530 eV, and a weak Cu 2p3 line at 934 eV. Carbon impurities were observed for the C1s orbital at 282 eV for carbidic impurities262 and 284.5 eV for graphitic carbon.263 Graphitic carbon was present in th e XPS spectra due to its use as the sample substrate and its peaks were utilized as anchors to identify peak shifting in the La 3d5 and O1s regions264 265 through each NOx adsorption spectrum. Three weeks of thermal cycling changed the surface composition only slightly. Peaks corresponding to the La 3d5 orbital increased sli ghtly in intensity, indi cating relaxation of the La2O2 layer to the surface of the electrode grains. Thermal cyc ling causes the lanthanum layer to establish itself more strongly on the surface,266 267 though the actual surface structure likely does not establish itself fully into this ideal st ructure. Correspondingly, peaks from the CuO2 layer lost their intensity, and the signal from the Cu 2p3 or bital was reduced to a level just above noise. The Cu 2p3 peak is not especially large even wh en the sample is fresh, so its suppression was reasonable if the lanthanum la yer became more prominent.
71 The XPS spectra of the electrode revealed that short term e xposure to thermal cycling in sensor testing conditions did not fundamentally change the character of the electrode surface, reflected in the superposition of the spectra take n before and after aging. Local manipulation of oxygen partial pressure by the use of reducing gases causes no irreversible e ffects to the structure or composition of the electrode. Furthermore, obtaining XPS spectra in room temperature vacuum did not reduce La2CuO4 to its component oxides even at oxygen partial pressures far below the high temperature decomposition thre shold in mechanically induced vacuum.134 From these discoveries, it can be presumed that ad sorption of gases on the surface will not cause permanent compositional changes that preven t examination of adsorbates via XPS. The powder samples were examined after adsorp tion of gas mixtures on oxidized samples. The powder displayed a survey baseli ne spectrum essentially identical to that of the solid sample after thermal cycling. The spectrum reflected an O1s peak at 530 eV, La3p5 peak at 837 eV, and C1s peak at 281 eV, but showed no signs of Cu p eaks throughout the survey This is possibly in part due to the additional pro cessing required to prepare the sa mples, where increased roughness of the powder may have obscured the copper signal,268 but the reconstruction of the surface to favor lanthanum exposure may also have played a role. NO adsorption induces splitting of the composite O 1s peak at 530 eV into a its component peaks representing oxygen unaffected by the adsorp tion of NO to the surface, and increases in the region near 534 eV (Figure 4-4). This shif t was caused by the bi nding of surface lattice oxygen in the oxide by NO bonding.269 270 The relative intensity of the two peaks indicates the extent to which NO affects the oxygen in the obse rvable volume. The differe nt chemical state of oxygen bonding is seen in the binding energy shift, which for the formation of the complexes on the surface, will result in a lower Fermi energy, and thus a higher binding energy of the core
72 oxygen electrons. The formation of complexes on the surface tied up the oxygen and accounted for the strengthening of the binding energy in the split peak. Since the powder sample is overall electrically isolated during NO adsorption, the amount of NO ad sorbed cannot fully cover the surface of the powder, and so the XPS results show a combination of the La2CuO4 surface oxygen as well as the oxygen pulled into the surface complexes. A stronger peak splitting was observed for the O1s orbital with the adsorption of NO2 than in the NO adsorption case. (Figure 4-5) The large peak at 530 was observed to split into a peak at 527 and a larger peak at 534, with the total area underneat h the peaks greate r than in the unadsorbed or NO adsorbed case. This additiona l area can be explained by the stronger bonding behavior of NO2 in relation to NO. The greater separati on in this case corresponds to the greater perturbation of the observable volume by the adsorption of NO2. The large shift of area to the 534 eV peak showed the oxygen chemistry of the observable volume shifts to the production of complexes on the surface. However, unlike the NO adsorption case, the oxygen rearrangement was compensated by the formation of charge carr ying species (electron holes) evidenced by the peak shift to 527 eV. We reported that La2CuO4 powder TPR experiments for NO and NO+O2 showed no compositional changes due to homogeneous gas-phase or heterogeneous gassolid reactions over the 50-800 oC range of the experiment.66 239 271 TPD results also show no dissociation of NO to N2 and O2, or oxidation from NO to NO2, only a very small conversion of NO to N2O, due to direct bonding to Cu+ sites during adsorption.225 Since desorption of NO almost exclusively produces species chemically identical to th e adsorbed NO, the catalytic production of electrochemical products for generation of volta ges was unlikely. Instead, it seems that the formation of the surface complexe s facilitated the formation of a charged layer at the surface.
73 Thus, instead of full and irreversible electr ochemical production, a reversible but stable equilibrium occurs at the interface. Since NO l acks the oxygen to form the nitrate and nitrito surface complexes on its own, it draws oxygen fro m the surface to complete the compound. The proposed complexing mechanism for NO ad sorption can be seen in Figure 4-6. Upon encountering the La2CuO4 surface, the molecule had the possibil ity of attaching to either a metal or oxygen site. For the sake of simplicity, the NO molecule is assumed to orient in the N-up or N-down configuration for bonding wi th the surface. In the case where NO bonds to a metal site, bonding N-down creates covalent or ionic NO which can interact with other NO complexes, but are otherwise stable. In the N-up c onfiguration, the nitros yl complex can form as an intermediate before re-orienting to a chelated nitrite group. Th is group is functionally e quivalent to the nitrito clearly seen in the IR spectra, and with sufficient thermal perturbation, can likely relax to the nitrito configuration. In the case of NO bonding to an oxygen site, N-down bonding produces a nitrito group after the lattice oxygen retains one bond to a metal site, then relaxes away from the surface. The N-up bonding configuration would force an oxygen-oxygen bond, which does not form any complexes seen in the IR spectra. Unidentate nitrate forms from NO complexi ng to two oxygen sites. The mechanism for nitrate formation is not entirely clear; how ever, the bridged nitrate complex forms on La2O3, which realigns to form the unidentate nitrate.257Although the bridged nitrat e cannot be reliably observed at room temper ature on the surface of La2CuO4 after NO adsorption, its presence as an intermediate can be dependably postulated, si nce the nitrito group would be a precursor to bridged nitrate formation.
74 As the complexes tie up a greater amount of oxygen, the surface layers of La2CuO4 became oxygen deficient: '' ()2()x OgadOONONOVe [4-1] '''' 2()3()23x OadOadOONOVeNOVe [4-2] Due to the net addition of electrons to the bulk, there is a shift in the Fermi level at the surface, raising the potential to pus h electrons to the counter-electr ode. This Fermi change for the full sensor cannot be measured directly via XPS, due to the need for vacuum, but a Kelvin probe capable of atmospheric pressure operation co uld quantify the shift by measuring the work function while the sensor is operating.177 182 By monitoring the ionisorb ed layer, the extent of charge conversion could be determined over an area.173 At the time of writing, the ability to make this measurement was unavailable. The oxygen deficiency lowers the availabil ity of electron holes, due to the p-type semiconductive nature of La2CuO4, leading to a drop in electronic conductivity, consistant with observed resistive changes.125 The electron binding was seen in the XPS spectra of the O1s orbital following NO adsorption. When compared directly to the fully oxidized La2CuO4 sample, a significant shift upwards in the binding energy is clear. This increased energy results from the integration of the oxygen electron into th e unpaired orbital of the nitrogen in NO.272 This shift in potential due to surface complexing is reversib le, as the potential across a full sensor will disappear with removal of NO. Upon heating, there is no evidence of low temperature oxygen emission, only NO desorption. The lack of oxygen emission meant the complexes that formed by oxygen removal from the surface replace the oxygen before disass ociating. As a result, the surface retains its original state after low temperature NO co mplex breakdown and desorption. One possible
75 explanation for this behavior is the rearrangement of the comp lexes on the surface to favorable compositions for O2 retention and NO desorption (i.e. n itrito). In these configurations, oxygen bonds to the surface are sufficiently strong to a llow separation from a nitrogen atom without removal from the surface. Reversibility is unique to the NO adsorption due to its weak association with surface oxygen, seen in temperature programmed desorption experiments in earlier chapters. It does not retain surface oxyge n when it desorbs, as opposed to CO, which retains the composition of CO2 while desorbing. The adsorption behavior of NO2, by contrast, was much more complicated. The adsorption of NO2 caused a stronger effect of peak splitting than in the NO adsorption case. The large peak at 530 eV was observed to split in to a small peak at 527 eV and a relatively larger peak at 534 eV, with a greater total area than in either th e unadsorbed or NO adsorbed case. This can be explained by the strong er bonding behavior of NO2 in relation to NO. Desorption products of NO2 from the surface were much more complex th an those of NO desorption, and the quantities evolved from the surface were much greater.66 The possible mechanism of NO2 adsorption can be explaine d similarly to NO adsorption (Figure 4-7). Upon encountering the La2CuO4 surface, the NO2 molecule had the possibility of attaching to either a metal or oxygen site. Again, the NO2 molecule is assumed to orient in the Nup or N-down configuration for bonding w ith the surface. In the case where NO2 bonds whole to a metal site, bonding N-down forms nitro groups. These groups can ei ther remain in the nitro configuration, which desorbs as NO2, or realign to form ionic or covalent NO, and releases the second oxygen atom to the oxide surface. In th e N-up configuration, oxygen bonds to metal and directly forms the nitrito group, which can reconstruc t into a chelated nitrite or interact with the surface oxygen to form a bridged nitrate in termediate. For oxygen sites, the N-up bonding
76 configuration would force an oxygen-oxygen bond, which does not form any complexes seen in the IR spectra. N-down bonding to the oxygen dir ectly forms the unidentate nitrate once the oxygen relaxes away from the surface. At that point, the nitrate groups can shift from one construction to another. NO2, in contrast to NO, does not draw oxyge n from the surface to complete all the complexesfor nitrite formation, it enriches the local lattice oxygen concentration by removing an electron to form ion complexes without removing oxygen, causing valence changes in the subsurface copper atoms, while increasing the conductivity due to electron hole formation. 2()2() x x M gadM M NONOhM [4-3] This behavior is analogous to electron hole creation caused by dopi ng on the A-site of La2CuO4, such as Sr.273 275 Electron hole formation was s hown by the lessening of the O1s electron binding energy peak from 530 eV to 527 eV. NO2 has more available oxygen than NO, so when forming an NO3 negative complex, it does not need to bond to two oxygen sites, requir ing only one. This oxygen can be obtained in a number of ways. The adsorbed nitro complex ma y dissociate to an NO molecule and adsorbed oxygen, which provides for the formation of th e nitrate complex wit hout oxygen removal from the lattice by a second adsorption of NO2. If this occurs, the nitrate forms in a similar manner to the nitrite group. 2()()3() x x M gadadM M NOONOhM [4-4] If adsorbed oxygen is unavailable, the nitrate can be formed by NO2 bonding to an oxygen site, where the complex can relax into one of several configurations. '' 2()3()x OgadOONONOVe [4-5]
77 This complex desorbed identically to th e analogue formed by NO. The oxygen bonded to the metal site at the surface remains attach ed, and the NO desorbs, whereupon the oxygen proceeds to bond with another metal-bonded oxygen or surface oxygen, and desorb as an oxygen molecule to maintain charge balance. This process occurs both wh en the third oxygen was removed from the surface and when the third oxy gen results from nitro group decomposition. It accounts for the dissociation of NO2 into NO and O2 seen in the mass spectrometry desorption profiles and the raised binding energy seen in the XPS results. 4.4 Conclusions It can be noted that while the intermediate s on the surface due to adsorption are for all intents similar, the method in which they form greatly affects the charge the surface layer generates, and leads to the diffe rences in potential across the electrolyte. By examining the working electrode material i ndependently of the sensor, th e adsorptive component can be examined without compensation for possible ch arging effects on the se nsor electrodes. NO adsorption generates charge deple tion layers by the formation of NO2 and NO3 at the gas/solid interface, where electrons are ta ken from the bulk and held. These depleted layers form a potential gradient that (in a fu ll sensor) draws electrons from the counter-electrode, for a positive voltage. Th is phenomenon was observed in the absence of catalytic conversion of NO to N2 or NO2, thus advancing the theory that NO sensing via the LCO electrode is entirely adsorptive. NO2 forms similar complexes, but due to the extra oxygen present for some of these complexes, electron holes are generated, which produced a negative potential gradient across the se nsor couple in a full device.
78 Table 4-1. IR Peak a ssignments of the NO/NO2 adsorbed samples of La2CuO4 NOx species Structure Vibration Inorganic compound or metal oxide225 243 257 NO Wavenumber (cm-1) NO2 Wavenumber (cm-1) Ionic NO NO+ / NO(NO) 2100-2400 2400 2396 Ionic NO3 Transitional (NO2 ,as) 1350-1450 1435 1414 Covalent NO M-NO 1600-1800 1766 1763 Chelated Nitrite M O N O (NO2 ,as) (NO2 ,s) ONO 1270-1390 1360-1260 840-860 1355 1338 833 1351 1325 825 Unidentate Nitrate M O N O O (NO2 ,as) (NO2 ,s) (NO) 1450-1570 1250-1330 970-1035 1510 1310 1046 1500 1305 1039 Nitrito MON=O (N=O) (NO) ONO 1400-1485 1050-1100 820-840 1466 1046 833 1463 1039 825 Nitro O O M N (NO2 ,as) (NO2 ,s) ONO (NO2 ,as) Conjugated Nitro 1370-1470 1320-1340 820-850 1350-1400 1436 1351 825 1390
79 Figure 4-1a. IR spectra of La2CuO4/KBr mulled mixture after NO adsorption. Figure 4-1b. IR spectra of La2CuO4/KBr mulled mixture after NO2 adsorption
80 Figure 4-2a. Fourier Deconvolution of 1550-1200 cm-1 spectrum of NO-adsorbed La2CuO4 (Labeled line corresponds to the Fourier deconvolution) Figure 4-2b. Fourier Deconvolution of 1550-1200 cm-1 spectrum of NO2 adsorbed La2CuO4 (Labeled line corresponds to the Fourier deconvolution)
81 Figure 4-3. Full range XPS spectra of solid La2CuO4 samples in the (a) freshly sintered case and (b) after 3 weeks of exposure to sensor testing conditions.
82 Figure 4-4. O1s XPS spectra of NO adsorbed La2CuO4 versus the unadsorbed case. Figure 4-5 O1s X PS spectra of NO2 adsorbed La2CuO4 versus the unadsorbed case.
83 Adsorption Step N O MOMOMOMOM Bonds Oxygen to Metal Metal Site M Oxygen Site O N O MOMOMOMOMOM Bonds Oxygen to Oxygen O N MOMOMOMOM Bonds Nitrogen to Metal O N MOMOMOMOM Covalently Bonded NO O+ N MOMOMOMOM Ionic NO+ N OMOMOMOMOM Nitrosyl Intermediate M O MON=O V M Nitrito Group O N MOMOMOMOMOM Bonds Nitrogen to Oxygen M O O M N V O M Chelated Nitrite M V O M--O --N V O M Unidentate Nitrate O M O V N--O M O V Bridged Nitrate M V O M N O V O M Bidentate Nitrate PossiblePossible Seen via IR Seen via IR IR vibrations do not occur in peroxide regime Seen via IR Seen via IR Seen via IRProposed RearrangementProposed Rearrangement Proposed Rearrangement Proposed Rearrangement Proposed Rearrangement Observed Rearrangement Seen via IR Figure 4-6. Flowchart of proposed NO adsorption mechanism
84 2 Adsorption Step Metal Site M Oxygen Site O O N O MOMOMOMOMOM Bonds Oxygen to Oxygen O N MOMOMOMOM Covalently Bonded NO O+ N MOMOMOMOM Ionic NO+ M O MON=O O M Nitrito Group O O N MOMOMOMOM Bonds Nitrogen to Metal O O N MOMOMOMOMOM Bonds Nitrogen to Oxygen M O O M N O O M Nitro Group O N O MOMOMOMOM Bonds Oxygen to Metal M O O M N O O M Chelated Nitrite O M O V N--O M O O Bridged Nitrate M V O M--O --N O O M Unidentate Nitrate M V O M N O O O M Bidentate Nitrate PossiblePossible Observed via IR and TPD Observed via IR Observed via IR Observed via IR Observed via IR Observed via IR Only possible method of conversion Proposed RearrangementProposed Rearrangement Proposed Rearrangement Proposed Rearrangement Only possible method of conversion Figure 4-7. Flowchart of proposed NO2 adsorption mechanism.
85 CHAPTER 5 ISOTOPICALLY LABELED OXYGEN STUD IES OF THE EXCHANGE BEHAVIOR OF NOX OVER LA2CUO4 5.1 Introduction Isotope exchange studie s utilizing heavy oxygen (18O2) have proven useful in the study of catalysis, and by use on sensor materials, perm it further exploration of the adsorption and reaction mechanism of NOx involved in solid-state gas sensor re sponse. Previous research in this topic focused on diffusion, using se condary ion mass spectrometry (SIMS)276 279 and gas-phase mass spectrometry to track exchange with surface oxygen.219 280 281 Gas phase mass spectrometry (MS) using heavy oxygen has also been used to examine reactions at the gas/solid interface through temperature programmed experiments.218 282 285 Standard temperature programmed reaction (TPR) and temperature programmed de sorption (TPD) experiments provide important information about the composition of reaction/de sorption products, but can only inform of the presence of compounds, not necessarily their method of formation. We demonstrated in past chapters, using infrared spectroscopy and X-ray photoelectron spectroscopy experiments, that a substantial amount of nitrate a nd nitrite complexes form on the surface of La2CuO4 powder as a result of NOx adsorption.286 Through MS analysis of the TPR/TPD effluent we determined that NO remained unconverted to either NO2 or N2, but NO2 will decompose to NO over La2CuO4. 66 102 NO adsorption resulted in a mix of adsorbate complexes that utilize adsorbed gas-phase or lattice oxygen. These complexes decomposed on the surface to re-form the NO molecule, and de sorbed NO while desorbing the sequestered oxygen in both the purely adsorptive and in the reactive case. While these experiments confirmed the uncatalytic nature of La2CuO4 to the NONO2 and NON2 reactions, the mechanism by which the NO reformed as it desorbed was still unknown.
86 Similarly, NO2 formed nitrate, nitrite, and nitr o complexes on the powder surface that decomposed to NO, NO2, and O2. The products of desorption and reaction were easily characterized, but there again remained only a suggestion of the process by which these gases formed during separation from the powder surface. By using heavy 18O in place of the naturally occurring 16O, the origin of molecules desorbing from the surface could be determined. Mass spectra of the effluent gas from the reactor can separate N or C containing molecules desorbed with their original atoms from those produced from scrambling 16O/18O compounds on the surface.110 By utilizing the reaction mechanis m constructed in Chapter 4 for NOx adsorption, a sequence of all possible pathways from ad sorption to desorption was obtained. Using 18O2 as the gas-phase oxygen, the possible reaction pathways can be probed experimentally. By examining the results of TPD, TPR, and isothermal reaction, the most likely methods of complex formation can be obtained. If the compounds desorbing from the surface show a large amount of 18Ocontaining products, the gas-phase adsorbates of oxygen will play a much larger role than previously considered. If the ma ss of the products is the same as the naturally occurring type, then the oxygen in NOx/COx compounds results from a comple te reversal to the same atoms during desorption or the oxygen in th e compounds is a result of exch ange with the lattice. This mechanism allowed the separation of the relativ e contribution of electro chemical (i.e. Mixed Potential) reactions to the sensor response to NOx gas. To determine the mechanism, it proved necessary to anneal the sample in order to exchange 18O into the lattice, and then repeat experi ments in order to see whether the powder exchanges lattice oxygen by measuring for desorp tion products of increased mass. The possible sequence for NO adsorption can be seen in Figure 5-1.
87 The simplest possible path for NO is to bond in the nitrogen-down configuration to a metal site, and desorb identically by simply break ing the metal-nitrogen bond. These bonds can be expressed as either ionic or covalent in char acter, and the adsorption/desorption reactions are shown in Equations 5-1 and 5-2. 16 ()()gNONOe [5-1] 16 ()()gNONO [5-2] A possible branch of this adsorption is for the upward extended oxygen to exchange with oxygen on the surface, shown in Equation 5-3 and 5-4. 181816 11 22 2()2()()() s urfacesurfaceNOONOO [5-3] 181816 11 22 2()2()()() s urfacesurfaceNOONOO [5-4] The covalent and ionic NO complexes are th e least strongly bonded to the surface of La2CuO4, so if this exchange occurs during the ad sorption process, low temperature desorption peaks corresponding to N18O will be visible during heating. However, the nature of the ionic bond generated by NO occurs due to its re sonance in the N-O pseudo-triple bond.287 As a result, oxygen exchange cannot occur without destroying the bond to the surface, and the mechanism proceeds according to Equations 5-1 and 5-2. Adsorption of NO as a nitrite (NO2 -) or nitrate (NO3 -) species requires extra oxygen atoms, and these can be drawn from the adsorbed surface or the bulk. The key step was to determine the source of the exchange, if a ny occurred. If NO exchanged it s oxygen, the MS profile would change to reflect the scramble d oxygen product at m/z 34, as well as an increase in m/z 32 that would almost exclusively correspond to N18O. These mechanisms are represented by Equation 55 for nitrite species, and Equati ons 5-6 to 5-10 for nitrates.
88 Nitrite species form by bonding in the nitr ogen-down configuration to a lattice oxygen, pulling an electron from the bulk to comple te the complex, seen in Equation 5-5. 181616 1 2 2()latticeNOOeNO [5-5] Following nitrite formation, the complex can desorb by reversing the adsorption mechanism, or can further alter its structure. By combining with surface oxygen, the nitrite can become a nitrate, as seen in Equation 5-6 and 5-7. 161616 1 2 22()3()()surfaceNOONO [5-6] 16181618 1 2 22()2()()surfaceNOONOO [5-7] This nitrate can remain on the surface, or re verse the formation to revert to the nitrite configuration. By retaining the 18O atom added in Equation 5-7, the nitrite can effectively exchange the extended oxygen, and this change would be reflected upon desorption. These steps can be seen in Equation 5-8 and 5-9. 1618161816 1 2 22()()() s urfaceNOONOOO [5-8] 18181816()latticeNOONOOe [5-9] The exchanged nitrite can again complex with an adsorbed 18O to reform the nitrate, seen in Equation 5-10. 1618181816 ()2()()surfaceNOOONOO [5-10] Once the nitrates have formed, they deco mpose upon heating to form the highest temperature desorption peak via Equations 5-11 to 5-14 below. 16161616 3()()gassurfaceNONOOOe [5-11]
89 1618161816 2()()gassurfaceNOONOOOe [5-12] 1618181616 2()()gassurfaceNOONOOOe [5-13] 1816181816 2()()gassurfaceNOONOOOe [5-14] From previous studies, it has been demonstrat ed that the reaction a nd desorption profiles are identical in the case of NO exposur e. As a result, the TPD and TPR using 18O should look quite similar. However, while the chemical composition of the effluent gas is the same as the adsorbate, the question remains if the desorbing species are identic al to those that adsorb. Since the desorption spectra was fairly simple, th e steps in the adsorption mechanism can be determined readily. As observed in Chapter 3, the NO desorp tion spectra has three distinct peaks, corresponding roughly to the decomposition and deso rption of ionic/covalent NO, nitrite, and nitrate species. The weakly bonded ionic/covalent NO species desorb with the lowest amount of activation energy, and can be identified at ~150 oC during 30 oC per minute heating. Thus, if exchange occurs between the adsorbed NO comple xes, there should be an increase in 32 m/z intensity during the TPD or TPR in that temperature range. Above the desorption threshold for the ioni c and covalent species, there are peaks corresponding to the decomposition of other NOx complexes. Nitrate and nitrite species are more stable on the surface than the ionic and covalent complex, and so they desorb at a higher temperature. Thus, changes due to 18O acquisition can be resolved by monitoring the level of m/z 32 relative to m/z 30. If the only method of acquiring 18O is by integrating adsorbed surface oxygen, then the higher 32 peak corresponding to nitrate decomposition should be a larger fraction of the total NO desorption at that temperat ure range relative to th e fraction of 32/ 30 due
90 to nitrite decomposition. If the 18O can substitute for the 16O already in the nitrite and nitrate complexes by oxygen exchange with the surface, th e signal for m/z 32 should be much larger than the signal for m/z 30. If the co mplexes still desorb identically as N16O, then one of two paths occurs. Either the adsorption favors the N-16O bond to the point that it always remains (which rules out 18O/16O exchange), or the complex forms by bonding exclusively to lattice oxygen. This can be further resolved by repeatin g the experiments with a LCO powder that was annealed in 18O to promote lattice oxygen exch ange, then adsorbed under NO and 16O2. If N18O desorption can be stimulated by this method, then the lattice oxyge n plays a major role. NO2 has a much more complicated desorption profile due to the extra oxygen present during adsorption, and as such, the possible mech anisms for adsorption and exchange are more extensive, as can be seen in Figure 5.2. Ionic and covalent complexes we re observed to form on the surface, which necessitates a dissociative adsorption of NO2 in order to form the (NO) complexes. The formation of these groups is described by Equations 5-15 and 5-16. 161616 1 2 22()()surfaceNONOOe [5-15] 161616 1 2 22()() s urfaceNONOO [5-16] Similarly to NO adsorption, the ionic and covalent groups are likely incapable of exchanging their extended oxygen with one from the surface. These st eps are identical to Equation 5-3 and 5-4. NO2 can also adsorb non-dissociatively on a metal site in order to form the nitro group, shown in Equation 5-17. 1616 22() NONO [5-17] The possible methods of exchange for nitro oxygen with surface oxygen are similar to that of the ionic and covalent NO groups, and can be seen in Equations 5-18 and 5-19.
91 1618161816 11 22 22()2()()() s urfacesurfaceNOONOOO [5-18] 1618181816 11 22 2()22()()() s urfacesurfaceNOOONOO [5-19] Nitro group desorption accounts for part of the NO2 effluent in the gas phase during desorption, and the mechanism can be seen in Equations 5-20 to 5-22. 1616 22()() g NONO [5-20] 1618181618 ()() g NOONOO [5-21] 1818 22()() g NONO [5-22] Nitrite formation for NO2 adsorption does not require oxygen from the surface or lattice to form. By bonding one oxygen atom to a metal site NO2 directly forms a nitrite ion, shown by Equation 5-23. 1616 22() NOeNO [5-23] The nitrite ion can decompos e to desorb NO, as shown in Equation 24, desorb in combination with another nitrite ion, shown in Equation 5-25, or simply reverse the formation reaction seen in Equation 5-23. 161616 1 2 22()()surfaceNONOOe [5-24] 161616 2222()2 NONOOe [5-25] The nitrite may also use an oxygen atom at th e surface in order to form the nitrate ion, shown in Equations 5-26 and 5-27. 161616 1 2 22()3()()surfaceNOONO [5-26] 1618161618 1 2 2()2()()surfaceNOOONOO [5-27]
92 By reconfiguring according to Equation 5-27, th en returning to the nitrite ion via Equation 5-28, the complex could then return to the nitr ate configuration and pick up a second isotopically labeled oxygen, shown in Equation 5-29. 1618161816 1 2 22()()() s urfaceNOONOOO [5-28] 1618181618 1 2 2()2()()surfaceNOOONOO [5-29] From this point, the nitrates can either revert to nitrite io ns or decompose as seen in Equations 5-11 to 5-14. Since the nitrate-nitrite conversion can resu lt in oxygen substitution, this may result in desorption of a N16O18O molecule. 16161616 3()()gaslatticeNONOOOe [5-11] 1618161816 2()()gaslatticeNOONOOOe [5-12] 1618181616 2()()gaslatticeNOONOOOe [5-13] 1816181816 2()()gaslatticeNOONOOOe [5-14] NO2 adsorption is somewhat more complex due to the extra oxygen that must be accounted for, but fortunately the desorption spectrum br eaks down into three regions. The NO desorption from ionic and covalent complexes is noticeably absent in the spectrum. In that temperature range there is a small increase in m/z 44 signa l, which corresponds to a desorption of N2O. This is likely a low-temperature precursor to the larger peak later, with the N2O formed by combination of weakly bonded ionic NO+ on Cu sites. Since NO2 must dissociate in order to form the NO ionic/covalent complexes, they are especially weakly bound, due to the excess oxygen.
93 5.2 Experimental 5.2.1 Powder Preparation A new batch of La2CuO4 powder was prepared by th e autoignition technique221 similarly to the method detailed in Chapter 3. A stoichiometric mixture of La(NO3)3H2O (Kanto Chemical Company, 99.99%) and Cu(NO3)2xH2O (Alfa Aesar, 99.999%) was prepared in deionized water. After dissolving the metal nitrates, citric acid (Alfa Aesar, 99.5%) was added to the mixture until 0.25 mole percent was achieved, ma king this a fuel-lean solution. The solution was heated to ~60 C with constant stirring until the solution thickened. The gel was held near 70 oC for 7 hours, until the gel hardened from water lo ss. Following the drying step, the crusty gel was heated ~3 oC/minute until the precipitate autoignited due to heating. The powder generated by the autoignition process was cal cined at 600 C for 10 hours. Samples weighing nominally 12 milligrams were placed in the TPR/TPD reactor and cycled to 700 oC under 21% O2 to clean the sample of atmosp heric adsorbates and fully oxidize the sample. After cooling the sample, it was stored in a flowing helium atmosphere at room temperature. 5.2.2 Isotopically Labeled TPR/ TPD The procedure of the temperature-progra mmed desorption (TPD) and reaction (TPR) experiments was described previously.66 68 69 205 Samples of La2CuO4 containing nominally 12 mg of powder (BET area 10 m2/g) were initially purged of at mospheric adsorbed species by cycling the sample to 700 oC and back to room temperature under an atmosphere of 1 % 16O2.The process was repeated until spectra did not show any desorption peaks. Once clean, the sample was cooled under 1 % 16O2 and then maintained under the helium atmosphere until further experiments were performed. Concentrat ions for the reactor were maintained by MKS
94 mass flow controllers (MFC) for all standard gases, and 18O2 was controlled by a custom-built Alicat MFC. For all mass spectrometry experiments, specific mass to charge ratios were examined: m/z = 14 (nitrogen species), m/z = 16 (oxygen species), m/z = 28 (C16O and N2), m/z = 30 (N16O and C18O), m/z = 32 (N18O and 16O2), m/z = 34 (16O18O), m/z = 36 (18O2), m/z = 44 (C16O2 and N2 16O), m/z = 46 (N2 18O, C18O2 and NO2 16) and m/z=48 (N16O18O). The ratios of m/z 14 and m/z 16 were monitored to distinguish between gase s with the same m/z ratio, in particular m/z =32, corresponding to N18O and 16O2. To distinguish between the two gases, a matrix approach was used. The characteristic signal strength ratio of m/z 14 and 16 was isolated and compared to the identifying m/z of a nitrogen or oxygen cont aining species, for each gas present in the gas stream. Using these ratios, the contribution of every identifiable specie s was subtracted from signal intensity at m/z 14 and 16, and so the rema inder of m/z 14 and 16 after subtraction served as the contribution to the undetermined species. This process allowed the determination of each gas concentration, but had the unde sired effect of injecting some noise into the measurement for N18O, as that concentration was typically low. For TPR experiments, the sample was exposed to 30 cubic centimeters per minute (ccm) of a reaction mixture at 30 oC, and held until gas flow and mass spectrometer signal remained steady. The sample was heated to 700 oC at 30 oC / minute, and the efflue nt spectra recorded to determine the extent of reaction and reaction products. Experimental procedure for TPD was modified slightly to conserve isotopic oxygen. The sample was cooled to 30 oC from 300 oC at 10 oC / minute without a dwelling period at elevated temperature. At 30 oC, the sample was purged with 30 ccm helium until a stable baseline for mass analysis was achieved. The sample was heated to 700 oC at 30 oC /minute beneath 30 ccm
95 flowing helium atmosphere. During heating, the effluent gas was analyzed to determine the quantity and nature of desorbed products. Meas urements were normalized to the initial surface area of the powder sample measured following ca lcination step. This surface area was assumed to change during heating cycles, but samples used for the experiments were too small to accurately determine the change through BET analys is. Table 5-1 shows the list of experiments. TPR and TPD experiments were carried out on separate powder samples in order to preserve the sequence and maximize similarity between experiments. Primary experiments maintained an atmosphere of N16Ox and 16O2. After the initial run, the sample pretreatment consisted of annealing in 16O2 at 500 oC followed by cooling to room temperature. After this pretreatment, runs were performed in N16Ox and 18O2 Subsequent pretreatments of the sample were performed in 1% 18O2 in order to exchange lattice oxyg en for isotopic oxygen to produce La2Cu18O4 in lightly and heavily 18O-enriched forms. The lightly enriched sample was prepared by annealing La2Cu16O4 in 1% 18O2 for 1 hour at 500 oC, and the heavily enriched sample by a further 2 hours under 1% 18O2 for at 500 oC. The final runs in the sequence contained N16Ox and 18O2 and N16Ox and 16O2 in the gas phase, respectively. Additionally, a TPR of 1000 ppm 18O2 was performed on a freshly prepared sample in order to examine the oxygen exchange behavior of pure La2Cu16O4. 5.3 Results and Discussion 5.3.1 18O2 Temperature Programmed Reaction A 12 milligram La2CuO4 sample was exposed to 1000 ppm of 18O2 at room temperature and heated to examine the exchange behavior of oxygen on La2CuO4. The mass spectrum for this experiment can be seen in Figure 5-3. Near 400 oC, 18O2 begins to exchange with 16O2 from the La2CuO4 lattice. The concentration of 18O2 in the gas phase begins to decrease and the concentration of 16O2 increases correspondingly so that the overall oxygen concentration in the
96 gas phase remains essentially unchanged. Interest ingly, the exchange is mostly between whole molecules. The consumed oxygen is 18O2, and the emitted oxygen remains mostly as 16O2 or 18O2, with only ~40 ppm of 16O18O formed at 700 oC. This is a small amount compared to the 500 ppm of 16O2 emitted at that temperature. Since the potentiometric sensors operate at temperatures above 400 oC,66 271 the experiment revealed that the La2CuO4 was capable of oxygen exchange for all temperature re gions currently tested for the sensors. Further, due to the la rge exchange even in dilute oxygen, it can be assumed that at temperatures above 400 oC, gas-phase oxygen can be considered to integrate into the La2CuO4 lattice. 5.3.2 Determination of NOx Adsorption via TPD/TPR Since lattice oxygen begins to exchange at temperatures approaching 400 oC, TPD was performed by first adsorbing 1% NO and 1% 18O2 from 300 oC down to room temperature over La2Cu16O4. The subsequent desorption profiles of this experiment can be seen in Figure 5-4 The desorption profile reveals no evidence of N18O throughout the temperature range, and only a very small amount of 16O18O on the high edge of the 16O2 desorption peak from 400 to 500 oC, seen in Figure 5-4b. This range falls above the oxygen exchange threshold of the powder, and so it can be concluded that it is the reaction product of residual gas-phase 18O2 exchanging with the La2Cu16O4 lattice during the experiment, or simply a result of naturally occurring 18O (~0.4%) during La2CuO4 synthesis. As such, it was shown that no 18O was present in the surface complexes, and so no gas phase O2 reacted with the adsorbed NO. The desorption peaks observed in the first TPD (Figure 5-4a and 5-4b) are very large compared to subsequent runs. Some aging effect s were expected in la ter tests, as observed below, however, the high capacity of the fresh sample provided the best opportunity to produce N18O, and so this mass spectrum was displayed. An additional TPD run on a more aged sample
97 is shown in Figure 5-4c. The magnitude of gas adsorption and desorption decreases with age, most likely due to the coarsening of the grains. In addition, the aging resulted in an increase in O2/NO desorption as well as a continued evolution of O2 at high temperatures above 500 oC in Figure 2c, is most likel y bulk evolution of O2 from the lattice and contributed to greater O2/NO ratios in subsequent TPD experiments. When the oxygen sources are reversed (i.e. 16O2 in the gas phase and highly enriched La2Cu18O4 lattice), a much different situati on emerges. For the TPD of 1% N16O / 1% 16O2 over La2Cu18O4, (Figure 5-5a) scrambled oxygen products were observed even below the bulk oxygen exchange temperature. Above the oxygen exchange threshold, N18O tracked a similar desorption to N16O until all NO was exhausted on the surface. It was previously shown that the bulk supplies oxygen for surface complexes, so 18O was present in the complexes. To determine the effect of 18O enrichment on the system, TPD of N16O and 18 O2 were performed on a partially substituted La2Cu18O4 lattice. The TPD (Figure 5-5b) showed oxygen scrambling behavior intermediate to the previous two experiments. Concentrations of 16O2 remained the highest, followed by scrambled 16O18O at concentrations near N16O, and 18O2 levels were the lowest. In this partially enriched lattice, N18O production occurred above 400 oC, similarly to the highly enriched La2Cu18O4 case, but at a much lower level. Thus, 18O in the bulk is a necessary but not sufficient factor for N18O production. In addition, energy must be provided to decompose surface complexes in a way to produce N18O. Deconvolution of NO TPD results We subsequently used MATLAB code adapted from OHaver288 to perform Gaussian analysis and deconvolute the TPD spectra into i ndividual desorption peaks. The deconvolutions can be seen in Figure 5-6. The identified peaks of N18O (Figure 5-6b) served as the initial
98 guesses for the peak positions, which line up well with the middle peaks in the NO desorption spectra. The N18O peaks are noisy due to the effects of matrix subtraction of mass spec signal. This effect is lessened when N18O levels are large, and does not impact peak placement since the program accounts for this during analysis. Previous evaluation studies of desorption energies for NOx without labeled oxygen could only generate approximate results due to the lack of anchor peaks in the TPD. Substituted NOx products provided basis peak values for nonlinear Gaussian curve fitting, and these results allow quantitative analysis of the scrambling process as well as better peak separation for desorption regions. The products of 16O2 (Figure 5-6c), 16O18O (Figure 5-6d), and 18O2 (Figure 5-6e) were analyzed to determine first guesses for the larger desorption peaks. Using these first guesses, the program evaluated larger convolut ed peaks and output curve data and peak temperature. Peaks were evaluated to minimize overall error. The program removed the high temperature scrambled oxygen data in the TPD spectrum since it levels off at high temperatures and corresponds to lattice oxygen emission. As such, it was not included in the area calculations in the deconvolution tables. The peak values and desorption energies for these regions can be seen in Table 5-2. Integration of the peaks compile s the total quantity of desorpti on product due to a particular complex environment formed on the powder surface, and these values can also be seen in Table 5-2. The separated NO signal breaks into four separa te peak groupings, identifiable as species found on the powder surface at room temperature.286 These complexes were identified, in terms of increasing stability as: ionic and c ovalent NO, nitrite, and nitrate groups.225 243 250 255 257 289 290 The peak at 315 oC corresponded to simple ionic and covale nt complexes, as desorption of these
99 complexes was not accompanied by oxygen desorpti on of any type. As e xpected, there is not concurrent desorption of N18O, since this region is far belo w the oxygen exchange threshold in the surface complexes. The 16O2 peak at 383 oC also likely corres ponds to loosely bound adsorbed oxygen from the gas phase, and its contri bution is not included in the analysis of the desorption data. The next, largest, NO desorption peak at 420 oC should correspond to the formation of nitrite groups, identified by the fairly large amount of NO release. The ratio of total O2 to total NO is approximately 1:1. This region is accompanied by desorption of a large amount of all three oxygen species, with labeled oxygen species representing about 60% of total oxygen in the effluent stream, indicating that much of the oxyge n is resulting from the bulk and the complexes in this region. When NO desorption is compared to isotopically labeled oxygen emission (which will almost exclusively result from complex desorption and bulk emission), the ratio of O2 to NO rose to 1.5:1. This ratio of total O2 to total NO is still too large for even nitrate formation, let alone nitrite, thus the data suggested that bulk oxygen emission was obscuring the representative ratio. Additionally, N18O desorption occurs following rearrangement of oxygen in the surface complexes, representing approximately 1/7 of total NO. Eighty degrees higher at ~496 oC, the third NO desorption peak occurred. This desorption represents the mixed nitrates formed from c onversion of nitrite groups, which ideally would show an O2 to NO ratio of ~1:1. However, the labeled O2 to NO ratio remains 2:1, which indicated a much larger amount of oxygen coming from the complexes, suggesting nitrate group decomposition due to the shift of the oxygen to favor labeled groups, as well as a substantial emission of bulk oxygen as gas. The groups inte rconvert from nitrite to nitrate at these temperatures, and this behavior can be observed by the substa ntial overlap between the two
100 peaks of N16O and N18O, with N18O representing about 1/5 of tota l NO in this region. Again, desorption of oxygen species of each type is centere d in this region, and is much larger than NO desorption, emphasizing the breakdown of nitrate complexes and bulk emission. The highest temperature re gion for NO desorption at 582 oC represents the unidentate nitrate complex, and is almost exclusively N16O. In this region, very little NO remains on the surface and lattice oxygen emissions are very low, evidenced by the extremely low ratio of O2 to NO emission. NO 2 TPD Adsorption of N16O2 and 18O2 over La2Cu16O4 resulted in the desorption pattern seen in Figure 5-7a. The spectrum exhibits no evidence of N18O or N16O18O throughout the entire temperature range. The spectrum exhib its matching peak placement to the N16O2 and 16O2 TPD repeated for this study previously (Fi gure 5-7b). Previous studies showed NO2 formed nitro, nitrite, and nitrate comp lexes on the powder surface.286 Deconvolution of NO 2 TPD results Analysis of the NO2 peaks in Figure 5b in comparison with XPS and IR results286 separates the spectrum into nitro desorption, nitrite desorption, nitrite decomposition, and nitrate decomposition in order of ascending temperatur e. The peak corresponding to nitro desorption shows minor oxygen substitution in the corres ponding temperature range centered at 339 oC. Above this range, in the region wh ere nitrite desorption may produce N16O2 (Figure 6a), the scrambled product N16O18O (Figure 6b) and its decomposition product N18O (Figure 6c) were observed to occur from 325-450 oC. This is reasonable due to the slightly greater thermal stability of the nitrite complex.
101 At higher temperatures than the peak assume d for nitrite desorption, the mechanism for NO2 formation was assumed to be decompositi on during the nitrite-n itrate conversion. N16O18O emission decreased simultaneously with N16O2, replaced by increasing concentration of oxygen products 16O2 (Figure 5-8d), 18O2 (Figure 5-8e), and 16O18O (Figure 5-8f) with N16O (Figure 5-8g) and N18O resulting from nitrate decomposition. Total amounts of desorption from each species can be seen in Table 5-3. The spectrum following N16O2 and 16O2 adsorption over La2Cu18O4 revealed a complicated desorption, seen in Figure 5-7c. The four zones de scribed above are expressed as a series of five regions, described below. The peak at 340-350 likely corresponding to ni tro desorption shows no substitution in the corresponding temperature range of 275-375 oC, only a small amount of apparent NO desorption from ionic and covalent NO, with no oxyge n emission occurring in this range. N16O2 and N18O16 desorbed in a 10:1 ratio, demonstrating oxygen exchange even at temperatures below the La2CuO4 oxygen exchange threshold. This region re sults from the dissolution of the relatively weak bond between a metal site on the powder surface to the nitrogen of NO2. Just above this region, peaks appeared at 362 and 382 oC, corresponding to N16O2 and N18O16O, respectively. These peaks result from n itrite desorption, where the adsorption of NO2 as a nitrite complex reverses itself. Desorption in this region was 50% great er than for the nitro region (4.078 vs. 2.74 mol), but the ratio of N16O2 to N18O16O remained steady at 10:1. These complexes were more stable on the surface than th e nitro complex, and this was reflected in the higher temperature emissions range from 325-450 oC. The next region starts with a N16O2 desorption peak at 404 oC, accompanied by desorption of 16O2 starting at 392 oC. This threshold likely corr esponds to the beginning of
102 nitrate/nitrite transitioning, and oxygen from the original N16O2 loosed as a result. At 434 oC, N16O peaks, accounting for the oxygen released during the final N16O2 desorption, with the total NO and O2 ratio established at about 4: 1. This region also contains 16O18O and 18O2 emission overlapped with the NO desorption section, wh ich accounts for approximately 20% of total oxygen. Above this temperature range, nitrate deco mposition began to dominate, and this region was where the majority of N18O was seen. Just above 450 oC, N18O reaches its final peak, accompanied by oxygen products 16O2, 18O2 and 16O18O in larger quantities than in lower regions. 16O2 dominated the release, representing ~7/10 of total oxygen in this region corresponding to complex decomposition, with the NO to O2 ratio shifting to 1:2. This indicated that the majority of oxygen in the gas phase at th is temperature comes as a resu lt of complex decomposition, as 16O2 can only result from the original oxygen adsorbed as N16O2. N16O levels resulting from nitrate decomposition remained large, finally trailing off to zero ppm just below 650 oC. Oxygen levels from the decomposition of the complexes remain very small, though some constant lat tice oxygen emission occurs as 16O18O and 18O2. This contribution was removed in the deconvoluti on process, since it corresponds to oxygen enrichment from previous NO2 decomposition on the surface of the powder. Another partially substituted lattice of La2Cu18O4 was supplied for N16O2 experiments. The TPD (Figure 5-7d) revealed identical peaks to the TPD over the highly en riched lattice, but as expected, substituted products due to exchange on the nitrogen complexes were significantly lessened by the relative unavailability of 18O from the lattice. Additionally, scrambled oxygen complexes were reduced in gas-phase concentration.
103 NO TPR The TPR of 700 ppm N16O and 600 ppm 16O2 over heavily enriched La2Cu18O4 displayed nearly immediate conversion of N16O to N18O even at temperatures lower than the oxygen exchange threshold of th e lattice (Figure 5-9a). Above the threshold, N18O conversion continues to increase until ~50% conversion, stabilized near 550 oC. Oxygen exchange occurs freely, and scrambled 16O18O is the major oxygen product above 400 oC. N18O concentrations were determined by a total nitrogen balance (N16O/N18O) throughout the system referenced to a matrix of characteristic signal ratios from m/z 14, 16, 30, and 32 to determine levels of 16O2 and N18O. This balance accounted for each species in the sp ectrum, but generates noise in the data as a result. TPR under 700 ppm N16O / 300 ppm 16O2 over lightly enriched La2Cu18O4, shown in Figure 5-9b, reveals a similar pattern of reaction as the N16O / 16O2 TPR over highly enriched La2Cu18O4, though N18O levels and conversion are significantly lower across the entire temperature range. The lesser degree of 18O enrichment was shown by 16O2 concentration dominating the total oxygen level at temper atures above the exchange threshold. NO 2 TPR The TPR for 700 ppm N16O2 and 250 ppm 18O2 over La2Cu16O4 showed no signs of oxygen exchange at temperatures below 400 oC, shown in Figure 5-10a. Decomposition of N16O2 was observed beginning near 300 oC, with corresponding rise of 16O2 from the lattice and as a decomposition product. At higher temperatures, N18O was seen due to gas phase 18O2 promotion into the lattice, in similar fashion to the N16O and 18O2 TPR. TPR of 700 ppm N16O2 and 700 ppm 16O2 over heavily substituted La2Cu18O4 produces a similar spectrum to the TPD, shown in Figure 5-10b. N16O18O forms from 275-450 oC, tracking
104 desorption and decomposition of N16O2 as it disappears from the spectrum. N18O rose as N16O18O began to decompose, and stabilized near 200 ppm above the oxygen exchange threshold of 400 oC. Scrambled 16O18O was observed to be a major oxygen pr oduct of the nitrate decomposition above the exchange threshold. The TPR of 700 ppm N16O2 and 250 ppm 18O2 (Figure 5-10c) was almost exactly the same as the highly enriched La2Cu18O4 TPR of the same N16O2 concentration, in both peak placement and peak intensity. Only oxygen concentrations differ between the two TPR, which was most likely due to aging effects from the se quential testing of the fresh sample. NO x Adsorption Mechanism From these experiments, the sequence of adsorption and exchange for NO on La2CuO4 can be determined. The framework established in a previous study provides an adequate starting point to explain the mechanism.286 NO formed an initial bond with the La2CuO4 surface. If oriented on a metal site, ionic or covalently bonded NO results. If the ionic bond forms on a Cu site, the ionic complex may serve as a reaction intermediate for N2O formation, but this mechanism is minor, only occurring during low te mperature adsorption and so does not play a role in the sensing mechanism. On the La site, the covalent/ionic complex re versibly desorbs at low temperatures and should not be observed at the operati ng temperature of the sensor (>400 oC). The covalent and ionic NO complexes are the least strongly bonded to the surface of La2CuO4, and the nature of the ionic bond generated by NO occurs due to its resonance in the N-O pseudo-triple bond.287 As a result, oxygen exchange cannot occur without destroying the bond to the surface, and the mechanism proceeds according to Equati ons 1 and 2 at temperatures near 250 oC. The formation of these complexes is the likely reason for saturation of the sensor at low temperatures, and the
105 decomposition point for ionic and covalent speci es is likely the lowest possible operation temperature for the sensor itself. 16 ()()gNONOe  16 ()()gNONO  For oxygen-down bonding to the metal site, the nitrosyl ionic complex forms. This complex has the possibility of desorption at lo w temperatures, due to the dominant bonding between N and O, but if sufficient tilt was induc ed to allow bonding to a lattice O, the chelated nitrite complex may form, and may relax to fo rm the nitrite complex, seen in Figure 5-11. From the spectra of NOx adsorption on La2CuO4 powder, it was determined that the amount of heavy NOx species was dependant on the level of oxygen substitution in the lattice. Even in the presence of 18O2, La2Cu16O4 lattice with only partial substitution of 18O2 produced less N18Ox species than the fully substituted La2Cu18O4 in the presence of gas phase 16O2. This supports the conclusion that the reaction during adsorption occurs solely with bulk oxygen. This was important since the reaction due to nitr ate/nitrite group decomposition is not mixed potential, since it is not a true electrocatalytic reaction. Thus by proving the interaction of NOx with oxygen is only with the bulk, it has been prov en that the adsorption of gas onto the sensor, and the formation of complexes is the dominant potential generator for voltage establishment across the La2CuO4/Pt couple. Since oxygen integration to the lattice at high temperature is important to the mechanism, it is possible that the over-step of voltage generated by gas steps (when it occurs) is caused by the removal or ad dition of oxygen, and those contributions relax out as equilibrium is reestablished with oxygen in the gas phase. The ability of complexes to interconvert means that the adsorption signal will also be a function of the ratio of complexes formed on the surface at a given temperature.
106 In the TPDs, the bond between N and the original O dominates all complexes, and so in the nitrite configuration, it will relax to a confi guration where the original oxygen extends away from the surface as the sample cools, according to Figure 5-12. If NO bonds nitrogen down to an oxygen lattice site, the distinction be comes irrelevant, as the original O atom is already oriented upwards. In formation of the nitrate complex, the third oxygen in the complex was taken from the bul k, so all complexes forming on the pure La2Cu18O4 surface contain at least one 18O atom, and usually two. Sin ce NO reacts with lattice oxygen during adsorption and desorption and not gas-phase O2, this means that the complexes withdraw oxygen from the bulk to form positively charged oxygen vacancies, creating a voltage expressed across the sensor as a whole. However, even in the heavily enriched surface, the N16O still dominates the effluent stream, i ndicating that the orig inal oxygen was retained when the nitrate or nitrite complex decom posed. In the TPD of N16O / 18O2 over partially substituted La2Cu18O4 and N16O / 16O2 over La2Cu18O4 a very small amount of N18O was seen in the nitrate decomposition region, which supports that conclusion. Oxygen in the original NO molecule is bound via double bonds and this bond retains its character in the adsorption pro cess during nitrite formation. Fu rther, it retains enough of that character that during nitrate formation, it remains the retained oxygen when the complex decomposes. The only way to get around the ch aracter of the bond is to provide sufficient thermal energy to undergo the nitrit e-nitrite conversion repeatedly to increase the probability for exchange of the primary oxygen. The stability of su rface complexes is at most as stable as bulk oxygen, so even heating during adsorption (as in the TPR case) supplies sufficient thermal excitement to destabilize the bonding be tween the nitrogen and primary oxygen.
107 In the TPR, adsorption and nitrite complex formation were combined with additional energy since the sample was heated during NO ad sorption. This destabilized the complex, allowing realignment of bulk oxygen into the prim ary position. This behavior was seen during TPR of the 18O enriched lattices. In those experiments, peaks near 300 oC corresponding to nitrite complex decomposition were observed to produce N18O, though this was not a product seen in the TPD. To form the nitrate, the ni trite complex must procure one more oxygen atom, and with energy from heating in addition to oxygen exchange into the bulk, allows destabilization and substitution. This required destabilizati on was likely also why NO2 does not result from NO adsorptionthe amount of energy required to make the transiti on from NO to nitrate and nitrite destabilizes the complex too much for the NO2 molecule to form, on this material. Additionally, the nitrate and nitrite are stable eno ugh to retain at least one oxygen atom and prevent N2 from being the product. From these results, a simplification of the possible NO adso rption mechanism shown earlier can be established. This si mplification is shown in Figure 5-13. For each of the six NO2 TPR and TPD, there is no conc urrent desorption of scrambled NO2 or N18O in the nitro desorption range, so it can r easonably be concluded the exchange of oxygen in that complex does not occur. The second peak corresponding to NO2 resulted in the high end of the nitrite desorption region, which indicate d that the complex dissociated as it rearranged through the unstable bridged nitrate configuration, s een in Figure 5-14. In the case of NO2 adsorption, the majority of complexes formed result in the injection of negative charge into the electrode, since bonds to the surface occurs for this adsorption without simultaneous oxygen removal from the bulk to produce charge from oxygen vacancies. The remainders of the complexes (nitrate) require ox ygen from the bulk to form, as in the NO case.
108 These nitrates cancel the voltage contribution of the nitrite formation, and so the total charge movement is determined by the relative amount of each complex. This accounts for the voltage generated across the sensor during NO2 exposure. This also occurs in TPD of N16O2 and 18O2 over La2Cu18O4 and N16O2 and 18O2 over La2Cu18O4 so it suggests the rearrangement is necessar ily less energetic than in the NO case for nitrite/nitrate rearrangement, but operates similarly in all other respects. The analysis of the qualitative data allows for the simplification of the earlier reaction mechanism, and this can be seen in Figure 5-15. 5.4 Summary Labeled oxygen studies were performed on La2CuO4 powder samples. The main techniques utilized were Temperature Pr ogrammed Reaction and Temperature Programmed Desorption experiments. The material was examined under multiple gas conditions of NOx and 16O2/18O2 atmospheres, as well as varied levels of 18O enrichment in the lattice itself. Through these studies, it was determined that the formation of charged surface complexes occurs solely through the use of lattice oxygen. Gasphase oxygen exchange with the La2CuO4 bulk occurs at temperatures above 400 oC. This substitution of the lattice allows gas phase oxygen to indirectly integrate into surface complexes, shown in TPR experiments. In TPD experiments, the concentration of scrambled O2 and NOx products was noted to increase with greater 18O2 pretreatment. Substitution of 18O for 16O in surface NOx complexes was postulated to occur due to the repeated transformation between the nitrit e and nitrate complexes, as well as resonance between multiple nitrate configurations. Oxygen-18 retention from the gas phase occurred due to substitution in the surface complex combined wi th thermally induced vibration leading to the destabilization of the bond between nitrogen and the primary (original) oxygen in the complex. Since the surface complexes formed solely du e to lattice oxygen, the Mixed Potential
109 electrochemical reaction doe s not occur as such on La2CuO4, and so that theory of sensor response does not adequately explain the mech anism of gas sensing. A more comprehensive model that includes the shift in Fermi level due to adsorption and charged complex formation can be found in Differential Electrode Equilibria.
110 NO Adsorption Complex Formation Nitrate Nitrite Ionic/Covalent NO Third Oxygen Second Oxygen No exchange Desorbs as NO16 Gas-phase (adsorbed on surface) Bulk (Surface/subsurface lattice) O16 N O16 (O bonded up or down) O18 N O16 (O bonded down) Desorbs as NO16 Desorbs as NO18 Nitrate Third Oxygen Bulk (Surface/subsurface lattice) Gas-phase (adsorbed on surface)Gas-phase (adsorbed on surface) Bulk (Surface/subsurface lattice) Gas Phase Oxygen assumed to be O18 molecules O16 O18 N O16 O16 O16 N O16 O18 O18 N O16 O18 O16 N O16 Decomposes to NO16 and O16 Desorbs as NO16/ O16O16 Desorbs as NO16/ O16O18 Decomposes to NO16 and O18 Decomposes to NO18 and O16 Desorbs as NO16/ O18O16 Desorbs as NO16/ O18O18 Desorbs as NO18/ O16O16 Desorbs as NO18/ O16O18 Decomposes to NO18 and O18 Desorbs as NO18/ O18O16 Desorbs as NO18/ O18O18 After exchange Desorbs as NO18 Figure 5.1 Flowchart of possible isotopic oxyge n exchange mechanisms during NO adsorption on La2CuO4.
111 NO2Adsorption Complex Formation Nitrate Nitrite Ionic/Covalent NO (O16 liberated) Third Oxygen Desorbs as NO16 / adsorbed O16Bulk (Surface/subsurface lattice)Gas-phase (adsorbed on surface) Recombination with Gas-phase O18 (adsorbed on surface) Gas Phase Oxygen assumed to be O18 molecules O16 O18 N O16 O16 O16 N O16 O18 O18 N O16 Decomposes to NO16 and O16 Desorbs as NO16/ O16O16 Desorbs as NO16/ O16O18 Decomposes to NO16 and O18 Decomposes to NO18 and O16 Desorbs as NO16/ O18O16 Desorbs as NO16/ O18O18 Desorbs as NO18/ O16O16 Desorbs as NO18/ O16O18 Decomposes to NO18 and O18 Desorbs as NO18/ O18O16 Desorbs as NO18/ O18O18 Nitro No exchange or exchange with bulk Desorbs as NO16 After exchange Desorbs as NO18 Partial exchange with gas Desorbs as NO16O18 Total exchange with gas Desorbs as NO18O18 No exchange or exchange with bulk Desorbs as NO16O16 O16 N O16 O18 N O16 Recombination with Gas-phase O18 (adsorbed on surface) Desorbs as NO18 / adsorbed O16 Desorbs as N2O16 / adsorbed O16O16 Desorbs as NO18 / adsorbed O16 Figure 5.2 Flowchart of possible oxygen exchange mechanisms during NO2 adsorption on La2CuO4.
112 Table 5-1. List of Experiments Performed Gas Mixture (ppm) Type Initial Lattice Oxygen Gas Mixture (ppm) 18O2 (1000) TPR La2Cu16O4 18O2 (1000) NO (10000) + 16O2 (10000), NO2 (10000) +16O2 (10000) TPD La2Cu16O4 18O partially enriched 18O fully enriched NO (10000) + 16O2 (10000), NO2 (10000) +16O2 (10000) NO(10000) +18O2 (10000), NO(10000) 2+18O2 (10000) TPD La2Cu16O4 18O partially enriched 18O fully enriched NO(10000) +18O2 (10000), NO(10000) 2+18O2 (10000) NO(700) +16O2(600), NO2(700) +16O2(600) TPR La2Cu16O4 18O partially enriched 18O fully enriched NO(700) +16O2(600), NO2(700) +16O2(600) NO(700) +18O2(600), NO2(700) +18O2(600) TPR La2Cu16O4 18O partially enriched 18O fully enriched NO(700) +18O2(600), NO2(700) +18O2(600)
113 0 200 400 600 800 1000 100200300400500600700Concentration (PPM)Temperature18O18O18O16O16O16O Figure 5-3. Temperature pr ogrammed reaction of 1000 ppm 18O2 over La2Cu16O4 0 500 1000 1500 2000 2500 3000 200300400500600700Concentration (PPM/m2)Temperature (oC) N18O16O18O18O2 16O2N16O 0 20 40 60 80 100 200300400500600700Concentration (PPM/m2)Temperature (oC) N18O16O18O18O2 16O2N16O 0 100 200 300 400 500 600 700 200300400500600700Concentration (PPM/m2)N16O16O2 16O18O Temperature (oC) (c). Figure 5-4 (a). 1% NO+ 1% 18O2 TPD over La2Cu16O4 Lattice, (b) Low Concentration Detail, (c).TPD of aged La2CuO4 sample
114 0 200 400 600 800 1000 200300400500600700Concentration (PPM/m2)Temperature16O18O N18O N16O16O2 18O2 (a) 0 100 200 300 400 500 600 700 200300400500600700Concentration (PPM/m2)Temperature18O2N18O16O18O16O2N16O (b) Figure 5-5. (a) 1% NO+ 1% 16O2 TPD over heavily enriched La2Cu18O4 Lattice; (b) 1% NO+ 1% 18O2 TPD over lightly enriched La2Cu18O4 Lattice 0 100 200 300 400 500 600 200300400500600700091607_#03_LCOXXIII_NOO216TPD_NO16Concentration N16O (PPM)Temperature (oC) (a) 0 50 100 150 200 200300400500600700091607_#03_LCOXXIII_NOO216TPD_NO18Concentration N18O (PPM)Temperature (oC) (b) 0 100 200 300 400 500 600 700 800 200300400500600700Concentration 16O2 (PPM)Temperature (oC) (c) 0 100 200 300 400 500 200300400500600700091607_#03_LCOXXIII_NOO216TPD_O16O18Concentration 16O18O (PPM)Temperature (oC) (d) 0 100 200 300 400 500 600 700 200300400500600700091607_#03_LCOXXIII_NOO216TPD_O18O18Concentration 18O2 (PPM)Temperature (oC) (e) Figure 5-6. Gaussian Analysis of (a) Desorption Peak of N16O; (b) Desorption Peak of N18O; (c) Desorption Peak of 16O2; (d) Desorption Peak of 16O18O; (e) Desorption Peak of 18O2
115 Table 5-2: Desorption Peaks for 1% NO/1% 16O2 Adsorption on La2Cu18O4 Species Temperature (oC) Desorption Energy (kJ/mol) Peak mol/m2 Adsorbed Total mol/m2 Adsorbed Peak Fit % Error N16O 315 543 0.626 4.12 1.15 420 644 1.54 497 718 1.48 582 802 0.475 N18O 442 665 0.255 0.717 9.86 507 728 0.462 16O2 383 608 1.57 2.95 2.31 420 644 1.38 16O18O 433 656 1.17 3.18 2.87 513 734 2.01 18O2 427 651 1.14 2.94 2.31 486 708 1.80
116 0 500 1000 1500 2000 2500 3000 3500 4000 200300400500600700Concentration (PPM/m2)Temperature16O2N16O N16O2(a). 0 500 1000 1500 2000 2500 3000 3500 4000 200300400500600700Concentration (PPM/m2)Temperature N16O2N16O16O18O18O2 16O2 (b). 0 500 1000 1500 2000 2500 200300400500600700Concentration (PPM/m2)Temperature16O2N16O18O N16O16O18O18O2N16O2N18O (c). 0 500 1000 1500 2000 200300400500600700Concentration (PPM/m2)Temperature N16O18O16O18O N18O16O2N16O N16O2 18O2 (d). Figure 5-7. (a) 1% NO2 + 1% 18O2 TPD over La2Cu16O4 Lattice; (b) 1% NO2 + 1% 16O2 TPD over La2Cu16O4 Lattice; (c) 1% NO2 + 1% 16O2 TPD over heavily enriched La2Cu18O4 Lattice; (d) 1% NO2 + 1% TPD over lightly enriched La2Cu18O4 Lattice.
117 0 500 1000 1500 2000 2500 200300400500600700091707_#03_LCOXXIV_NO2O216TPD_NO16O16Concentration N16O2 (PPM/m2)Temperature (oC) (a) 0 50 100 150 200 250 200300400500600700091707_#03_LCOXXIV_NO2O216TPD_NO16O18Concentration N16O18O (PPM/m2)Temperature (oC) (b) 0 50 100 150 200 250 300 200300400500600700091707_#03_LCOXXIV_NO2O216TPD_NO18Concentration N18O (oC)Temperature (oC) (c) 0 100 200 300 400 500 600 700 800 200300400500600700091707_#03_LCOXXIV_NO2O216TPD_O16O16Concentration 16O2 (PPM/m2)Temperature (oC) (d) 0 20 40 60 80 100 200300400500600700091707_#03_LCOXXIV_NO2O216TPD_O18O18Concentration 18O2 (PPM/m2)Temperature (oC) (e) 0 20 40 60 80 100 120 200300400500600700091707_#03_LCOXXIV_NO2O216TPD_O16O18Concentration 16O18O (PPM/m2)Temperature (oC) (f)
118 0 200 400 600 800 1000 1200 200300400500600700Concentration N16O (PPM/m2)Temperature (oC) (g) Figure 5-8. Gaussian Analysis of: (a) Desorption Peak of N16O2; (b) Desorption Peak of N16O18O; (c) Desorption Peak of N18O; (d) Desorption Peak of 16O2; (e) Desorption Peak of 18O2; (f) Desorption Peak of 16O18O; (g) Desorption Peak of N16O Table 5-3: Desorption Peaks for 1% N16O2 and 1% 16O2 Adsorption on La2Cu18O4 Species Temperature (oC) Desorption Energy (kJ/mol) Peak mol/m2 Adsorbed Total mol/m2 Adsorbed Peak Fit % Error N16O 434 658 5.28 7.66 2.77 549 769 2.39 N18O 347 574 0.357 1.17 6.51 457 680 0.812 N16O2 340 567 2.47 7.48 1.27 362 589 3.81 404 629 1.20 N16O18O 350 577 0.268 0.650 6.81 382 608 0.381 16O2 392 617 1.04 2.88 2.18 441 664 1.84 16O18O 422 647 0.150 0.227 4.24 446 669 0.0770 18O2 425 649 0.0897 0.212 3.09 452 675 0.122
119 0 100 200 300 400 500 600 700 300400500600700Concentration (PPM)Temperature (oC)16O2 16O18O18O2N18O N16O Total NO (a). 0 100 200 300 400 500 600 700 300400500600700(b).18O2 16O18O N18O N16O Total NOConcentration (PPM)Temperature (oC) Figure 5-9. (a) 700 ppm N16O+600 ppm 16O2 TPR over heavily enriched La2Cu18O4 Lattice; (b) 700 ppm NO+ 300 ppm 18O2 TPR over lightly enriched La2Cu18O4 Lattice. 0 200 400 600 800 1000 200300400500600700Concentration (PPM)Temperature N16O2 18O2 16O2N16O16O18O N18O (a). 0 200 400 600 800 1000 200300400500600700Concentration (PPM)Temperature18O2N18O N16O18O16O18O16O2N16O N16O2(b). 0 200 400 600 800 1000 200300400500600700Concentration (PPM)Temperature N16O2 16O18O N18O18O2N16O N16O18O (c). Figure 5-10. (a) 600 ppm NO2 + 250 ppm 18O2 TPR over La2Cu16O4 Lattice; (b) 700 ppm NO2+ 700 ppm 16O2 TPR over La2Cu18O4 Lattice; (c) 700 ppm NO2 + 250 ppm 18O2 TPR over heavily enriched La2Cu18O4 Lattice
120 Figure 5-11. Formation of nitrito via nitr osyl ion/ chelated nitrite intermediate O N M O M O M O M O M O M M O M O N= O V M O = OO = O O = OO = OO = OO = OO = O Figure 5-12. Nitrogen-down bonding to 16O surface to form one-step nitrito charged complex. Recombination with Bulk O18Recombination with Bulk O18 Figure 5-13. Simplified Mechanism for NO Adsorption on La2CuO4.
121 Figure 5-14. Reversible rearrangement of the nitrite to nitrate formed by NOx adsorption. NO2Adsorption Complex Formation Nitrate Nitrite Ionic/Covalent NO (O16liberated) Third Oxygen Desorbs as NO16/ adsorbed O16Recombination with Bulk O18 O16O18N O18 O16O16N O18 O18O18N O18 Decomposes to NO16and O16 Desorbs as NO16/ O16O18 Decomposes to NO16and O18 Decomposes to NO18and O16 Desorbs as NO16/ O2 18 Desorbs as NO18/ O2 16 Desorbs as NO18/ O16O18 Decomposes to NO18and O18 Desorbs as NO18/ O2 18 Nitro No exchange with gas phase or bulk Desorbs as NO16 No exchange with gas phase or bulk Desorbs as NO2 16 O16N O16 O16N O18 Recombination with Bulk O18 Desorbs as NO16/ adsorbed O18 Desorbs as NO2 16 Desorbs as NO16O18 Figure 5-15.Simplified Mechanism for NO2 Adsorption on La2CuO4.
122 CHAPTER 6 MASS SPECTROMETRY AND ELECTRICAL ANALYSIS OF A LA2CUO4/YSZ/PT FIELD MODIFIED DESORPTION SENSOR 6.1 Introduction Temperature Programmed Desorption and React ion (TPD/TPR) has been an effective method for examining the storage and reacti on behavior of gases on the surface of a material.72,215,234,291-295 The techniques are very useful in determining the processes and parameters of reaction for catalysis and sensi ng mechanisms of gas sensors. However, the requirements of heating to provide energy fo r desorption or reacti on makes the technique inapplicable to in situ observation of potentiometric sensor s at a constant temperature. One method of navigating this obstacle is to provide the energy for deso rption in a different manner. A separate method used for desorption of mo lecules from a substrate is termed Pulse Programmed Desorption (PPD), as this technique uses voltage pulses to provide the necessary energy for removing molecules from the surface. The PPD technique has its basis in high vacuum studies, where high voltage was used to ionize organic samples off of an emitter tip surface.296 Its utility was extended to metal microe lectrodes for the purpose of constructing functionalized proteins,297 and also on gold electrodes to reverse the adsorption of thiols.298 These techniques examined fundamental patterns of adsorption using direct mass spectrometry at low temperatures. In sensor experiments utilizing a potentiometric gas sensor with heating elements, another member of the research group noticed large sens or responses to small applied heater voltages after a small crack formed in one of the heaters.299 This crack formed an effective capacitor which produced an electric field that penetrated the sensing electrodes and caused a change in the response. This researcher then made a sens or with capacitors purposef ully constructed on the opposite side of the YSZ from the sensing electr odes. The capacitors consisted of an alumina
123 layer with gold contacts, all deposited on top of the YSZ. Experiments from this sample verified that an electric field from the capacitors ha s an effect on the EMF formed at the sensing electrodes, and hence on the sensor response. Using these results and the field desorption concept as a basis, it was expected that the application of direct or externally ge nerated electronic fields could be used to analyze and influence a poten tiometric sensor during operation at elevated temperatures. There exist important differences between these techniques applied to the vacuum condition versus ambient pressure. The energy supp lied by the voltage pulse will be insufficient in the macro-scale to desorb gas complexes in the absence of another driving force. In the microscale, pure desorption requires electric fields of tenths of a volt per angstrom to supply the desorption energy,300-302 which would be difficult to s upply for sensors previously used. Fortunately, the application of an external fiel d on the sensor couple for gas desorption has the effect of shifting the equilibrium of the sensor complexing reaction, which does not require as strong an electric field.303 In the macroscale case, desorption energy will be provided by thermal energy, while the field only modifies the c onditions for formation and decomposition of the charged complexes. This effect has been obser ved and modeled for the simpler case of charged colloid particles in liquid.304,305 The formation of charged complexes on the surface of the sensing electrode of the La2CuO4/YSZ/Pt has previously been demonstrated to provide most of the asymmetric potentia l between the sensing and counter electrodes.223 Therefore, polarization of the electrode field in a direc tion that assists the reversal of the formation reaction should allow a substantial fraction of the adsorbed gases to desorb, and change the sensor signal. Additionally, the effect of the voltage pulse on the response time of the sens or can be observed, possibly opening new methods of sensor ope ration at low temperatures. Conversely, the adsorption can be
124 assisted by maintaining an electric field that favors complex formation,306 which can lead to greater sensitivity at low ppm concentrations and higher temperatures. In both cases, the field should lead to shifts in the bonding or bitals, changing the bonding behavior.307 By modifying the system in this manner, the sensitivity of the sensor can be altered fo r multiple concentration ranges and temperatures. Modification of the sensor response and sensitivity using a perpendicularly applied magnetic field by this method has been previously observed in a resistive, MOSFET-type SnO2 sensor.308 We have also observed changes in the potentiometric sensor response due to both perpendi cular and parallel electric fields. Previous desorption studies in this La2CuO4 work utilized powder samples, where the surface area of the sample allows significant quantities of gases to adsorb. By desorbing these species, it was possible to begin constructing a mechanism in order to explain the individual contributions to the potential of a La2CuO4/YSZ/Pt potentiometric couple. One significant limitation of the mass spectrometry technique used thus far was the inability to observe the quantities adsorbed on the surface of the La2CuO4 electrode correspondi ng to the potential generated by NOx gases. By examining the electrode while in operation, the amount of catalytic reaction due to the electrode can be quantified, and this provides useful data when examining the relative contribution of el ectrochemical reaction to the overall potential. 6.2 Reactor Construction In order to conduct mass spectrometry experime nts for field modified devices, it proved necessary to construct two additional reactor setups. The reactors were designed individually to provide maximum flexibility in experimental technique. To perform TPD device analysis, a variant of the original powder reactor tube was constructed using a longer section of 2 mm inner diameter tubing welded to a short section of 1 mm inner diameter tube. This configuration allowed for the insertion of a full device into the tube center for rapid heating, while maintaining
125 minimal dead volume downstream of the sample into the mass spectrometer. Samples loaded into the wide end, and were then connected to gold leads supported in a small alumina tube which remains in the upstream volume. The alumin a tube cuts down the dead volume and serves to isolate the leads from each other. This arra y of leads extends to a pass-through at the UltraTorr seal to allow connection to a voltage source outside the reactor. The reactor used for device testing at constant temperature was fabricated according to the following procedure. Two concentric alumina t ubes were obtained from Ortech Ceramics. The outer tube was wound with resistive nichrome h eating element and placed within a section of insulating pipe. The area between the ceramic a nd pipe was coated with a ceramic cement to stabilize the ceramic wool insulation. The inne r pipe was machined to provide a gas channel through the length of the pipe, and a small window was ground from the tube side in order to provide a mounting point for sensor samples. A sc hematic of the reactor can be seen in Figure 6.1.309 6.3 Experimental 6.3.1 Sample Preparation Two configurations of electrochem ical cells were fabricated to test changes in desorptive properties and changes in potenti ometric sensor behavi or during exposure to varying strengths and directions of externally applied electric fields. To construct La2CuO4 electrodes for devices used in this chapter, a portion of the powder pr epared for Chapter 5 experiments was pressed into a solid porous bar for the purpose of in creasing surface area. Two grams of La2CuO4 powder was combined with graphite to produce a 97/3 La2CuO4/C ratio, mixed with a 5 milliliters of 10% polyvinyl buytral (PVB, Alfa Aesar, Ward Hill, MA) in ethanol and then uniaxially pressed at 2000 psi for 2 minutes. The bar was sintered for 10 hours at 800 oC for 10 hours. Electrode slices for each device were cut from the bar and ground to size usi ng fine grit silicon carbide on a
126 polishing wheel. Alumina slices were prepared by combining 5 grams alumina powder (Alcoa) with 5 milliliters 10% PVB in ethanol, pressing isotatically to 2000 psi for 1 minute, and then sintering the resulting bar at 1250 oC for a half-hour. The bar was cut using a diamond wheel, and thin slices were polished down using a polishing wheel. To study the electric-field effect on desorption, a La2CuO4capacitor type sample was constructed with a 1.26 mm La2CuO4 section attached between two thin alumina slices using an alumina paste (ESL 450) as cement, followed by drying for 4 hours at 115oC. After drying, Au paste (Engelhard, Newark, NJ) was applied to th e outer areas of the alumina slices to form electrodes. The device was again dried for 1 hour. Following this drying step, Au wire leads were attached to the electrodes on the alumin a and the entire device was sintered at 800oC for 10 hours. A schematic of the device can be seen in Figure 6-2. The external electric-field sensor was construc ted in a similar configuration to sensors built by other members of the research group.66,67,69,118,204,205,235 In this basic configuration, a sensing electrode and a pseudo-reference are aligned on o pposite sides of a yttrium-stabilized zirconia (YSZ) substrate. The modified sensor device co nsisted of a rectangular ring of alumina paste applied to a 100 m YSZ (Marketech) substrate dried at 115 oC for one hour, and then sintered at 1000 oC for 3 hours. Next, a gold cont act layer was deposited onto the alumina and a gold lead was attached. After drying, a second alumina la yer was applied over the gold layer, the sample was dried, and the device was sintered for 10 hours at 800 oC. A platinum air-reference electrode was applied to the YSZ inside the alumina ring us ing the paste described above, then dried. A La2CuO4 section was attached to the opposite side of the YSZ substrate by means of a La2CuO4 / poly-ethylene glycol (PEG 800) solu tion of ~15% solid load. A thin Pt wire was attached to the top of the electrode with the PEG based pa ste. Both the top face of the porous La2CuO4 electrode
127 and the Pt wire were capped with a thin alumin a section. A gold contact was painted on the top of the alumina section, and dried for an addi tional hour. A gold lead wa s attached to the gold contact. The entire device was dried for one hour, and then sintered at 800 oC for 10 hours. 6.3.2 Testing Setup For all experiments in the tubular alumin a reactor, gas environment was provided by a bank of mass flow controllers (M100B, MKS, Wilmington, MA) used for sensor measurements. This system utilized a manifold of four gas cylinders: 1000 ppm NO (N2 balance), 1000 ppm NO2 (N2 balance), 100% O2. and 100% N2, controlled by a LabVIEW program310 to provide a total flow rate of 50 cubic centimeters per mi nute. Differences in electronic potential were measured with a Keithley 2000 digital multimeter outfitted with a scanning card in order to monitor sensor signal and applied field bias. Vo ltage for field testing was provided by a HewlettPackard 6681A DC power supply, interf aced with a separate LabVIEW program311 or manually set to deliver a constant potential. 6.3.3 Electric-field Effect on Desorption The La2CuO4 capacitor-type sample was exposed to 1% NOx/1% O2 gas mixtures at 300 oC, subsequently cooled at 5 oC/minute to 25 oC under gas atmosphere from the mass flow controller setup used for mass spectrometry measurements. The sample was then outgassed under pure helium for 8 hours to remove all residual NOx species from the reactor chamber. Following outgassing, the helium flow rate was increased to 30 cubic centimeters per minute, and the sample was heated at 10 oC/minute to 600 oC. During heating, the effluent from the reactor was analyzed via mass spectrometry to determine the desorption points of the NOx from the La2CuO4 surface. This experiment was performed at 0 V, 2 V, 5 V, and 8 V across the gold electrodes attached to the alumina sections.
128 6.3.4 Electric-Field Effect on Sensing Behavior During testing of the external electric-field sensor, the potentiometric voltage signal between the La2CuO4 and platinum electrodes was measured for NOx concentrations of 0, 50, 100, 200, 400, and 650 parts per million. These conditi ons were repeated for electric-field biases of 0 V, V, V, and V between th e gold/alumina ring surrounding the platinum airreference and the gold electrode atop the La2CuO4 sensing electrode. Flow rates were held at 50 ccm total flow. The effluent gases from the reactor during testing we re analyzed by mass spectrometry. 6.4 Results and Discussion 6.4.1 Testing of La2CuO4 Capacitor-Type Sample The effect of the electric fiel d on the desorption profile of La2CuO4 was measured in a small dead-volume reactor tube. The initia l desorption series performed on this La2CuO4 capacitor-type sample was 1% NO/1% 16O2 for conditions of zero, 2, 5, and 8 V applied to the electrically isolated electrodes. The first deso rption profile under zero volt bias shows an NO spectra similar to the powder samples in the pr evious chapters, but at lower temperatures. The sample exhibits easy desorption of NO from th e surface, with 3 peaks in the region below 400 oC. The NO spectrum slowly tapers downward above these temperatures, as the pores begin to exhaust NOx species from the pellet interior. Subseque nt experiments increase the voltage across the sample, and the result of this change can be seen in Figure 6-3. As the electric-field voltage setpoint increased (polarity being irrelevant due to the device confi guration), the desorption peaks for NO shifted to higher temperatures, while maintaining similar initial peak shape. The solid sample was also compared to earlier da ta on powder samples under similar experimental conditions (Figure 6-4). The peak s from desorption under the effect of an electric field more closely resemble the spectrum of slightly aged La2CuO4 powder than the virgin solid sample,
129 though the solid sample exhibited a more prom inent nitrate decomposition peak. The peak temperatures in the solid sample were slightly higher than in the powder sample. In addition to the upward temperature shift of the peaks, the lingering desorption of NO at high temperatures was expressed as more defined peaks. These peak s sharply reduce to zero concentration as the system approaches 600 oC. This shift was accompanied by the appearance of substantial NO2 desorption peaks at the same temperatures as NO desorption, shown in Figure 6.5. The area underneath both the NO and NO2 peaks increased due to the voltage step from 0 V to 2 V during desorption. The subsequent ap plication of 5 V duri ng the desorption cycle following NO adsorption resulted in a minor increase in peak temperatures, but was accompanied by a small decrease in the intensity of the low temperature NO peak at ~300 oC, a doubling of the peak sizes for NO2 desorption and a decrease in the high-temperature NO peak roughly equivalent to the increase in NO2 emission. The 8 V desorption field left peak placement for NO unchanged, but the high temperature NO p eak returned to a larger area, and the NO2 desorption peaks returned to the 2 V level, albe it with slightly higher peak temperatures. The desorption experiment was immediately repeated at 0 V to determine the residual effect of the field on the sample. The sample was observed to gain a small amount of area in the low temperature NO desorption region, and lose a subs tantial amount in the hi gh temperature region. Peak placement for NO remained si milar to the 8 V case, but NO2 desorption levels rose to the rough midpoint between the 5 V and 8 V case. The quantity of gases desorbed from the sample can be found in Table 6.1. The behavior of the sa mple can most likely be explained as a result of two factors: polarization of the lattice and the electrostatic interaction of the complexes, and orientation of the NO molecules for adsorption.
130 Since La2CuO4 is paramagnetic at temperatures of interest312,313 it shows some degree of susceptibility to polarization by electromagnetic fields314, owing to the genera l polarizability of its perovskite layers.315 As a result, during the de sorption step, the orienta tion of electron orbitals interacting with the charged su rface complexes may, in zones, allow the release of an oxygen from the surface to the nitrite complexes duri ng decomposition. This mechanism was discussed in Chapters 4 and 5. The modification of bonding in these zones due to the orientation changes might allow for the separation of an NO2 molecule during the nitrate/nitrite transition, which in the non-modified case decomposes without oxygen re lease. A second factor to consider is the simple electrostatic forces exerted on th e ionic complexes as they sit on the surface.316 Complexes remaining on the electrode surface at moderate to high temperatures in the nonmodified case all possessed negative charge, and while the adsorbate layer was not entirely independent of the surface, a preferential orie ntation of electrons to wards the surface of a complex would, by simple repulsion, tend to force the charged complex further from the surface, and make it more likely to desorb intact. Th e possibility of gas-pha se orientation of NOx may also play a role, as the adsorption bonding direc tion can play a role in the statistical complex formation due to the different formation m echanisms. As polar molecules, both NO and NO2 would show some degree of rotational shift in the field generated from the electrodes on the insulating alumina layer. Identical experiments were subsequently performed on the sample, substituting 1% NO2 for 1% NO in the adsorption step. These spectr a can be seen in Figures 6-6 and 6-7. The desorption spectra obtained from 0 V approximates the results from powder samples, though the amount of NO desorbed at the higher temper ature peak is larger compared to the NO2 desorption peak for the solid sample experiment. Increase in field voltage from 0 to 2 V caused a 300 ppm
131 drop in NO2 peak desorption at 325 oC, but the peak at 425 oC was roughly equivalent to the 0 V desorption. NO concentration at the low temperat ure peak was slightly reduced from the 0 V sample, but again remains roughly constant ov er the higher temperature region. Desorption amounts can be seen in Table 6-2. Stepping the voltage to 5 V resu lts in an increase of NO and NO2 at the low temperature peak, and a decrease for both gases at the high temperature peak against the 2 V case. This experiment produced the least amou nt of NO desorption of all the NO2 adsorption experiments. At 8 volts, the NO2 signal dips below the concentration at 2 V for the lower peak, and the 5 V level for the high peak. Overall, this expe riment generated the lowest level of NO2 emission. NO emission at the low temperature peak is slightly above the 2 V case, and slightly above the 5 V case at the high temperature peak. In both the NO and NO2 adsorption cases, the desorption at a 5 V electric bias setpoint produced a higher relative amount of NO2 versus the other bias conditions. In the NO adsorptive case, the difference is substantia l, exhibiting a much lower NO desorption than any of the other experiments, and emitting the most NO2 of any experiment under those adsorption conditions. This same effect of higher NO2/lower NO emission was observed in the NO2 adsorption case. This suggests that the bias region corresponding to th is field strength must be examined in more detail during future testing. Unfortunately, space constraints on these initial samples forced the construction of a field generating device that doe s not lend itself to a simple modeling of the field strength. Though the area of the gold electrodes was sufficientl y large to allow for coverage of the La2CuO4 layer, the ratio between the areas of the gold electrodes to the square of the distance between the electrodes was only ~2. Th is situation does not well resemble the archetypal case of two infinite parallel plates,317 and as a result, Laplaces equation
132 222 2220dVdVdV dxdydz cannot reduce to the simplificati on of change only in the z di rection (i.e. fringing of the electric field occurs near the edges). However, the concentrations of gas desorbed during the experiment were sufficient to allow for a ten-fold reduction in volume of the La2CuO4 pellet, in future designs. This change would allow for a diffe rent configuration of th e device and result in a much better (solvable) approximation of an ideal multiple dielectric condenser.318 Fringing and non-ideal fields have advantag es in certain applications, but by reduction to a better approximation of ideality allows for easier m odeling when the device voltage is stepped. 6.4.2 Testing of External Electric-Field Sensor The sensor was evaluated for multiple NOx concentrations in 3% O2 under a bias of 0, 2, 5, and 8 V in the positive (high potential on La2CuO4 side) and backward (high potential on Pt side) directions. These experiments were perfor med at temperatures of 450, 500, and 550 oC and repeated twice. The sensors under bias voltage were compared to the unbiased sample for each temperature. The unbiased case was evaluated first to avoid the effects of pol arization seen in the La2CuO4 capacitor-type sample desc ribed above. In similar fashion to previous studies,223 the unbiased La2CuO4 electrode behaves with positive response to NO2 gas and negative response to NO when compared to Pt maintained under a reference atmosphere at 550 oC, but exhibits reversed response at 500 and 450 oC. The field-assisted deso rption studies shown above demonstrated that an indirectly applied el ectric field can affect the behavior of NOx on the surface of La2CuO4 even at relatively small field strengths. As such, this indicates that the response of a sensor under field bias has at least three component s. The first component is the EMF generated from the Nernstian response, as th e counter-electrode in th is case is a true airreference electrode (Equation 6-1).
133 2 2[()] ln 4[()]O OPUnknown RT V FPReference [6-1] Since this contribution comes out to be a result of the constant difference between the test chamber and the outside atmosphere, it will remain constant at a given te mperature and electricfield bias voltage setpoint. The second effect acting on the syst em is the voltage induced via polarization due to the presence of the field. Th is will vary depending on the voltage applied to the outer electrodes, but will be constant for each condition, and the deviation from 0V baseline for each case was observed by comparison of the baselines at 0ppm NOx. The third effect comes from the influence of the field on the bonding of NOx to the surface of La2CuO4 and its influence on the formation of the nitrite and nitrate complexes that produce the potential seen by the voltmeter. The effects of these changes were e xpressed by the differences as the slope of the signal/concentration curves. The comparison of the sensitivity of the device to NO atmospheres can be seen in Figure 6-8, and r2 values for logarithmic curve fitting are shown in Figure 6-9.299 At 550 oC the signal showed sensitivity of ~0.25 mV/decade for 0 V bias, but even this limited sensitivity was reduced further by the application of field in both directions. Field ap plication appeared to induce instability in the sensor signa l, evidenced by the large drop in r2. Recorded data from the voltage source indicated that th e actual voltage across the oute r electrodes did not remain constant, which could be re sponsible for noise in th e inner electrode couple.299 At 550 oC, the signal even in the unbiased case was small, and fu rther decreases in the sensitivity due to the electric field made the signal to noise ratio sufficiently small to cause a poor logarithmic fit to data. Data fitting was too poor to draw any signi ficant conclusions from the trends shown by the sample at this temperature, other than that the response to increasi ng NO concentration was a small negative voltage.
134 Previous sensor series showed an increased sensitivity as temperature was decreased, and this effect was also observed on this sample. At 500 oC, the sensor was observed to respond positively to increasing NO concentration. This response was found to increase in sensitivity under applied-field conditions, increasing with bi as voltage. Bias in the backwards direction displayed greater NO sensitivity than in the forward bias. At 8 volt backwards bias, the sample jumped from a positive sensitivity of 1 mV/decade ppm to 24 mV/decade ppm. At 450 oC, sensitivity also increased with increas ing electric-field strength. The effect of the field was more pronounced, displaying an increase from 3.2 mV/decade at 0 V to 5.2 mV/decade at 2 V forward bias. This sensitivity in creased slightly when biased in the backwards direction. Forward bias from 5 V showed very li ttle difference from the 2 V backwards case, but when bias was reversed in the 5 V case, se nsitivity stepped from ~5.6 mV/decade to 40 mV/decade. Increasing bias to 8 V forward decreased sensitivity to 4.8 mV/ decade, and reversing polarity at 8 V rebounded the signal to 28 mV/decade. The sensor response of La2CuO4 was shown previously to exhibit greater sensitivity to NO2 relative to NO, even at 550 oC. This had the effect of impr oving the signal to noise ratio of sensing behavior, allowing for mean ingful analysis of the data.299 Sensitivity plots for NO2 can be seen in Figure 6-10 and plots of r2 in Figure 6-11.299 The non-modified case for La2CuO4 against platinum reference produced 0.55 mV/decade ppm at 550 oC, and this was increased slightly to 0.65 mV/decade at 2V forward bias. Revers ing bias at 2V resulted in sensitivity of 0.9 mV/decade for NO2. A maximum sensitivity was reached at 5V forward bias of ~1 mV/decade, and this was only slightly reduced in the case of 8V forward bias Reverse bias at both 5 and 8 V caused marked decreases of sensitivity, to 0.65 and 0.8 mV/decade, respectively. The data displayed good logarithmic fit, with r2 values of ~0.9 for every testing condition.
135 At 500 oC, sensitivity of the sample in the unbi ased case was ~0.9 mV/decade. Forward bias at 2, 5, and 8 V increased this value to 1.2 mV/decade with logarithmic fitting, comparable to the unbiased case. Reverse bias at 2 V lowere d sensitivity to 0.8 mV/decade, raised to ~1.2 mV/decade for 5 V, and to 1.4 mV/decade at 8 V. As the electric-field st rength increased, the r2 value of the fit decreased, but rema ined comparable to the cases at 550 oC (~0.9) except for the case with 8 V backward. The most promising results of the study came for the NO2 case at 450 oC. Initial sensitivity of 2 mV/decade for the unbiased case was found to increase to about 3 mV/decade for both the forward and reverse case for 2 V bias. At 5 V fo rward bias, that increased to 4 mV/decade, but upon testing at 5 V backwards bias, the sensitivity was found to reve rse direction and increase in magnitude to -24 mV/decade. At 8 V forward bi as, the maximum sensitivity of -29 mV/decade was achieved. Reverse bias of 8 V reduced this value to -18 mV/decade. All data fitting of the bias conditions at 450 oC were as good as or better th an the unbiased case with r2 of 0.8. The effluent gases from the reactor were sampled for analysis by an Extrel 5220 Mass Spectrometer. The effluent profiles for the 450 oC case are displayed in Figures 6-12 to 6-15, staggered slightly in order to separate the spectra. To mimic the sensor test ing setup as closely as possible, the sensor test gas ma nifold was used to provide atmosphere for the reactor chamber. Since those gases contain nitrogen as a balance gas, the iden tifier m/z ratios of 14 and 28 were unavailable to determine conversion of NO2 to NO or the reverse. As a result, the only definitive conclusions can be drawn using the major m/z ra tio for the gas being tested. For forward bias conditions, a weak field generated by th e 2 V bias showed no conversion of NONO2 in the NO test case, but did decr ease the conversion of NO2NO in the NO2 testing case. Higher forward biases decrease the concentra tion of NO below 0 V baseline leve ls during NO testing, indicating
136 some conversion is taking place. For NO2 testing, the levels of NO2 decrease further below baseline with increasing bias strength. In the reverse bias case, NO te sting shows a trend of decreasing NO concentration with increasing field strength, with the 5 V case showing an outlier of baseline NO concentration. This outlier result was also seen in the field desorption study at 5 V, suggesting that the electric-fiel d strength at that voltage may be near a critical point. In the NO2 test case, increasing field strength lowers the concentration of NO2 in the gas phase, indicating greater conversion of NO2NO. 6.5 Summary Temperature programmed desorption experiments were performed on a La2CuO4 pellet. Initial bias and aging effects of the sample were shown to increase the desorption temperature of NO from the virgin case to subseque nt runs. Increases in field stre ngth were shown to change the desorption profile of NO and NO2 gases from the surface of the La2CuO4 pellet. A sensor utilizing a La2CuO4 pellet with similar field generating structures was also constructed. The sensor displayed changes in sensitivity to NOx gases with changes in field voltage and bias. This effect was the most pronounced at the lowest temperature in the study, 450 oC. Analysis of the data in this study provided a few factors to guide further experiments in the work. Firstly, the improvement on the sensitivity comes mainly at low temperatures for the low field strengths. To more fully evaluate the effect of fields at high temp eratures, a voltage source capable of producing voltages up to kV levels will be needed. The second improvement for further experiments is in the design of the de vice itself. As noted a bove, neither the field desorption device, nor the sensor device, appr oximated the ideal case of a parallel plate capacitor. Desorption experiments demonstrated th at the electrode pellets used for both studies were large enough to be significantly reduced in volume while maintaining sufficient surface area for further desorption experiments. As a resu lt, a better fabrication process that produces a
137 larger area/ electrode distance ration will produce a satisfactory sample. In this way, a field can be established to apply equally throughout the La2CuO4 electrode. Figure 6.1 Tubular reactor for sensor reference studies
138 Figure 6.2 La2CuO4 capacitor device for determini ng E-field effects on desorption 0 100 200 300 400 500 600 100200300400500600Concentration NO (PPM)Temperature (oC) 0 Vvirgin state 2V 0 V -final test 8V 5V Figure 6.3 NO desorption from the La2CuO4 capacitor following NO adsorption
139 0 5 10 15 20 25 30 35100200300400500600Intensity (arb. units)Temperature (oC) Figure 6.4 NO desorption from La2CuO4 powder following NO adsorption 0 100 200 300 400 500 600 100150200250300350400450500 0 Vvirgin state 2V 0 V -final test 5V 8VConcentration NO2 (PPM)Temperature (oC) Figure 6.5 NO2 desorption from the La2CuO4 capacitor following NO adsorption
140 Table 6.1 NO adsorption TPD of device Species Voltage Total mol Adsorbed NO 0 V-initial 13.08 2 V 13.45 5V 6.706 8 V 11.04 0 V-repeat 10.11 NO2 0 V-initial 2 V 5.165 5V 8.591 8 V 4.994 0 V-repeat 7.848 0 100 200 300 400 500 600 700 100200300400500600Concentration NO (PPM)Temperature (oC) 0 V 2V 5V 8V Figure 6.6. NO desorption from the La2CuO4 capacitor following NO2 adsorption
141 0 200 400 600 800 1000 100200300400500600Concentration NO2 (PPM)Temperature (oC) 0 V 2V 5V 8V Figure 6.7. NO2 desorption from the La2CuO4 capacitor following NO2 adsorption. Table 6.2. NO2 adsorption TPD of device Species Voltage Total mol Adsorbed NO 0 V 15.22 2 V 14.55 5V 13.29 8 V 14.00 NO2 0 V 13.24 2 V 11.47 5V 10.38 8 V 9.105
142 0 0.8 1.6 2.4 3.2 4 4.8 5.6 25 30 35 40 450C500C550C 0V 2V(f) 2V(b) 5V(f) 5V(b) 8V(f) 8V(b)Sensitivity (mV/decade) Figure 6-8. Sensitivity plot of NO for the sensor 0 0.2 0.4 0.6 0.8 1 1.2 450C500C550C 0V 2V(f) 2V(b) 5V(f) 5V(b) 8V(f) 8V(b)Sensitivity (mV/decade) Figure 6-9. R2 values of data fitting to linear plots for NO
143 -30 -25 -20 -15 -10 -5 0 5 450C500C550C 0V 2V(f) 2V(b) 5V(f) 5V(b) 8V(f) 8V(b)Sensitivity (mV/decade) Figure 6-10. Sensitivity plot of NO2 for the sensor 0 0.2 0.4 0.6 0.8 1 1.2 450C500C550C 0V 2V(f) 2V(b) 5V(f) 5V(b) 8V(f) 8V(b)Sensitivity (mV/decade) Figure 6-11. R2 values of data fitting to linear plots for NO2
144 0 100 200 300 400 500 600 700 800 600650700750800850900NO Concentration (PPM)Time (minutes)NO NO28 V 5 V 2 V 0 V Figure 6-12. Mass spectrometry comparison of NO in reactor effluent from forward biased sensor testing
145 0 50 100 150 200 250 300 600650700750800850900Concentration NO2 (PPM)Time (minutes)NO NO28 V 5 V 2 V 0 V Figure 6-13. Mass spectrometry comparison of NO2 in reactor effluent from forward biased sensor testing
146 0 100 200 300 400 500 600 700 800 600650700750800850900Concentration NO (PPM)Time (minutes)NO NO28 V 5 V 2 V 0 V Figure 6-14. Mass spectrometry comparison of NO in reactor effluent from reverse biased sensor testing
147 0 50 100 150 200 250 600650700750800850900Concentration NO2 (PPM)Time (minutes)NO NO28 V 5 V 2 V 0 V Figure 6-15. Mass spectrometry comparison of NO2 in reactor effluent from reverse biased sensor testing
148 CHAPTER 7 CONCLUSIONS Temperature Programmed Desorption and Temperature Programmed Reactions (TPD/TPR) were performed on La2CuO4. Experiments on La2CuO4 show quantifiable adsorption and reaction of NOx and COx mixtures on the surface. Examination of the desorption peaks for Temperature Programmed Desorption allowed for th e calculation of desorption energy of gas species adsorbed on the powder, and gas phase c onversion with respect to temperature could be tracked by examining concentration profiles dur ing Temperature Programmed Reaction. NO was found to be unreactive for conversion to other NOx gas species even in the presence of oxygen, indicating that its respon se in the sensor is more strongly influenced by adsorption rather than mixed potential. NO2 decomposed catalytically at sensor operating temperatures, and was found to fragment into several products at low temp erature, indicating mixed potential may play a significant role for the sensing behavior of NO2 on the potentiometric sensor. To determine the composition of surface specie s, the techniques of diffuse reflectance infrared spectroscopy (DRIFT) and X-ray photoelectron spectr oscopy (XPS) were used to investigate the adsorptive behaviors of La2CuO4 for NOx gases. In both the NO and NO2 case, the presence of ionically charge d nitrite and nitrate species wa s detected by IR. While these complexes appear under both gas adsorptions, the response of a full sensor was different for NO than for NO2. Thus, while the complexes on the surface due to adsorption are for all intents similar, the method in which they form greatly aff ects the charge the surfa ce layer generates, and leads to the differences in poten tial across the electrolyte. By examining the working electrode material by XPS independently of the sensor, the adsorptive component can be examined without compensation for possible charging effects on the sensor electrodes. Peak splitting to higher ejection energies from surface oxygen
149 demonstrated NO adsorption generates charge depletion layers by the formation of NO2 and NO3 at the gas/solid interface, where electrons ar e taken from the bulk and held. These depleted layers form a potential gradient that (in a full sensor) attempted to draw electrons from the counter-electrode, for a positive voltage. This phenomenon was observed in the absence of catalytic conversion of NO to N2 or NO2, thus advancing the theory that NO sensing via the La2CuO4 electrode is entirely adsorptive. NO2 forms similar complexes, but due to the extra oxygen present for use in these complexes, el ectron holes are generated, which produced a negative potential gr adient across the sensor couple in a full device. This phenomena was demonstrated by peak splitting towards the lesse ning of electron ejection energy from oxygen in XPS studies. Labeled oxygen studies were performed on La2CuO4 powder samples. The main techniques utilized were Temperature Pr ogrammed Reaction and Temperature Programmed Desorption experiments. The material was examined under multiple gas conditions of NOx and 16O2/18O2 atmospheres, as well as varied levels of 18O enrichment in the lattice itself. Through these studies, it was determined that the formation of charged surface complexes occurs solely through the use of lattice oxygen. Gasphase oxygen exchange with the La2CuO4 bulk occurs at temperatures above 400 oC. This substitution of the lattice allows gas phase oxygen to indirectly integrate into surface complexes, shown in TPR experiments. In TPD experiments, the concentration of scrambled O2 and NOx products was noted to increase with greater 18O2 pretreatment. Substitution of 18O for 16O in surface NOx complexes was postulated to occur due to the repeated transformation between the nitrit e and nitrate complexes, as well as resonance between multiple nitrate configurations. Oxygen-18 retention from the gas phase occurred due to substitution in the surface complex combined wi th thermally induced vibration leading to the
150 destabilization of the bond between nitrogen and the primary (original) oxygen in the complex. Since the surface complexes formed solely du e to lattice oxygen, the Mixed Potential electrochemical reaction doe s not occur as such on La2CuO4, and so that theory of sensor response does not adequately explain the mechan ism of gas sensing for this material. A more comprehensive model that includes the shift in Fermi level due to adsorption and charged complex formation can be found in Differential Electrode Equilibria. In an attempt to modify the adsorption equilibrium to modify sensor response, Temperature Programmed Desorption experiments unde r electric field bias were performed on a La2CuO4 pellet. Initial bias and aging effects of the sample were shown to increase the desorption temperature of NO from the virgin case to subsequent runs. Increases in field strength were shown to change the de sorption profile of NO and NO2 gases from the surface of the La2CuO4 pellet. A sensor utilizing a La2CuO4 pellet with similar fiel d generating structures was also constructed. The sensor disp layed changes in sensitivity to NOx gases with changes in field voltage and bias. This effect wa s the most pronounced at the lowe st temperature in the study, 450 oC. Analysis of the data in this study provided a few factors to guide further experiments in the work. Firstly, the improvement on the sensitivity comes mainly at low temperatures for the low field strengths. To more fully evaluate the effect of fields at high temp eratures, a voltage source capable of producing voltages up to kV levels will be needed. The second improvement for further experiments is in the design of the devi ce itself. As noted above neither did the field desorption device, nor the sensor device, approxi mate the ideal case of a capacitor dielectric. Desorption experiments demonstrated that the electrode pellets us ed for both studies were large enough to be significantly reduced in volume while maintaining sufficient surface area for
151 further desorption experiments. As a result, a be tter fabrication process that produces a larger area/electrode distance ratio wi ll produce a satisfactory sample. In this way, a field can be established to apply e qually throughout the La2CuO4 electrode, and facilitate the optimization of the electric field.
152 APPENDIX A CALIBRATION AND CORRECTION OF MASS SPECTROMETRY DATA Calibration of the mass spectrometer takes two ma in factors into account while obtaining a calibration profile. These factors consist of the primary m/z ratio, or the largest peak unique to that species of gas, and the fragmentary peaks caused by the decomposition of gas under ionization as the analyte passes into the mass sp ectrometer. Ideally, each gas species would have an identifier m/z independent of all other gas sp ecies, but since this outc ome is not possible, the contributions of each gas must be rem oved to obtain the true concentration. Experiments in Chapter 3 have a common set of searched m/z ratios, and these were analyzed differently than results in later wo rk. The initial TPD/TPR experiments monitor m/z ratios of 12, 14, 28, 30, 32, 44, and 46. The list of primary and secondary items is listed in Table A-1. Table A-1. Primary and secondary i ons found in Chapter 3 experiments Gas Species Primary m/z Secondary m/z NO 30 14 NO2 46 14, 30, 32 N2 O 44 14, 30 CO 28 12, 14 CO2 44 12, 14, 28, 32 O2 32 16 Instances where the secondary ions overlap with primary ions can be resolved by a concentration balance. This can be achieved w ith careful experimental control to allow a known concentration of gas to enter the reaction chamber. For most expe riments, the task of separating the individual gases is simply to work from the m/z calibration profile and set up a series of equations to relate the primary m/z intensity to a concentration profile. Once the concentration
153 profile has been established for the primary m/z of each gas, it is a si mple matter to relate concentration of the gas species to its fragments. Thus, the concentration of all components of a mixture can be found by starting with a unique non-overlapping primary m/z ratio and constructing a series of equations to subtract signal intensity pr ior to fitting the gas calibration profile. In cases where the prim ary m/z ratio is identical (CO2/N2O), a second series of equations must be instituted to assure balance of nitr ogen and carbon. Fortunately, since the fragmentation of these gases is small due to their stability, a useful approximation can be obtained even without additional analysis during a TPR. The most important overlap results from the fragmentation of the NO2 molecule to NO during the ionization process. Th e primary m/z ratio of 46 was us ed to determine the actual concentration of NO2 in the gas phase, and the attendant m/z of 30 was plotted in order to provide a second set of equations to compensate for the falsely inflated intensity of m/z 30 during TPD and TPR experiments. 246 NOCI 304630/4646 fromIAI 3030/4646() N ONONOCPIAIB In later work, the 18O isotope of oxygen was introduced into TPR and TPD of NO and NO2 gases. As a result of oxygen exchange of the ga s with the bulk, and in terface reactions between the 16O lattice and NOx gas, N16O (m/z=30), N18O (m/z=32), N16O2 (m/z=46), N16O18O (m/z=48), and N18O2 (m/z=50) are present in the effluent stream. At temperatures above 400 oC, the gas phase 18O2 (m/z=36) interacts with the bulk to exchange 16O2 (m/z=32), and 16O18O (m/z=34), into the gas phase. 18O2 and 16O18O each have a unique m/z ratio, and so the concentrations of
154 these species are easily monitored without resorting to a series of matrices. The m/z ratio of 32, however, is common to the atmospherically prevalent 16O2 as well as the isotopically exchanged N18O. To separate these two gases, it is necessary to employ a series of equations, shown below. 21414/4646 fromNOIAI 1814/3030 14 fromNO I AI 1816 21414 1414 totalfromNO fromNOfromNOIIII 181814 NO N OfromNOCPI 1618 2 23232/14 14()Ototal OfromNOCPIAI C = concentration I = intensity P = proportionality constant B= background constant A= fragmentation ratio
155 APPENDIX B TEMPERATURE PROGRAMMED DESORP TION DATA FOR RELEVANT GAS COMBINATIONS 0 5 10 15 0100200300400500600700Concentration (ppm)Temperature (oC)NO NO212+3a 200250300350400450500550600Arbitrary UnitsTemperature (oC) 2 3b Figure B-1. Temperature Programmed Desorption of 1% NO on La2CuO4 at 30 oC per minute (a.) and 5 oC per minute (b.). Table B-1. Temperature Programmed De sorption Characteristics of 1% NO on La2CuO4 for the 30 oC per minute desorption Species Peak Temperature (oK) Desorption Energy (kJ/mol) -moles Desorbed -moles Desorbed Per m2 -moles Desorbed Per g 1 423 73.78 1.87 E-02 0.125 0.434 2 3 559 650 110.23 127.56 6.22 E-02 0.415 1.45 NO Total 8.22 E-02 0.548 1.91
156 0 10 20 30 40 50 60 0100200300400500600700800Concentration (ppm)Temperature (oC) NO2NO O2 N2O Figure B-2. Temperature Programmed Desorption of 500 ppm NO2 on La2CuO4. Table B-2. Temperature Programmed Desorption Characteristics of 1% NO2 on La2CuO4 Species Peak Temperature (oK) Desorption Energy (kJ/mol) -moles Desorbed -moles Desorbed Per m2 -moles Desorbed Per g NO 1 2 Total 586 672 115.2 131.575 1.12 E-01 8.79 E-02 2.21 E-01 0.747 0.586 1.47 2.60 2.04 5.14 NO2 1/Total 586 115.2 1.10 E-01 0.733 2.56 N2O 1 2 586 741 115.2 152.79 N/A N/A N/A O2 1 2 Total 586 672 115.2 131.575 1.71 E-02 4.53 E-02 6.90 E-02 0.114 0.302 0.460 0.397 1.05 1.60
157 0 10 20 30 40 50 60 0100200300400500600700800Concentration (ppm)Temperature (oC) NO CO2/ N2O NO2CO O2 Figure B-3. Temperature Programmed Deso rption of 0.67% NO and 0.33% CO on La2CuO4. Table B-3. Temperature Programmed Desorption Characteristics of 0.67% NO and 0.33% CO on La2CuO4 Species Peak Temperature (oK) Desorption Energy (kJ/mol) -moles Desorbed -moles Desorbed Per m2 -moles Desorbed Per g NO 1 2 3 4 Total 436 560 795 859 104.74 135.7 146.8 159.2 3.48 E-02 6.06 E-02 6.60 E-01 7.56 E-01 0.232 0.404 4.40 5.04 0.809 1.41 15.3 17.6 CO 1 Total 731 134.5 2.94 E-02 1.23 E-01 0.196 0.820 0.684 2.86 CO2 / N2O 1 / Total ---4.41 E-01 2.94 10.3
158 0 5 10 15 20 25 0100200300400500600700800Concentration (ppm)Temperature (oC) CO2/ N2O NO NO2O2CO Figure B-4. Temperature Programmed Desorp tion of 500 ppm NO, 500 ppm CO, and 1% on La2CuO4.
159 Table B-4. Temperature Programmed Desorptio n Characteristics of 500 ppm NO, 500 ppm CO, and 1% O2 on La2CuO4. Species Peak Temperature (oK) Desorption Energy (kJ/mol) -moles Desorbed -moles Desorbed Per m2 -moles Desorbed Per g NO 1 2 3 4 Total 466 575 763 859 84.0 104.6 140.6 159.2 9.48 E-02 6.33 E-02 4.91 E-02 8.70 E-02 2.94 E-01 0.632 0.422 0.327 0.580 1.96 2.20 1.47 1.14 2.02 6.84 CO2 1 2 3 4 Total 450 576 664 858 81.0 104.8 121.7 159.0 7.27 E-03 7.94 E-02 1.02 E-01 3.43 E-02 2.23 E-01 0.048 0.529 0.680 0.229 1.49 0.169 1.85 2.37 0.798 5.19 CO 1 2 3 Total 452 578 669 81.4 105.2 122.6 5.72 E-03 5.32 E-03 1.65 E-02 3.23 E-02 0.0381 0.0354 0.110 0.215 0.133 0.124 0.384 0.751
160 0 10 20 30 40 50 60 70 100200300400500600700Concentration (ppm)Temperature (oC) NO2O2 NO CO2/ N2O CO Figure B-5. Temperature Progr ammed Desorption of 500 ppm NO2, 250 ppm CO and 1000 ppm O2 on La2CuO4.
161 Table B-5. Temperature Programmed De sorption Characteristics of 500 ppm NO2, 250 ppm CO and 1000 ppm O2 on La2CuO4. Species Peak Temperature (oK) Desorption Energy (kJ/mol) -moles Desorbed -moles Desorbed Per m2 -moles Desorbed Per g NO 1 2 3 Total 580 685 871 105.6 125.7 161.5 1.25 E-01 9.17 E-02 1.61 E-01 3.77 E-01 0.833 0.611 1.07 2.51 2.91 2.13 3.74 8.77 NO2 1 564 102.6 1.84 E-01 1.23 4.28 CO 1 2 Total 474 562 85.5 102.2 6.88 E-03 2.02 E-02 4.13 E-02 0.0458 0.135 0.275 0.160 0.469 0.960 CO2 1 2 3 4 Total 481 578 743 877 86.8 105.2 136.8 162.7 1.50 E-02 1.29 E-01 5.18 E-02 1.03 E-01 2.92 E-01 0.100 0.860 0.345 0.687 1.95 0.349 3.00 1.20 2.40 6.79 O2 1 2 Total 567 685 103.1 125.7 2.81 E-02 2.29 E-01 5.00 E-01 0.187 1.53 3.33 0.653 5.33 11.6
162 0 5 10 15 20 25 30 35 0100200300400500600700800Concentration (ppm)Temperature (oC)CO2NO Figure B-6. Temperature Programmed Desorptio n of 500 ppm NO and 1000 ppm CO2 on La2CuO4. Table B-6. Temperature Programmed Desorp tion Characteristics of 500 ppm NO and 1000 ppm CO2 on La2CuO4. Species Peak Temperature (oK) Desorption Energy (kJ/mol) -moles Desorbed -moles Desorbed Per m2 -moles Desorbed Per g NO 1/Total 428 76.9 2.35 E-02 0.157 0.547 CO2 1/Total 621 113.4 3.25 E-01 2.17 7.56
163 APPENDIX C SILICA CONTAMINATION OF TH E SENSOR ELECTRODE INTERFACE The development of a long term model for behavi or of a potentiometric sensor requires an understanding of the modification of the electrode and electrolyte surfaces, as well as accounting for changes at the interface betw een electrode and electrolyte. Many factors contribute to changes over time in the output of a potentiometric sensor.319 Most of these factors can be mitig ated by altering processing conditions320 and operating environment.321 One variable that proves very difficult to remove is the presence of silicon oxide in Yttrium Stabilized Zirconia (YSZ). This oxide rests in the grain boundaries and di ffuses to the surface of the YSZ, forming an oxygen diffusion barrier of its own accord and by reaction with material from the electrode may form silicates.322-324 Additionally, siloxane decomposition may occur due to heating of the vacuum grease used to seal the reac tor, as seen in the literature.325 The aim of this study is to examine the structure of the contamination of si lica on the electrode, elec trolyte, and interfaces of a screen-printed La2CuO4/YSZ/Pt potentiometric sensor that occurs over time. X-ray photoelectron spectroscopy (XPS-Perkin -Elmer PHI 5100 ESCA System) was used to examine the composition of the electrode, bare electrolyte, and exposed electrolyte surfaces. Scanning electron microscopy (SEM-JEOL JSM-6335F ) was utilized to examine the surface and interface regions to detect the presence a nd structure of any amorphous silicon-based contaminants. A quartz tube reaction chamber was used to cycle the sample thermally and environmentally. The paste for screen-printing sensor electrodes was prepared by ball-milling the La2CuO4 powders with poly ethylene glycol (Avocado, PEG 400) and excess ethyl alcohol for 2 hours. After evaporating the ethyl alcohol, the paste was screen-printed onto commercial dense YSZ
164 substrates (Marketech ) of dimension (20 mm 0.1 mm 100 mm). The devi ce was sintered at 800C for 10 hours. The sensor was then loaded into a glass te sting chamber for exposure testing. Gas mixing was carried out using MKS mass fl ow controllers. The total fl ow was kept constant at 300 standard cubic centimeters with background gas containing 3% O2 and a balance of N2. The concentration of either NO or NO2 was stepped up and down usi ng a cycle of 0 ppm, 50 ppm, 100 ppm, 200 ppm, 400 pp m and 650 ppm steps, at temperatures of 450, 500, 550, and 600 oC. SEM and XPS of the sensing electrode s howed no deposition of silica on the La2CuO4 surface. Silicon compounds appeared on the YSZ surface, however but their presence was lessened by the electrode layer a bove. The presence of these compounds was expected, and so it was concluded that the aging of the sample does not produce silicon co mpounds that interfere with sensor operation over the us eful testing lifetime of sensor s used in the research group. Figure C-1. La2CuO4 electrode on 8% YSZ substrate used for heat and gas cycling tests. 8%-Yittrium Stabilized Zirconia Screen-printed La2CuO4 electrode
165 Figure C-2. XPS of La2CuO4 electrode surface following 3 weeks of cycling under sensor testing conditions. After 3 weeks in a thermally cycled environment and exposure to NO, NO2, CO, and CO2 gas mixtures, the structur e and composition of the La2CuO4 electrode remains constant according to XPS, indicating no silica deposition from siloxane occurs in the short term. 0 2 1044 1046 1048 1041 1051.2 1050 200 400 600 800 1000 BE (ev) Fresh Sample Aged Sample La 3d La MNV O KVV Si 2p3Not Present O 1s La 4d Binding Energy (eV)
166 Figure C-3. XPS spectra of YSZ electrolyte on th e just-sintered sample. Some trace peaks of the La2CuO4 electrode can be seen due to slu rry wetting during the screen printing process. Figure C-4. XPS spectra of the YS Z electrolyte after thermal cycli ng in sensor testing conditions. The peak for the Si 2p3 orbital was seen to increase due to diffusion of Si to the grain boundaries of the YSZ. Binding Energy (eV) N(E)Min: 3547Max: 213417 10009008007006005004003002001000 Cu 2p3 0.3 % Na KVV Y 3p1 Y 3p3 4.1 % Zr 3p1 Zr 3p3 Zr 3d 6.2 % Si 2s Y 3d La 3d3 0.8 % La 3d5 Si 2p3 17.6 % O 1s 61.4 % O KVV C 1s 9.6 % Binding Energy (eV) N(E)Min: 3790Max: 241973 10009008007006005004003002001000 Si 2p3 20.9 % Si 2s Y 3d Zr 3d 4.2 % Na KVV C 1s 4.6 % Y 3p 2.0 % Zr 3p3 Zr 3p1 O 1s 66.7 O KVV La 3d5 1.2 % La 3d3 Cu 2p3 0.4 %
167 Figure C-5. XPS spectra of th e YSZ electrolyte below the La2CuO4 electrode after thermal cycling. The Si 2p3 peak was observed to increase over the uncycled sample, but remained lower than the uncovered and cycled YSZ sample. Figure C-6. SEM photograph of the electrode/e lectrolyte interface after sintering at 800 oC for 10 hours. The La2CuO4/YSZ interface was well-adhe red, and showed no sign of auxiliary phases. Binding Energy (eV) N(E)Min: 2573Max: 169953 10009008007006005004003002001000 Cu 2p3 0.5 % Si 2p3 18.9 % Si 2s Y 3d Zr 3d 3.4 % Na KVV C 1s 15.9 % Y 3p1 Y 3p3 2.6 % Zr 3p3 Zr 3p1 La 3d3 La 3d5 0.1 % O KVV O 1s 58.7
168 Figure C-7. The surface of the La2CuO4 electrode shows no visible deposits of silica products after 3 weeks of thermal cycling under sensor testing conditions.
169 APPENDIX D ADSORPTION XPS OF NO/NO2/CO ON LA2CUO4 POWDER Figure D-1. C1s XPS spectra for unadsor bed (blue) and NO adsorbed (green) La2CuO4 powder. The counts from the carbon peak are somewhat reduced by the presence of NO, with the peak near 285 flattening out, and near 280 keepin g the similar shape but lesser intensity. This effect is likely due to the prevailing adsorption of NO to the surfaces in the absence of gas phase CO.
170 Figure D-2. La3d5 XPS spectra of unadsorbe d (blue) and NO adsorbed (green) La2CuO4 powder. NO heavily damps the peaks at 857, 842, and 837. Figure D-3. C1s XPS spectra of unadsorbed (blue) and NO2 adsorbed (red) La2CuO4 powder.
171 For the case of NO2 adsorption, the C 1s spectrum does not vary too much from the unadsorbed spectrum. The intensity is damped down, but there appears to be no shifting of peaks, and no appreciable splitting. Figure D-4. La3d5 XPS spectra of unadsorbed (blue) and NO2 adsorbed (red) La2CuO4 powder. The qualities of the La peaks seem to be almost identical for the unadsorbed and NO2 adsorbed cases, but there seems to be a very minor lessening of intensity in the adsorbed case.
172 Figure D-5. O1s XPS spectra of unadsorbed (blue) and NO+CO adsorbed (purple) La2CuO4 powder. For NO+CO, we see a very similar pattern to NO; the peak at ~330 remains unshifted, and there is a stronger binding energy peak sitting at 535 eV. Notable difference comes as a result of the CO interference on the surface. The more st rongly bound peak is larger than the original peak, since CO causes NO to bond more thoroughl y to the surface. The secondary weak peak near 522 drops in intensity alsowhether this is a consequence of CO stealing the oxygen to form a CO2 molecule or CO3 ion is not certain.
173 Figure D-6. C1s XPS spectra of unadsorbed (blue) and NO+CO adsorbed (purple) La2CuO4 powder. Here, the two peaks assigned to carbon are lesser in intensity, but are now equally intense, likely as a result of the CO interference.
174 Figure D-7. La3d5 XPS spectra of unadsorbed (blue) and NO+CO adsorbed (purple) La2CuO4 powder. In the La spectra, the peaks and intensities match upthough the adsorbed sample has much lower counts.
175 Figure D-8. O1s XPS spectra of unadsorbed (blue) and NO2+NO adsorbed (dark blue) La2CuO4 powder. Here peak spitting was present for both NO and NO2. The peak shape was much more characteristic of NO than for NO2. The larger peak appears to be another doublet, split between the NO peak at 529 eV, and the NO2 shifted peak at 527. The intensity of the peak at 529 has been mostly preserved, indicating that the adsorb ed species on the surface are mostly composed of NO, implying that the presence of NO on the surface assists the dissociation of NO2, or at least the NO is more prone to bond. The former seems more likely, since TPD experiments show the NO2 has more gas adsorbed on the surface. The more strongly bound peak is slightly less intense than in pure NO adsorption, likely owing to the hybrid adsorption of NO vs. NO2. The reduction and oxidation oppose each other, which mi ght account for this reduction in intensity.
176 Figure D-9. C1s XPS spectra of unadsorbed (blue) and NO2+NO adsorbed (dark blue) La2CuO4 powder. The structures of these peaks were very similar to that of the NO2 adsorption case, but were slightly less intense overall. Keep in mi nd, however, that the NO-adsorbed peaks looked very similar to that of the unadsorbed sample (with lesser intensity over the whole area), so NO2 will be the only adsorbate seen in this spect ra. The more weakly bound peak near 280 eV is slightly shifted to lower bindi ng energy from the unadsorbed case, which differentiates it from the individual adsorptions of NO and NO2.
177 Figure D-10. La3d5 XPS spectra of unadsorbed (blue) and NO2+NO adsorbed (dark blue) La2CuO4 powder. The peaks from the mixed adsorption resembled the unadsorbed and NO2 cases, but the NO adsorption is shown by the damping down of intensity across the whole peak range.
178 Figure D-11. O1s XPS spectra of unadsorbe d (blue) and CO adsorbed (orange) La2CuO4 powder. Figure D-12. C1s XPS spectra of unadsorbe d (blue) and CO adsorbed (orange) La2CuO4 powder.
179 Figure D-13. La3d5 XPS spectra of unadsorbed (blue) and CO adsorbed (orange) La2CuO4 powder.
180 Figure D-14. Full range XPS comparison sp ectra of unadsorbed and adsorbed La2CuO4 powder samples. Figure D-15. O1s XPS comparison spect ra of unadsorbed and adsorbed La2CuO4 powder samples.
181 Figure D-16. C1s XPS comparison spect ra of unadsorbed and adsorbed La2CuO4 powder samples. Figure D-17. La3d5 XPS comparison spect ra of unadsorbed and adsorbed La2CuO4 powder samples.
182 APPENDIX E DECONVOLUTED MASS SPECTRA OF ISOTOPICALLY LABELED NOX/LA2CUO4 TPD EXPERIMENTS Table E-1: Desorption Energy of Peaks for 1% NO and 1% 16O2 Adsorption on La2Cu16O4 Species Temperature (oC) Desorption Energy (kJ/mol) Peak mol/m2 Adsorbed Total mol/m2 Adsorbed Peak Fit % Error NO 313 129.62 1.81 16.4 0.9385 386 146.10 4.40 453 161.51 9.15 550 183.90 1.03 16O2 367 141.65 0.629 4.18 1.165 445 159.69 3.55 16O18O 444 159.31 0.037 0.037 5.567 0 500 1000 1500 2000 2500 200300400500600700N18O Concentration (PPM/m2)Temperature (oC) Figure E-1. Desorption Peak of N16O after NO adsorption on La2CuO4 16 with Gaussian Analysis
183 0 200 400 600 800 1000 1200 200250300350400450500550600Concentration 16O2 (PPM/m2)Temperature (oC) Figure E-2. Desorption Peak of 16O2 after NO adsorption on La2Cu16O4 with Gaussian Analysis 0 2 4 6 8 10 12 14 100200300400500600700Concentration O16O18 (PPM/m2)Temperature (oC) Figure E-3. Desorption Peak of 18O16O after NO adsorption on La2Cu16O4 with Gaussian Analysis
184 Table E-2: Desorption Energy and Adso rption Capacity for 1% NO and 1% 18O2 Adsorption on La2Cu16O4 Partially Substituted with 18O Species Temperature (oC) Desorption Energy (kJ/mol) Peak mol/m2 Adsorbed Total mol/m2 Adsorbed Peak Fit % Error NO 309 128.45 0.496 3.49 1.334 419 153.57 1.74 505 173.51 1.25 16O2 423 154.60 2.39 3.43 3.178 516 176.09 1.04 16O18O 422 154.28 1.20 2.49 1.375 496 171.36 1.29 18O2 411 510 151.77 174.63 1.16 0.247 1.41 1.411 0 100 200 300 400 500 600 100200300400500600700Concentration N16O (PPM/m2)Temperature (oC) Figure E-4. Desorption Peak of N16O after NO adsorption on partially substituted La2Cu18O4 with Gaussian Analysis
185 0 100 200 300 400 500 600 700 100200300400500600700Concentration 16O2 (PPM/m2)Temperature (oC) Figure E-5. Desorption Peak of 16O2 after NO adsorption on partially substituted La2Cu18O4 with Gaussian Analysis 0 100 200 300 400 500 350400450500550600650Concentration O16O18 (PPM/m2)Temperature (oC) Figure E-6. Desorption Peak of 16O 18O after NO adsorption on partially substituted La2Cu18O4 with Gaussian Analysis
186 0 50 100 150 200 250 300 250300350400450500550600Concentration 18O2 (PPM/m2)Temperature (oC) Figure E-7. Desorption Peak of 18O2 after NO adsorption on partially substituted La2Cu18O4 with Gaussian Analysis Table E-3: Desorption Energy a nd Adsorption Capacity of La2Cu18O4 for Adsorption of 1% NO and 1% 16O2 Species Temperature (oC) Desorption Energy (kJ/mol) Peak mol/m2 Adsorbed Total mol/m2 Adsorbed Peak Fit % Error N16O 315 543 0.626 4.12 1.15 420 644 1.54 497 718 1.48 582 802 0.475 16O2 383 608 1.57 2.95 2.31 420 644 1.38 16O18O 433 656 1.17 3.18 2.87 513 734 2.01 18O2 427 651 1.14 2.94 2.31 486 708 1.80 N18O 442 665 0.255 0.717 9.86 507 728 0.462
187 0 100 200 300 400 500 600 200300400500600700Concentration N16O (PPM/m2)Temperature (oC) Figure E-8. Desorption Peak of N16O after NO adsorption on La2Cu18O4 with Gaussian Analysis 0 50 100 150 200 350400450500550600650Concentration N18O (PPM/m2)Temperature (oC) Figure E-9. Desorption Peak of N18O after NO adsorption on La2Cu18O4 with Gaussian Analysis
188 0 100 200 300 400 500 600 700 800 200300400500600700Concentration 16O2 (PPM)Temperature (oC) Figure E-10. Desorption Peak of 16O2 after NO adsorption on La2Cu18O4 with Gaussian Analysis 0 100 200 300 400 500 200300400500600700Concentration O16O18 (PPM/m2)Temperature (oC) Figure E-11. Desorption Peak of 16O 18O after NO adsorption on La2Cu18O4 with Gaussian Analysis
189 0 100 200 300 400 500 600 700 300350400450500550600Concentration 18O2 (PPM/m2)Temperature (oC) Figure E-12. Desorption Peak of 18O2 after NO adsorption on La2Cu18O4 with Gaussian Analysis Table E-4: Desorption Energy and Adsorption Capacity for 1% NO2 and 1% 18O2 Adsorption on La2Cu16O4 Species Temperature (oC) Desorption Energy (kJ/mol) Peak mol/m2 Adsorbed Total mol/m2 Adsorbed Peak Fit % Error N16O 370.9 142.59 0.217 5.03 1.530 431.7 156.58 3.45 485.4 168.98 1.36 16O2 426.1 155.29 4.16 4.89 0.6498 433.5 156.99 0.214 497.6 171.79 0.516 16O18O 427.0 155.51 0.102 0.114 3.928 512.8 175.32 0.0118 18O2 416.9 153.18 0.065 0.065 8.580 N16O2 333.9 134.10 1.46 14.54 0.8011 343.3 136.25 7.56 398.2 148.87 5.52
190 0 500 1000 1500 300350400450500550Concentration N16O (PPM/m2)Temperature (oC) Figure E-13. Desorption Peak of N16O after NO2 adsorption on La2Cu16O4 with Gaussian Analysis 0 500 1000 1500 2000 2500 3000 3500 4000 200250300350400450500550Concentration N16O2 (PPM/m2)Temperature (oC) Figure E-14. Desorption Peak of N16O2 after NO2 adsorption on La2Cu16O4 with Gaussian Analysis
191 0 200 400 600 800 1000 1200 1400 300350400450500550Concentration 16O2 (PPM)Temperature (oC) Figure E-15. Desorption Peak of 16O2 after NO2 adsorption on La2Cu16O4 with Gaussian Analysis 0 5 10 15 20 25 30 300350400450500550600Concentration O16O18 (PPM)Temperature (oC) Figure E-16. Desorption Peak of 16O 18O after NO2 adsorption on La2Cu16O4 with Gaussian Analysis
192 0 5 10 15 20 25 30 35 40 370380390400410420430440450Concetration 18O2 (PPM/m2)Temperature (oC) Figure E-17. Desorption Peak of 18O2 after NO2 adsorption on La2Cu16O4 with Gaussian Analysis Table E-5: Desorption En ergy of Peaks for 1% NO2 and 1% 18O2 Adsorption on La2Cu16 O4 Partially Substituted with 18O Species Temperature (oC) Desorption Energy (kJ/mol) Peak mol/m2 Adsorbed Total mol/m2 Adsorbed Peak Fit % Error N16O 443.9 159.40 5.18 5.85 2.531 542.5 182.20 0.665 16O16O 311.8 129.03 0.279 3.72 1.719 381.9 145.13 0.473 445.3 159.73 2.28 526.0 178.38 0.686 16O18O 452.2 161.32 1.41 1.80 1.930 529.6 179.21 0.390 18O18O 453.3 161.57 0.598 0.764 1.956 510.4 174.76 0.166 N18O 349.3 137.63 0.357 0.679 9.500 466.2 164.55 0.322 9.861 N16O16O 344.1 136.45 1.94 4.66 1.636 367.6 141.84 2.72 N16O18O 360.8 140.28 0.122 0.122 6.731
193 0 200 400 600 800 1000 1200 300350400450500550600Concentration N16O (PPM/m2)Temperature (oC) Figure E-18. Desorption Peak of N16O after NO2 adsorption on partially substituted La2Cu18O4 with Gaussian Analysis 0 500 1000 1500 2000 250300350400450Concentration N16O2 (PPM/m2)Temperature (oC) Figure E-19. Desorption Peak of N16O2 after NO2 adsorption on partially substituted La2Cu18O4 with Gaussian Analysis
194 0 20 40 60 80 100 250300350400450500Concentration NO16O18 (PPM/m2)Temperature (oC) Figure E-20. Desorption Peak of N16O18O after NO2 adsorption on partially substituted La2Cu18O4 with Gaussian Analysis 0 50 100 150 200 250 300350400450500550Concentration N18O (PPM/m2)Temperature (oC) Figure E-21. Desorption Peak of N18O after NO2 adsorption on partially substituted La2Cu18O4 with Gaussian Analysis
195 0 100 200 300 400 500 600 700 200300400500600700Concentration 16O2 (PPM/m2)Temperature (oC) Figure E-22. Desorption Peak of 16O2 after NO2 adsorption on partially substituted La2Cu18O4 with Gaussian Analysis 0 50 100 150 200 250 300 350 350400450500550600650700750Concentration O16O18 (PPM/m2)Temperature (oC) Figure E-23. Desorption Peak of 16O18O after NO2 adsorption on partially substituted La2Cu18O4 with Gaussian Analysis
196 0 50 100 150 200 350400450500550600650700Concentration 18O2 (PPM/m2)Temperature (oC) Figure E-24. Desorption Peak of 18O2 after NO2 adsorption on partially substituted La2Cu18O4 with Gaussian Analysis
197 Table E-11: Desorption Energy and Adsorption Capacity for 1% NO2 and 1% 16O2 Adsorption on La2Cu18O4 Species Temperature (oC) Desorption Energy (kJ/mol) Peak mol/m2 Adsorbed Total mol/m2 Adsorbed Peak Fit % Error N16O 434 658 5.28 7.66 2.77 549 769 2.39 16O2 392 617 1.04 2.88 2.18 441 664 1.84 16O18O 422 647 0.150 0.227 4.24 446 669 0.0770 18O2 425 649 0.0897 0.212 3.09 452 675 0.122 N18O 347 574 0.357 1.17 6.51 457 680 0.812 N16O2 340 567 2.47 7.48 1.27 362 589 3.81 404 629 1.20 N16O18O 350 577 0.268 0.650 6.81 382 608 0.381
198 0 200 400 600 800 1000 1200 300350400450500550600650Concentration N16O (PPM/m2)Temperature (oC) Figure E-25. Desorption Peak of N16O after NO2 adsorption on La2Cu18O4 with Gaussian Analysis 0 500 1000 1500 2000 2500 200250300350400450500550Concentration N16O2 (PPM/m2)Temperature (oC) Figure E-26. Desorption Peak of N16O2 after NO2 adsorption on La2Cu18O4 with Gaussian Analysis
199 0 50 100 150 200 250 320340360380400420440Concentration NO16O18 (PPM/m2)Temperature (oC) Figure E-27. Desorption Peak of N16O18O after NO2 adsorption on La2Cu18O4 with Gaussian Analysis 0 50 100 150 200 250 300 250300350400450500550Concentration N18O (PPM/m2)Temperature (oC) Figure E-28. Desorption Peak of N18O after NO2 adsorption on La2Cu18O4 with Gaussian Analysis
200 0 100 200 300 400 500 600 700 800 300350400450500550Concentration 16O2 (PPM/m2)Temperature (oC) Figure E-29. Desorption Peak of 16O2 after NO2 adsorption on La2Cu18O4 with Gaussian Analysis 0 20 40 60 80 100 120 360380400420440460480Concentration O16O18 (PPM/m2)Temperature (oC) Figure E-30. Desorption Peak of 16O18O after NO2 adsorption on La2Cu18O4 with Gaussian Analysis
201 0 20 40 60 80 100 380400420440460480Concentration 18O2 (PPM/m2)Temperature (oC) Figure E-31. Desorption Peak of 18O2 after NO2 adsorption on La2Cu18O4 with Gaussian Analysis
202 LIST OF REFERENCES 1. de Sa, R. A. B.; da Silva, R. M. C. F.; Dauzacker, E. R. Mathematical Model to Predict the Otto Cycle Vehicles Exhaust Emissions From Automotive Gasoline Consumption. In Emissions Measurement and Tes ting, Advanced Catalysts and Substrates, and Diesel Gaseous Emissions, Society of Automotive Engineers, Inc.: Warrendale, PA, 2004; pp 27-54. 2. H.Oshima, M.Tatemichi, and T.Sawa "Chemical Basis of Inflammation Induced Carcinogenesis," Archives of Biochemistry and Biophysics,  3-11 (2003). 3. J. N. Seinfeld, and S. N. Pandis, Atmospheric Chemistry and Physics From Air Pollution to Climate Change John Wiley and Sons, New York, 1997. 4. E. Gutt, V. Patton, and N. Spencer. Bu ilding on 30 Years of Clean Air Act Success: The Case for Reducing NOX Air Pollution. Retrieved January 5, 2007, from www.environmentaldefense.org 5. W. J. Fleming, "Physical Principles Governing Nonideal Beha vior of Zirconia Oxygen Sensor," Journal of the Electrochemical Society, 123  C237 (1976). 6. W. J. Fleming, "Physical Principles Governing Nonideal Beha vior of Zirconia Oxygen Sensor," Journal of the Electrochemical Society, 124  21-28 (1977). 7. B. M. White, E. Macam, F. M. Van A ssche, E. Traversa, and E. D. Wachsman, "A Theoretical Framework for Predicting So lid-State Potentiometric Gas Sensor Behavior" pp 179. in Proceedings of th e 210th Meeting of The Electrochemical Society, Vol. 3, Chemical Sensors 7 and MEMS/NEMS 7, ECS Transactions. Edited by G. Hunder, S. Akbar, S. Bhan sali, O. Brand, C. Bruckner-Lea, M. Carter, J. Davidson, E. Enikov, P. Hesket h, R. Hillman, C. Kranz, J. Li, R. Maboudian, R. Mukundan, C. Roper, S. Sho ji, A. Simonian, and M. Tabib-Azar. The Electrochemical Society, Pennington, NJ, 2006. 8. S. Akbar, P. Dutta, and C. H. Lee, "H igh-Temperature Ceramic Ga s Sensors: A Review," International Journal of Applied Ceramic Technology, 3  302-311 (2006). 9. A. D. Brailsford, M. Yussouff, and E. M. Logothetis, "Theory of Gas Sensors," Sensors and Actuators B-Chemical, 13 [1-3] 135-138 (1993). 10. A. D. Brailsford, M. Yussouff, a nd E. M. Logothetis, "Steady State Model of Electrochemical Gas Sensors With Multiple Reactions," Sensors and Actuators BChemical, 36 [1-3] 392-397 (1996). 11. A. D. Brailsford, M. Yussouff, and E. M. Logothetis, "Theory of Gas Sensors: Response of an Electrochemical Sensor to Multi-Component Gas Mixtures," Sensors and Actuators B-Chemical, 34 [1-3] 407-411 (1996).
203 12. A. D. Brailsford, M. Yussouff, E. M. Logothetis, T. Wang, and R. E. Soltis, "Experimental and Theoretical Study of the Response of ZrO2 Oxygen Sensors to Simple One-Reducing-Gas Mixtures," Sensors and Actuators B-Chemical, 42  15-26 (1997). 13. E. L. Brosha, R. Mukundan, R. Lujan, and F. H. Garzon, "Mixed Potential NOX Sensors Using Thin Film Electrodes and Electrolyt es for Stationary Reciprocating Engine Type Applications," Sensors and Actuators B-Chemical, 119  398-408 (2006). 14. L. Chevallier, M. L. Grilli, E. Di Bartolomeo, and E. Traversa, "Non-Nernstian Planar Sensors Based on YSZ With Ta (10 at.%)Doped Nanosized Titania As a Sensing Electrode for High-Temper ature Applications," International Journal of Applied Ceramic Technology, 3  393-400 (2006). 15. L. Chevallier, E. Di Bartolomeo, M. L. Grilli, M. Mainas, B. M. White, E. D. Wachsman, and E. Traversa, "Non-Nerstian Planar Sensors Based on YSZ With an Nb2O5 Electrode: Discussion on Sensing Mechan ism" pp 163. in Proceedings of the 208th Meeting of The Electroch emical Society, Vol. 1, Solid State Ionic Devices IV, ECS Transactions. Edited by E. D. Wachsman, V. Birss, F. H. Garzon, R. Mukundan, and E. Traversa. The Electroch emical Society, Pennington, NJ, 2006. 16. E. Comini, G. Faglia, and G. Sberveglie ri, "UV Light Activation of Tin Oxide Thin Films for NO2 Sensing at Low Temperatures," Sensors and Actuators B-Chemical, 78 [1-3] 73-77 (2001). 17. E. Comini, G. Faglia, G. Sberveglieri, D. Calestani, L. Zanotti, and M. Zha, "Tin Oxide Nanobelts Electrical and Sensing Properties," Sensors and Actuators B-Chemical, 111 2-6 (2005). 18. E. Di Bartolomeo, E. Traversa, M. Baroncini, V. Kotzeva, and R. V. Kumar, "Solid State Ceramic Gas Sensors Based on In terfacing Ionic Conductors With Semiconducting Oxides," Journal of the European Ceramic Society, 20  2691-2699 (2000). 19. E. Di Bartolomeo, M. L. Grilli, and E. Traversa, "Sensing Mechanism of Potentiometric Gas Sensors Based on Stabilized Zirconia With Oxide Electrodes: Is It Always Mixed Potential?" pp 222. in Proc eedings of the 202nd Meeting of The Electrochemical Society, Vol. 2002-26, Solid State Ionic Devices III. Edited by E. D. Wachsman, K. Swider-Lyons, M. F. Ca rolan, F. H. Garzon, M. Liu, and J. R. Stetter. The Electrochemical Society, Pennington, NJ, 2003. 20. E. Di Bartolomeo, M. L. Grilli, J. W. Yoon, and E. Traversa, "Zirconia-Based Electrochemical NOX Sensors With Semiconducting Oxide Electrodes," Journal of the American Ceramic Society, 87  1883-1889 (2004). 21. E. Di Bartolomeo, M. L. Grilli, and E. Traversa, "Sensing Mechanism of Potentiometric Gas Sensors Based on Stabilized Zirconia W ith Oxide Electrodes Is It Always
204 Mixed Potential?," Journal of the Electrochemical Society, 151  H133-H139 (2004). 22. E. Di Bartolomeo, N. Kaabbuathong, M. L. Grilli, and E. Traversa, "Planar Electrochemical Sensors Based on Tape-C ast YSZ Layers and Oxide Electrodes," Solid State Ionics, 171 [3-4] 173-181 (2004). 23. E. Di Bartolomeo, N. Kaabbuathong, A. D' Epifanio, M. L. Grilli, E. Traversa, H. Aono, and Y. Sadaoka, "Nano-Structured Per ovskite Oxide Electrodes for Planar Electrochemical Sensors Using Tape Casted YSZ Layers," Journal of the European Ceramic Society, 24  1187-1190 (2004). 24. E. Di Bartolomeo, M. L. Grilli, N. An tonias, S. Cordiner, and E. Traversa, "Testing Planar Gas Sensors Based on Yttria-Stabili zed Zirconia With Oxide Electrodes in the Exhaust Gases of a Spark Ignition Engine," Sensor Letters, 3  22-26 (2005). 25. A. Dutta, N. Kaabbuathong, M. L. Grilli, E. Di Bartolomeo, and E. Traversa, "Study of YSZ-Based Electrochemical Sensors With WO3 Electrodes in NO2 and CO Environments," Journal of the Electrochemical Society, 150  H33-H37 (2003). 26. A. Dutta, D. Das, M. L. Grilli, E. Di Bartolomeo, E. Traversa, and D. Chakravorty, "Preparation of Sol-Gel Nano-Compos ites Containing Copper Oxide and Their Gas Sensing Properties," Journal of Sol-Gel Science and Technology, 26 [1-3] 1085-1089 (2003). 27. A. Dutta and T. Ishihara, "Amper ometric NOX Sensor Based on Oxygen Pumping Current by Using LaGaO3-Based Solid El ectrolyte for Monitoring Exhaust Gas," Sensors and Actuators B-Chemical, 108 [1-2] 309-313 (2005). 28. P. Elumalai, J. Wang, S. Zhuiykov, D. Terada, M. Hasei, and N. Miura, "Sensing Characteristics of YSZ-Based Mixed-Potential-Type Planar NOX Sensors Using NiO Sensing Electrodes Sintered at Different Temperatures," Journal of the Electrochemical Society, 152  H95-H101 (2005). 29. P. Elumalai and N. Miura, "Performances of Planar NO2 Sensor Using Stabilized Zirconia and NiO Sensing Elect rode at High Temperature," Solid State Ionics, 176 [31-34] 2517-2522 (2005). 30. F. H. Garzon, R. Mukundan, R. Lujan, and E. L. Brosha, "Solid State Ionic Devices for Combustion Gas Sensing," Solid State Ionics, 175 [1-4] 487-490 (2004). 31. W. Gopel, G. Reinhardt, and M. Rosc h, "Trends in the Development of Solid State Amperometric and Potentiometric High Temperature Sensors," Solid State Ionics, 136 519-531 (2000).
205 32. M. L. Grilli, E. Di Bartolomeo, and E. Traversa, "Electrochemical NOX Sensors Based on Interfacing Nanosized LaFeO3 Perovskite-Type Oxide and Ionic Conductors," Journal of the Elect rochemical Society, 148  H98-H102 (2001). 33. M. L. Grilli, N. Kaabbuathong, A. Dutta, E. Di Bartolomeo, and E. Traversa, "Electrochemical NO2 Sensors With WO3 Electrodes for High Temperature Applications," Journal of the Ceramic Society of Japan, 110  159-162 (2002). 34. M. L. Grilli, E. Di Bartolomeo, and E. Traversa, "NOX Sensors Based on Interfacing Yttria Stabilized Zirconia With P and N-Type Semiconducting Oxides," Euro Ceramics Vii, Pt 1-3, 206-2 1243-1246 (2002). 35. M. L. Grilli, E. Di Bartolomeo, A. Lunard i, L. Chevallier, S. Cordiner, and E. Traversa, "Planar Non-Nernstian Electrochemical Sens ors: Field Test in the Exhaust of a Spark Ignition Engine," Sensors and Actuators B-Chemical, 108 [1-2] 319-325 (2005). 36. H. Kaneko, H. Taimatsu, Y. Miyoshi, K. Kawanaka, and T. Kusano, "YSZ/Ag Potentiometric Sensor for Reducing Gas-Detection," Sensors and Actuators BChemical, 13 [1-3] 151-154 (1993). 37. G. Y. Lu, N. Miura, and N. Yamazo e, "High-Temperature Hydrogen Sensor Based on Stabilized Zirconia and a Metal Oxide Electrode," Sensors and Actuators BChemical, 35 [1-3] 130-135 (1996). 38. G. Y. Lu, N. Miura, and N. Yamazoe, "Mixed Potential Hydrogen Sensor Combining Oxide Ion Conductor With Oxide Electrode," Journal of the Electrochemical Society, 143  L154-L155 (1996). 39. C. M. Mari, "Non-Nernstian Solid St ate Gas Sensors: Operating Principles and Materials," Ionics, 9 [5-6] 365-369 (2003). 40. N. Miura, S. Yao, Y. Shimizu, and N. Yamazoe, "Development of High-Performance Solid-Electrolyte Sensors for NO and NO2," Sensors and Actuators B-Chemical, 13 [1-3] 387-390 (1993). 41. N. Miura, H. Kurosawa, M. Hasei, G. Y. Lu, and N. Yamazoe, "Stabilized ZirconiaBased Sensor Using Oxide Electrode for Detection of NOX in High-Temperature Combustion-Exhausts," Solid State Ionics, 86-8 1069-1073 (1996). 42. N. Miura, G. Y. Lu, N. Yamazoe, H. Kurosawa, and M. Hasei, "Mixed Potential Type NOX Sensor Based on Stabilized Zirconia and Oxide Electrode," Journal of the Electrochemical Society, 143  L33-L35 (1996). 43. N. Miura, G. Y. Lu, and N. Yamazoe, "High-Temperature Potentiometric/Amperometric NOX Sensors Combining Stabilized Zirconia With Mixed-Metal Oxide Electrode," Sensors and Actuators B-Chemical, 52 [1-2] 169-178 (1998).
206 44. N. Miura, M. Ono, K. Shimanoe, and N. Yamazoe, "A Compact Solid-State Amperometric Sensor for Detection of NO2 in Ppb Range," Sensors and Actuators B-Chemical, 49 [1-2] 101-109 (1998). 45. N. Miura, G. Y. Lu, and N. Yamazoe, "Progress in Mixed-Poten tial Type Devices Based on Solid Electrolyte for Sensing Redox Gases," Solid State Ionics, 136 533-542 (2000). 46. N. Miura, S. Zhuiykov, T. Ono, M. Hasei, and N. Yamazoe, "Mixed Potential Type Sensor Using Stabilized Zirconia and ZnFe2O4 Sensing Electrode for NOX Detection at High Temperature," Sensors and Actuators B-Chemical, 83 [1-3] 222-229 (2002). 47. N. Miura, J. Wang, M. Nakatou, P. Elumalai, S. Zhuiykov, and M. Hasei, "HighTemperature Operating Characteristi cs of Mixed-Potential-Type NO2 Sensor Based on Stabilized-Zirconia Tube and NiO Sensing Electrode," Sensors and Actuators B-Chemical, 114  903-909 (2006). 48. J. Mizusaki, K. Yashiro, T. Kawada, a nd H. Ohama, "A Concept of Chemical Potential Pumping Effect With the NOx Sensing Mechanism of Perovskite-Type Oxide Mixed Conductors" pp 236. in Procee dings of the 202nd Meeting of The Electrochemical Society, Vol. 2002-26, Solid State Ionic Devices III. Edited by E. D. Wachsman, K. Swider-Lyons, M. F. Ca rolan, F. H. Garzon, M. Liu, and J. R. Stetter. The Electrochemical Society, Pennington, NJ, 2003. 49. R. Moos, "A Brief Overview on Automotive Exhaust Gas Sensors Based on Electroceramics," International Journal of Applied Ceramic Technology, 2  401-413 (2005). 50. R. Mukundan, E. L. Brosha, D. R. Brown, and F. H. Garzon, "A Mixed-Potential Sensor Based on a Ce0.8Gd0.2O1.9 Electrolyte and Platinum and Gold Electrodes," Journal of the Electrochemical Society, 147  1583-1588 (2000). 51. R. Mukundan, K. Teranishi, E. L. Brosha, and F. H. Garzon, "Nitrogen Oxide Sensors Based on Yttria-Stabilized Zirconia Electrolyte and Oxide Electrodes," Electrochemical and So lid State Letters, 10  J26-J29 (2007). 52. M. Ono, K. Shimanoe, N. Miura, and N. Yamazoe, "Amperometric Sensor Based on NASICON and NO Oxidation Catalysts for Detection of Total NOx in Atmospheric Environment," Solid State Ionics, 136 583-588 (2000). 53. M. Ono, K. Shimanoe, N. Miura, an d N. Yamazoe, "Reaction Analysis on Sensing Electrode of Amperometric NO2 Sensor Based on Sodium Ion Conductor by Using Chronopotentiometry," Sensors and Actuators B-Chemical, 77 [1-2] 78-83 (2001).
207 54. T. Ono, M. Hasei, A. Kunimoto, and N. Miura, "Improvement of Sensing Performances of Zirconia-Based Total NOX Sensor by Attachment of Oxidation-Catalyst Electrode," Solid State Ionics, 175 [1-4] 503-506 (2004). 55. C. O. Park, S. A. Akbar, and W. We ppner, "Ceramic Electrolytes and Electrochemical Sensors," Journal of Materials Science, 38  4639-4660 (2003). 56. C. O. Park and N. Miura, "Absolute Pote ntial Analysis of the Mi xed Potential Occurring at the Oxide/YSZ Electrode at High Temperature in NOX-Containing Air," Sensors and Actuators B-Chemical, 113  316-319 (2006). 57. Park, J. Solid-State Electrochemical Properties of Oxygen-Ion Conducting Ceramic Materials And Their Applications. Ph.D. Dissertation, University of Florida, 2004. 58. P. Schmidt-Zhang, K. P. Sandow, F. A dolf, W. Gopel, and U. Guth, "A Novel Thick Film Sensor for Simultaneous O2 and NO Monitoring in Exhaust Gases," Sensors and Actuators B-Chemical, 70 [1-3] 25-29 (2000). 59. Y. Shimizu and K. Maeda, "Solid Electrolyte-Based NOX Sensor Using Auxiliary Phase of Metal Oxide," Chemistry Letters,  117-118 (1996). 60. Y. Shimizu and K. Maeda, "Solid Electrolyte NOX Sensor Using Pyrochlore-Type Oxide Electrode," Sensors and Actuators B-Chemical, 52 [1-2] 84-89 (1998). 61. Y. Shimizu, H. Nishi, H. Suzuki, and K. Maeda, "Solid-State NOx Sensor Combined With NASICON and Pb-Ru-Based Pyrochlore-Type Oxide Electrode," Sensors and Actuators B-Chemical, 65 [1-3] 141-143 (2000). 62. Y. Shimizu, N. Nakashima, T. Hyodo, and M. Egashira, "NOx Sensing Properties of Varistor-Type Gas Sensors Consis ting of Micro P-n Junctions," Journal of Electroceramics, 6  209-217 (2001). 63. E. D. Wachsman and P. Ja yaweera, "Selective Detection of NOX by Differential Electrode Equilibria" pp 298-304. in Proceedings of the 202nd Meeting of The Electrochemical Society, Vol. 2000-32, Solid State Ionic Devices II Ceramic Sensors. Edited by E. D. Wachsman, W. We ppner, E. Traversa, M. Liu, P. Vanysek, and N. Yamazoe. The Electroc hemical Society, Pe nnington, NJ, 2001. 64. E. D. Wachsman, "Selective Potentiometric Detection of NOX by Differential Electrode Equilibria" pp 215. in Proceedings of the 202nd Meeting of The Electrochemical Society, Vol. 2002-26, Solid State Ionic Devices III. Edited by E. D. Wachsman, K. Swider-Lyons, M. F. Carolan, F. H. Ga rzon, M. Liu, and J. R. Stetter. The Electrochemical Society, Pennington, NJ, 2003. 65. D. L. West, F. C. Montgomery, and T. R. Armstrong, ""NO-Selective" NOX Sensing Elements for Combustion Exhausts," Sensors and Actuators B-Chemical, 111 8490 (2005).
208 66. J. Yoo, S. Chatterjee, F. M. Van Assc he, and E. D. Wachsman, "Influence of Adsorption and Catalytic Reaction on Sensing Properties of a Potentiometric La2CuO4/YSZ/Pt Sensor," Journal of the Electrochemical Society, 154  J190J195 (2007). 67. J. Yoo, H. Yoon, and E. D. Wachsman, "Sensing Properties of MOX/YSZ/Pt (MOX = Cr2O3, SnO2, CeO2) Potentiometric Sensor for NO2 Detection," Journal of the Electrochemical Society, 153  H217-H221 (2006). 68. J. Yoo, F. M. Van Assche, and E. D. Wachsman, "Temperature-Programmed Reaction and Desorption of the Sensor Elements of a WO3/YSZ/Pt Potentiometric Sensor," Journal of the Elect rochemical Society, 153  H115-H121 (2006). 69. J. Yoo, S. Chatterjee, F. M. Van Assc he, and E. D. Wachsman, "Influence of Adsorption and Catalytic Reaction on Sensing Properties of a Potentiometric La2CuO4/YSZ/Pt Sensor," Journal of the Electrochemical Society (accepted), (2007). 70. J. Yoo and E. D. Wachsman, "NO2/NO Response of Cr2O3and SnO2-Based Potentiometric Sensors and TemperatureProgrammed Reaction Evaluation of the Sensor Elements," Sensors and Actuators B: Chemical, In Press, Corrected Proof (2007). 71. J. W. Yoon, M. L. Grilli, E. Di Ba rtolomeo, R. Polini, and E. Traversa, "The NO2 Response of Solid Electrolyte Sens ors Made Using Nano-Sized LaFeO3 Electrodes," Sensors and Actuators B-Chemical, 76 [1-3] 483-488 (2001). 72. S. Benard, L. Retailleau, F. Gaillard P. Vernoux, and A. Giroir-Fendler, "Supported Platinum Catalysts for Nitrogen Oxide Sensors," Applied Catalysis B: Environmental, 55  11-21 (2005). 73. N. Imanaka, A. Oda, S. Tamura, and G. y. Adachi, "Total Nitrogen Oxides Gas Sensor Based on Solid Electrolytes With Refract ory Oxide-Based Auxiliary Electrode," Journal of The Electrochemical Society, 151  H113-H116 (2004). 74. J. Tamaki, A. Hayashi, Y. Yamamoto, and M. Matsuoka, "Detection of Dilute Nitrogen Dioxide and Thickness Effect of Tu ngsten Oxide Thin Film Sensors," Sensors and Actuators B: Chemical, 95 [1-3] 111-115 (2003). 75. W. Xiong and G. M. Kale, "Novel High-Selectivity NO2 Sensor Incorporating MixedOxide Electrode," Sensors and Actuators B: Chemical, 114  101-108 (2006). 76. S. Yao and J. R. Stetter, "Modifi cation of NASICON Solid Electrolyte for NOx Measurements," Journal of The Electrochemical Society, 151  H75-H80 (2004).
209 77. S. Zhuiykov, T. Nakano, A. Kunimoto, N. Yamazoe, and N. Miura, "Potentiometric NOx Sensor Based on Stabiliz ed Zirconia and NiCr2O4 Sensing Electrode Operating at High Temperatures," Electrochemistry Communications, 3  97-101 (2001). 78. S. Zhuiykov, "Mathematical Modelling of YSZ-Based Potentiometric Gas Sensors With Oxide Sensing Electrodes: Part I: Model of Interactions of Measuring Gas With Sensor," Sensors and Actuators B: Chemical, 119  456-465 (2006). 79. S. Zhuiykov and N. Miura, "Develop ment of Zirconia-Based Potentiometric NOx Sensors for Automotive and Energy Industries in the Early 21st Century: What Are the Prospects for Sensors?," Sensors and Actuators B: Chemical, 121  639-651 (2007). 80. E. L. Brosha, R. Mukundan, D. R. Brown, F. H. Garzon, J. H. Visser, M. Zanini, Z. Zhou, and E. M. Logothetis, "CO/HC Sensors Based on Thin Films of LaCoO3 and La0.8Sr0.2CoO3-[Delta] Metal Oxides," Sensors and Actuators B: Chemical, 69 [1-2] 171-182 (2000). 81. E. L. Brosha, R. Mukundan, D. R. Brown, and F. H. Garzon, "Mixed Potential Sensors Using Lanthanum Manganate and Terbium Yttrium Zirconium Oxide Electrodes," Sensors and Actuators B-Chemical, 87  47-57 (2002). 82. E. L. Brosha, R. Mukundan, D. R. Brown, F. H. Garzon, and J. H. Visser, "Development of Ceramic Mixed Potential Sensor s for Automotive Applications," Solid State Ionics, 148 [1-2] 61-69 (2002). 83. T. Hibino, Y. Kuwahara, S. Q. Wa ng, S. Kakimoto, and M. Sano, "Nonideal Electromotive Force of Zirconia Sensor s for Unsaturated Hydrocarbon Gases," Electrochemical and So lid State Letters, 1  197-199 (1998). 84. T. Hibino, S. Tanimoto, S. Kakimoto and M. Sano, "High-Temperature Hydrocarbon Sensors Based on a Stabilized Zirconia El ectrolyte and Metal Oxide Electrodes," Electrochemical and So lid State Letters, 2  651-653 (1999). 85. T. Hibino, S. Kakimoto, and M. Sano, "Non-Nernstian Behavior at Modified Au Electrodes for Hydrocarbon Gas Sensing," Journal of the Electrochemical Society, 146  3361-3366 (1999). 86. T. Hibino, A. Hashimoto, S. Kakimoto, and M. Sano, "Zirconia-Based Potentiometric Sensors Using Metal Oxide Electrodes for Detection of Hydrocarbons," Journal of the Electrochemical Society, 148  H1-H5 (2001). 87. R. Mukundan, E. L. Brosha, D. R. Brown, and F. H. Garzon, "Ceria-Electrolyte-Based Mixed Potential Sensors for the De tection of Hydrocarbons and Carbon Monoxide," Electrochemical and So lid State Letters, 2  412-414 (1999).
210 88. N. Miura, Y. T. Yan, G. Y. Lu, and N. Yamazoe, "Sensing Characteristics and Mechanism of Hydrogen Sulfide Sensor Using Stabilized Zirconia and Oxide Sensing Electrode," Sensors and Actuators B-Chemical, 34 [1-3] 367-372 (1996). 89. Y. Yan, N. Miura, and N. Yamazoe, "Construction and Working Mechanism of Sulfur Dioxide Sensor Utilizing Stabili zed Zirconia and Metal Sulfate," Journal of the Electrochemical Society, 143  609-613 (1996). 90. Y. T. Yan, Y. Shimizu, N. Miura, and N. Yamazoe, "Chara cteristics and Sensing Mechanism of SOX Sensor Using Stabilized Zirconia and Metal Sulfate," Sensors and Actuators B-Chemical, 12  77-81 (1993). 91. Y. T. Yan, Y. Shimizu, N. Miura, and N. Yamazoe, "High-Performance Solid-Electrolyte SOX Sensor Using MgO-Stabilized Zirconia Tube and Li2SO4-CaSO4-SiO2 Auxiliary Phase," Sensors and Actuators B-Chemical, 20 [2-3] 81-87 (1994). 92. A. Vogel, G. Baier, and V. Schule, "N on-Nernstian Potentiometric Zirconia Sensors Screening of Potential Working Electrode Materials," Sensors and Actuators BChemical, 15 [1-3] 147-150 (1993). 93. E. Kanazawa, G. Sakai, K. Shimanoe, Y. Kanmura, Y. Teraoka, N. Miura, and N. Yamazoe, "Semiconductor Gas Sens or Using SrO-Modified SnO2 for the Detection of N2O in Air," Electrochemical and Solid State Letters, 3  572-574 (2000). 94. E. Kanazawa, G. Sakai, K. Shimanoe, Y. Kanmura, Y. Teraoka, N. Miura, and N. Yamazoe, "Metal Oxide Semiconductor N2O Sensor for Medical Use," Sensors and Actuators B-Chemical, 77 [1-2] 72-77 (2001). 95. E. Kanazawa, M. Kugishima, K. Shimanoe Y. Kanmura, Y. Teraoka, N. Miura, and N. Yamazoe, "Mixed-Potential Type N2O Sensor Using Stabilized Zirconiaand SnO2-Based Sensing Electrode," Sensors and Actuators B-Chemical, 75 [1-2] 121-124 (2001). 96. K. Tanaka, S. Morimoto, S. Sonoda, S. Matsuura, K. Moriya, and M. Egashira, "Combustion Monitoring Sensor Using Tin Dioxide Semiconductor," Sensors and Actuators B-Chemical, 3  247-253 (1991). 97. D. E. Williams, P. McGehin, and B. C. Tolfiled, "Solid Electrolyte Mixed Potential Phenomena" pp 275. in Solid State Chemistry, 1982 proceedings of the Second European Conference, 1982. 98. F. H. Garzon, R. Mukundan, and E. L. Br osha, "Solid-State Mixed Potential Gas Sensors: Theory, Experiments and Challenges," Solid State Ionics, 136-137 633-638 (2000). 99. N. Miura, S. Zhuiykov, T. Ono, M. Hasei, and N. Yamazoe, "Mixed Potential Type Sensor Using Stabilized Zirconia an d ZnFe2O4 Sensing Electrode for NOx
211 Detection at High Temperature," Sensors and Actuators B: Chemical, 83 [1-3] 222-229 (2002). 100. F. Menil, V. Coillard, and C. Lucat, "C ritical Review of Nitrogen Monoxide Sensors for Exhaust Gases of Lean Burn Engines," Sensors and Actuators B-Chemical, 67 1 (2000). 101. B. K. Narayanan, S. A. Akbar, and P. K. Dutta, "A Phosphate-Based Proton Conducting Solid Electrolyte Hydrocarbon Gas Sensor," Sensors and Actuators B: Chemical, 87  480-486 (2002). 102. E. D. Wachsman, P. Jayaweera, G. Krishnan, and A. Sanjurjo, "Electrocatalytic Reduction of NOx on La(1-x)A(x)B(1-y)B(y)O(3-): Evidence of Electrically Enhanced Activity," Solid State Ionics, 136 775-782 (2000). 103. E. Di Bartolomeo, N. Kaabbuathong, M. L. Grilli, and E. Traversa, "Planar Electrochemical Sensors Based on Tape-C ast YSZ Layers and Oxide Electrodes," Solid State Ionics, 171 [3-4] 173-181 (2004). 104. V. V. Kharton, E. V. Tsipis, A. A. Yaremchenko, and J. R. Frade, "Surface-Limited Oxygen Transport and Electrode Properties of La2Ni0.8Cu0.2O4+," Solid State Ionics, 166 [3-4] 327-337 (2004). 105. J. Mizsei, "How Can Sensitive and Se lective Semiconductor Gas Sensors Be Made?," Sensors and Actuators B: Chemical, 23 [2-3] 173-176 (1995). 106. N. S. Baik, G. Sakai, K. Shimanoe, N. Miura, and N. Yamazoe, "Hydrothermal Treatment of Tin Oxide Sol Solution for Preparation of Thin-Film Sensor With Enhanced Thermal Stability and Gas Sensitivity," Sensors and Actuators BChemical, 65 [1-3] 97-100 (2000). 107. C. Xu, J. Tamaki, N. Miura, and N. Ya mazoe, "Grain-Size Effect s on Gas Sensitivity of Porous SnO2-Based Elements," Sensors and Actuators B-Chemical, 3  147-155 (1991). 108. T. Anukunprasert, C. Saiwan, and E. Tr aversa, "The Development of Gas Sensor for Carbon Monoxide Monitoring Usi ng Nanostructure of Nb-TiO2," Science and Technology of Advanced Materials, 6 [3-4] 359-363 (2005). 109. J. Tamaki, Z. Zhang, K. Fujimori, M. Akiyama, T. Harada, N. Miura, and N. Yamazoe, "Grain-Size Effects in Tungsten OxideBased Sensor for Nitrogen-Oxides," Journal of the Elect rochemical Society, 141  2207-2210 (1994). 110. A. Bielanski, and J. Haber, Oxygen in Catalysis Marcel Dekker, New York, 1991. 111. E. Traversa, G. Gnappi, A. Montenero, and G. Gusmano, "Ceramic Thin Films by SolGel Processing As Novel Materials fo r Integrated Humidity Sensors," Sensors and Actuators B-Chemical, 31 [1-2] 59-70 (1996).
212 112. Nowotny, J. Interface Electri cal Phenomena in Ionic Solids. In CRC Handbook of Solid State Ionics, Gellings, P. J., Bouwmeester, H. J. M., Eds.; CRC Press: Boca Raton, 1997; pp 121-169. 113. P. B. Weisz, "Effects of Electronic Ch arge Transfer Between Adsorbate and Solid on Chemisorption and Catalysis," The Journal of Chemical Physics, 11  (1953). 114. Z. Y. Can, H. Narita, J. Mizusaki, and H. Tagawa, "Detection of Carbon-Monoxide by Using Zirconia Oxygen Sensor," Solid State Ionics, 79 344-348 (1995). 115. A. Dutta, N. Kaabbuathong, M. L. Grilli, E. Di Bartolomeo, and E. Traversa, "Study of YSZ-Based Electrochemical Sensor s With Oxide Electrodes for High Temperature Applications," Bulletin of Materials Science, 25  451-453 (2002). 116. N. Miura, G. Lu, and N. Yamazoe, "Pr ogress in Mixed-Potentia l Type Devices Based on Solid Electrolyte for Sensing Redox Gases," Solid State Ionics, 136-137 533-542 (2000). 117. N. Barsan, J. R. Stetter, M. Findlay, and W. Gopel, "High Performance Gas Sensing of CO: Comparative Tests for Semiconducting (SnO2-Based) and for Amperometric Gas Sensors," Analytical Chemistry, 71  2512-2517 (1999). 118. J. Yoo, S. Chatterjee, and E. D. Wach sman, "Sensing Properties and Selectivities of a WO3/YSZ/Pt Potentiometric NOx Sensor," Sensors and Actuators B: Chemical, 122 644-652 (2007). 119. J. E. Anderson and Y. B. Graves, "Stea dy-State Characteristics of Oxygen Concentration Cell Sensors Subjected to NonEquilibrium Gas-Mixtures," Journal of the Electrochemical Society, 128  294-300 (1981). 120. A. Dutta, T. Ishihara, and H. Nishiguc hi, "An Amperometric Solid-State Gas Sensor Using a LaGaO3-Based Perovskite Oxide Electr olyte for Detecting Hydrocarbon in Exhaust Gas. A Bimetallic Anod e for Improving Sensitivity at Low Temperature," Chemistry of Materials, 16  5198-5204 (2004). 121. T. Hyodo, Y. Tominaga, T. Yamaguchi, A. Kawahara, H. Katsuki, Y. Shimizu, and M. Egashira, "NOX Sensing Properties of WO3-Based Semiconductor Gas Sensors Fabricated by Slide-Off Transfer Printing," Electrochemistry, 71  481-484 (2003). 122. T. Ishihara, H. Fujita, H. Nishiguchi, and Y. Takita, "SrSnO3-WO3 As Capacitive-Type Nitrogen Oxide Sensor for Mon itoring at High Temperature," Sensors and Actuators B-Chemical, 65 [1-3] 319-324 (2000). 123. T. Maekawa, J. Tamaki, N. Miura, N. Yamazoe, and S. Matsushima, "Development of SnO2-Based Ethanol Gas Sensor," Sensors and Actuators B-Chemical, 9  63-69 (1992).
213 124. J. Mizusaki, H. Tagawa, Y. Miyaki, S. Yamauchi, K. Fueki, I. Koshiro, and K. Hirano, "Kinetics of the Electrod e-Reaction at the CO-CO2, Porous Pt/Stabilized Zirconia Interface," Solid State Ionics, 53-6 126-134 (1992). 125. B. M. White, F. M. Van Assche, E. Traversa, and E. D. Wachsman, "Electrical Characterization of Semiconducting La2CuO4 For Potentiometric Gas Sensor Applications" pp 109. in Proceedi ngs of the 208th Meeting of The Electrochemical Society, Vol. 1, Solid State Ionic Devices IV, ECS Transactions. Edited by E. D. Wachsman, V. Birss, F. H. Garzon, R. Mukundan, and E. Traversa. The Electrochemical Society, Pennington, NJ, 2006. 126. J. W. Yoon, T. Sasaki, N. Koshizaki, and E. Traversa, "Prepara tion and Characterization of M/TiO2 (M = Ag, Au, Pt) Nanocomposite Thin Films," Scripta Materialia, 44 [8-9] 1865-1868 (2001). 127. S. Zurek, M. Mosialek, P. Tomczyk, a nd D. Oblakowska, "Inves tigation of the Partial Oxidation of Methane at the Interface Metal (Pt, Au)Vertical Bar YSZ," Journal of Electroceramics, 15  83-93 (2005). 128. E. L. Brosha, R. Mukundan, D. R. Brown, F. H. Garzon, J. H. Visser, M. Zanini, Z. Zhou, and E. M. Logothetis, "CO/HC Sensors Based on Thin Films of LaCoO3 and La0.8Sr0.2CoO3Metal Oxides," Sensors and Actuators B-Chemical, 69 [1-2] 171-182 (2000). 129. Bhattacharjee, S. The Expert System for Thermodynamics (TEST). Bhattacharjee, S. 2007. San Diego State University. Ref Type: Data File 130. S. Bhattacharjee, "TEST The Expert System for Thermodynamics" in Proceedings of the 2003 American Society for Engineer ing Education Annu al Conference & Exposition,American Society for Engineering Education, 2003. 131. D. C. Johnston, J. P. Stokes, D. P. Goshorn, and J. T. Lewandowski, "Influence of Oxygen Defects on the Physical-Properties of La2CuO4-y," Physical Review B, 36  4007-4010 (1987). 132. J. D. Jorgensen, H. B. Schuttler, D. G. Hinks, D. W. Capone, K. Zhang, M. B. Brodsky, and D. J. Scalapino, "Lattice Instabil ity and High-Tc Supe rconductivity in La2XBaxCuO4," Physical Review Letters, 58  1024-1027 (1987). 133. J. D. Jorgensen, B. Dabrowski, S. Y. Pei, D. G. Hinks, L. Soderholm, B. Morosin, J. E. Schirber, E. L. Venturini, and D. S. Ginley, "Superconducting Phase of La2CuO4+A Superconducting Composition Resulting From Phase-Separation," Physical Review B, 38  11337-11345 (1988). 134. H. Kanai, J. Mizusaki, H. Tagawa, S. Hoshiyama, K. Hirano, K. Fujita, M. Tezuka, and T. Hashimoto, "Defect Chemistry of La(2-x)SrxCuO4: Oxygen Nonstoichiometry
214 and Thermodynamic Stability," Journal of Solid State Chemistry, 131 150-159 (1997). 135. J. Maier and G. Pfundtner, "Defect Chemistry of the High-Tc Superconductors," Advanced Materials, 3  292-297 (1991). 136. G. Cao, J. W. O'Reilly, J. E. Crow, and L. R. Testardi, "Enhanced Electric Polarizability at the Magnetic Ordering Temperature of La2CuO4+x," Physical Review B, 47  11510-11511 (1993). 137. A. Lappas and K. Prassides, "Layered C uprates With the T* Structure: Structural and Conducting Properties," Journal of Solid State Chemistry, 115  332-346 (1995). 138. J. M. Longo and P. M. Raccah, "The Structure of La2CuO4 and LaSrVO4," Journal of Solid State Chemistry, 6  526-531 (1973). 139. A. Shukla, M. Calandra, M. Taguchi, A. Kotani, G. Vanko, and S. W. Cheong, "Polarized Resonant Inelastic X-Ray S cattering As an Ultrafine Pr obe of Excited States of La2CuO4," Physical Review Letters, 96  077006 (2006). 140. J. C. Bennett, M. Olfert, G. A. Schol z, and F. W. Boswell, "Synthesis of Bulk Superconducting La2CuO4 in an Electrochemical Cell," Physical Review B, 44  2727-2731 (1991). 141. Y. F. Zhu, R. Q. Tan, T. Yi, S. Gao, C. H. Yan, and L. L. Cao, "Preparation of Nanosized La2CuO4 Perovskite Oxide Using an Amor phous Heteronuclear Complex As a Precursor at Low-Temperature," Journal of Alloys and Compounds, 311  16-21 (2000). 142. L. Bapat, G. N. Natu, J. Kher, and M. Bhide, "Reactivity of Binary Mixture of La(III) Oxide and Cu(II) Oxalate or Nitrate and Synthesis of La(III) Cuprate," Journal of Thermal Analysis and Calorimetry, 62 285-294 (2000). 143. J. Mahia, J. L. Martinez-Lorenzo, M. C. Blanco, and J. Lopez-Quintela, "Structural Characterization of Electro chemically Oxidized La2CuO4+ Particles Prepared by a Sol-Gel Method," Journal of Solid State Chemistry, 131 246-251 (1997). 144. L. M. Gan, L. H. Zhang, H. S. O. Chan, C. H. Chew, and B. H. Loo, "A Novel Method for the Synthesis of Perovskite-Type Mixed Metal Oxides by the Inverse Microemulsion Technique," Journal of Materials Science, 31 1071-1079 (1996). 145. A. A. Galkin, B. G. Kostyuk, V. V. Lunin, and M. Poliakoff, "Continuous Reactions in Supercritical Water: A New Route to La2CuO4 With a High Surface Area and Enhanced Oxygen Mobility," Angewandte Chemie-International Edition, 39  2738-2740 (2000).
215 146. B. L. Kirkwood, T. S. Luhman, R. R. Stephenson, and M. M. Strasik, Method for Producing Superconducting Oxide Compounds. USA 6,559,103, May 6, 2003. 147. Vayenas, C. G.; Bebelis, S. I.; Yentekak is, I. V.; Neophytides, S. N. Electrocatalysis and Electrochemical Reactors. In CRC Handbook of Solid State Ionics, Gellings, P. J., Bouwmeester, H. J. M., Eds.; CRC Press: Boca Raton, 1997; pp 445-480. 148. M. Akiyama, Z. Zhang, J. Tamaki, N. Miura, N. Yamazoe, and T. Harada, "Tungsten Oxide-Based Semiconductor Sensor for Detection of Nitrogen-Oxides in Combustion Exhaust," Sensors and Actuators B-Chemical, 14 [1-3] 619-620 (1993). 149. Y. G. Choi, G. Sakai, K. Shimanoe, N. Miura, and N. Yamazoe, "Wet Process-Prepared Thick Films of WO3 for NO2 Sensing," Sensors and Actuators B-Chemical, 95  258-265 (2003). 150. Y. G. Choi, G. Sakai, K. Shimanoe, Y. Teraoka, N. Miura, and N. Yamazoe, "Preparation of Size and Habit-Controlled Nano Cr ystallites of Tungsten Oxide," Sensors and Actuators B-Chemical, 93 [1-3] 486-494 (2003). 151. T. Hyodo, T. Mori, A. Kawahara, H. Katsuki, Y. Shimizu, and M. Egashira, "Gas Sensing Properties of Semiconductor Hete rolayer Sensors Fabricated by Slide-Off Transfer Printing," Sensors and Actuators B-Chemical, 77 [1-2] 41-47 (2001). 152. X. S. Wang, G. Sakai, K. Shimanoe, N. Miura, and N. Yamazoe, "Spin-Coated Thin Films of SiO2-WO3 Composites for Det ection of Sub-Ppm NO2," Sensors and Actuators B-Chemical, 45  141-146 (1997). 153. X. S. Wang, N. Miura, and N. Yamazoe, "Study of WO3-Based Sensing Materials for NH3 and NO Detection," Sensors and Actuators B-Chemical, 66 [1-3] 74-76 (2000). 154. C. N. Xu, N. Miura, Y. Ishida, K. Ma tsuda, and N. Yamazoe, "Selective Detection of NH3 Over NO in Combustion Exhausts by Using Au and MoO3 Doubly Promoted WO3 Element," Sensors and Actuators B-Chemical, 65 [1-3] 163-165 (2000). 155. G. Lu, N. Miura, and N. Yamazoe, "Stabilized Zirconia-Based Sensors Using WO3 Electrode for Detection of NO or NO2," Sensors and Actuators B-Chemical, 65 [1-3] 125-127 (2000). 156. M. C. Carotta, G. Martinelli, L. Crem a, M. Gallana, M. Merli, G. Ghiotti, and E. Traversa, "Array of Thick Film Sensors for Atmospheric Pollutant Monitoring," Sensors and Actuators B-Chemical, 68 [1-3] 1-8 (2000). 157. M. C. Carotta, C. Martinelli, L. Crema, C. Malagu, M. Merli, G. Ghiotti, and E. Traversa, "Nanostructured Thick-Film Gas Sensors for Atmospheric Pollutant Monitoring: Quantitative Analysis on Field Tests," Sensors and Actuators B-Chemical, 76  336-342 (2001).
216 158. M. C. Carotta, M. Ferroni, S. Gherardi, V. Guidi, C. Malagu, G. Martinelli, M. Sacerdoti, M. L. Di Vona, S. Licoccia, and E. Tr aversa, "Thick-Film Gas Sensors Based on Vanadium-Titanium Oxide Powders Prepared by Sol-Gel Synthesis," Journal of the European Ceramic Society, 24  1409-1413 (2004). 159. M. C. Carotta, V. Guidi, C. Malagu, B. Vendemiati, A. Zanni, G. Martinelli, M. Sacerdoti, S. Licoccia, M. L. Di Vona, and E. Traversa, "Vanadium and Tantalum-Doped Titanium Oxide (TiTaV): a Novel Material for Gas Sensing," Sensors and Actuators B-Chemical, 108 [1-2] 89-96 (2005). 160. M. Ferroni, M. C. Carotta, V. Guidi, G. Martinelli, F. Ronconi, M. Sacerdoti, and E. Traversa, "Preparation and Characterizati on of Nanosized Titania Sensing Film," Sensors and Actuators B-Chemical, 77 [1-2] 163-166 (2001). 161. M. L. Frank, M. D. Fulkerson, B. R. Patton, and P. K. Dutta, "TiO2-Based Sensor Arrays Modeled With Nonlinear Regression Anal ysis for Simultaneously Determining CO and O2 Concentrations at High Temperatures," Sensors and Actuators BChemical, 87  471-479 (2002). 162. N. O. Savage, S. A. Akbar, and P. K. Dutta, "Titanium Dioxide Based High Temperature Carbon Monoxide Selective Sensor," Sensors and Actuators B-Chemical, 72 239248 (2001). 163. E. Traversa, G. Gusmano, and A. Mont enero, "Innovative Humidity Sensitive ElectricalProperties of Sol-Gel Processed Ceramic Thin-Films," European Journal of Solid State and Inorganic Chemistry, 32 [7-8] 719-729 (1995). 164. E. Traversa, M. L. Di Vona, S. Licoccia M. Sacerdoti, M. C. Ca rotta, M. Gallana, and G. Martinelli, "Sol-Gel Nanosized Semi conducting Titania-Based Powders for Thick-Film Gas Sensors," Journal of Sol-Gel Science and Technology, 19 [1-3] 193-196 (2000). 165. E. Traversa, Y. Sadaoka, M. C. Carotta, and G. Martinelli, "Environmental Monitoring Field Tests Using Screen-Printed Thic k-Film Sensors Based on Semiconducting Oxides," Sensors and Actuators B-Chemical, 65 [1-3] 181-185 (2000). 166. E. Traversa, M. L. Di Vona, S. Licoccia M. Sacerdoti, M. C. Carotta, L. Crema, and G. Martinelli, "Sol-Gel Processed TiO2-Based Nano-Sized Powders for Use in Thick-Film Gas Sensors for Atmo spheric Pollutant Monitoring," Journal of SolGel Science and Technology, 22 [1-2] 167-179 (2001). 167. J. Wu, K. Mobley, and R. L. McCreery "Electronic Characteri stics of Fluorene/TiO2 Molecular Heterojunctions," Journal of Chemical Physics, 126  024704 (2007). 168. C. Y. Yu, T. Hyodo, Y. Shimizu, and M. Egashira, "Preparation of Thermally Stable Mesoporous TiO2 Powder and Its Gas Sensor Application," Electrochemistry, 71  475-480 (2003).
217 169. N. S. Baik, G. Sakai, N. Miura, and N. Yamazoe, "Hydrothermally Treated Sol Solution of Tin Oxide for Thin-Film Gas Sensor," Sensors and Actuators B-Chemical, 63 [1-2] 74-79 (2000). 170. N. Barsan, M. Schweizer-Berberich, and W. Gopel, "Fundamental and Practical Aspects in the Design of Nanoscaled SnO2 Gas Sensors: A Status Report," Fresenius Journal of Analytical Chemistry, 365  287-304 (1999). 171. S. C. Chang, "Thin-Film Semiconductor NOx Sensor," IEEE Transactions on Electron Devices, 26  1875-1880 (1979). 172. W. Gopel and K. D. Schierbaum, "SnO2 Sensors Current Status and Future-Prospects," Sensors and Actuators B-Chemical, 26 [1-3] 1-12 (1995). 173. T. Sahm, A. Gurlo, N. Barsan, and U. Weimar, "Basics of Oxygen and SnO2 Interaction; Work Function Change and Conductivity Measurements," Sensors and Actuators B-Chemical, 118 [1-2] 78-83 (2006). 174. G. Sakai, N. S. Baik, N. Miura, and N. Yamazoe, "Gas Sensing Properties of Tin Oxide Thin Films Fabricated From Hydrotherm ally Treated Nanoparticles Dependence of CO and H2 Response on Film Thickness," Sensors and Actuators B-Chemical, 77 [1-2] 116-121 (2001). 175. G. Sberveglieri, S. Groppelli, and P. Nelli, "Highly Sensitive and Selective NOX and NO2 Sensor Based on Cd-Doped SnO2 Thin-Films," Sensors and Actuators BChemical, 4 [3-4] 457-461 (1991). 176. G. Sberveglieri, G. Faglia, S. Groppelli, and P. Nelli, "Methods for the Preparation of NO, NO2 and H2 Sensors Based on Tin Oxide Thin-Films, Grown by Means of the RF Magnetron Sputtering Technique," Sensors and Actuators B-Chemical, 8  79-88 (1992). 177. K. D. Schierbaum, U. Weimar, W. G opel, and R. Kowalkowski, "Conductance, Work Function and Catalytic Activity of SnO2-Based Gas Sensors," Sensors and Actuators B-Chemical, 3  205-214 (1991). 178. Y. Shimizu, A. Jono, T. Hyodo, and M. Egashira, "Preparation of Large Mesoporous SnO2 Powder for Gas Sensor Application," Sensors and Actuators B-Chemical, 108 [1-2] 56-61 (2005). 179. J. Tamaki, T. Maekawa, N. Miura, and N. Yamazoe, "CuO-SnO2 Element for Highly Sensitive and Selective Detection of H2S," Sensors and Actuators B-Chemical, 9  197-203 (1992). 180. H. Yamaura, T. Jinkawa, J. Tamaki, K. Moriya, N. Miura, and N. Yamazoe, "Indium Oxide-Based Gas Sensor for Se lective Detection of CO," Sensors and Actuators B-Chemical, 36 [1-3] 325-332 (1996).
218 181. X. Zhou, Q. Cao, Y. Hu, J. Gao, a nd Y. Xu, "Sensing Behavior and Mechanism of La2CuO4-SnO2 Gas Sensors," Sensors and Actuators B-Chemical, 77 443-446 (2001). 182. S. Bredikhin, J. Liu, and W. Weppner, "Solid Ionic Conductor Semiconductor Junctions for Chemical Sensors," Applied Physics A-Materials Science & Processing, 57  37-43 (1993). 183. H. Aono, E. Traversa, M. Sakamoto, a nd Y. Sadaoka, "Crystall ographic Characterization and NO2 Gas Sensing Property of LnFeO3 Prepared by Thermal Decomposition of Ln-Fe Hexacyanocomplexes, Ln[Fe(CN)6]NH2O, Ln = La, Nd, Sm, Gd, and Dy," Sensors and Actuators B-Chemical, 94  132-139 (2003). 184. M. C. Carotta, M. A. Butturi, G. Martin elli, Y. Sadaoka, P. Nunziante, and E. Traversa, "Microstructural Evolution of Nanosized LaFeO3 Powders From the Thermal Decomposition of a Cyano-Complex for Thick Film Gas Sensors," Sensors and Actuators B-Chemical, 44 [1-3] 590-594 (1997). 185. E. Traversa, S. Matsushima, G. Okada, Y. Sadaoka, Y. Sakai, and K. Watanabe, "NO2 Sensitive LaFeO3 Thin-Films Prepared by RF-Sputtering," Sensors and Actuators B-Chemical, 25 [1-3] 661-664 (1995). 186. H. Aono, M. Sato, E. Traversa, M. Sakamoto, and Y. Sadaoka, "Design of Ceramic Materials for Chemical Sensors: Effect of SmFeO3 Processing on Surface and Electrical Properties," Journal of the American Ceramic Society, 84  341-347 (2001). 187. M. C. Carotta, G. Martinelli, Y. Sadaoka P. Nunziante, and E. Traversa, "Gas-Sensitive Electrical Properties of Perovskite-Type SmFeO3 Thick Films," Sensors and Actuators B-Chemical, 48 [1-3] 270-276 (1998). 188. E. Traversa, S. Villanti, G. Gusmano, H. Aono, and Y. Sada oka, "Design of Ceramic Materials for Chemical Sensors: SmFeO3 Thick Films Sensitive to NO2," Journal of the American Ceramic Society, 82  2442-2450 (1999). 189. E. L. Brosha, R. Mukundan, D. R. Brown, F. H. Garzon, J. H. Visser, M. Zanini, Z. Zhou, and E. M. Logothetis, "CO/HC Sensors Based on Thin Films of LaCoO3 and La0.8Sr0.2CoO3Metal Oxides," Sensors and Actuators B: Chemical, 69 [1-2] 171-182 (2000). 190. C. Baratto, G. Sberveglieri, A. Onischuk, B. Caruso, and S. di Stasio, "Low Temperature Selective NO2 Sensors by Nanostructured Fibres of ZnO," Sensors and Actuators B-Chemical, 100 [1-2] 261-265 (2004). 191. T. Ishihara, S. Sato, and Y. Takita, "C apacitive-Type Sensors for the Selective Detection of Nitrogen-Oxides," Sensors and Actuators B-Chemical, 25 [1-3] 392-395 (1995).
219 192. E. Traversa, M. Miyayama, and H. Yanagida, "Gas Sensitivity of ZnO/La2CuO4 Heterocontacts," Sensors and Actuators B-Chemical, 17  257-261 (1994). 193. E. Traversa, A. Bearzotti, M. Miyayama and H. Yanagida, "The Electrical-Properties of La2CuO4/ZnO Heterocontacts at Different Relative Humidities," Journal of Materials Research, 10  2286-2294 (1995). 194. E. Traversa, A. Bearzotti, M. Miya yama, and H. Yanagida, "Study of the Conduction Mechanism of La2CuO4-ZnO Heterocontacts at Diffe rent Relative Humidities," Sensors and Actuators B-Chemical, 25 [1-3] 714-718 (1995). 195. E. Traversa, A. Bearzotti, M. Miyayama and H. Yanagida, "Inf luence of the Electrode Materials on the Electrical Respons e of ZnO-Based Contact Sensors," Journal of the European Ceramic Society, 18  621-631 (1998). 196. N. Q. Wu, M. H. Zhao, J. G. Zheng, C. B. Jiang, B. Myers, S. X. Li, M. Chyu, and S. X. Mao, "Porous CuO-ZnO Nanocomposite for Sensing Electrode of HighTemperature CO Solid-State Electrochemical Sensor," Nanotechnology, 16  2878-2881 (2005). 197. W. Y. Chung, G. Sakai, K. Shimanoe N. Miura, D. D. Lee, and N. Yamazoe, "Preparation of Indium Oxide Thin Film by Spin-Coating Method and Its GasSensing Properties," Sensors and Actuators B-Chemical, 46  139-145 (1998). 198. W. Y. Chung, G. Sakai, K. Shimanoe, N. Miura, D. D. Lee, and N. Yamazoe, "SpinCoated Indium Oxide Thin Film on Alumina and Silicon Substrates and Their Gas Sensing Properties," Sensors and Actuators B-Chemical, 65 [1-3] 312-315 (2000). 199. H. Yamaura, J. Tamaki, K. Moriya, N. Miura, and N. Yamazoe, "Selective CO Detection by Using Indium Oxide-Based Semiconductor Gas Sensor," Journal of the Electrochemical Society, 143  L36-L37 (1996). 200. K. Obata, S. Kumazawa, K. Shimanoe, N. Miura, and N. Yamazoe, "Potentiometric Sensor Based on NASICON and In2O3 for Detection of CO2 at Room Temperature Modification With Foreign Substances," Sensors and Actuators BChemical, 76 [1-3] 639-643 (2001). 201. J. Zosel, D. Westphal, S. Jakobs, R. Mueller, and U. Guth, "Au-Oxide Composites As HC-Sensitive Electrode Material for Mixed Potential Gas Sensors," Solid State Ionics, 152 525-529 (2002). 202. A. Kohli, C. C. Wang, and S. A. Akbar, "Niobium Pentoxide As a Lean-Range Sensor," Sensors and Actuators B-Chemical, 56 121 (1999). 203. J. Zosel, K. Ahlborn, R. Mueller, D. Westphal, V. Vashook, and U. Guth, "Response Behavior of Perovskites a nd Au/Oxide Composites As HC -Electrodes in Different Combustibles," Solid State Ionics, 169 115-119 (2004).
220 204. E. R. Macam, F. M. Van Assche, J. Yoo, and E. D. Wachsman, "Optimization of La2CuO4 Sensing Electrodes for a NOx Potentiometric Sensor," ECS Transactions, 1  121 (2006). 205. F. M. Van Assche, J. Yoo, S. Chatterjee, and E. D. Wachsman, "Heterogeneous Catalytic Evaluation of Potentiometric La2CuO4 Sensor Electrodes" pp 185. in Proceedings of the 208th Meeting of The Elect rochemical Society, Vol. 1, Solid State Ionic Devices IV, ECS Transactions. Edited by E. D. Wachsman, V. Birss, F. H. Garzon, R. Mukundan, and E. Traversa. The Electrochemical Society, Pennington, NJ, 2006. 206. N. A. Besley, "Theoretical Study of the Electronic Spectrosc opy of CO Adsorbed on Pt(111)," Journal of Chemical Physics, 122  184706 (2005). 207. P. Hu, D. A. King, M. H. Lee, and M. C. Payne, "Orbital Mixing in Co Chemisorption on Transition-Metal Surfaces," Chemical Physics Letters, 246 [1-2] 73-78 (1995). 208. G. A. Somorjai, K. M. Bratlie, M. O. Montano, and J. Y. Park, "Dynamics of Surface Catalyzed Reactions; the Roles of Surf ace Defects, Surface Diffusion, and Hot Electrons," Journal of Physical Chemistry B: Condensed Matter, 110  2001420022 (2006). 209. A. R. Balkenende, J. W. Geus, A. J. H. M. Kock, and R. J. van der Pas, "The Effect of a Coverage-Dependent Heat of Adso rption on Temperature-Programmed Desorption and Adsorption in Porous Catalysts," Journal of Catalysis, 115  365-375 (1989). 210. J. M. Kanervo, K. M. Reinikainen, and A. O. I. Krause, "Kinetic Analysis of Temperature-Programmed Desorption," Applied Catalysis A: General, 258  135-144 (2004). 211. J. M. Kanervo, T. J. Keskitalo, R. I. Slioor, and A. O. I. Krause, "TemperatureProgrammed Desorption As a Tool to Extract Quantitative Kinetic or Energetic Information for Porous Catalysts," Journal of Catalysis, 238  382-393 (2006). 212. H. E. Kissinger, "Reaction Kinetics in Differential Thermal Analysis," Analytical Chemistry, 29  1702-1706 (1957). 213. L. Olsson, B. Westerberg, H. Persson, E. Fridell, M. Skoglundh, and B. Andersson, "A Kinetic Study of Oxygen Adsorption/Desorption and NO Oxidation Over Pt/Al2O3 Catalysts," Journal of Physical Chemistry B, 103  10433-10439 (1999). 214. L. Olsson, H. Persson, E. Fridell, M. Skoglundh, and B. Andersson, "Kinetic Study of NO Oxidation and NOx Storage on Pt/Al2O3 and Pt/BaO/Al2O3," Journal of Physical Chemistry B, 105  6895-6906 (2001).
221 215. A. C. van Veen, D. Farrusseng, M. Rebe illeau, T. Decamp, A. Holzwarth, Y. Schuurman, and C. Mirodatos, "Acceleration in Catalyst Development by Fast Transient Kinetic Investigation," Journal of Catalysis, 216 [1-2] 135-143 (2003). 216. O. J. Wimmers, "Analytical Expre ssions for Temperature Programmed Reduction Patterns Using Solid-State Kinetics," Thermochimica Acta, 95  67-72 (1985). 217. O. J. Wimmers, P. Arnoldy, and J. A. Moulijn, "Determination of the ReductionMechanism by Temperature-Programmed Reduction Application to Small Fe2O3 Particles," Journal of Physical Chemistry, 90  1331-1337 (1986). 218. J. Baltrusaitis and V. H. Grassian, "Sur face Reactions of Carbon Dioxide at the Adsorbed Water-Iron Oxide Interface," Journal of Physical Ch emistry B: Condensed Matter, 109  12227-12230 (2005). 219. D. Duprez, "Study of Surface Reaction Mechanisms by 16O/18O and H/D Isotopic Exchange," Catalysis Today, 112 [1-4] 17-22 (2006). 220. P. S. Lambrou and A. M. Efstathiou, "The Effects of Fe on the Oxygen Storage and Release Properties of Model Pd-Rh/CeO2-Al2O3 Three-Way Catalyst," Journal of Catalysis, 240  182-193 (2006). 221. S. Roy, W. Sigmund, and F. Aldinge r, "Nanostructured Yttria Powders Via Gel Combustion," Journal of Materials Research, 14  1524-1531 (1999). 222. D. Bhattacharya, L. C. Pathak, S. K. Mishra, D. Sen, and K. L. Chopra, "Pyrophoric Synthesis Technique for Multicomponent High-Temperature Superconductors," Applied Physics Letters, 57  2145-2147 (1990). 223. White, B. M. Solid-State Potentriometric NOx/CO Sensors and Differential Electrode Equilibria. Ph.D. Dissertation, University of Florida, 2007. 224. P. A. Redhead, "Thermal Desorption of Gases," Vacuum, 12  203-211 (1962). 225. A. A. Davydov, Infrared Spectroscopy of Adsorbed Species on the Sur face of Transition Metal Oxides John Wiley and Sons, Chichester, 1990. 226. J. A. Rodriguez, T. Jirsak, G. Liu, J. Hrbek, J. Dvorak, and A. Maiti, "Chemistry of NO2 on Oxide Surfaces: Formation of NO3 on TiO2(110) and NO2<-->O Vacancy Interactions," Journal of the Americ an Chemical Society, 123  9597-9605 (2001). 227. D. L. S. Nieskens, A. P. van Bavel, and J. W. Niemantsverdriet, "The Analysis of Temperature Programmed Desorption Expe riments of Systems With Lateral Interactions; Implications of the Compensation Effect," Surface Science, 546  159-169 (2003).
222 228. F. H. Garzon, R. Mukundan, and E. L. Br osha, "Solid-State Mixed Potential Gas Sensors: Theory, Experiments and Challenges," Solid State Ionics, 136-137 633-638 (2000). 229. W. Gopel, G. Reinhardt, and M. Rosc h, "Trends in the Development of Solid State Amperometric and Potentiometric High Temperature Sensors," Solid State Ionics, 136-137 519-531 (2000). 230. N. Miura, G. Lu, and N. Yamazoe, "Pr ogress in Mixed-Potentia l Type Devices Based on Solid Electrolyte for Sensing Redox Gases," Solid State Ionics, 136-137 533-542 (2000). 231. N. Miura, S. Zhuiykov, T. Ono, M. Hasei, and N. Yamazoe, "Mixed Potential Type Sensor Using Stabilized Zirconia and ZnFe2O4 Sensing Electrode for NOx Detection at High Temperature," Sensors and Actuators B: Chemical, 83 [1-3] 222-229 (2002). 232. E. Di Bartolomeo, N. Kaabbuathong, M. L. Grilli, and E. Traversa, "Planar Electrochemical Sensors Based on Tape-C ast YSZ Layers and Oxide Electrodes," Solid State Ionics, 171 [3-4] 173-181 (2004). 233. K. Hadjiivanov and L. Dimitr ov, "IR Spectroscopy Study of CO and NOx Adsorption on a Cu/Zr-HMS Catalyst," Microporous and Mesoporous Materials, 27  49-56 (1999). 234. C. Drouet, P. Alphonse, J. L. G. Fierro, and A. Rousset, "Adsorp tion of Nitric Oxide and Temperature Programmed Desorption on Nonstoichiometric Nickel-Copper Manganites," Applied Surface Science, 174 [3-4] 289-295 (2001). 235. B. M. White, F. M. Van Assche, IV, E. Traversa, and E. D. Wachsman, "Electrical Characterization of Semiconducting La2CuO4 for Potentiometric Gas Sensor Applications," ECS Transactions, 1  109 (2006). 236. N. J. Crane, E. G. Bartick, R. S. Perlman, and S. Huffman, "Infrared Spectroscopic Imaging for Noninvasive Detecti on of Latent Fingerprints," Journal of Forensic Science, 52  48-53 (2007). 237. D. G. Lumsdon and A. R. Fraser, "Infrared Spectroscopi c Evidence Supporting Heterogeneous Site Binding Models for Humic Substances," Environmental Science and Technology, 39  6624-6631 (2005). 238. L. Cho, J. A. Reffner, B. M. Gatewood, and D. L. Wetzel, "Single Fiber Analysis by Internal Reflection Infr ared Microspectroscopy," Journal of Forensic Science, 46  1309-1314 (2001). 239. F. M. Van Assche, J. Yoo, S. Chatterjee, and E. D. Wachsman, "Heterogeneous Catalytic Evaluation of Potentiometric La2CuO4 Sensor Electrodes," ECS Transactions, 1  185 (2007).
223 240. Potassium Bromide MSDS. 2007. Mallinckrodt Baker. Ref Type: Data File 241. B. Stuart, Infrared Spectroscopy : Fu ndamentals and Applications John Wiley & Sons, Ltd.., Chichester, England, 2004. 242. A. V. Bazhenov, T. N. Fursova, V. B. Timofeev, A. S. Cooper, J. P. Remeika, and Z. Fisk, "Infrared Reflectivity Spectra of Single Crystals of La2CuO4," Physical Review B. Condensed Matter, 40  4413-4418 (1989). 243. Ch. Sedlmair, K. Seshanc, A. Jentys, and J. A. Lercher, "Elementary Steps of NOx Adsorption and Surface Reaction an a Comm ercial Storage-Reduction Catalyst," Journal of Catalysis, 214 308-316 (2003). 244. G. Busca and V. Lorenzelli, "Infrare d Study of the Adsorption of Nitrogen Dioxide, Nitric Oxide and Nitrous Oxide on Hematite," Journal of Catalysis, 72  303313 (1981). 245. K. Nakamoto, Infrared and Raman Spectra of Inorganic Coordination Compounds, 5th ed. Wiley, New York, 1997. 246. M. Schraml-Marth, A. Wokaun, and A. Baikert, "Surface Structure of Crystalline and Amorphous Chromia Catalysts for the Sele ctive Catalytic Re duction of Nitric Oxide IV. Diffuse Reflectance FTIR Study of NO Adsorption and Reaction," Journal of Catalysis, 138  306-321 (1992). 247. G. Ghiotti and A. Chiorino, "An IR Study of NO Adsorption on a CrOx/ZrO2 Catalyst," Spectrochimica Acta Part A: Molecular Spectroscopy, 49  1345-1359 (1993). 248. M. J. D. Low and R. T. Yang, "Reactions of Gaseous Pollutants With Solids : V. Infrared Study of the Sorption of NO on CaO," Journal of Catalysis, 34  479-489 (1974). 249. L. Cerruti, E. Modone, E. Guglielm, and E. Borello, "Infrared Study of Nitric-Oxide Adsorption on Magnesium-Oxide," Journal of the Chemical Society-Faraday Transactions I, 70  729-739 (1974). 250. L. H. Little, Infrared Spectra of Adsorbed Species, Academic Press, London, 1966. 251. M. Shelef and J. T. Kummer, Chemical Engineering Progress Symposium Series, 67  (1971). 252. A. Martinezarias, J. Soria, J. C. Conesa X. L. Seoane, A. Arcoya, and R. Cataluna, "NO Reaction at Surface Oxygen Vacancies Generated in Cerium Oxide," Journal of the Chemical Society-Faraday Transactions, 91  1679-1687 (1995). 253. J. Laane and J. R. Ohlsen, Progress in Inorganic Chemistry, 27 465 (1980).
224 254. A. Snis and I. Panas, "Surface Comple xes of Nitric Oxide: Adsorption of Anionic NO and N2O2 on CaO," Surface Science, 413 477-488 (1998). 255. L. M. Roev, and A. V. Alekseev, Elementary Photoprocesses in Molecules Consultants Bureau, New York, 1968. 256. Miscellaneous. In CRC Handbook of Chemistry and Physics, 57 ed.; CRC Press, Cleveland, 1976; p F-245-F-264. 257. S.-J. Huang, A. B. Walters, and M. A. Vannice, "Adsorption and Decomposition of NO on Lanthanum Oxide," Journal of Catalysis, 192  29-47 (2000). 258. F. M. Capece, V. Dicastro, C. Furlani, G. Mattogno, C. Fragale, M. Gargano, and M. Rossi, Journal of Electron Spectroscopy Related Phenomenon, 27 119 (1982). 259. N. S. Mcintyre and M. G. Cook, "X-R ay Photoelectron Studies on Some Oxides and Hydroxides of Cobalt, Nickel, and Copper," Analytical Chemistry, 47  22082213 (1975). 260. C. D. Wagner, W. M. Riggs, L. E. Davis, J. F. Moulder, and G. E. Muilenberg, Handbook of X-Ray Photoelectron Spectroscopy Perkin-Elmer Corporation, Physical Electronics Division, Eden Prairie, Minnesota, 1979. 261. Y. Uwamino, Y. Ishizuka, and H. Yamatera, Journal of Electron Spectroscopy Related Phenomenon, 34 69 (1984). 262. Kaufmann R., Klewe-Nebenius H., Moers H., Pfennig G., Jenett H., and Ache H.J., Surface Interface Analysis, 11 502 (1988). 263. Johansson G., Hedman J., Berndtsson A., Klasson M., and Nilsson R., Journal of Electron Spectroscopy Related Phenomenon, 2  (1973). 264. Y.S.Meng Combining Ab Initio Com putation with Experiments for Designing/ Understanding High Energy Density Electr ode Materials for Advanced Lithium Batteries. Ph.D Massachusetts Institute of Technology, 2005. 265. B. Glorieux, R. Berjoan, M. Matecki, A. Kammouni, and D. Pera rnau, "XPS Analyses of Lanthanides Phosphates," Applied Surface Science, 253  3349-3359 (2007). 266. N. Mizuno, "Supported Perovskites," Catalysis Today, 8 221-230 (1990). 267. J. Zhu, Z. Zhao, D. Xiao, J. Li, X. Yang, and Y. Wu, "Effect of Valence of Copper in La2XThxCuO4 on NO Decomposition Reaction," Catalysis Communications, 7  2932 (2006). 268. B. Lesiak, J. Zemek, P. Jiricek, and A. Jozwik, "Investigation of CoPd Alloys by XPS and EPES Using the Patt ern Recognition Method," Journal of Alloys and Compounds, 428 [1-2] 190-196 (2007).
225 269. S. Lizzit, A. Baraldi, D. Cocco, G. Come lli, G. Paolucci, R. Rosei, and M. Kiskinova, "NO Adsorption on the Rh(110) Surface: Kinetics and Composition of the Adlayer Studied by Fast XPS," Surface Science, 410 [2-3] 228-236 (1998). 270. J. F. Zhu, M. Kinne, T. Fuhrmann, R. Denecke, and H.-P. Steinruck, "In Situ HighResolution XPS Studies on Adsorp tion of NO on Pt(1 1 1)," Surface Science, 529  384-396 (2003). 271. E. R. Macam, F. M. Van Assche, J. Yoo, and E. D. Wachsman, "Optimization of La2CuO4 Sensing Electrodes for a NOx Potentiometric Sensor," ECS Transactions, 1  121 (2006). 272. J. A. Connor, Handbook of X-Ray and Ultravio let Photoelectron Spectroscopy, 2nd ed. Heyden, 1978. 273. S. Hfner, Photoelectron Spectroscopy, 2nd ed. Springer-Verlag Berlin Heidelberg, Berlin, 1996. 274. N. Nucker, J. Fink, B. Renker, D. Ewert, C. Politis, P. J. W. Weijs, and J. C. Fuggle, "Experimental Electronic-S tructure Studies of La2-XSrxCuO4," Zeitschrift fur Physik B-Condensed Matter, 67  9-14 (1987). 275. H. Romberg, N. Nucker, M. Alexander, J. Fink, D. Hahn, T. Zetterer, H. H. Otto, and K. F. Renk, "Density and Symmetry of Unoccupied Electronic States of Ti2Ba2CaCu2O8," Physical Review B, 41  2609-2611 (1990). 276. T. Horita, K. Yamaji, N. Sakai, Y. Xiong, T. Kato, H. Yokokawa, and T. Kawada, "Determination of Proton and Oxygen Movements in Solid Oxides by the Tracer Gases Exchange Technique and Se condary Ion Mass Spectrometry," Applied Surface Science, 203-204 634-638 (2003). 277. T. Kawada, M. Sase, M. Kudo, K. Yashiro, K. Sato, J. Mizusaki, N. Sakai, T. Horita, K. Yamaji, and H. Yokokawa, "Microscopic Observation of Oxygen Reaction Pathway on High Temperature Electrode Materials," Solid State Ionics, 177  3081-3086 (2006). 278. R. A. De Souza and R. J. Chater, "Oxygen Exchange and Diffusion Measurements: The Importance of Extracting the Correct Initial and Boundary Conditions," Solid State Ionics, 176 [23-24] 1915-1920 (2005). 279. N. Sakai, K. Yamaji, T. Horita, H. Kishimoto, M. E. Brito, H. Yokokawa, and Y. Uchimoto, "Application of SIMS Anal yses on Oxygen Transport in SOFC Materials," Applied Surface Science, 252  7045-7047 (2006). 280. J. M. Bassat, M. Petitjean, J. Fouletier, C. Lalanne, G. Caboche, F. Mauvy, and J. C. Grenier, "Oxygen Isotopic Exchange: A Useful Tool for Characterizing Oxygen Conducting Oxides," Applied Catalysis A: General, 289  84-89 (2005).
226 281. H. Wang, R. G. Tobin, G. B. Fisher C. L. DiMaggio, and D. K. Lambert, "Oxygen Exchange Between NO Gas and Adsorbed Atomic Oxygen on Pt(335)," Surface Science, 440  429-437 (1999). 282. A. Danon, A. Saig, Y. Finkelstein, and J. E. Koresh, "A New Route of Oxygen Isotope Exchange in the Solid Phase: Demonstration in CuSO4.5H2O," Journal of Physical Chemistry B: Condensed Matter, 109  21197-21201 (2005). 283. K. Krau[ss], A. Drochner, M. Fehlings, J. Kunert, and H. Vogel, "Oxygen Exchange at Mo/V Mixed Oxides: A Transient and 18O Isotope Study Under Technical Conditions," Journal of Molecular Ca talysis A: Chemical, 177  237-245 (2002). 284. A. J. Fletcher, Y. Yuzak, and K. M. Thomas, "Adsorption and Desorption Kinetics for Hydrophilic and Hydrophobic Vapors on Activated Carbon," Carbon, 44  9891004 (2006). 285. E. Boehm, J. M. Bassat, P. Dordor, F. Mauvy, J. C. Grenier, and P. Stevens, "Oxygen Diffusion and Transport Propertie s in Non-Stoichiometric Ln2-XNiO4+ Oxides," Solid State Ionics, 176 [37-38] 2717-2725 (2005). 286. F. M. Van Assche, J. C. Nino, and E. D. Wachsman, "Adsorbate Studies on La2CuO4 Using Infrared Spectroscopy and XRay Photoemisson Spectroscopy to Determine Potentiometric NOx Sensor Response Mechanism," J. Electrochemical Society, accepted (2008). 287. L. Pauling, The Nature of the Chemical Bond and the Structure of Molecules and Crystals, Second ed. Oxford University Press, Ithica, NY, 1948. 288. O'Haver, T. Interactive Peak Fitter, version 1.6. 2007. Ref Type: Data File 289. E. Foldes, Z. Szabo, A. Janecska, G. Na gy, and B. Pukanszky, "Quantitative Analysis of Functional Groups in HDPE Powder by DRIFT Spectroscopy," Macromolecular Symposia, 202 97-115 (2003). 290. S.-J. Huang, A. B. Walters, and M. A. Vannice, "Adsorption and Decomposition of NO on Lanthanum Oxide," Journal of Catalysis, 192  29-47 (2000). 291. M. Bhattacharya, W. G. Devi, and P. S. Mazumdar, "Determination of Kinetic Parameters From Temperature Pr ogrammed Desorption Curves," Applied Surface Science, 218 [1-4] 1-6 (2003). 292. P. Garcia, A. Molina, and F. Mondra gon, "Desorption Activa tion Energy Distribution Function of Nitric Oxide Chemisorbe d on Carbonaceous Materials at 373 K," Carbon, 43  1445-1452 (2005).
227 293. T. Ishihara, M. Ando, K. Sada, K. Tak iishi, K. Yamada, H. Nishiguchi, and Y. Takita, "Direct Decomposition of NO into N2 and O2 Over La(Ba)Mn(In)O3 Perovskite Oxide," Journal of Catalysis, 220  104-114 (2003). 294. A. Kotsifa, D. I. Kondarides, and X. E. Verykios, "Comparative Study of the Chemisorptive and Catalytic Properties of Supported Pt Catalysts Related to the Selective Catalytic Redu ction of NO by Propylene," Applied Catalysis B: Environmental, In Press, Corrected Proof. 295. R. Zhang, A. Villanueva, H. Alamdari, and S. Kaliaguine, "SCR of NO by Propene Over Nanoscale LaMn1-XCuxO3 Perovskites," Applied Catalysis A: General, 307  85-97 (2006). 296. R. W. Odom, S. E. Buttrill, R. H. Fl eming, M. Rossi, L. N. Goeller, and W. Gohl, "Pulsed Field Desorption Mass Spectrometry : A Technique for Investigating Field Desorption Ion Formation Mechanisms," International Journal or Mass Spectrometry and Ion Physics, 49  319-335 (1983). 297. T. Wilhelm and G. Wittstock, "Pattern s of Functional Proteins Formed by Local Electrochemical Desorption of Self-Assembled Monolayers," Electrochimica Acta, 47 [1-2] 275-281 (2001). 298. C. A. Widrig, C. Chung, and M. D. Porter, "The Electrochem ical Desorption of NAlkanethiol Monolayers From Polycr ystalline Au and Ag Electrodes," Journal of Electroanalytical Chemistry, 310 [1-2] 335-359 (1991). 299. Blackburn, B. 2007. 300. C. Carretti, P. Manini, and G. L. Grotto, "Mass Spectrometr ic Evaluation of the Desorption Properties of Some Glasses Under High Applied Voltage Conditions," Vacuum, 46 [8-10] 769-772 (1995). 301. N. Mingo and F. Flores, "Lateral Forces and Atomic Desorption Induced by the Electric Field Created by STM Tips on Metal Surfaces," Surface Science, 395 [2-3] 342355 (1998). 302. D. Sengupta, "Effect of External Fi eld on Bond Energy and Activation Barrier for Surface Diffusion," Journal of Crystal Growth, 286  91-95 (2006). 303. C. Tang, L. Feller, P. Rossbach, B. Keller, J. Voros, S. Tosatti, and M. Textor, "Adsorption and Electrically Stimulated Desorption of the Triblock Copolymer Poly(Propylene Sulfide-Bl-E thylene Glycol) (PPS-PEG) From Indium Tin Oxide (ITO) Surfaces," Surface Science, 600  1510-1517 (2006). 304. M. Zembala, "Electrokinetics of Heterogeneous Interfaces," Advances in Colloid and Interface Science, 112 [1-3] 59-92 (2004).
228 305. R. W. O'Brian and L. R. White, "Elect rophoretic Mobility of a Spherical Colloidal Particle," Jourtnal of the Chemical So ciety, Faraday Transactions, 2  16071626 (1978). 306. N. Golovan and V. Smyntyna, "The Sensitization of Semiconductor Gas Sensors," Sensors and Actuators B: Chemical, 6 [1-3] 289-292 (1992). 307. S. Gonzalez, C. Sousa, and F. Illas, "El ectric Field Effects in the Chemisorption of CO on Bimetallic RhCu Surface Models," Surface Science, 548 [1-3] 209-219 (2004). 308. W. Hellmich, G. Muller, C. Braunmuhl, T. Doll, and I. Eisele, "Field-Effect-Induced Gas Sensitivity Changes in Metal Oxides," Sensors and Actuators B: Chemical, 43  132-139 (1997). 309. Blackburn, B. Rendered image of PPD experimental reactor setup. 2007. Ref Type: Data File 310. Blackburn, B. Sensor Automation Protocol. . 2007. Ref Type: Data File 311. Blackburn, B. HP Voltage St ep Automation Protocol. [1.01]. 2007. Ref Type: Data File 312. M. Matsuda, M. Fujita, and K. Yamada, "Static Spin Correlations in Lightly-Doped La2CuO4+[Delta]," Physica C: Superconductivity, 408-410 810-811 (2004). 313. A. Du, G. Z. Wei, Q. L. Hu, and Z. Xianyu, "Neel Temperature of Orthorhombic La2CuO4," Journal of Magnetism and Magnetic Materials, 146 [1-2] 201-202 (1995). 314. J. Shimizu, M. Nakagawa, and Y. Oda, "The Normal State Magne tic Susceptibility of La2CuO4+[Delta] With Excess Oxygen," Physica C: Superconductivity, 388-389 263-264 (2003). 315. N. Shah, M. A. Green, and D. A. Neumann, "Phase Stability in the LnCa2Mn2O7 (Ln=Pr, Nd, Sm and Gd) Ruddlesden-Popper Series," Journal of Physics and Chemistry of Solids, 63  1779-1786 (2002). 316. D. di Caprio, J. Stafiej, and J. P. Ba diali, "Field Theory for Ionic Systems. From Fluctuations and Structure at a Hard Wall to Thermodynamics," Electrochimica Acta, 48 [20-22] 2967-2974 (2003). 317. J. C. Maxwell, A Treatise on Electricity and Magnetism, 3rd ed. Dover Publications, New York, 1891. 318. S. S. Atwood, Electric and Magnetic Fields, 3rd ed. New York, 1932.
229 319. S. Capone, M. Epifani, L. Francioso, S. Kaciulis, A. Mezzi, P. Siciliano, and A. M. Taurino, "Influence of Electrodes Agei ng on the Properties of the Gas Sensors Based on SnO2," Sensors and Actuators B: Chemical, 115  396-402 (2006). 320. J. Nowotny, T. Bak, and C. C. Sorrell, "Charge Transfer at Oxygen/Zirconia Interface at Elevated Temperatures Part 7: Effect of Surface Processing," Advances in Applied Ceramics, 104  195-199 (2005). 321. J. J. Sprague, O. Porat, and H. L. Tu ller, "Mixed Conducting Gas Sensors: Atmosphere Dependent Electrode Impedance," Sensors and Actuators B: Chemical, 36 [1-3] 348-352 (1996). 322. D. Kuscer, J. Holc, M. Hrovat, S. Bern ik, Z. Samardzija, and D. Kolar, "Interactions Between a Thick Film LaMnO3 Cathode and YSZ SOFC Electrolyte During High Temperature Ageing," Solid State Ionics, 78 [1-2] 79-85 (1995). 323. J. Nowotny, T. Bak, M. K. Nowotny, and C. C. Sorrell, "Charge Transfer at Oxygen/Zirconia Interface at Elevated Temperatures Part 3: Segregation Induced Interface Properties," Advances in Applied Ceramics, 104  165-173 (2005). 324. J. Nowotny, T. Bak, and C. C. Sorrell, "Charge Transfer at Oxygen/Zirconia Interface at Elevated Temperatures Part 5: Reactivity Models," Advances in Applied Ceramics, 104  181-187 (2005). 325. M. M. Viitanen, R. G. Welzenis, H. H. Brongersma, and F. P. F. van Berkel, "Silica Poisoning of Oxygen Membranes," Solid State Ionics, 150 [3-4] 223-228 (2002
230 BIOGRAPHICAL SKETCH Frederick Martin Van Assche IV was born on September 26, 1980 in Winter Park, Florida. After a short and fulfilling stint as an enfant te rrible, he progressed through Winter Park High School, garnering a reputation for dogged (alm ost ludicrous) persiste nce in academics and athletics, winning four state championships in rowing before graduating with honors in 1998. He was accepted to the chemical engineering program at the University of Florida, and following five years of late nights, early mornings and the occasional injury, was awarded his undergraduate degree in the spring of 2003. Duri ng his undergraduate tenure, he studied the Diels-Alder reactions of tetrachlo rocyclopropene under the direction of Dr. Merle Battiste in the Chemistry department. His research experience in chemistry convinced him to further his education through graduate study, though in a slightly less flammable concentration. Following graduation, he was accepted into the University of Florida graduate school in both materials science and civil engineering. As he was never properly informed about the civil engineering admission, he accepted a position in Dr. Eric Wachsmans sensor group in materials. His initial work in mass spectrometry with La2CuO4, WO3, and other sensor materials inspired him to further work in the subject following hi s masters in 2005. In the final two years of doctoral pursuit, Martin has pr esented at conferences in Los Angeles, Daytona Beach, and Washington DC, and served as the Student Chapte r President of the Electrochemical Society at the University of Florida. At time of writing, he eagerly awaits the next opportunity to apply his can-do spirit and ridiculous ly exuberant character.