Mixed Conduction and Defect Chemistry of Manganese and Molybdenum Substituted Gadolinium Titanate Pyrochlore
Total Page:16
File Type:pdf, Size:1020Kb
Mixed Conduction and Defect Chemistry of Manganese and Molybdenum Substituted Gadolinium Titanate Pyrochlore By John Jason Sprague B.S. Materials Science and Engineering, Johns Hopkins University, Baltimore (1993) Submitted to the Department of Materials Science and Engineering in Partial Fulfillment of the Requirements for the Degree of Doctor Of Philosophy in Electronic Materials at the Massachusetts Institute of Technology February 1999 @ Massachusetts Institute of Technology 1999 Signature of Author [J Department of Materials Science and Engineering January 8, 1999 Certified by Harry L. Tuller Professor of Ceramics and Electronic Materials Thesis Supervisor Accepted by_ Linn W. Hobbs .1 John F. Elliott Professor of Materials MASSCHUSETT Chair, Departmental Committee on Graduate Students MAR LIBRARIES I 2 Mixed Conduction and Defect Chemistry of Manganese and Molybdenum Substituted Gadolinium Titanate Pyrochlore by John Jason Sprague Submitted to the Department of Materials Science and Engineering on January 8th, 1999 in Partial Fulfillment of the Requirements of the Degree of Doctor of Philosophy in Electronic Materials Abstract The Solid Oxide Fuel Cell (SOFC) is an electrochemical device that converts chemical energy directly to electrical energy. This device by passes the pollution problems and relatively low efficiencies of conventional fossil fuels generators. The main drawback to SOFC utilization has been the need for high operating temperatures (1000 'C) to provide adequate efficiencies. The primary goal of SOFC research has been the reduction in this temperature. As operating temperatures around 700 - 800 0C have been achieved, it has become clear that the primary efficiency loss comes from the slow kinetics of the electrode reactions. This thesis will pursue mixed ionic and electronic conducting oxide ceramics that could be used to enhance the slow kinetics. Electrical conductivity measurements were made on the pyrochlore Gd2 ((MoIyMny)xTi1-) 2 07- (GMMT) as a function of oxygen partial pressure (P0 2 ) (1025 0C T 1000 0C), and composition (x,y). Particular < P0 2 < 1 atm), temperature (600 emphasis was placed on the materials value as a mixed ionic and electronic conductor (MIEC). Previous work on the Gd2 (Ti1 xMox) 2 0 7 system has shown it to have high electronic conductivity (102 S/cm) with a high minority ionic conductivity (up to 101 S/cm) for high Mo values under anodic conditions. However, the material decomposes at higher P0 2, due to a sharp increase in oxygen interstitial concentration accompanying the oxidation of Mo4 + to Mo6 +. In this study, we investigated the ability of variable valent Mn to compensate the oxidation of Mo and thereby stabilizing the material to much higher P0 2 while retaining the high conductivity. The conductivity of GMMT was found to increase by over 4 orders of magnitude by increasing x from 0.01 to 0.3 with y = 0.66. The peak value in air of 0.2 S/cm was obtained at 1000 C for the x = 0.33, y = 0.66 material. The conductivity was weakly dependent on y, increasing by about 2 order of magnitude under oxidizing conditions when increasing y from 0.33 to 0.66 (x = 0.1). The activation energy for the conduction in the characteristic P0 2 independent plateau decreased systematically with x from nearly 1.5 eV at x = 0.01, y = 0.66 to 0.64 eV at x = 0.3, y = 0.66. All of the compositions were found to be stable pyrochlore up to 1000 'C, the highest temperature used in this study. 3 A detailed defect model was developed to model and explain the electrical conductivity of GMMT. Our analysis indicates that the material is primarily electronic at all x,y, with some potential ionic conduction under reducing conditions. The conductivity conforms well to a dilute defect model for x 0.05, while at x > 0.1, the material is suspected to experience defect band hopping conduction within the Mo and Mn levels. The incorporation of defect bands into the defect model is necessary to explain the large jump in conductivity and decrease in activation energy with increasing x. Modeling of the conductivity data for x = 0.01 and 0.05 (y = 0.66) materials yields an estimate for the Mn ionization energy of- 2.6 eV as well as expressions for the reduction 1 2 9 constant, K= / cm )exp(-6.4eV/kT), and Frenkel constant, Kf = 10(cm-6 )exp(-2.9eV/kT). Similar modeling for the x = 0.3, y = 0.66 material yields defect band hopping energies in the Mn and Mo defect bands of 0.19 eV and 0.22 eV respectively. Independent measurements utilizing an electronic blocking cell to separate the ionic and electronic contributions to the conductivity were made for the x = 0.1, y = 0.66 composition. The measurements confirm that the material is electronic with a transference number of 0.9 - 0.95 under oxidizing environments. The level of the ionic conductivity is on the order of 10-2 - 10-3 S/cm at 900 C. The electrode impedance of Pt electrodes on Ca doped GT was monitored as a function of P0 2 and temperature for both 2% and 10% Ca levles in order to provide a reference for later studies with mixed conducting electrodes. The magnitude of the electrode conductance (I/Reectrode) is maximized under oxidizing regimes (Po 2 = 0.21 atm) and high temperatures (1000 0C) with a value of 0.1 S/cm. The conductance decreases in oxidizing environments as P0 2 decreases, reaching a minimum of 4 approximately 10- S/cm at intermediate Po 2 (~ 10 10 atm), which coincides with a switch from 0 2/Ar gas mixtures to CO/CO 2 gas mixtures. Under reducing conditions, the conductance increases as P0 2 decreases (CO concentration increases). The slopes of log-log plots of the electrode conductance vs. P0 2 were compared to those predicted by theoretical models of the electrode reaction for the purpose of identifying a likely microscopic rate limiting mechanism. In oxidizing environments, both materials are modeled as being largely limited by surface diffusion of adsorbed oxygen on the Pt surface, with a shift to control by dissociative adsorption of 02 on the Pt at lower temperatures for the 10% Ca material (not exhibited by the 2% Ca material). The mechanism under reducing environments is unclear, but seems likely to involve the reaction of CO on the Pt surface with oxygen to form CO 2. Thesis Supervisor: Harry L. Tuller Title: Professor of Ceramics and Electronic Materials 4 TABLE OF CONTENTS TITLE..............................................................................................................................................................1 ABSTRA CT..................................................................................................................................................3 TABLE O F CO N TEN TS...............................................................................................................................5 LIST O F FIG URE S........................................................................................................................................7 LIST O F TABLES........................................................................................................................................12 A CK N O WLED GM ENTS............................................................................................................................13 1 INTR O D U CTION ............................................................................................................................. 14 2 BA CK GR O UN D ................................................................................................................................ 19 2.1 APPLICATIONS UTILIZING M IXED CONDUCTORS........................................................................... 19 2.1.1 Solid Oxide Fuel Cell ............................................................................................................ 20 2 .1. 1.1 B asic P rin cip les ............................................................................................................................... 2 2 2.1.1.2 M aterials Requirements and Choices........................................................................................... 28 2.1.1.3 The M onolithic Pyrochlore SOFC............................................................................................... 32 2.1.2 Oxygen Separation Membranes ....................................................................................... 35 2.2 CONDUCTION IN OXIDES..................................................................................................................36 2.2.1 Ionic Conduction...................................................................................................................37 2.2.2 Electronic Conduction...........................................................................................................41 2.2.3 Mixed Ionic and Electronic Conduction............................................................................ 45 2.3 THE PYROCHLORE SYSTEM ........................................................................................................... 47 2.3.1 Crystal Structure ................................................................................................................... 48 2.3.2 Phase Stability.......................................................................................................................51 2.3.3 Conductive Properties...........................................................................................................56 2.3.3.1 The Gd 2Ti2O 7 - Gd 2Zr 2O7 System ...............................................................................................