<%BANNER%>

Strategies to Improve the Electrochemical Performance of Electrodes for Li-Ion Batteries

Permanent Link: http://ufdc.ufl.edu/UFE0044044/00001

Material Information

Title: Strategies to Improve the Electrochemical Performance of Electrodes for Li-Ion Batteries
Physical Description: 1 online resource (189 p.)
Language: english
Creator: Yang, Ming-Che
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: battery -- electrode -- film -- flake -- ion -- laser -- lithium -- pulsed -- thin
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Lithium-ion batteries are widely used in consumer market because of their light-weight and rechargeable property. However, for the application as power sources of hybrid electric vehicles (HEVs), which need excellent cycling performance, high energy density, high power density, capacity, and low cost, new materials still need to be developed to meet the demands. In this dissertation work, three different strategies were developed to improve the properties of the electrode of lithium batteries. First, the voltage profile and lithium diffusion battier of LiM1/2Mn3/2O4 (Ti, V, Cr, Fe, Co, Ni and Cu) were predicted by first principles theory. The computation results suggest that doping with Co or Cu can potentially lower Li diffusion barrier compared with Ni doping. Our experimental research has focused on LiNixCuyMn2-x-yO4 and we found that the amount of Cu will affect the lattice parameters, the cation disorder in the spinel lattice, the particle morphology, as well as the electrochemical properties. With detailed electrochemical measurements and in situ XAS experiments of LiNi0.25Cu0.25Mn1.50O4, the proposed explanation of the voltage profile by the first principles computation was proven, a second plateau at 4.2 V originates from the oxidation of Cu2+ to Cu3+ and the plateau at 4.95V originates from extra electrons provided by oxygen ions. Although the reversible discharge capacity decreases with increasing the Cu amount, optimized composition such as LiCu0.25Ni0.25Mn1.5O4 exhibits high capacities at high rates. Second, titanium dioxide flakes have been synthesized through a simple spreading method that is easily scalable. The calcined titanium dioxide flakes exhibit larger reversible charge/discharge capacity, better rate capability and excellent cycling stability compared to anatase titanium dioxide nanoparticles. The smaller grain size in the flakes most likely enables the formation of the new LiTiO2 phase during the lithiation process attributing to the improved reversible charge/discharge capacity. The larger surface area of the flakes leads to a larger contact area between electrode and electrolyte, shorter diffusion lengths for the transfer of ions and electrons results in the better rate capability. The cycling performance was significantly improved by the porous structure of the calcined titanium dioxide flakes. Finally, the all-solid state thin film batteries were deposited by pulsed laser deposition method to explore the intrinsic properties of electrode itself. The cathode of LiNi0.5Mn1.5O4 thin film was fabricated at 300mtorr 600 degree C on SiO2/Si and show good electrochemical properties. The thin film electrodes offer the ability to probe the surface of the material without the need of a conductive agent and polymer binder, typically used in composite electrodes. The results suggest that neither oxidation of PF6 to POF5 nor the decomposition of ethylene carbonate or dimethylene carbonate occurs on the surface of the spinel material. These results confirm the enhanced cycling stability and rate capability associated with the high voltage spinel material and suggests that the SEI layer forms due to the extra electrochemically inactive components in the composite electrode. LVSO ceramic was chosen as the solid electrolyte and deposited by PLD technique. The analysis results show the LVSO thin film exhibit uniform element distribution with exact stoichiometry and amorphous phase that is suitable for the application of all-solid state thin film batteries. Finally, two different thin films were fabricated as the anodes for all solid state thin film batteries. The Li4Ti5O12 thin film/Au deposited at 700 degree C shows excellent electrochemical properties but high temperature deposition limits its application because there are differences in thermal expansion coefficients along with inter-layer diffusion during the final layer deposition process. The TiO2 thin film deposited at 350-450 degree C show high capacity coupled with excellent cycling performance. The smaller grain size in the thin films most likely enables the formation of the new LiTiO2 phase during the lithiation process. The porous thin film deposited at 350 degreeC leads to a larger electrode/electrolyte contact area results in the better rate capability. Finally, the all solid state thin film battery with TiO2 / LiNi0.5Mn1.5O4 /L3.4V0.6S0.4O4/SiO2/Si stack was fabricated successfully only by PLD process
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Ming-Che Yang.
Thesis: Thesis (Ph.D.)--University of Florida, 2012.
Local: Adviser: Meng, Ying Shirley.
Local: Co-adviser: Singh, Rajiv K.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044044:00001

Permanent Link: http://ufdc.ufl.edu/UFE0044044/00001

Material Information

Title: Strategies to Improve the Electrochemical Performance of Electrodes for Li-Ion Batteries
Physical Description: 1 online resource (189 p.)
Language: english
Creator: Yang, Ming-Che
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: battery -- electrode -- film -- flake -- ion -- laser -- lithium -- pulsed -- thin
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Lithium-ion batteries are widely used in consumer market because of their light-weight and rechargeable property. However, for the application as power sources of hybrid electric vehicles (HEVs), which need excellent cycling performance, high energy density, high power density, capacity, and low cost, new materials still need to be developed to meet the demands. In this dissertation work, three different strategies were developed to improve the properties of the electrode of lithium batteries. First, the voltage profile and lithium diffusion battier of LiM1/2Mn3/2O4 (Ti, V, Cr, Fe, Co, Ni and Cu) were predicted by first principles theory. The computation results suggest that doping with Co or Cu can potentially lower Li diffusion barrier compared with Ni doping. Our experimental research has focused on LiNixCuyMn2-x-yO4 and we found that the amount of Cu will affect the lattice parameters, the cation disorder in the spinel lattice, the particle morphology, as well as the electrochemical properties. With detailed electrochemical measurements and in situ XAS experiments of LiNi0.25Cu0.25Mn1.50O4, the proposed explanation of the voltage profile by the first principles computation was proven, a second plateau at 4.2 V originates from the oxidation of Cu2+ to Cu3+ and the plateau at 4.95V originates from extra electrons provided by oxygen ions. Although the reversible discharge capacity decreases with increasing the Cu amount, optimized composition such as LiCu0.25Ni0.25Mn1.5O4 exhibits high capacities at high rates. Second, titanium dioxide flakes have been synthesized through a simple spreading method that is easily scalable. The calcined titanium dioxide flakes exhibit larger reversible charge/discharge capacity, better rate capability and excellent cycling stability compared to anatase titanium dioxide nanoparticles. The smaller grain size in the flakes most likely enables the formation of the new LiTiO2 phase during the lithiation process attributing to the improved reversible charge/discharge capacity. The larger surface area of the flakes leads to a larger contact area between electrode and electrolyte, shorter diffusion lengths for the transfer of ions and electrons results in the better rate capability. The cycling performance was significantly improved by the porous structure of the calcined titanium dioxide flakes. Finally, the all-solid state thin film batteries were deposited by pulsed laser deposition method to explore the intrinsic properties of electrode itself. The cathode of LiNi0.5Mn1.5O4 thin film was fabricated at 300mtorr 600 degree C on SiO2/Si and show good electrochemical properties. The thin film electrodes offer the ability to probe the surface of the material without the need of a conductive agent and polymer binder, typically used in composite electrodes. The results suggest that neither oxidation of PF6 to POF5 nor the decomposition of ethylene carbonate or dimethylene carbonate occurs on the surface of the spinel material. These results confirm the enhanced cycling stability and rate capability associated with the high voltage spinel material and suggests that the SEI layer forms due to the extra electrochemically inactive components in the composite electrode. LVSO ceramic was chosen as the solid electrolyte and deposited by PLD technique. The analysis results show the LVSO thin film exhibit uniform element distribution with exact stoichiometry and amorphous phase that is suitable for the application of all-solid state thin film batteries. Finally, two different thin films were fabricated as the anodes for all solid state thin film batteries. The Li4Ti5O12 thin film/Au deposited at 700 degree C shows excellent electrochemical properties but high temperature deposition limits its application because there are differences in thermal expansion coefficients along with inter-layer diffusion during the final layer deposition process. The TiO2 thin film deposited at 350-450 degree C show high capacity coupled with excellent cycling performance. The smaller grain size in the thin films most likely enables the formation of the new LiTiO2 phase during the lithiation process. The porous thin film deposited at 350 degreeC leads to a larger electrode/electrolyte contact area results in the better rate capability. Finally, the all solid state thin film battery with TiO2 / LiNi0.5Mn1.5O4 /L3.4V0.6S0.4O4/SiO2/Si stack was fabricated successfully only by PLD process
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Ming-Che Yang.
Thesis: Thesis (Ph.D.)--University of Florida, 2012.
Local: Adviser: Meng, Ying Shirley.
Local: Co-adviser: Singh, Rajiv K.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044044:00001


This item has the following downloads:


Full Text

PAGE 1

1 STRATEGIES TO IMPROVE THE ELECTROCHEMICAL PERFORMANCE OF ELECTRODES FOR LI ION BATTERIES By MING CHE YANG A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREM ENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Ming Che Yang

PAGE 3

3 To my lovely d ad and mom

PAGE 4

4 ACKNOWLEDGMENTS I would like to first thank my advisor Dr. Ying Shirley Meng, who shows her incredibly passion, and ene rgetic working attitude to support my research. I would also like to thank my co advisor, Dr. Rajiv Singh and my committee member Dr. Valentin Craciun, Dr. Kevin Powers Dr. David Arnold and Dr. Chang Yu Wu for their constructive comments. Moreover, I wou ld like to acknowledge all the past and current members in Dr. Bo Xu for the computation work, Dr. Kyler J. Carroll for XPS analysis, Chris Fell for the help of refinement work about XRD patterns Dr. Danijel Gostov ic, Dr. Dhamodaran Santhanagopalan and Mr. Thomas McGilvray for the in situ TEM work, I also want to thank Yuan Wang and Jui Ting Hung to build the LESC group in University of Florida. I also want to acknowledge Dr. Yang Yao Lee for the synthesis work of the TiO 2 flakes, Dr. Sushant Gupta at the U niversity of Florida and Dr. Xia Hui at the National University of Singapore for teaching me the operation of PLD system I am also grateful the assistantship of Mr. Ju Hsiang Cheng and Mr. Chun Jern Pan in Natio nal Taiwan University of Science &Technology for the in situ XAS work I am also appreciated the help of the staff of Materials Analyt ical Instrument Center(MAIC), Particle Engineering Research Center ( PERC) at University of Florida, Oak Ridge National la b oratory Brookhaven National Lab oratory and Argonne National Laboratory regarding the characterization work. I am extremely grateful to my parents for their love and unselfish support throughout my study in United States. Without them, this dissertation would have never b een accomplished.

PAGE 5

5 Finally I would like to acknowledge my funding source, the Florida Energy System Consortium through University of Florida for the financial support to make this research possible

PAGE 6

6 TABLE OF CONTENTS page A CKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ .......... 10 LIST OF FIGURES ................................ ................................ ................................ ........ 11 LIST OF ABBRE VIATIONS ................................ ................................ ........................... 15 ABSTRACT ................................ ................................ ................................ ................... 16 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 19 1.1 Challenges an d Motivation ................................ ................................ ................ 19 1.2 Research Objectives ................................ ................................ ......................... 20 1.3 Organization of the Dissertation ................................ ................................ ........ 21 2 BACKGROUND ................................ ................................ ................................ ...... 24 2.1 Principles of Li Ion Batteries ................................ ................................ ............. 24 2.2 Materials of Cathode Electrodes ................................ ................................ ....... 26 2.2.1 Layered LiCoO 2 ................................ ................................ ....................... 27 2.2.2 Spinel LiMn 2 O 4 ................................ ................................ ........................ 27 2.2.3 Olivine LiFePO 4 ................................ ................................ ....................... 28 2.3 Materials of Anode Electrodes ................................ ................................ .......... 29 2.3.1 Spinel Li 4 Ti 5 O 12 ................................ ................................ ........................ 29 2.3.2 Anatase and R utile TiO 2 ................................ ................................ .......... 31 2.3.3 Nanoalloys ................................ ................................ ............................... 31 2.4 Solid State Electrolyte ................................ ................................ ....................... 32 2. 4.1 LiPON Electrolyte ................................ ................................ .................... 33 2.4.2 LVSO electrolyte ................................ ................................ ..................... 33 3 EXPERIMENT PROCEDURES ................................ ................................ .............. 41 3.1 Principles of Sol Gel Synthesis Method ................................ ............................ 41 3.1.1 Selection of Precursor ................................ ................................ ............. 42 3.1.2 Conversion of Sol to Gel ................................ ................................ .......... 42 3.1.3 Drying of gels[50] ................................ ................................ .................... 43 3.1.4 Gel Densification During Sintering ................................ ........................... 46 3.2 Principles of Pulsed Laser Deposition ................................ ............................... 46 3.2.1 Mechanism of Pulsed Laser Deposition ................................ .................. 47 3.2.1.1 The interaction of the laser beam and target ................................ .. 47

PAGE 7

7 3.2.1.2 Plume formation and propagation ................................ .................. 49 3.2.1.3 Nucleation and growth ................................ ................................ ... 49 3.3 Set up of Pulsed Laser Deposition System ................................ ....................... 51 3.4 Experiment Procedure for PLD ................................ ................................ ......... 52 3.5 Elect rochemical Measurements ................................ ................................ ........ 52 3.5.1 Configuration of Cells ................................ ................................ .............. 52 3.5.2 Potentiostatic Intermittent Titration Technique (PITT) method ................ 53 4 ELECTRONIC, STRUCTURALAND ELECTROCHEMICAL Properties OF LiNi x Cu y Mn 2 x y O 4 (0
PAGE 8

8 6.1.3 Structural Changes of Delithiated Lithium Manganese Spinel; .............. 123 6.1.4 Chemical Dissolution Reaction; ................................ ............................. 124 6.1.5 Oxidation o f Electrolyte Components And Surface Film Formation ....... 124 6.2 Experiment Method ................................ ................................ ......................... 126 6.2.1 Synthesis of Powders ................................ ................................ ............ 126 6.2.2 Synthesis of Thin Films ................................ ................................ ......... 126 6.2.3 Characterization ................................ ................................ .................... 126 6.2.4 Electrochemistry ................................ ................................ .................... 127 6.3 Results and Discussion ................................ ................................ ................... 1 28 6.3.1 Characterization ................................ ................................ .................... 128 6.3.2 Electrochemical Measurement ................................ .............................. 129 6.3.3 X ray Photoelectron Spectroscopy ................................ ........................ 130 6.4 Summary ................................ ................................ ................................ ........ 132 7 SOLID ELECTROLYTE DEPOSITION BY PLD ................................ ................... 138 7.1 Introduction ................................ ................................ ................................ ..... 138 7.2 Experiment Method ................................ ................................ ......................... 139 7.2.1 Thin Films Preparation ................................ ................................ .......... 139 7.2.2 Characterization ................................ ................................ .................... 139 7.3 Results And Discuss ion ................................ ................................ .................. 140 7.4 Summary ................................ ................................ ................................ ........ 142 8 ELECTROCHEMICAL PROPERTIES OF TiO 2 and Li 4 Ti 5 O 12 THIN FILMS AND ALL SOLID STATE THIN FILM BATTERIES PREP ARED BY PULSED LASERDEPOSITION ................................ ................................ ............................ 148 8.1 Introduction ................................ ................................ ................................ ..... 148 8.2 Experimental Methods ................................ ................................ .................... 149 8.2.1 Preparation of Li 4 Ti 5 O 12 Materials ................................ .......................... 149 8.2.2 Preparation of TiO 2 Materials ................................ ................................ 150 8.2.3 Preparation of All So lid State Thin Film Battery ................................ .... 150 8.2.4 Characterization ................................ ................................ .................... 151 8.2.5 Electrochemical Testing ................................ ................................ ........ 151 8.3 Results and Discussion ................................ ................................ ................... 151 8.3.1 Materials Characterization and Electrochemical Properties o f Li 4 Ti 5 O 12 151 8.3.2 Materials Characterization of TiO 2 Materials ................................ .......... 153 8.3.3 Charge Discharge Studies of TiO 2 Thin Films ................................ ....... 155 8.3.4 Rate Ca pability Studies of TiO 2 Thin Films ................................ ............ 157 8.3.5 Cycling Performance of TiO 2 Thin Films ................................ ................ 158 8.3.6 Characterization of All Solid State Thin Film Battery ............................. 159 8.4 Summary ................................ ................................ ................................ ........ 159 9 CONCLUSIONS AND SUGGESTIONS FOR FUTURE WORK ............................ 172 9.1 Conclusion ................................ ................................ ................................ ...... 172 9.2 Suggestions for Future Work ................................ ................................ .......... 174

PAGE 9

9 LIST OF REFERENCES ................................ ................................ ............................. 176 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 189

PAGE 10

10 LIST OF TABLES Table page 2 1 Properties of various electrodes ................................ ................................ ......... 36 4 1 Chemical composition of LiNi x Cu y Mn 2 x y O 4 ................................ ........................ 80 4 2 I(400)/(311) Integrated intensity ratios of XRD spectrum of LiNi x Cu y Mn 2 x y O 4 .. 80 4 3 Rietveld Refinement Fitting Results of LiNi x Cu y Mn 2 x y O 4 ................................ .. 81 4 4 Charge capacity (mAh/g)of LiNi x Cu y Mn 2 x y O 4 at low voltage range .................. 82 5 1 Particle diameter statistics for F TiO 2 and CF TiO 2 ................................ .......... 113 5 2 Surface area and pore size of TiO 2 particles, F TiO 2 and CF TiO 2 ................... 113

PAGE 11

11 LIST OF FIGURES Figure page 1 1 Comparison of the different battery technologies in terms of volumetric and gravimetric energy density ................................ ................................ .................. 23 1 2 Cycle life characteristics of energy storage technologies ................................ .... 23 2 1 Schematics o f a typical Lithium ion cell ................................ .............................. 36 2 2 The crystal structure of Layered structure (Blue: transition metal ions;Red Li Ions) ................................ ................................ ................................ ................... 37 2 3 The crystal structure of Spinel struct ure(Blue: transition metal ions;Red Li Ions) ................................ ................................ ................................ ................... 37 2 4 The crystal structure of Olive structure(Green: transition metal ions;Orange: P ions;Red : Li Ions) ................................ ................................ ........................... 38 2 5 The crystal structure of anatase TiO 2 (Green: transition metal ions;Blue Li Ions) ................................ ................................ ................................ ................... 39 2 6 The crystal structure of rutile TiO 2 (Green: transition metal io ns;Blue Li Ions) ..... 39 2 7 The voltage profiles of various electrode materials ................................ ............. 40 2 8 Conductivity of Li 4 SiO 4 Li 3 VO 4 solid so lution. Numbers refer to % Li 4 SiO 4 content ................................ ................................ ................................ ................ 40 3 1 The routes that indicate the process of sol gel ................................ .................. 56 3 2 Stage of dry ing ................................ ................................ ................................ .... 56 3 3 Schematic illustration of shrinkage of the gel ................................ ...................... 57 3 4 Mechanism of PLD ................................ ................................ .............................. 57 3 5 Schematic illustration of nucleation of a cluster on a substrate surface .............. 58 3 6 Lambda physic LPX300.. ................................ ................................ .................... 58 3 7 Chamber design of the in house PLD system ................................ ..................... 59 3 8 The design of vacuum system ................................ ................................ ............. 59 3 9 The cells for electrochemical measurement ................................ ........................ 60 3 10 The equipment for electrochemical measurement ................................ .............. 60

PAGE 12

12 4 1 Structure of LiNi 0.5 Mn 1.5 O 4 ................................ ................................ ................... 82 4 2 Local environments of transition metal ion rings ring; ................................ ......... 83 4 3 Li diffusion barriers in LiM 0.5 Mn 1.5 O 4 (M= Cr, Fe, Co, Ni, Cu, Mn) calculated by GGA ................................ ................................ ................................ ............... 83 4 4 Calculated DOS of Cu in Li x Cu 1/2 Mn 3/2 O 4 ( (a) x=1 (b) x=1/2 (c) x=0 ) ................. 84 4 5 X ray diffraction patterns of LiNi x Cu y Mn 2 x y O 4 ................................ ................... 85 4 6 Refinement Results of LiNi x Cu y Mn 2 x y O 4 ................................ ............................ 86 4 7 Lattice parame ters of LiNi 0.5 x Cu x Mn 1.5 O 4 (x=0,0.05,0.1,0.25,0.5) ....................... 87 4 8 FESEM images of LiNi x Cu y Mn 2 x y O 4 ................................ ................................ .. 88 4 9 Charge / discharge c urves of LiNi x Cu y Mn 2 x y O 4 ................................ .................. 89 4 10 Rate Capability of LiNi x Cu y Mn 2 x y O 4 ................................ ................................ ... 90 4 1 1 The cycling performance of of LiNi x Cu y Mn 2 x y O 4 ................................ ................ 90 4 12 Charge curve of LiNi 0.25 Cu 0.25 Mn 1.5 O 4 for XANES experiment ............................. 91 4 13 The Cu K XANES spectra of LiCu 0.25 Ni 0.25 Mn 1. 5 O 4 measured as a function of voltage ................................ ................................ ................................ ................ 92 4 14 The Ni K XANES spectra of LiCu 0.25 Ni 0.25 Mn 1.5 O 4 measured as a function of voltage ................................ ................................ ................................ ................ 93 4 15 The Mn K XANES spectra of LiCu 0.25 Ni 0.25 Mn 1.5 O 4 measured as a function of voltage ................................ ................................ ................................ ................ 94 4 16 XPS plots of the Cu2p, Ni2p and Mn2p region scans for LiCu 0.25 Ni 0.25 Mn 1.5 O 4 ... 96 4 17 The XRD pattern of the LiNi 0.25 Cu 0.25 Mn 1.5 O 4 sample and t LiNi 0 5 Mn 1.5 O 4 sample at different voltage ................................ ................................ ................. 97 4 18 Diffusion Coefficie nt of LiNi 0.45 Mn 1 55 O 4 ,LiNi 0.5 Mn 1.5 O 4 and LiNi 0.25 Cu 0.25 Mn 1.5 O 4 ................................ ................................ ........................... 98 5 1 TG/DTA measurements of F TiO 2 ................................ ................................ ..... 114 5 2 X ray dif fraction patterns of TiO 2 samples. ................................ ........................ 114 5 3 FESEM images of TiO 2 samples ................................ ................................ ....... 115 5 4 Particle size distribution for F TiO 2 an d CF TiO 2 by laser diffraction ................. 116

PAGE 13

13 5 5 HR TEM images of TiO 2 samples (the SAD pattern as inset). .......................... 117 5 6 UV visible diffus e reflectance spectra of F TiO 2 and CF TiO 2 ........................... 118 5 7 Charge/discharge curves of TiO 2 samples ................................ ....................... 119 5 8 Neutron diffraction patterns of LixTiO 2 for different particle sizes andt he XRD patterns of Li 0 5 TiO 2 LiTiO 2 simulated by Crystal Diffraction ........................... 120 5 9 Rate capability comparison of CF TiO 2 and TiO 2 particles ................................ 121 5 10 The variation of specific discharge capacity with respect to the cycle number of CF TiO 2 and particles ................................ ................................ ................... 121 5 11 Lit hiation mechanism of bulk and nanosized material ................................ ....... 122 6 1 SEM images of thin film (a,b) and powder (c,d) NiMn spinel electrode materials. ................................ ................................ ................................ .......... 133 6 2 XRD patterns for thin film and powder composite NiMn spinel electrodes and t he refinement pattern of the composite powder NiMn ................................ ..... 134 6 3 Charge/discharge profiles an d rate testing for NiMn spinel powder (a,b) and thin film (c,d) electrodes. ................................ ................................ .................. 135 6 4 Discharge capacity performance as a function of cycle number for thin film and powder electrodes. ................................ ................................ .................... 135 6 5 XPS data collected using a Mg anode on pristine and post electrochemically cycled powder composite (a c) and (d f) thin film NiMn spinel electrodes. ....... 136 6 6 XPS plots of the Mn2pand O1s region scans for the composite (a,c) and thin film (b,d) electrodes. ................................ ................................ ......................... 137 7 1 XRD pattern of LVSO target ................................ ................................ .............. 143 7 2 XRD pattern of LVSO thin film deposited at different pressure ......................... 143 7 3 SEM of LVSO thin films deposited on different pressure ................................ .. 144 7 4 EDS of LVSO thin films/stainless steel deposited on 5 mtorr ............................ 144 7 5 The element distribution of LVSO investigated b y LEAP ................................ .. 145 7 6 XPS results of LVSO thin films ................................ ................................ .......... 146 7 7 SEM pictures of LVSO after 650 C ................................ ................................ ... 147

PAGE 14

14 8 1 XRD pattern of Li 4 Ti 5 O 12 powder synthesized by sol gel method and solid state method ................................ ................................ ................................ ..... 161 8 2 XRD pattern of L i 4 Ti 5 O 12 thin film deposited on SiO 2 /Si ................................ ..... 161 8 3 FESEM images of Li 4 Ti 5 O 12 thin film deposited at different temperatures on Au ................................ ................................ ................................ ..................... 162 8 4 Charge/Discharge curves of Li 4 Ti 5 O 12 ................................ ............................... 163 8 5 X ray diffraction patterns of the TiO 2 target. ................................ ...................... 164 8 6 X ray dif fraction patterns of TiO 2 thin films deposited onto a SiO 2 /Si substrate 164 8 7 X ray diffraction patterns of TiO 2 thin films deposited onto a stainless steel substrate ................................ ................................ ................................ ........... 165 8 8 FESEM images of TiO2 thin films deposited at different temperature of thin ................................ ................................ ................... 166 8 9 XPS plots of the Ti2p(a) and O1s(b) region scans for the TiO 2 thin films .......... 167 8 10 First charge/ discharge curves of TiO 2 thin films ................................ ................ 168 8 11 Rate capability comparison of TiO 2 thin films deposited at diffe rent temperatures. ................................ ................................ ................................ ... 169 8 12 Cycling performance of TiO 2 thin films deposited at different temperatures. ..... 170 8 13 Cross secti onal SEM image of the all solid state battery prepared by PLD and subsequent FIB processing. ................................ ................................ ...... 171 8 14 TEM image of the all solid state battery prepared by PLD and diffraction pattern of all the three layers ................................ ................................ ............ 171

PAGE 15

15 LIST OF ABBREVIATION S BET brunauer, emmett and t eller EDS e nergy dispersive spectroscopy FIB focus ion beam ICP induced couple d plasma LEAP l ocal electrode atom probe LIPON lithium phosphorous oxynitride LVSO Li 2 O SiO 2 V 2 O 5 PLD pulsed laser deposition PITT potentiost atic intermittent titration technique PVDF p olyvinylidene fluoride SAD select area diffraction SEM scanning electron microscope SPM scanning probe microscopy TEM t ransmission electron microscopy TG DTA thermogravimentic differential thermal analysis VASP v ienna ab initio simulation package XANES x ray absorption near edge s tructure XPS x ray photoelectron spectroscopy XRD x ray diffraction pattern

PAGE 16

16 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy STRATEGIES TO IMPROVE THE ELECTROCHEMICAL PERFORMANCE OF ELECTRODES FOR LI ION BATTERIES By Ming Che Yan g August 2012 Chair: Ying Shirley Meng Co chair : Rajiv Singh Major: Materials Science and Engineering Lithium ion batteries are widely used in consumer market because of their light weight and rechargeable property. However, for the application as power sources of hybrid electric vehicles (HEVs), which need excellent cycling performance, hig h energy density, high power density, capacity, and low cost, new materials still need to be developed to meet the demands. In this dissertation work three different strategies were developed to improve the properties of the electrode of lithium batteries. First, the voltage profile and lithium diffusion battier of LiM 1/2 Mn 3/2 O 4 ( M=Ti, V, Cr, Fe, Co, Ni and Cu) were predicted by first principles theory. The computatio n results suggest that doping with Co or Cu can potentially lower Li diffusion barrier compared with Ni doping Our experimental research has focused on LiNi x Cu y Mn 2 x y O 4 (0
PAGE 17

17 proven, a second plateau at 4.2V originates from the oxidation of Cu 2+ to Cu 3+ and the plateau at 4.95V originates from extra electrons provided by oxygen ions. Although the reversible discharge capacity decrea ses with increasing the Cu amount, optimized composition such as LiCu 0.25 Ni 0.25 Mn 1.5 O 4 exhibits high capacities at high rates. Second, titanium dioxide flakes have been synthesized through a simple spreading method that is easily scalable. The calcined titanium dioxide flakes exhibit larger reversible charge/discharge capacity, better rate capability and excellent cycling stability compared to anatase titanium dioxide nanoparticles The smaller grain size in the flakes most likely enables the formation o f the new LiTiO 2 phase during the lith i ation process attributing to the improved reversible charge/discharge capacity The larger surface area of the flakes leads to a larger contact area between electrode and electrolyte shorter diffusion lengths for the transfer of ion s and electron s results in the better rate capability The cycling performance was significantly improved by the p orous structure of the calcined titanium dioxide flakes Finally, the all solid state thin film batteries were deposited by pu lsed laser deposition method to explore the intrinsic properties of electrode itself. The cathode of LiNi 0.5 Mn 1.5 O 4 thin film was fabricated at 3 00mtorr 2 /Si and show good electrochemical properties. The thin film electrodes offer the ability t o probe the surface of the material without the need of a conductive agent and polymer binder, typically used in composite electrodes. The results suggest that neither oxidation of PF 6 to POF 5 nor the decomposition of ethylene carbonate or dimethylene carb onate occurs on the surface of the spinel material. These results confirm the enhanced cycling stability and rate capability associated with the high voltage spinel material and suggests that the

PAGE 18

18 SEI layer forms due to the extra electrochemically inactive components in the composite electrode. LVSO ceramic was chosen as the solid electrolyte and deposited by PLD technique The analysis results show the LVSO thin film exhibit uniform element distribution with exact stoichiometry and amorphous phase that is suitable for the application of all solid state thin film batteries. Finally, two different thin films were fabricated as the anode s for all solid state thin film batteries. T he Li 4 Ti 5 O 12 thin film /Au deposited at 700C shows excel lent electrochemical pro perties but high temperature deposition limits its application because there are differences in thermal expansion coefficients along with inter layer diffusion during the final layer deposition process. The TiO 2 thin film deposited at 350 show high c apacity coupled with excellent cycling performance. The smaller grain size in the thin films most likely enables the formation of the new LiTiO 2 phase during the lith i ation process. The leads to a larger electrode/electr olyte contact area results in the better rate capability Finally, the all solid state thin film battery with TiO 2 / LiNi 0.5 Mn 1.5 O 4 /L 3.4 V 0.6 S 0.4 O 4 /SiO 2 /Si stack was fabricated successfully only by PLD proces s.

PAGE 19

19 CHAPTER 1 INTRODUCTION 1.1 Challenges and Motivation Recent increases in the demand for oil, associated with the price increases and environmental issues continue to exert pressure on an already stretched world energy in frastructure. Significant efforts have been made in the development of renewable energy technologies such as solar cells, fuel cells and biofuels. E lectrochemical energy production is also considered as a different energy source such as t he lithium ion ba ttery system [1 4] Lithium ion batteries are widely used in consumer market because of its light weight and rechargeable property that is a suitable energy source for the laptop and camera application Recently t his system also ha s been considered as the power sources for hybrid electric vehicles (HEVs). Hybrid Electric Vehicles (HEVs) are a combination of a typical Internal Combustion Engine (ICE) vehicle and a Battery Electric Vehicle (BEV) with an electric mot or capable of supplying auxiliary power to the drive train. The combination provides better gas mileage compared to an ICE vehicle. The g asoline consumption of Hybrid Electric Vehicles can diminish about 30% compare to usual vehicle. Plug in Hybrid Ele ctric Vehicles (PHEVs) [5] take this concept one step further by adding additional batteries to the design. This allows the vehicle to be charge d at night and be powered solely from stored electric energy during the day. By using Plug in Hybrid Electric Vehicles the various energy sources can be used Li Ion battery has a higher energy density than a Ni Metal hydrid e system as show n in Figure1 1 that maybe an excellent candidate for the storage system of PHEV. However, for the commercial

PAGE 20

20 application, the battery system need to supply 4000 deep discharge cycles The 4000 discharge cycles only can reach if the batteries discharge around 50 % depth o f discharge as shown in Figure 1 2 Moreover, energy density, power density (rate capability), operating voltage, capacity, cost of materials, and safety issues of the lithium ion battery system still limits its commercial application. In this research, th ree different strategies were developed to improve the electrochemical performance of electrodes for LI ion batteries. 1. 2 Research Objectives The hypotheses of this study are set as follows In order to improve the rate capability of the high voltage spin el material, the electronic, structural and electrochemical properties of LiM 1/2 Mn 3/2 O 4 (M=Ti, V, Cr, Fe, Co, Ni and Cu) was predicted by first principles theory first. A series of bi doped spinel oxides of LiNi x Cu y Mn 2 x y O 4 (0
PAGE 21

21 (XAS) and electrochemical measurements including potentiostatic intermittent titration technique (PITT) and compare with the computation results. The crystal structure, surface area, pore size, electronic propertie s and electrochemical properties are compared for flakes and nanoparticles N ano size effects of flakes are di scussed i ncluding t he true size effects, which involve changes of actual materials, including the change of structure and phase stability, electr onic band gap and intrinsic diffusivity and the trivial size effects, which effect is largely relied on the surface to volume ratio the shape and the pores of the nanomaterials The electrochemical properties and the surface stability after cy c ling of LiM n 2 x Ni x O 4 (x= 0.5, 0.45) thin film and composite electrodes are compared. The thin film electrodes offer the ability to probe the surface of the material without the need of a conductive agent and polymer binder Solid state electrolyte s have been also f a bricated by PLD. The influence of deposition conditions on the microstructure, surface morphology and the electrochemical properties of the of TiO 2 and Li 4 Ti 5 O 12 thin film electrodes have been investigated 1.3 Organization of the Dissertation This disser tation consists of night chapters In Chapter 1, the problem and motivation of this research is addressed. Chapter 2 begins with the fundamental aspects of lithium ion batteries, and advances with the criteria for the commercial application. The materials u sed for lithium ion batteries are also introduced.

PAGE 22

22 Chapter 3 focus on the theory of sol gel method, principles of PLD, the setup of the PLD system and the electrochemical measure methods In Chapter 4, the electronic, structural and electrochemical propert ies of LiM 1/2 Mn 3/2 O 4 (M=Ti, V, Cr, Fe, Co, Ni and Cu) was predicted by first principles theory E xperimental research has focused on LiNi x Cu y Mn 2 x y O 4 (0
PAGE 23

23 Figure1 1 Comparison of the different battery technologies in terms of volumetric and gravimetric energy density [4] Figure1 2 Cycle l ife c haracteristics of e nergy s torage t echnologies [5]

PAGE 24

24 CHAPTER 2 BACKGROUND The purpose of this chapter is to provide readers with essential background knowledge of lithium ion batteries. Hence an overview of the principles of Li ion batteries and the materials used for Lithium batteries are presented. The chapter begins with the fundamental aspects of Lithium ion batteries, and advances with the criteria for lithium ion batteries for the commercial application. Since this research in volved the cathode, solid electrolyte and anodes, the various materials used for different part was also introduced 2. 1 Princ i ples of Li Ion Batteries Batteries are systems for electrochemical energy storage and conversion. In batteries, electrical ene rgy is generated by conversion of chemical energy by redox reactions at the anode and cathode. Because the reactions occur at anode usually take place at lower potentials than at the cathode, the terms negative and positive electrodes are used. Batteries a re closed systems that energy storage and conversion occur at the same compartment that means the electrodes are not only the charge transfer medium but also an active element. Various batteries were developed, such as Lead acid, Ni Cd, Ni matal hydride an d Lithium ion batteries. Among all batteries system s Li Ion batteries can supply higher energy density because Lithium is the most electropositive ( 3.04V v. s. standard hydrogen electrode) and lightest metal (M=6.94 g/mole) [4] A n electrochemical cell of batteries is composed of three parts, an anode (negative), a cathode (positive) and an electrolyte. During charge and discharge process, Li ions transfer across an electrolyte between the anode and cathode with oxidation and reduc tion occurring at the two electrodesas shown in Figure 2 1. The majority of Li ion ( d )

PAGE 25

25 batteries use liquid electrolytes containing a lithium salt, such as LiPF 6 LiBF 4 LiClO 4 LiBC 4 O 8 which dissolves in a mixture of organic alkyl carbonat e solvents like eth ylene(EC), dimethyl(DMC), diethyl(DEC) and ethylmethyl(EMC) carbonate that has high electronic and ionic conductivity. The two electrodes usually are separated by a porous plastic film (separator) to prevent the contact of electrode. The Li ion technolog ies started with the finding of intercalation compounds, such as Li x MO 2 that proposed by Goodenough [6] : In 1991, Sony commercialize the LiC 6 /Li 1 x CoO 2 [4] cells The energy storage in the cells can be expressed by following reaction: Anode: (2 1) Cathode ; ; ( 2 2) Cell reaction: ( 2 3) The li ion cells operate with 3.7 V with a capacity with 150mAh/g. The basic thermodynamic equations for a reversible electrochemical transformation are given as equation ( 2 4 ) (2 4) Where n is the number of electrons transferred, F is Faraday constant and E is the voltage of the cell. so the voltage of a lithium ion cell can be decided by equation (2 5): (2 5)

PAGE 26

26 For the commercial Li ion battery, the materials and reaction of the cell need to satisfy following criteria [2] : Mechanical and Chemical Stability : t he electrode must keep their mechanical properties and their chemical composition, structure and surface over time and temperature in operation environment. Mechanical and chemical stability limitations come from reaction of electrolyte, phase change, corrosion and poor conductivity of materials in the dis charge state. High Energy Density : t he voltage of the cell should achieve the maximum amount by selecti ng the anodes with the high lithium chemical potential or the cathodes with the low lithium chemical potential Moreover, the electrode material should a llow a large amount of lithium to insert and maximize the capacity. High power density : t he electrode should have high electronic conductivity and ionic conductivity in order to support high current density. Temperature range of operation : t he operation temperature of battery is between 50 to 80C for automotive and military applications. For a general application, the range is between 0C to 40C .In the hot summer or cold winter, the performance of batteries will face the challenge. The organic electr olyte and interface formed at interface limited the performance of battery operated at extreme condition. Shape of the discharge curve : f or operation of an electronic device, a flat voltage plateau is necessary. A sloping curve is preferred for applicatio ns when determining the state of charge is important. The shape of curve is determined by the one(slopping) or two phase reactions (plateau). Cost : t he cost of battery is determined by the materials used and the manufacturing and assembling process, The ch eaper Fe, Ni and Mn materials are preferred to replace the Co based material with high price. Safety : a ll consumer and commercial batteries must be safe in the normal operating environment or the mild abuse conditions. The battery should not leak, vent h azardous materials or explode. Cycle life : for the commercial application, the battery system need to supply 4000 deep discharge cycles This requires a very robust system and reversible electrode reactions. There should be min imal side reactions and stru cture transformation during the charg e /discharge process. 2. 2 Materials of Cathode E lectrode s Li intercalation compounds are ideal electrode that can accommodate lithium over large concentration ranges Good cathode electrode should have low lithium

PAGE 27

27 diffus ion barrier, high electronic conductivity, low lithium chemical potential, high capacity, good structure stability that can prevent significant structural change after charge/discharge process. The lithium transition metal oxides own above advantages that are widely used in Li ion Battery field [3] 2.2.1 Layered LiCoO 2 Layered l ithium transition metal oxides : LiMO 2 (M=Co,Ni or Mn) are intercalation compounds that have been widely studies as cathodes for lithium ion batteries. LiCoO 2 [6] has the NaFeO 2 structure with the oxygen in cubic closed packed arrangement. A typical layered structure of transition metal oxides can be expressed as Figure 2 2. The lithium and cobalt ions occupy octahedral sites in alternating layer .On complete removal of the lithium, the oxygen layers rearrange themselves to give hexagonal close packing of the oxygen in CoO 2 [7] .The theoretical capacity of LiCoO 2 is 272mAh/g but several phase transition happen when the cell is charged above 4 .2V which are accompanied significant structure change [7, 8] For the commercial application, the capacity of LiCoO 2 is 140mAh/g when cell charge to 4.2V that only extract about 0.5 mole lithium Although the LiCo O 2 cathode dominates the rechargeable lithium battery market now, there is a limited of cobalt. The high price of the cobalt and safety issue make the material only can be used for small cells. For application of the large scale such as plug in hybrid car an alternative cathode develop ment is necessary. 2.2.2 Spinel LiMn 2 O 4 Thackeray et al [9] proposed the spinel cathode LiMn 2 O 4 first and the materials has been extensively developed by Bellcore [10, 11] labs. LiMn 2 O 4 adopt the spinel struc ture with space group Fd3 m as shown in Figure 2 3. The edge shaped octahedral Mn 2 O 4 structure supply the stability and possesses a series of intersecting tunnels,

PAGE 28

28 formed by the face sharing of tetrahedral lithium (8a) sites and empty octahedral (16c) sit es, When Li ions diffuse in this structure, it first moves from 8a site to the neighboring 16c site, and then to the next 8a site in such a way that enables three dimensional lithium diffusion. The discharge of LiMn 2 O 4 proceeds in two steps, one around 4V and one around 3V. To avoid Jahn Tell er distortion, only the 4V plateau was used with a capacity about 120mAh/g. However, LiMn 2 O 4 tends to exhibit capacity fade even in the 4V region, particularly at elevated temperatures. Factors such as Mn dissolution [12] into the electrolyte and the development of micro strains [13] during cycling have been suggested t o be t he main sources of capacity fading The poor cycling performance could be improved by partial substitution of Mn with other metals, an approach of making the LiM x Mn 2 X O 4 (M = Co, Mg, Cr, Ni, Fe, Al, Ti, Cu, Zn etc.) electrode material [14 16] It has been found that a higher voltage plateau (> 4V) accompanies some transition metal doping 2.2.3 Olivine LiFePO 4 In 1997, Padhi et al [17] discover the excellent electrochemical properties of the olivine phase. This is the first cathode material with low, abundant amount and also environment friendly. The crystal structure of LiFePO 4 composed of LiO 6 and FeO 6 octahedral and PO 4 tetrahedral as described in the space group Pmna with the orthorhombic unit cell. LiO 6 octahedral share edges in order to form LiO 6 chains runni ng parallel to (111) direction, where the one dimensional lithium i on mobility in this framework is high as shown in Figure 2 4. The discharge plateau of LiFePO 4 is about 3.4V with capacity about 170 mAh/g and owns excellent cycling performance T he low intrinsic electrical conductivity of the material lead the theoretical capacity onl y can be

PAGE 29

29 achieved at low rate capability [18] or a high temperature [19] T he electrochemical properties of the material can be improved by t he carbon coating though enhanced electrical conductivity [ 20] Ch i ang et al [21] get excellent electrochemical behavior when the LiFePO 4 doped with different elements, such as niobium, which increased the conductivity by orders of ma gnitude. Malik et al [22] prove the dif fusion constant will increase with the decreasing of particle size for LiFePO 4. The existence of defects in the diffusion path blocks the diffusion of the lithium ion in bulk material while smaller particles with small amount defects facilitate the migra tion of lithium ions Critical to the use of LiFePO 4 is its reactivity and thermal stability [19] Fo r its charged product, FePO 4 will transfer to spinel phase under high pressure 2. 3 Materials o f A node E lectrode s Good anode electrode should have low lithium diffusion barrier, high lithium chemical potential, high electronic conductivity, high capacity, a nd good structure stability that can prevent significant structural change after charge/discharge process The graphite was used as anode for lithium battery now because its low cost and the low operation voltage. The formation of LiC 6 produces a maximum ca pacity about 372mAh/g. However the chemical potential of graphite is closed to 0V versus lithium, which causes serious safety issue. Diverse anodes were developed to solve the problem. 2.3.1 Spinel Li 4 Ti 5 O 12 Spinel Li 4 Ti 5 O 12 has been investigate as a prom oting anode material for lithium ion batters because of the extremely flat discharge and charge plateaus at 1.55V, excellent cycla bility, zero strain insertion and stable in moisture environment.

PAGE 30

30 Li[Li 1/3 Ti 5/3 ]O 4 has a defect spinelframework structure (Fd3 m:a=8.36A) For conventional spinel material, such as LiMn 2 O 4 c athode materials, extra lithium can be intercalated into the spinel LiMn 2 O 4 up to a maximum composition Li 2 Mn 2 O 4 with the cubic to tetragonal phase transformation The cubic to tetragonal transition, often referred as Jahn Teller distortion, is accompanied by 6.5% increase in the unit cell volume, which is too severe for the electrodes to maintain structural reliability during charge/discharge cycling and will result in rapid capacity fad i ng However, for Li[Li 1/3 Ti 5/3 ]O 4 materials, X ray diffraction measurements indicate that the lattice dimension did not change during the reaction and capacity fading due to the damage to the matrix material was not observed after 100 cycles. Ohzuku et al proposed a mechanism for this zero strain material [23] O 4 O 4 (2 6) 8(a) 16(d ) 32(e) 8(a) 16(c)16(d ) 32(e) O 4 (2 7) 16(c)16(d ) 32(e) The other advantage of Li[Li 1/3 Ti 5/3 ]O 4 is its higher voltage compared to lithium metal. I t could be interesting to shift the voltage window from 0 4V to 1 5V range with some of the high voltage cathode materi als(>4.5V). The zero strain during cycling, stable in moisture environment and flat discharge and charge plateaus at 1.55V make Li[Li 1/3 Ti 5/3 ]O 4 a good anode candidate for thin film battery.

PAGE 31

31 2.3.2 Ana tase and R utile TiO 2 For TiO 2 eight polymorphs are rec ognized but only three structures were discussed for their electrochemical performance : anatase(I4 1 /amd) as show in Figure 2 5 rutile(space group P4 2 /mmm) as shown in Figure 2 6, and TiO 2 B(bronze C2/m) Rutile are recognized as the most thermodynamic stable structure of eight polymorphs but only anatase [24] and TiO 2 (B) [25, 26] exhibit theoretical capacity ,. Due to the safety against overcharging and stable voltage plateau at 1.78V that corresponds to the occurrence of the two phase equilibrium of Lithium poor (Li 0.01 TiO 2 ) and Lithium rich (Li 0.6 TiO 2 ) domains [27] anatase titanium dioxide is a promis ing negative electrode material The theoretical capacity of anatase is 168mAh/g that corresp onding to insert half amount lithium. However various nanostructures had also fabricated and the improvement on electrochemical properties ha s been shown [28, 29] The smaller grain size in the anatase TiO 2 most li kely enables the formation of the new LiTiO 2 phase during the lith i ation process attributing to the improved reversible charge/discharge [30] Similar results were shown in nanosized rutile. Hu et al [31] proved that the capacity can been improved with the decreasing of the particle size The short diffusion length for both electrons and ions lead to the improved capacity. 2.3.3 Nanoalloys Lithium metal alloy have the ability to storage large amounts of lithium because it low molecular weight. Such alloys have capacity that exceed than graphite anode, such as Li 4.4 Sn(993 mAh/g ) and Li 4.4 Si(4200 mAh/g) [32] However, the large volume expan sion cause cracking and pulverization of the metal particles during the alloy formation inducing the deterioration of the electrode and limit the cycling performance.

PAGE 32

32 There have been many attempts to overcome the problem. First, tin based, SnO X [33] amorphous oxide was proposed: a irreversible reaction happen as follows [34] : Sn+2 ( 2 8 ) the fine Sn particles were we ll dispersed in the Li 2 O medium Second, the reversible mechanism was proposed ( 2 9 ) The free degree supplied by the Li 2 O mediu m provide s accommodation of the stresses formed during the alloying process Nanowire anodes also facilit ate the strain relaxation and maintain good electrical connections both maximize the capacity with excellent cycling performance. The Si [36] and Ge [37] nanowires were synthesized by Chan et al with capacity 3000 and 1141mAh/g after twenty cycles. The major properties of the cathodes and anodes used in this research are summarized in Table 2 1 and the voltage profile s of the materi als are shown in Figure 2 7 2.4 Solid State Electrolyte An proper solid electrolyte material must satisfy following requirements [38] : High lithium ionic conductivity of t he solid state electrolyte Low electrical conductivity No chemical reaction between a node cathode materials and solid state electrolyte. H ig h electrochemical voltage window E asily fabricated and high stability. In order to investigate the electrode/electrolyte interface by TEM ex situ or in ex situ, some extra condition must be satisfied

PAGE 33

33 Dense structure Stable in the electrolyte beam Stable in the ion beam (in the cutting process of Focus Ion Beam(FIB ) Solid electrolytes can be divided into sulphide based and oxide based compound or amorphous and crystalline structure. Although the ionic conductivity of the sulphide glass is higher than o xide glass, the less hygroscopic properties of oxide glass and high stability in moisture make oxide glass get more attention Moreover, t he amorphous electrolyte can supply isotropic lithium transportation path [39] O nl y limited literature s have been reported for the solid electrolyte films. R ecent research has focused on two: Li 2 O SiO 2 V 2 O 5 (LVSO) and Li 2 O P 2 O 5 +N 2 (LiPON) because of their high ionic conductivity and easily preparation 2.4.1 LiPON E lectrolyte Bates et a l [40] deposite d LiPON by Li 3 PO 4 with N 2 atmosphere using sputtering system. The Lithium Phosphorus Oxy Nitride (LIPON) films are widely used in solid state lithium batteries now because of its low reactivity with lithium matel [4 1] and the high voltage window make it is possible to use various anode electro de However, the lower conductivity and unstable under the electron beam and ion beam exposure limit its application. Moreover, the low deposition rate hinder the commercial ap plication if the film fabricated by sputtering PLD with high energy can deposit the films with high rate [42] However, rough surface morphology and high fluence 15 J/cm 2 requirement also make it is impossible to fabrica te the films by PLD. 2.4.2 LVSO electrolyte Other candidate is the Lithium Vanadium Silicon Oxide (LVSO) [43] .This compound has relatively higher ionic condu ctivity and dense morphology In order to increase the ionic conductivity of solid electrolyte Li 4 XO 4 (X = Si, Ge) tetrahedral

PAGE 34

34 structures such as Li 3 PO 4 Li 3 AsO 4 Li 3 VO 4 and Li 2 ZnGeO 4 (LISICON) were add to Li 4 XO 4 to form a solid solution The conductivity can be improved by formation of lithium vacancies or interstitials. The conductivity data were shown as fol lowing plot(Figure 2 8 ) [43] with different composition. The experiment result show s 0.4Li 4 SiO 4 0.6Li 3 VO 4 has a maximum conductivity The A facile explanation for the increase of the conductivity can be expressed as follows: The substitution reaction Li + + Si 4+ ---V 5+ will induce extra lithium vacancy The concentration of lithium vacancies is small in Li 4 SiO 4 compound B y adding the Li 3 VO 4 to form a solid solution lithium vacancies will increase Structural differences between Li 4 SiO 4 and Li(P, As, V) are also considered to be unimportant because both structure have a similar oxide packing and should su pply same 3dimensional ionic transportation path The mechanism of conduction can be expressed in terms of the following equation (2 10) o k, T and G are, the ionic jump distance, a constant depending on the number of possible jumps available to a particular ion, the vibrational frequency of the N is the number of equivalent sites and c is the concentration of lithium ions on these sites if c = l/2, the product has a maximum value. The activation energy for conduction decreases to a minimum at intermediate solid solution compositions. The reason s for this are not known. Possibly, cooperative interactions between the mobile Li + ions could act to reduce the activation energy for

PAGE 35

35 migration, Second, the Arrhenius pre exponential factor( c(l c.)), passes through a maximum at intermediate composition s. This is attributed to a variation with composition in the number of mobile Li+ ions. LVSO thin films were first deposited by Hideaki [44] b y using rf sputtering method it is partially crystallized even deposited at room temperature and display high ionic conductivity and low electronic conductivity. Zhao et al [45] also deposited the LVSO thin film by PLD successfully The LVSO thin film deposited by PLD shows amorphous state even deposited at 300 exhibiting relatively lower ionic cond uctivity and lower electronic conductivity compared to the thin films deposited by sputtering E lectrolyte decomposition were not observed in the potential range of 1.75 to + 1.75V, suggesting that this electrolyte thin film exhibits a electrochemical wind ow at least more than 3.5V. The electrochemical window of Li V Si O thin film electrolyte seems to be narrow comparing with the LIPON electrolyte.

PAGE 36

36 Table 2 1 Properties of various electrodes LiCoO 2 LiFePO 4 LiMn 2 O 4 LiNi 0.5 Mn 1.5 O 4 Li 4/3 Ti 5/3 O 4 TiO 2 Vol tage 4V 305V 4V 4.7V 1.5V 1.75V Cycling Good Good Poor Good Enough? Excellent Poor Excellent Safety Dangerous Safe Safe Safe Safe Safe Capacity 140 mAh/g 162mAh/g 148mAh/g 146 mAh/g 170mAh /g 165 330mAh/g Price High Moderate Low Low Low Low St ructure Layer Olivine Spinel Spinel Spinel Anatase, Rutile Figure 2 1. Schematics o f a typical Lithium ion cell

PAGE 37

37 Figure 2 2 The crystal structure of Layered structure (Blue: transition metal ions;Red Li Ions) Figure 2 3 The crystal structur e of Spinel s tructure (Blue: transition metal ions;Red Li Ions)

PAGE 38

38 Figure 2 4 The crystal structure of Olive structure (G reen : transition metal ions ; Orange: P ions;Red : Li Ions)

PAGE 39

39 Figure 2 5 The crystal structure of anatase TiO2 (Green: transition me tal ions;Blue Li Ions) Figure 2 6 The crystal structure of rutile TiO 2 (Green: transition metal ions;Blue Li Ions)

PAGE 40

40 Figure 2 7 The voltage profile s of various electrode material s Figure 2 8 Conductivity of Li 4 SiO 4 Li 3 VO 4 solid solution. Numbers refer to % Li 4 SiO 4 content [43]

PAGE 41

41 CHAPTER 3 EXPERIMENT PROCEDURE S This chapter focuses on the principles of experiment method and experiment procedures especially focus on the theory of sol gel method, principles of PLD, the design of the PLD system in UF and the electrochemical measure methods that will be used in this research. The sol gel theory has been expressed first, four different steps were used to fabricate ceramic. The mechanism of laser ablation will be also described, and the set up of PLD system and the deposition procedure are introduced Finally, the electrochemical measurement method and electrochemical measurement system are given 3.1 Principles o f Sol Gel Synthesis M ethod [46] A ttention for the sol gel process start as mid 1800s with Ebelman and Grahams [47] A sol is composed of a suspension of colloidal particles in a liquid, a gel is a semi rigid mass formed if colloidal particles are linked to form a network or the polymer molecules are cross linked. The sol ge l process is a method to ceramic fabricating. There are two type sol gel processes, including the colloidal gel route and polymeric route. The precursors for the sol gel process consist of inorganic salt or metal organic compounds but people focus on the m etal alkoxides. The precursor composition, the concentration of reactants, pH value and the temperature will affect the conversion of the sol gel process. After gelation process, a large amount of liquids still exist in the interconnected channels, the dry ing process must be performed to remove to water, including the conventional drying and the supercritical drying process. Finally, sintering process was performed to get final product. The whole sol gel process can be divided into following steps [46, 48, 49] as Figure 3 1 :

PAGE 42

42 3.1.1 Selection o f Precursor The physical properties of metal alkoxides major depend on the properties of the metal, such as the valence, the electro negativity and atomic radius. The al kyl group will also affect the volatility of the precursor, with the increasing of the number of methyl groups, the size of atom will decrease and the precursors will become sublimable solids or volatile liquids. The wide range properties can be achieved depend on the alkyl group and metal. Many metal alkoxides are sensitive to the moisture and need to be careful for handling or stora ge. The use of dry atmosphere or dehydrated solvents is necessary. 3.1.2 Conversion o f S ol to G el The basic reaction process sol gel process can be described as follows by using tetramethoxysilane(TMOS) as a precursor in limited water and alcohol solvent The overal l reaction of Si(OCH 3 ) 4 are described as follows hydrolysis reaction (3 1) condensation reaction (3 2) 2 o condensation reaction (3 3) Hydrolysis in excess water commonly lead s to the insoluble hydroxides or hydrated oxides However, if the limited amount of water was added, partial hydrolysis reaction happen and y ield soluble species that will take part in polymerization and form a by product. An acid (HCl) or a base(NH 3 ) usually used as a catalyst Under conditions of acid catalysis and low H 2 O/Si precursor mole ratio, the sol composed of weakly branch polymers. With base catalysis and high H 2 O/Si precursor mole ratio condensed particulate sols will form. Structural evolution of sol gel silicates has been investigated

PAGE 43

43 on different length scales using different techniques. On the scale of 1 to 299nm, the structur es are not uniform, a tenuous fractal structure will form. These objects have an interesting property, they are invariant under a change of magnification and can be characterized by fractal number (3 4) Where M is the mass of the object R is the radius of the object and D is the fractal dimension. The fractal dimension is a useful morphologica l descript or. A lower value of D indicate s a more open structure. A fractal object has interesting properties that object decrease as the fractal size increases. The density of a fractal tend zero as the object increase in size. 3.1.3 Drying of gels [50] Gels are composed of a weak amorphous solid structure containing an interconnected network of very fine pores filled with liquid. The gel must be dried before the final sintering process The conventional drying involv ed to remove the water by evaporation in the air or in a drying chamber and the gel produced by conventional drying process is referred as a xerogel. The gel produced by supercritical drying is referred as an aerogel. Conventional drying invol ves the interaction of three independent process, evaporation, shrinkage a nd fluid flow in the pores and can be expressed as two steps : a constant rate evaporation period(CPR) and a falling rate period as shown in Figure 3 2 During the CPR, the liquid vap or meniscus remains at the surface of the gel and the evaporation occurs at a rate close to a free liquid surface. In this stage, the amount of gel decreases is same as the amount of liquid take out The shrinkage occurs with the liquid evaporation. The dr iving force of shrinkage come s from the decrease of the

PAGE 44

44 interfacial energy. The process can be described by the following process. When evaporation occurs and a solid interface appear the solid vapor boundary will be substituted by the solid liquid bounda ry as shown in Figure 3 3 to decrease the surface ener gy if the contact angle <90 ( SV SL ) where SV SL, LV are the specific surface energy of the solid vapor, solid liq uid and liquid vapor interfaces The capillary tension in the liquid can be expressed as (3 5) P is the capillary pressure, r is the radius of curvature. If the radius of curvature is same as the pore, t he maximum pressure will take place For solution wi th cylindrical shape and a radius a the r can be expressed as (3 6) The maximum tension is (3 7) liquid imposes a compressive stress on the solid phase causing contraction of the gel. Osmotic pressure that produced by concentration gradient also can cause the shrinkage of the gel. For example, gel pr epared by the hydrolysis of metal alkoxides contains a solution of liquids that has different volatility. Evaporation creates a composition gradient and liquid diffuses to reduce the gradient. If pores are large, a counterflow of liquid occurs and no stres s will be developed. On the other hand, if the pores are small enough to inhibit the flow, diffusion away from the interior can produce a tension in liquid.

PAGE 45

45 When shrinkage stop s evaporation forces the liquid meniscus into the pores and the evaporation ra te decrease. This stage is called the falling rate period, During the FRP, the liquid recedes into the gel as shown in Figure 3 2(c) During drying the structure of gel will change. Polymeric ge ls will collapse and cross link. When the structure becomes stiff enough to resist the capillary stress, residual porosity will be formed. Silica gels prepared by acid catalyzed reactions own the porosities as 35 40% compared to base catal yzed that has the porosities as 60 to 70 %. For particulate gels, the structu re will fold under the action of the capillary stresses. Neck formation or an increase in the coordination number of the gel will c ontinue until the structure is strong enough to resist the capillary stress. Because the particulate structure of the gel can better withstand the smaller capillary stresses arising from the large pores, the shrinkage will be much smaller than the polymer gels. The porosity is typical in the range of 70 80 % In supercritical drying, the liquid in the pores is removed above the critical temperature, T c and the critical pressure P c In this situation, there is no difference between the liquid and vapor phases. There is no capillary pressure and no stresses formed during drying process. The aerogel with little shrinkage and high p orosity can be fabricated. The critical temperature and pressure of ethanol are 243 C and 6.4MPa. An autoclave needed for such high temperatu re and pressure makes the proces s time consuming dangerous and expensive. Carbon dioxide has lower critical point means the process can be performed in ambient temperature. Moreover, the cheap price make s it possible to use in commercial application.

PAGE 46

46 3.1.4 Gel Densifi cation D uring S intering The structure of xerogel has a high free energy that can act as a powerful dr i vi ng source for sintering. The surface area of the solid vapor interface was the main source, the reduced cross link density and the free volume of the solid skeletal phase are also expected to provide a contribution to the free energy. The particulate gel shrink only at elevated temperatures and the behavior is similar to the porous metal glass. For polymeric gel, the mechanisms contribute to the densification depend on the porosity, pore size and skeletal density as well as the heating rate and therma l history. A borosilicate gel that heated at a constant rate 0.5 C/min was used as a example here: the whole processes can be divided into three regions: Region I : weight loss with a small shrinkage 25 150 C : t he desorption of water and alcohol contributed to a small loss of weight. The small shrinkage was due to energy increase contributed by the desorption process of liquid. Region II: weight loss with concurrent shrinkage150 525 C : t he weight loss was attributed by two processes, the removement of the wa ter and the oxidation of carbonate The shrinkage in region II was attributed by polymerization reaction, which lead s to a higher cross link density of polymer chain and structural relaxation of the polymeric network as the structure approach the configura tion of the melt glass. Region III: shrinkage without weight loss( >525 C) : t he large and fast shrinkage was contributed by viscous flow. The pores in the gel will be removed in this stage. The driving force of densification is the decreasing the solid vapo r surface tension formed by pores. Densification also involves transport of matter and it is limited by the kinetics of the process, for example, the rate of heating have a significant effect for densification. High heat rate reduces the amount of viscous flow that can occur in a given temperature interval and decrease the amount of cross link and structure relaxation. 3.2 Principles o f Pulsed Laser Deposition Pulsed laser deposition is a simple, but versatile method that can fabricate diverse range of mat erials. In1962, Breech and Cross [51] used ruby laser to vaporize and ablate material from a solid(or liquid) target. In 1965 Smith and Turner [52] used

PAGE 47

47 ruby laser to deposit thin films. However, the availability of the types of laser, stability and low laser repetition rate limits it application With the rapid progress of the laser technology such a s the development of nanosecond pulsed Nd YAG excimer lasers and the availability of intense pico and femto second pulses high temperature superconducting material was synthesized successfully [53] Laser used in PLD process, including CO 2 laser(10.6um), Nd YAG laser(1064nm and 532nm), excimer laser(XeCL 308nm, KrF 248nm ArF(193nm)157nm(F2). The KrF excimer laser wa s used in this research. There are many advantages for PLD process. The laser system is outside of the chamber that make the highly freedom for design of the deposition system. The targets used in PLD are smaller than conventional sputtering method. Moreov er, the all solid state thin film battery with multi layers can be fabricated only by this technique. In addition, the stoichiometry of material can be keep since the extremely high heating rate supply by the laser This deposition method was already used for different field, including film deposition, patterning polymers and wide band gap materials or removal of biological tissue and surface clean 3.2.1 Mechanism o f Pulsed Laser Deposition The mechanism of the PLD process can be expressed in three steps [54 56] as shown in Figure 3 4 : 3.2.1.1 The interaction of the laser beam and target E lectronic excitation thermal conduction in solids and the removal of material (macroscopic sputtering) will be induced by t he in teraction of laser beam and target ;

PAGE 48

48 their relative importance depend on the nature of the target material and the excitation wavelength and the pulse duration [54 56] When a laser beam interacts with material, ele ctrons are excited by the absorption of photons. The electrons are heated to high temperature by absorbing laser energy through collisions. For materials such as metals, the band structure allows absorption of most low to moderate energy photons. For semic onductor or insulator, the band struc ture make the photon energy can not be absorbed by linear process. The photon absorption will happen by accompanying with one of two processes: electron avalanche [57] and strong multiphoton absorption. Electrons at defects sites, which have intermediate energies can be excited by single photon absorption. Moreover, small amounts of multiphoton absorption can lead to a small number of electrons enter to the conduction band. Collisions of these excited electrons with bound electrons lead to avalanche ionization. For thermal conduction in solids t he effects including electron lattice energy coupling, thermal conduction and phase transition For femtosecond pulsed, there is insufficient time for energy transfer to the lattice and can be considered as direct solid vapor transition make the thermal damage is minimal [58] For long pulsed lasers, ablation of materials occurs through melt expulsion driven by the vapor pressure and the recoil pressure of light. The melt layer is resolidified, resulting in geometri c change of the holes. With ultr a short pulse, the deposited energy is limited to a smaller depth and the absorbed energy heats the material very quickly past the melt to the vapor phase. The material is removed by direct vaporization away from the surface without formation a recast layer.

PAGE 49

49 3.2.1.2 Plume f ormation a nd p ropagation [55] Extensive experiments have been performed to study the plume propagation using optical absorption and emission spectroscopy. Neutral atoms, ions and the electrons travel at different velocities and strong interaction between the species of plasma and background gas are observed. The plasma plume e xpands like a pulsed jet from a nozzle at supersonic velocities in the direction of the target normal. The emitted light and the color of the plume are caused by fluorescence and r ecombination proces ses in the plasma. The pressure and the laser fluence bot h have significant effect on the shape, size of the flume. 3.2.1.3 Nucleation and growth [55] The quality of the film was establish ed by this step Different reactions happen in this stage including atom deposition on the substrate, atom re evaporation from substrates, atom diffusion over substrate and clusters dissociation If the real vapor pressure of the arriving part icle is larger than its equilibrium value film will grow on the substrate The total change of free energy for the nucleation of a cluster on a substrate can be expressed as (3 8) and show in Figure 3 5 where the a i is the shape factor r is the radius of the cluster f s is the surfa ce energy of the substrate per i is the interface energy between the film and the substrate per unit area, and G V is the change of volume free energy. The cluster is stable and will start to grow if the derivative o f the change in the tot al free energy to the number of atoms is negative

PAGE 50

50 The critical size of the nucleus can be derived from Eq. 3 8 as following equation [59] ( 3 9) If the free energy is negative and the total surface free energy is positive the cluster will continue growing until reaching the critical size of the nucleus T hree growth modes developed depend on the surface energy [60, 61] : In the island ( Volmer Weber ) mode small clusters are nucleated dire ctly on the substrate surface and then grow into islands of the condensed phase. This happen s when the atoms of the deposit are more strongly bound to each other than to the substrate ( ). The layer ( Frank van der Merwe ) mode displays the opposite characteristics. Because the atoms are more strongly bound to the substrate than to each other ( ) the first a toms to condense form a complete monolayer on the surface, which becomes covered with a somewhat less tightly bound second layer The layer plus island ( Stranski Krastanov ), mode is an interesting intermediate case. After forming the first monolayer, or a fe w monolayers su bsequent layer growth is unfavo rable and is lands are formed on top of this monolayer. The reason is an increase with increasing thickness in the stress due to the lattice mismatch [62] For the PLD process, because of the short laser pulsed duration(ns) and the small temporal spreas(<10m/s) of the ablated materials. The layer mode is preferred and smooth film can be fabricated. Moreover, the energetic ablation sp ecies lead to a lower deposition temperature for crystalline the film.

PAGE 51

51 3.3 Set up o f Pulsed Laser Deposition System The PLD system was built in University of Florida. The system is composed of a laser, a deposition chamber, several optic lenses and a vacu um system. The laser source is Lambda Physic LPX300 as shown in Figure 3 6 Because KrF is the highest gain system for electrically discharged pumped excimer laser system, the KrF excimer laser with wavelength 248nm was used in this research. The pulse ene rgy of laser system can be controlled between 200 to 550mJ and the repetition rate can be controlled between 1 and 50 Hz with pulsed width 25ns. Four different gases were used including Kr F 2 Ne ,O 2. The deposition chamber was brought from Excel Instr uments, Mumbai and was composed 5 parts 3 7 850 C Programmable controller can index the targets as well as rotate them in front of the laser beam. The number of targets is 6 which means multi layer deposition is possible Gas flow assembly for controlling the ambient gas as well as venting the chamber Lens holder attachment for X,Y, Z movement of lens in order to focus the laser beam. The Vacuum syst em was brought from Pfeiffer Vacuum, including TMU 261P Turbo molecular Drag Pump with TC 600, XtraDry 150 2, Dry Piston pump with solenoid actuated angle valve compact full range Pirani/Cold Cathode Gauge (Measurement range 1000 to 5x10 9 mbar)and DCU 200 Display Control Unit for gauge with power supply. The series design was used for the vacuum system as shown in Figure 3 8

PAGE 52

52 3.4 Experiment P rocedure f or PLD Several different substrates were used for the experiment, including, Stainless Steel 304 foil, gold foil, SiO 2 /Si and Pt/Ti/SiO 2 /Si substrate. The SiO 2 /Si substrate was supplied by National Nano Device Laboratory in Taiwan and the oxide with 150nm was fabricated by wet oxidation method. The Pt/Ti with 150nm and 20nm was prepared by sputtering deposition in Nano Facility in University of Florida. For metal substrate, the substrates were polished with 3, 1 and 0.3 um alumina suspension. Then, three steps were performed to clean the contaminants. First, the acetone solvent was used by to remove organic conta minations by ultrasonic cleaner for 10 minutes. Then ultrasonic bath in alcohol was carried out for rinsing contaminated acetone as well as removing particles from surfaces Finally, de ionized water was used for the final cleaning. Before the deposition p rocess, the sample must be pasted on the sample holder the solvent of the silver paste .The chamber was evacuated to 10 5 torr by starting the rotary and turbo pump. Af ter the base pressure arrived, the thermal controller was turn on to heat the substrate to desired temperature. After the specific temperature was reached, oxygen gas was introduced into the chamber and kept at a specific pressure during the deposition. 3. 5 Electrochemical Measurements 3.5.1 Configuration o f Cells Two types of cells were used for e lectrochemical measurements, including coin cell 2016 and Swagelok c ell as shown in Figure 3 9(a)(b) For the powder or the flakes samples, the coin cell 2016 was used. For the thin film samples, the Swagelok cell was used.

PAGE 53

53 The 2016 coin cell and Swagelok c ell were consisting of metallic lithium as an anode and an electrolyte comprised of 1M LiPF 6 in ethylene carbonate (EC)/dimethyl carbonate (DMC) with 1:1 volume ratio Celgard model C480 separator (Celgard Inc.USA) was used as separator. Cathode films for coin cell were prepared from slurries of powders or flake s with suitable polyvinylidene fluoride (PVDF) and acetylene carbon black dissolved in N methyl 2 pyrrol idinone (NMP). The mixtures were cast onto Al foil using a doctor blade and dried at 100 C overnight. Electrode discs were cut into cathodes containing 2 3 mg of active materials before storing them in an argon filled glove box (H 2 O level<1.6ppm). The Swag elok Cells for thin films samples were similar configuration compared to coin cell but the re is no carbon blac k or PVDF necessary for testing The cells were assembled in an argon filled glove box and tested using an Arbin BT2000 instrument in galvanos tati c mode as shown in Figure 3 1 0 (a)(b). 3.5.2 Potentiostatic I ntermittent T itration T echnique (PITT) method [63, 64] T he potentiostatic intermittent titration technique ( PITT) was used to abstract the lithium chemical diffusion coefficient This technique applied a constant voltage step to the cell and measure the current drop with time. A new voltage is applied when the current drops to a very small value. th one dimension (3 10) where x is the distance into the solid film from the electrolyte/el ectrode interface, C Li is the Li concentration at x and time t and is the chemical diffusion coefficient. We can assume the electrode initially has a uniform lithium concentration C 0 A new concentration C S is imposed on the electrode surface by applying a voltage step.

PAGE 54

54 Lithium must be continuously supplied by transport through the electrolyte in order to keep the surface concentration constant at the imposed C S until the electrode reaches the composition C S everywhere The boundary conditions of the PITT can be expressed as following equation (3 11) (3 12) (3 13) Where the L is the diffusion length, Cs is the surface concentration of lithium [63] (3 14) The current formed by the voltage step depend on the concentration gradient and can be expressed as following equation (3 15) (3 16) where a is the area of the interface, z is charge number, F is the Faraday constant,. Moreover, by using following relationship (3 17) (3 18) where V M is t he molar volume Q is the charge transferred nunber and n B molar number (3 19) For a long time the current can be expressed by a reasonable approximation (3 20)

PAGE 55

55 Fi nally, t he chemical diffusion coefficient can be abstracted by following equation. (3 21) PITT can be avoided the two phase reaction during measurement when the voltage steps are well controlled and the relationship between the current and time more directly and exactly reflect the diffusion property of Li a lthough the results were obtained under a slight deviation fro m the equilibrium conditions,

PAGE 56

56 Fi gure 3 1 T he routes that indicate the process of sol gel [48] Figure 3 2 Stage of drying [50]

PAGE 57

57 Figure 3 3 Schematic illustration of shrinkage of the gel Figure 3 4 Mechanism of PLD [55]

PAGE 58

58 Fi gure 3 5 Schematic illustration of nucleation of a cluster on a substrate surface Figure 3 6 Lambda physic LPX300 Photo(s) courtesy of Ming Che Yang

PAGE 59

59 Figure 3 7 Chamber design of the in house PLD system Figure 3 8 The design of vacuum sy stem

PAGE 60

60 Figure 3 9 The cells for electrochemical measurement (a)coin cell 2016(b)Swagelok Cell Photo(s) courtesy of Ming Che Yang Figure 3 10 The equipment for electrochemical measurement (a) Arbin machine and (b)glove box Photo(s) courtesy o f Ming Che Yang Poly(diallyldimethyl ammonium chloride) (PDAC ) b )

PAGE 61

61 CHAPTER 4 E LECTRONIC S TRUCTURALAND E LECTROCHEMICAL PROPERTIES OF LINI X CU Y MN 2 X Y O 4 (0 4V) accompanies some transition metal doping [13] Among all LiM x Mn 2 x O 4 materials, LiNi 0.5 Mn 1.5 O 4 is an attractive high voltage cathode material since it offers a flat voltage plateau at 4.7 V and demonstrate s a reversible capacity >135mAh/g [66, 67] Several strategies were developed to improve the rate capability of LiNi 0.5 Mn 1.5 O 4 J. C. Arrebola et al. obtained excellent rate properties by homogeneously mixing nano a nd micro size LiNi 0.5 Mn 1.5 O 4 powders [68] or by adding polyethyleneglycol(P EG) to increase the crystallinity and decrease the strain [67] The rate capability was also affected by the atomistic structure of the spinel material, as proposed by M. Kunduraci et al [69] It is found that nonstoichiometric LiNi 0.5 Mn 1.5 O 4 ( Fd 3 m ) as sho wn in Figure 4 1 exhibited better structural reversibility at high rate compared with stoichiometric LiNi 0.5 Mn 1.5 O 4 (P4 3 32). The oxygen deficiency

PAGE 62

62 leads to the presence of a small amount of Mn 3+ in the pristine materials, which leads to better electronic c onductivities in the material. The Cu doped spinel materials were investigated by a few research groups [70 73] Although the Cu rich spinel electrodes provide lower discharge capacity than Ni rich spinel electrodes it is more stable during electrochemical cycling. Moreover, LiCu x Mn 2 x O 4 has a higher electronic conductivity than other spinel LiM x Mn 2 x O 4 materials (Cr, Fe, Co, Ni) [74] because Cu can participate in the charge tra nsport process. All the previous work focused mainly on the electronic conductivity in the doped spinel LiM x Mn 2 x O 4 materials. Little has been explored to enhance the lithium ionic diffusivity in this family of materials. Using ab initio computational mode ling, it is possible to rapidly screen the effect of different doping elements on the lithium diffusion activation barrier, as well as the redox potentials in the spinel structure. Ab initio computational method was firstly used by Van der Ven et al. to de termine the energy barrier of Li diffusion in layered transition metal oxides [75] Kang et al. also used this method to identify the factors that limits Li + ion hopping in layered oxide and successfully synthesize an optimized layered material with superior rate capability [76] Many researchers have already shown that the cycling perfo rmance of LiNi 0.5 Mn 1.5 O 4 can be improved by doped with different transition metal ions [77 80] However, a systematic understanding on the effects of co doped elements is still lacking In this work, first principle s computation, based on density functional theory (DFT), is used to examine the voltage profile and electronic structures of the LiM x Mn 2 x O 4 (M = Ti, V, Cr, Fe, Co, Ni, Cu,). The Li diffusion activation barriers in each material are calculated and compared Based on the computational pre screening, our

PAGE 63

63 experimental research focuses on LiNi x Cu y Mn 2 x y O 4 (0
PAGE 64

64 calculated charge densities from the first step. The cell volume was fixed with internal relaxation of the ions in the second step calculation. A supercell with one vacancy out of eight Li sites (Li 7 M 4 Mn 12 O 32 ) was used to calculate the Li diffusion activation barriers in the Li rich phase with GGA approximation. The Hubbard U correction was introduced to describe the effect of localized d electron s of transition metal ions when investigating the valence states and electronic structures of the ions. Previous work has shown that the U values can be calculated in a self consistent way and the U values of Mn and Ni ions in the spinel structure are well established [86 88] These values are taken as the references. In the doped spinel system, the same transition metal ions with different valence states may co exist; therefore, small adjustments are made to the re ference U values, so that each transition metal ion has a unique effective U value applied in the rotationally invariant LSDA+U approach [89] In this work, the applied effective U value given to Mn ions is 5eV, to Ni ions is 5.96eV, and to Cu ions is 7eV. 4.2.2 Synthesis Methods Sample Preparation: Six samples were synthesized using the sol gel method: LiNi 0.5 Mn 1.5 O 4 LiNi 0.45 Mn 1.55 O 4 LiCu 0.05 Ni 0.45 Mn 1.5 O 4 LiCu 0.1 Ni 0.4 Mn 1.5 O 4 LiCu 0.25 Ni 0.25 Mn 1.5 O 4 and LiCu 0.5 Mn 1.5 O 4 The sol solutions were prepared from the stoichiometric mixtures of Li(CH 3 COO)2H 2 O (Fisher), Ni(CH 3 COO) 2 4H 2 O (Fis her), Cu(CH 3 COO) 2 H 2 O (Fisher) and Mn(CH 3 COO) 2 4H 2 O (Fisher) in distilled water. Then the solution was added drop wise to a solution of citric acid which was used as a chelating agent Ph value was adjusted to 6.5 by adding. ammonium hydroxide A gel was o btain ed by heated at 75 C overnight The gel were calcined at 450 C for 10 hours to remove the carbonate in air and calcined at 800 C for 10 hours.

PAGE 65

65 4.2.3 Characterization s Powder diffractions of all samples were collected using a high resolution synchr otron powder X ray diffractometer at beamline 11 BM at the Advanced Photon Source (APS), Argonne National Laboratory (USA). The conditions for the data collection were: continuous scanning of a detector covering an angular range from 0.5 to 40 with a ste p size of 0.0001 and wavelengths of =0.413964 The diffraction patterns were analyzed by the Rietveld refinement software, FullProf The Ex Situ XRD after electrochemical testing were collected by a laboratory x ray diffractometer equipped with a curv ed position sensitive detector (Inel CPS120, detection range 0 120), a Cu X ray tube source and a Gobel mirror on the incident beam. The collection time for each XRD spectrum was 20 minutes. The samples for ex situ XRD were recovered by disassembling cycl ed batteries in an argon filled glove box and the powder mixture was scraped from the aluminum disks. The chemical compositions of the samples were analyzed by inductive coupled plasma atomic emission spectroscopy (ICP AES Perkin Elmer Plasma 3200). The in strument was initially calibrated using commercial LiCoO 2 and LiFePO 4 powders. Field emission secondary electron microscopy (FEG SEM, JEOL JSM 6335F) was carried out to investigate the particle size and morphologies. Images were collected with an accelerat ing voltage of 15kV. X Ray absorption spectroscopy (XAS) at the Ni, Mn and Cu K edges were obtained in transmission mode at the National Synchrotron Radiation Research Center (NSRRC), Hsinchu, Taiwan to observe the changes in Nickel, Manganese and Copper o xidation states for LiCu 0.25 Ni 0.25 Mn 1.5 O 4 during the charging process. A Si (111) double crystal monochromator was employed for energy selection. The coin cell 2016 was charged to target voltage (4.1 to 5V) and rested for 40min (10minutes to reach equilibr ium, 30

PAGE 66

66 minutes to get spectrum) to get the spectroscopy. The spectrum was collected every 0.1V. 4.2.4 Electrochemistry The electrochemical properties of all cathode electrodes were measured using 2016 coin cells consisting of metallic lithium as an anode and an electrolyte comprised of 1M LiPF 6 in ethylene carbonate (EC)/dimethyl carbonate (DMC) with 1:1 volume ratio Celgard model C480 separator (Celgard Inc.USA) was used as separator. Cathode films were prepared from slurries of LiCu y Ni x Mn 2 x y O 4 with 7. 5% polyvinylidene fluoride (PVDF) and 7.5% acetylene carbon black dissolved in N methyl 2 pyrrolidinone (NMP). The mixtures were cast onto Al foil using a doctor blade and dried at 100 C overnight. Electrode discs were cut into cathodes containing 5 10mg o f active materials before storing them in an argon filled glove box (H 2 O level<2ppm). The coin cells were assembled in an argon filled glove box and tested using an Arbin BT2000 instrument in galvanostatic mode. The tests were conducted between 3.0V and 5. 0V at a constant current rate of C/20, C/10, C/5 and C/2. The samples for ex situ XRD were recovered by disassembling charged batteries and washed using acetonitrile to remove residual electrolyte in an argon filled glove box. T he potentiostatic intermitte nt titration technique (PITT) was performed by applying a voltage step of 10mV to measure the chemical diffusion coefficient W hen the current decrease below 3x10A (which is equivalent to C/200 rate) the next voltage step will be applied The voltage win dow was set from 4.5V 4.9V

PAGE 67

67 4.3 Results and D iscussion 4.3.1 Computation Results Previous research has shown that, when Li ions diffuse in the un doped LiMn 2 O 4 spinel structure, the diffusion activation barriers are reached when Li ions occupy the intermedi ate 16c sites which are surrounded by six Mn ions forming a Mn ring [88] A similar situation occurs in the doped spinel Li x M 1/2 Mn 3/2 O 4 (M= Ti, V, Cr, Fe, Co, Ni and Cu); however, the rings are now com posed of two types of metal ions Mn and the doped transition metal ion M. In the supercell used for this work, for each dopant, two distinct local environments of transition metal ion rings can be found. One is composed of three doped M ions and three Mn ions alternatively arranged (Figure 4 2 (a)). The other one is composed of one doped M ions and five Mn ions (Figure 4 2 (b)). The corresponding Li diffusion barriers with different dopants (M= Ti, V, Cr, Fe, Co, Ni and Cu) are calculated using GGA method ( Figure 4 3 ). The Li diffusion barrier in un doped Mn spinel [88] is also provided as the reference. Comparing to the un doped Mn spinel material, the average Li diffusion barrier remains largely unchanged when Ni and Fe ions are present. The average diffusion barrier increa ses with Ti, V, Cr doping and decreases with Co and Cu doping. Specifically when only one Cu or three Co ions appears in the ring, the Li diffusion barrier can be reduced to as low as 256 meV, respectively. As shown in previous work [88] the introduction of the Hubbard U correction accurately captures the charge/discharge voltages of the Li intercalation spinel materials. More importantly different valence states of the same transition metal ions can b e distinguished. Calculations using GGA+U method are performed on LiCu 1/2 Mn 3/2 O 4 spinel with different Li concentrations. In the first model, where Cu ions

PAGE 68

68 occupy 4b octahedral sites and Li ions occupy 8c tetrahedral sites, the calculated average voltage o f Li x Cu 1/2 Mn 3/2 O 4 (0.5< x < 1) is 4.44V, which is 7% higher than the experimental low voltage plateau. In addition, the calculated average voltage ofLi x Cu 1/2 Mn 3/2 O 4 (0 < x < 0.5) is 5.36V, which is 9% higher than the experimental high voltage plateau. The projected electronic density of states (DOS) of Cu 3d orbitals in Li x Cu 1/2 Mn 3/2 O 4 (x=1, 1/2, 0) are calculated and presented in Figure 4 4 (a), Figure 4 4(b) and Figure 4 4 (c), respectively. In the proposed spinel model, the octahedral crystal field splits the Cu 3d orbitals into two parts. d xy d yz and d xy orbitals belong to the t 2g orbitals with lower energy levels while d z 2 and d x 2 y 2 belong to the e g orbitals with a higher energy level [90] Figure 4 4 (a) shows the projected DOS of Cu ions in the fully lithiated phase L iCu 1/2 Mn 3/2 O 4. Both spin up and spin down states of the three t 2g orbitals are below the Fermi energy, which means the t 2g orbitals are completely occupied. For e g orbitals, the d z 2 orbital is occupied, but the spin up state of the d x 2 y 2 orbital is above the Fermi energy indicating that the d x 2 y 2 orbital is half empty. The DOS plot is consistent with the t 2g 6 e g 3 electron configuration of Cu 2+ ions. The projected DOS of Cu ions in the half delithiated phase Li 1/2 Cu 1/2 Mn 3/2 O 4 (Figure 4 4 (b)) and in the fully delithiated phase Cu 1/2 Mn 3/2 O 4 (Figure 4 4 (c)) are generally the same minus trivial differences in shape. It suggests that Cu ions are in the same valence state in the two compositions. The spin up states of the e g orbitals are completely empty while all of the other electronic states are occupied. The DOS plots indicate the electron configuration of t 2g 6 e g 2 consistent with Cu 3+ ions. Based on these results from first principles calculation, the Cu valence in doped spinel LiCu 1/2 Mn 3/2 O 4 is Cu 2+ at fully l ithiated state. The Cu 2+ ions can be oxidized to Cu 3+ ions when half of the Li ions are extracted.

PAGE 69

69 Valence higher than 3+ cannot be obtained and the calculated voltage of the Cu 2+ /Cu 3+ redox couple is 4.44V. When more than half of the Li ions are extracted the charge densities of oxygen ions change significantly, suggesting that the extra electrons may be provided by oxygen ions. If one exchanges the sites of Cu ions with half of the Li ions, a defect spinel model can be created with tetrahedral Cu and oc tahedral Li included. For the defect spinel system, its total energy is 493 meV per formula higher than the perfect spinel system and therefore the structure should be thermodynamically less stable. However, the experimental synthesis is often conducted at relatively high temperatures (>700C) and a certain amount of transition metal and lithium ion site mixing occurs. The change of Cu valence versus Li concentration is investigated using a similar method as described above. Tetrahedral Cu 2+ can also be oxi dized to Cu 3+ but no higher valence can be obtained. In the defect spinel model, Li ions in octahedral sites are preferred to be extracted first and the calculated voltage is 4.58V. The calculated voltage step for removing the remaining tetrahedral Li is 4 .99V, a surprisingly high value. Compared to the NiMn spinel LiNi 1/2 Mn 3/2 O 4, in which Ni 2+ can be oxidized to Ni 4+ providing two electrons per ion, the reversible capacity of Cu doped spinel may be reduced as Cu ions can only provide one electron per ion. However, when doped with a small amount of Cu, it is possible to lower the Li diffusion activation barriers so that better rate capability can be obtained. To further understand the crystal structure, electronic structure and electrochemical properties in a new series of bi dopant spinel materials, we successfully synthesized and studied the LiNi x Cu y Mn 2 x y O 4 with different amount of dopants (x=0, 0.05, 0.10, 0.25 and 0.5).

PAGE 70

70 4.3.2 Experimental R esults The atomic compositions of LiNi x Cu y Mn 2 x y O 4 electrode m aterials calcinated at 800C obtained from ICP AES analysis Although a Nickel oxide and a Copper oxide impurity phase appear respectively in LiNi 0.5 Mn 1.5 O 4 and LiCu 0.5 Mn 1.5 O 4 samples, the molar ratio of Li, Ni and Mn are in good agreement with the theore tical ratio of all samples. X ray diffraction patterns for LiNi x Cu y Mn 2 x y O 4 are depicted in Figure 4 5 Detailed Rietveld Refinement Fitting Results of LiNi x Cu y Mn 2 x y O 4 were shown in 4 6(a)(b)(c). The spinel structure with space group Fd 3 m can be identi fied according to the indexed major peaks. In some samples, small amounts of impurities are present, such as nickel oxide in LiNi 0.5 Mn 1.5 O 4 and copper oxide in LiCu 0.5 Mn 1.5 O 4 as shown in an inset in Figure 4 5 The impurities may be tied with the solubili ty limit of nickel or copper in the spinel LiNi x Cu y Mn 2 x y O 4 as nickel or copper approach 0.5 which supports the findings of other research groups [72, 91] It is also consistent with previous reports that oxygen defi ciency is usually found in LiCu 0.5 Mn 1.5 O 4 .The results show that the impurity phase can be eliminated by slightly changing the Ni Mn stoichiometry (LiNi 0.45 Mn 1.55 O 4, Figure 4 (a)) or by doping Cu into the NiMn spinel (Figure 4 5 (c)(d)(e)). According to Ohzu ku et al., the integrated intensity ratios of the (400)/(311) peaks in spinel structure represents the extent of occupancy of the substituent ions on the tetrahedral site ( 8a ) [92, 93] From Table 4 2 the integrate d intensity ratio of (400)/(311) peaks generally decreases with the increase of Cu content. Exceptions are found in LiNi 0.5 Mn 1.5 O 4 and LiCu 0.5 Mn 1.5 O 4 where the oxide impurities are involved and may affect the related peak intensities. The decrease of the i ntegrated intensity ratio of

PAGE 71

71 (400)/(311) peaks indicates that the occupancy of the substituent Cu in the tetrahedral sites ( 8a ) increases when the Cu amount increases. For LiNi 0.5 y Cu y Mn 1.5 O 4 samples, refinements are performed and Rietveld fitting results of LiNi x Cu y Mn 2 x y O 4 are shown in Table 4 3 and the trend of lattice parameters change with resp ect to y is depicted in Figure 4 7 From the plot, the lattice parameter decreases from y=0 to y=0.05, and then increases with the increase of y. The trend gen erally agrees with the results reported in previous literature. The site refinements also suggest that some Cu ions occupy the 8a tetrahedral sites instead of the 16d octahedral sites. SEM was performed to examine the particle size and morphologies of LiNi x Cu y Mn 2 x y O 4 after 800C calcination. Figure 4 8 (a), (b) and(c) show that in samples with no Cu or small amounts of Cu doping, spherical particles with 100nm diameter were formed. It is also noted that when the amount of doped Cu increases(Figure 4 8 (d), (e) and(f)), the particle size increases and the shape of the particles becomes more faceted. For the LiCu 0.5 Mn 1.5 O 4 sample, the particle size is in the range of 300nm 500nm, while the particle size for NiMn spinel sample is in the range of 50 100nm. The electrochemical properties of LiNi x Cu y Mn 2 x y O 4 (0
PAGE 72

72 around 4.7V, which corresponds to the Ni 2+ Ni 4+ redox potential. A relatively small plateau can be found in the region around 4.0V. This is attributed to the Mn 3+ Mn 4+ redox couple. P revious work has shown that calcination temperature may cause oxygen loss and can lower the average oxidation state of Mn and introduce Mn 3+ ions [69] For the LiNi 0.45 Mn 1.55 O 4 sample, the average oxidation state of mang anese decreases further due to the decrease of Ni/Mn ratio, therefore the capacity of the LiNi 0.45 Mn 1.5 O 4 sample at the 4.0V region is larger than that of the LiNi 0.5 Mn 1.5 O 4 sample. When Cu ions are doped, the voltage plateau shifts from 4.7V to 4.95V and a new plateau appears between 4.2V to 4.6V. When the doped Cu amount increases, the total capacity decreases while the capacity between 4.0V and 4.6V increases. Capacities contributed by the plateau around 4.0V and by the plateau between 4.2V to 4.6V in di fferent materials ar e listed respectively in Table 4 4 To study the rate capability of LiNi x Cu y Mn 2 x y O 4 (0
PAGE 73

73 X ray Absorption Near Edge Structure (XANES) measurements was performed on the LiCu 0.25 Ni 0.25 Mn 1.5 O 4 cathode material to study the active redox couple(s) in different voltage r ange. Figure 4 12 shows the charge curve for a LiCu 0.25 Ni 0.25 Mn 1.5 O 4 half cell that is used for in situ XAS experiment. The Cu, Ni, and Mn K edge XANES spectra of LiCu 0.25 Ni 0.25 Mn 1.5 O 4 measured as a function of voltage are shown in Figure 4 13, Figure 4 1 4 and Figure4 15 In each figure, (a) is the original spectra and its second order derivative is shown in (b) to investigate the distinct energy shifts. The spectra of related oxides are also provided as references for comparison. The pre peak region of K edge absorption peak of pristine sample is in the same position as the pre edge peak for the CuO reference. This indicates that the oxidation states of Cu in the two materials are the same, thus Cu is +2 in the pristine LiCu 0.25 Ni 0.25 Mn 1.5 O 4 During the ch arging process, the position of Cu K edge absorption peak remains unchanged until the voltage reaches 4.2V. After that, the peak begins to shift to the higher energy side. The shift stops at around 4.7V and no significant energy shift can be observed at hi gh voltage region between 4.7 V to 5.0V. The Cu spectra shows that the Cu valence only changes between 4.2V and 4.7V. Similar methodology was applied on Ni and Mn spectra. Ni is Ni 2+ in the pristine sample and it starts to be oxidized from 4.7V and the K e dge energy shift continues until 4.9V is reached. The Mn valence is +4 in the pristine sample and it remains unchanged during the entire charging process. X ray ph otoelectron spectroscopy was used to examine surface features on the LiCu 0.25 Ni 0.25 Mn 1.5 O 4 cathode. Figure 4 16(a)(b)(c) shows the spectra observed for the Cu2p, Ni2p and Mn2p regions for pristine cathode. The binding energy of the Cu2p and Ni2p shown in Figure 4 16(a)(b) indicate the oxidation states of Cu and Ni are two. De

PAGE 74

74 convolution of the Mn2p spectra reveals two peaks, a minor peak at 641.5 eV(Mn 3+ ) and major peak at 642.5eV(Mn 4+ ) confirming the presence of a small amount Mn 3+ that is similar to previous electrochemical measurement results. Ex situ XRD patterns for LiNi 0.25 Cu 0.25 Mn 1.5 O 4 a t different state of charge were measured between 4.2V to 5.0V, in order to investigate the possible phase transition during Li de intercalation. Figure 4 17(a) shows the XRD patterns of the LiNi 0.25 Cu 0.25 Mn 1.5 O 4 sample at different charged voltages during the second charging process. At low voltages between 4.2V to 4.6V, the diffraction peaks shift to higher angles with increasing voltage, indicating that the lattice parameters are decreasing. However, at high voltage, no shifts in the diffraction peaks we re observed, indicating the lattice parameters remain mostly constant. Although the amount and direction of shift may change with state of charge, no additional peaks appear during the entire charging process, suggesting the reaction remains single phase t hroughout the charging. The potentiostatic intermittent titration technique [94] was performed to identify the lithium diffusion coefficient associated with the high voltage plateau observed in the second charging cycle of the electrochemical curve. Figure 4 18 (a) shows the PITT profile for LiNi 0.25 Cu 0.25 Mn 1.5 O 4 sample between 4.5V a nd 4.9V.Within this region, the current decays quickly at first, then takes seven hours to reach the limit value. The trends are consistent when different sets of samples are tested. Figure 4 18 (b) shows the variation of chemical diffusion coefficients of Li ions in LiNi 0.45 Mn 1.55 O 4 LiNi 0.5 Mn 1.5 O 4 and LiCu 0.25 Ni 0.25 Mn 1.5 O 4 The lithium diffusion coefficient is determined from the following equation where I is the current and L is the particle size. L is estimated from SEM images (Figure 4 8 ).

PAGE 75

75 (4 1) D Li changes with the lithium content. This trend shown in Figure 4 18 (b) is similar to previous li terature but the values of the diffusion coefficient from our results are smaller than results acquired by Xia et al. [95] The main source of discrepancy may come from the errors in estimiation of the diffusion length for po wered composite electrode as oppose to thin film electrode. The minimum of D Li appears between 4.7V and 4.8V. These coincide with the oxidation of Ni 2+ to Ni 4+ Among the three samples D Li of LiCu 0.25 Ni 0.25 Mn 1.5 O 4 is one order of magnitude higher than tha t of LiNi 0.5 Mn 1.5 O 4 and LiNi 0.45 Mn 1.55 O 4 4.3.3 Discussion 4.3.3.1 The crystal structure of LiNi x Cu y Mn 2 x y O 4 Although the results from first principles calculations performed at 0K suggest that tetrahedral Cu is energetically less favored, results from X RD indicate that Cu ions do occupy tetrahedral sites in materials synthesized at high temperature. According to Shannon et al., the effective ionic radii of Cu 2+ (0.57 ) is slightly smaller than that of Li + (0.59 ) if the atoms are tetrahedrally coordina ted [96] When a small amount of Cu ions are doped into the NiMn spinel materials, most of the Cu 2+ ions may occupy the tetrahedral sites, leading to a smaller lattice parameter than un doped materials. However, when more Cu ions are doped, C u 2+ will also occupy the octahedral sites that were originally occupied by Ni 2+ ions. Because the effective ionic radii of octahedral Cu 2+ (0.73 ) is larger than that of octahedral Ni 2+ (0.69 ) [96] the decrease of lattice parameter is gra dually compensated, and the lattice parameters will finally increase with the increase of doped Cu amount.

PAGE 76

76 4.3.3.2 The voltage profile and capacities of LiNi x Cu y Mn 2 x y O 4 Although in previous research [71, 73] th e high voltage plateau of 4.95V were attributed to the oxidization of Cu 2+ to Cu 3+ our results show that this may not be true. The calculated voltage corresponding to the oxidization from Cu 2+ to Cu 3+ in non defect spinel LiCu 0.5 Mn 1.5 O 4 is only 4.44V, muc h lower than the experimentally measured high voltage plateau at around 4.95V but quite close to the low voltage step at around 4.2V. Cu 3+ ions cannot be further oxidized to higher valences and the extra electrons needed in the high voltage may be provided by oxygen densities. This conclusion is also supported by our XANES results obtained from LiNi 0.25 Cu 0.25 Mn 1.5 O 4 in which Cu ions are only oxidized at low voltages between 4.2V and 4.7V. Mn ions are inactive and the high voltage plateau is caused by the N i 2+ oxidization, together with possible oxygen charge loss mechanism as suggested by first principles computation in this work, but cannot be detected by XANES. In regards to the electrochemical charge discharge results of LiNi x Cu y Mn 2 x y O 4 the capacity contributed by the second plateau between 4.2 to 4.6V increases with the increase of doped Cu amount (Table 4 4 ). This trend is consistent with the above conclusion, as more Cu 2+ ions are present and oxidized to Cu 3+ ions. However, when more Cu ions are do ped, the total capacity of LiNi x Cu y Mn 2 x y O 4 keeps on decreasing. The capacity of LiCu 0.5 Mn 1.5 O 4 is just slightly higher than half of the capacities of un doped NiMn spinel materials. It suggests that in Cu doped NiMn spinel, not all Li ions can be electr ochemically extracted. Two reasons may be attributed to this phenomenon. 1) As Cu ions can only provide one electron per atom, extra electron sources are not absolutely confirmed from experiments. Calculation results suggest a possibility that extra electr ons may be extracted from oxygen ions, though further investigation is

PAGE 77

77 necessary to confirm the exact electron sources. 2) The first principles calculations show that when tetrahedral Cu ions are presented, Li in octahedral sites prefer to be extracted fir st. However, with tetrahedral Li and Cu ions blocking the diffusion paths, some octahedral Li ions may never be extracted. On the other hand, it is possible that voltage plateaus higher than 5V may exist but cannot be obtained due to cut off voltage limit imposed by the electrolyte de composition. 4.3.3.3 Rate capabilities of LiNi x Cu y Mn 2 x y O 4 Although the reversible capacity is partially affected, a suitable amount of Cu doping in NiMn spinel may significantly improve the rate capability of the materials As shown in figure 4 10 LiCu 0.25 Ni 0.25 Mn 1.5 O 4 shows the best rate performance among all the materials. The discharge capacity of LiNi 0.25 Cu 0.25 Mn 1.5 O 4 at C/20 is around 95 mAh/g and 97% of them can be retained at C/2. In the other hand, the discharge ca pacity of LiNi 0.5 Mn 1.5 O 4 at C/20 is 122mAh/g, but it drops to 76mAh/g at C/2. The improvement of rate performance of LiNi 0.25 Cu 0.25 Mn 1.5 O 4 may come from two major sources. 1) The ionic conductivity of the materials is improved by the enhancement of Li mob ility. As shown in Figure 4 18 (b), the measured lithium diffusion coefficient of LiCu 0.25 Ni 0.25 Mn 1.5 O 4 is significantly larger than that of LiNi 0.5 Mn 1.5 O 4 or LiNi 0.45 Mn 1.55 O 4 A possible reason is that the cation distribution of LiNi x Cu y Mn 2 x y O 4 varies wi th the change of cation composition. When the amount of Cu doping is in a certain range, the corresponding cation arrangements can provide more local environments that are similar to Figure 4 2 (b), therefore may significantly reduce the Li diffusion barrie rs and improve the overall Li mobility. 2) The LiCu 0.25 Ni 0.25 Mn 1.5 O 4 system experiences smaller strains during the Li extraction than the non doped NiMn spinel. For LiNi 0.5 Mn 1.5 O 4 and LiMn 2 O 4 a second or third cubic phase will form during the charging sta ge [97] As the

PAGE 78

78 co existence of two phases usually causes lattice mismatch as shown in Figure4 17(b) high stress and strain may be introduced into the non doped NiMn spinel and affect the stabilities of the systems, especially at high rate. For LiCu 0.25 Ni 0.25 Mn 1.5 O 4 however, the ex situ XRD results (Figure 4 17(a) ) shown that, only single cubic phase is present during the charging stage. No phase transition occurs inside the materials, the experienced s trains are much less and the system is therefore more stable at high rate. Besides the above two sources, the effect of electrical conductivity may also play a role. It was already reported by M. Kunduraci et al [69] tha t cation ordered P 4 3 32 spinel exhibits less capacity retention than disordered spinel at fast discharge rates. This phenomenon is attributed to the higher electronic conductivity of the disordered spinel. J.Molenda et al. [74] also showed that the electrical conductivity of LiCu x Mn 2 x O 4 would increase with the Cu amount. There are factors that may negatively affect the rate capability of the materials as well, such as the increased particle sizes for Cu doped NiMn spinel a nd the presence of unmovable tetrahedral Cu that may block the Li diffusion channels. From the experimental results obtained in this work, however, these factors can be compensated by the positive effects discussed previously. 4.4 Summary In this work, a series of bi doped spinel oxides of LiNi x Cu y Mn 2 x y O 4 (0
PAGE 79

79 from extra electrons provided by oxygen ions. Li diffusion activation barriers of LiM 1/2 Mn 3/2 O 4 (M= Ti, V, Cr, Fe, Co, Ni and Cu) are calculated. The results show that different dopants can have signifi cant effects on the Li diffusion barriers. Although the capacity of the doped spinel materials decreases with the increasing doped Cu amount, LiCu 0.25 Ni 0.25 Mn 1.5 O 4 spinel oxideexhibits higher capacity than undoped LiNi 05 Mn 1.5 O 4 spinel at high rates. The go od rate capability of LiCu 0.25 Ni 0.25 Mn 1.5 O 4 spinel oxideis attributed to the single phase transformation during charging, the lower Li diffusion barrier induced by Cu doping, and possibly higher electronic conductivity contributed by Cu doping.

PAGE 80

80 Table 4 1 Chemical composition of LiNi x Cu y Mn 2 x y O 4 Chemical Composition Li:Ni:Cu:Mn Ratio LiNi 0.45 Mn 1.55 O 4 1.02: 0.45 :0.00: 1.55 LiNi 0.5 Mn 1.5 O 4 1.00: 0.51 :0.00: 1.49 LiCu 0.05 Ni 0.45 Mn 1.5 O 4 0.96:0.45 :0.05 :1.50 LiCu 0.1 Ni 0.40 Mn 1.5 O 4 0.97:0.39: 0.10 :1.51 LiCu 0. 25 Ni 0.25 Mn 1.5 O 4 0.97:0.25: 0.26 :1.49 LiCu 0.5 Mn 1.5 O 4 0.96:0.00: 0.49 :1.51 Table 4 2 I(400)/(311) Integrated intensity ratios of XRD spectrum of LiNi x Cu y Mn 2 x y O 4 I(400)/(311) Integrated Intensity LiNi 0.45 Mn 1.55 O 4 1.354 LiNi 0.5 Mn 1.5 O 4 1.278 LiCu 0.05 Ni 0.45 Mn 1.5 O 4 1.294 LiCu 0.1 Ni 0.40 Mn 1.5 O 4 1.121 LiCu 0.25 Ni 0.25 Mn 1.5 O 4 0.854 LiCu 0.5 Mn 1.5 O 4 0.983

PAGE 81

81 Table 4 3 Rietveld Refinement Fitting Results of LiNi x Cu y Mn 2 x y O 4 Materials Results LiNi 0. 5 Mn 1.5 0 O 4 a=8.18090.0001 Z(O)= 0.262 50.0001 Rwp =6.35 RB= 2.32 NiO a=4.15170.0001 Rwp =5.15 RB= 4.25 LiNi 0.45 Mn 1.55 O 4 a=8.17390.0001 Z(O)= 0.26230.0001 Rwp =4.62 RB= 4.87 LiCu 0.05 Ni 0.45 Mn 1.5 O 4 a=8.16930.0001 Z(O)= 0.26180.0001 n Cu(in Li site)= 0.0120.001 Rwp =5. 02 RB= 4.44 LiCu 0.1 Ni 0.40 Mn 1.5 O 4 a=8.17950.0001 Z(O)= 0.26210.0001 n Cu(in Li site)= 0.0400.001 R wp = 8.03 RB= 4.08 LiCu 0.25 Ni 0.25 Mn 1.5 O 4 a= 8.1991 0.0001 Z(O)= 0.26330.0001 n Cu(in Li site)= 0.1330.010 Rwp = 8.56 RB= 4.10 LiC u 0.5 Mn 1.5 O 4 a=8.2131 0.0001 Z(O)= 0.26460.0001 n Cu(in Li site)= 0.1120.010 Rwp =7.24 RB= 5.62

PAGE 82

82 Table 4 4 Charge capacity (mAh/g)of LiNi x Cu y Mn 2 x y O 4 at low voltage range Capacity (mAh/g) of 4V Capacity (mAh/g) from 4.2V 4.6V LiNi 0.45 Mn 1.55 O 4 23.18 0 LiNi 0.5 Mn 1.5 O 4 17.63 0 LiCu 0.05 Ni 0.45 Mn 1.5 21.27 0 LiCu 0.1 Ni 0.40 Mn 1.5 O 4 21.31 10.54 LiCu 0.25 Ni 0.25 Mn 1.5 O 4 23.12 20.83 LiCu 0.5 Mn 1.5 O 4 23.14 28.46 Figure 4 1 Structure of LiNi 0.5 Mn 1.5 O 4

PAGE 83

83 Figure 4 2 Local envi ronments of transition metal ion rings (a) Three M ions in the ring ; (b) One M ion in the ring Figure 4 3 Li diffusion barriers in LiM 0.5 Mn 1.5 O 4 (M= Cr, Fe, Co, Ni, Cu, Mn) calculated by GGA

PAGE 84

84 (a) ( b ) Figure 4 4 Calculated DOS of Cu in Li x Cu 1/2 Mn 3/2 O 4 ( (a) x=1 (b) x=1/2 (c) x=0 )

PAGE 85

85 (c) Figure 4 4 Continued Figure 4 5 X ray diffraction patterns of LiNi x Cu y Mn 2 x y O 4

PAGE 86

86 (a) (b) Figure 4 6 Refinement Results of (a)LiNi 0.45 Mn 1.55 O 4 (b) LiNi 0.5 Mn 1.5 O 4 (c) LiCu 0.25 Ni 0 25 Mn 1.5 O 4

PAGE 87

87 (c) Figure 4 6 Continued Figure 4 7 Lattice parameters of LiNi 0.5 x Cu x Mn 1.5 O 4 (x=0,0.05,0.1,0.25,0.5)

PAGE 88

88 (a) (b) (c) (d) (e) (f) Figure 4 8 FESEM images of LiNi x Cu y Mn 2 x y O 4 (a) LiNi 0.45 Mn 1.55 O 4 (b) LiNi 0.5 Mn 1.5 O 4 (c) LiNi 0.45 Cu 0.05 Mn 1.5 O 4 (d ) LiNi 0.4 Cu 0.1 Mn 1.5 O 4 (e) LiNi 0.25 Cu 0.25 Mn 1.5 O 4 (f) LiCu 0.5 Mn 1.5 O 4

PAGE 89

89 (a) (b) Figure 4 9 (a) Charge Curves of LiNi x Cu y Mn 2 x y O 4 (b) Discharge Curves of LiNi x Cu y Mn 2 x y O 4. Voltage window of 5.0 3.0V were used.

PAGE 90

90 Figure 4 10 Ra te Capability of LiNi x Cu y Mn 2 x y O 4 Figure 4 1 1 The cycling performance of of LiNi x Cu y Mn 2 x y O 4

PAGE 91

91 Figure 4 12 Charge c urve of LiNi 0.25 Cu 0.25 Mn 1.5 O 4 for XANES experiment

PAGE 92

92 (a) (b) Figure 4 13 (a) The Cu K XANES spectra of LiCu 0.25 Ni 0.25 Mn 1 .5 O 4 measured as a function of voltage (b) 2nd order derivative of Cu K edge XANES spectra of LiCu 0.25 Ni 0.25 Mn 1.5 O 4 measured as a function of voltage

PAGE 93

93 (a) (b) Figure 4 14 (a) The Ni K XANES spectra of LiCu 0.25 Ni 0.25 Mn 1.5 O 4 measured as a function of voltage(b) 2nd order derivative of Ni K edge XANES spectra of LiCu 0.25 Ni 0.25 Mn 1.5 O 4 measured as a function of voltage

PAGE 94

94 (a) (b) Figure 4 15 (a) The Mn K XANES spectra of LiCu 0.25 Ni 0.25 Mn 1.5 O 4 measured as a function of voltage (b) 2nd order derivat ive of Mn K edge XANES spectra of LiCu 0.25 Ni 0.25 Mn 1.5 O 4 measured as a function of voltage

PAGE 95

95 (a) ( b) Figure 4 16. XPS plots of the Cu2p, Ni2p and Mn2p region scans for LiCu 0.25 Ni 0.25 Mn 1 5 O 4 (a) Cu(b)Ni(c)Mn

PAGE 96

96 (c) Figure 4 16 Continued

PAGE 97

97 (a) (b) Fig ure 4 17 (a)The XRD pattern of the LiNi 0.25 Cu 0.25 Mn 1.5 O 4 sample at different voltage during the second charge process (b) The XRD pattern of the LiNi 0 5 Mn 1.5 O 4 sample at different voltage during the second charge process [97]

PAGE 98

98 (a) (b) Figure4 18 (a) PITT profile for LiNi 0.25 Cu 0.25 Mn 1.5 O 4 sample (b) Diffusion Coefficient of LiNi 0 4 5 Mn 1 55 O 4 ,LiNi 0.5 Mn 1.5 O 4 and LiNi 0.25 Cu 0.25 Mn 1.5 O 4

PAGE 99

99 CHAPTER 5 TIO 2 FLAKESAS ANODE MATER IALS FOR LI ION BATTERIE S 5.1 Introduction Anatase titanium dioxide is a promising negative electrode material for Li ion batteries due to its safety against overcharging and stable voltage plateau at 1.78 V, which made it possible to be paired with high voltage (5 V) cathode materials such as LiNi 0.5 Mn 1.5 O 4 [13, 67 98] However, the low electrical conductivity, poor rate capability and poor cycling performance caused by the structural changes during the lithiation process have limited its applications. Several different methods have been developed to address these issues. Wagemaker [99] investigated the dependence of insertion reactions to titanium dioxide particle sizes and revealed progressively increasing Li capacity and Li ion solubility with decreasing particle sizes. This result was also confir med by other groups [100 102] Besides the nanosized particle samples, various of titanium dioxide with different phases had been fabricated by several groups Bruce et al [9 8, 103 105] synthesized TiO 2 B nanowires by a hydrothermal reaction to increase the discharge capacity and improve the cycling performance. Hu et al [106] proved that difference occurs in the Li incorporation amount between bulk and nanosized TiO 2 with rutile phase Nevertheless, the difficulty of fabricating the nano sized rutile TiO 2 limits its validity. Moreover, various of nanostructures had al so been fabricated, such as nanofibers [107] nanorods [108] nanotubes [109 111] mesoporous films [112, 113] laminar TiO 2 [114] and hollow ribbon nano network structures [115] which were characterized with different surface areas, porositie s and shapes that can significantly enhance the kinetics and improve electrochemical properties [116 118] Using these low dimensional nanostructures as

PAGE 100

100 the building blocks, new electron and ion transportation pathw ays can be established as a new strategy. Although many different synthesis methods were developed, these processes are complex, time consuming and difficult to be applied in large scale synthesis. To reduce the complexity and the price of precursors of conventional synthesis process, such as non ionic surfactant template method and sol gel method [119, 120] a simple spreading method was used in this paper to make novel TiO 2 flakes. In particular, the porous Ti O 2 flakes exhibit larger reversible charge/discharge capacity, better rate capability and excellent cycling stability compared to a natase titanium dioxide nano particles. The thermal reaction process of the flakes was investigated by thermogravimentic diff erential thermal analysis (TG/DTA) .The phase, structure morphology and size distribution were examined by X ray diffraction (XRD), high resolution transmission electron microscopy (TEM) and scanning electron microscopy (SEM). The surface area and pore siz es were monitored by physisorption measurements. The r elations between crystal structure, surface area, pore volume and electrochemical properties of a natase TiO 2 flakes were examined The nano sized grains of TiO 2 flakes, the pore volume and the surface a rea are all key factors contributing to the enhanced electrochemical properties. 5.2 Experimental M ethods 5.2. 1 Sample P reparation Titanium dioxide flakes (F TiO 2 ) were fabricated by spreading a mixture of stearic acid (Fisher, 99%), low surface tension hy drocarbon (Fisher, 97 %) and titanium n butoxide (Aldrich, 97 %) on the surface of high purity water (Barnstead Nanopure Infinity, 18 M /cm 1 ). Stearic acid and the hydrocarbon were used to decrease the

PAGE 101

101 viscosity of titanium dioxide precursor and enhance t he spreadability of the mixture. Typically, a ratio of 1:8 of titanium n butoxide to hydrocarbon was used to produce titanium dioxide flakes with thickness about 40 nm. The resulting slurry was washed with Nanopure water and then centrifuged (Begman BH 2) at 3000 rpm for 15 minutes to concentrate the slurry. The precursor was then suspended in isopropanol and centrifuged to remove further impurities. This process was repeated for a total of five times. After solvent exchanging with isopropanol alcohol ((CH 3 ) 2 CHOH), the flakes were dried by a supercritical fluid drying process [121] The 40 0 C heat treatments were applied using a programmable electric furnace with a desired heating rate of 1 C /min for two hours in the air to make the calcined flakes (CF TiO 2 ) 5.2.2 Characterization Differential thermal analysis thermogravimetry TG/DTA m easurements were carried out by a Pyris diamond TG/DTA at a heating rate of 20 C /min with a 200 ml/min flow rate in N 2 to understand the thermal reaction process such as weight loss and phase formation temperature of the F TiO 2 X ray powder diffractions o f all samples were taken using XRD Philips APD 3720 using a Cu Kalpha radiation source operated at 45 k V and 20 m A. The scan speed was 3 sec/step with a step size of 0.02 degree in two theta. A Quantachrome NOVA 1200 was used to perform the physisorption m easurements S urface area and average pore size can be determined based on the measurement Field emission scanning electron microscopy (FEG SEM, JEOL JSM 6335F) was carried out to investigate the particle sizes and morphologies. Images were collected with an accelerating voltage of 10 kV. The sizes of the flake s distribution were examined by Particle Size Analyzer Coulter LS13320, The flakes were suspended on a copper grid with lacey carbon and investigated by using a field emission gun JEOL

PAGE 102

102 2010F high res olution transmission electron microscopy (TEM ) with an accelerating voltage of 200 kV. The band gap of the titania samples were estimated by UV visible diffuse reflectance spectroscopy using a Perkin Elmer Lambda 800 UV/Vis spectrometer in the wavelength r ange of 250 500 nm. 5. 2.3 Electrochemical T esting The electrochemical properties of anatase TiO 2 particles (Fisher, 99.9%, average size 50 nm diameter seen by SEM), F TiO 2 and CF TiO 2 electrodes were measured using 2016 coin cells consisting of metallic li thium as an anode and an electrolyte comprised of 1M LiPF 6 ethylene carbonate (EC)/dimethyl carbonate (DMC) with 1:1 volume ratio A Celgard model C480 separator (Celgard Inc.USA) was used. Anode films were prepared from slurries of TiO 2 nanomaterials with 20 % polyvinylidene fluoride (PVDF) and 20 % acetylene carbon black dissolved in N methyl 2 pyrrolidinone (NMP). The mixtures were cast onto Al foil using a doctor blade and dried at 100 C overnight. The cast film was cut into cathode discs containing 2 3 mg of active materials before storing them in argon filled glove box (H 2 O level<1.6 ppm). The coin cells were assembled in the glove box and tested using an Arbin BT2000 instrument in galvanostatic mode. The tests were conducted between 3.5 V and 1 V at a constant current rate of C/20, C/10, C/5, C/2 C and 2C. For cycling performance, tests were conducted with a voltage window between 3.5 V and 1 V at a constant current rate of C/20. 5. 3 Results and D iscussion 5. 3.1 Materials C haracterization Figure 5 1 shows the TG DTA curves for the F TiO 2 with a weight loss of about 9 % at the temperature range from room temperature to 400 C which was resulted

PAGE 103

103 from either the desorption of water and isopropanol or the decomposition of organic compounds. According to the DTA data, a broad endothermic peak around 150 C is present and is attributed to the loss of water and the organic residues absorbed at the surface of the flakes There is a relatively small exothermic peak at 255 C that was contributed to the decomp osition of titanium n butoxide in the pristine F TiO 2 The observed peaks shown at 400 C and 600 C were contributed by the phase transformations from amorphous to crystalline anatase phase, and from anatase phase to the rutile phase, respectively. C rystalline structural changes of the F TiO 2, CF TiO 2 and particles ( Fisher, average 50 nm diameter) were monitored by XRD. As shown in Figure 5 2, the F TiO 2 shows broadened and weak Bragg peaks, which indicate that the flakes are partially amorphous wit h presence of the crystalline anatase phase. After a 2 hr heat treatment at 400 C the expected phase transformation from the amorphous to crystalline anatase titanium dioxide was confirmed by the seven distinct characteristic diffraction peaks. The XRD r esults were identical to the TG DTA results confirming that the heat treated flakes were converted to a pure crystalline anatase phase. There is no indication of any rutile phase by XRD. Compared to the powder sample, the diffraction peaks of the flakes ar e broadening and overlap with each other due to the smaller grain size. The average crystal grain size can be calculated by the Scherrer equation (5 1) Where d is the calculated grain size, is the wavelength of X ray (Cu K 1.54 ), B is the full width at half maximum intensity, and B is the Bragg diffraction angle. The grain sizes of F TiO 2 and CF TiO 2 are 4 nm and 9 nm respectively.

PAGE 104

104 SEM was performed to examine the particle size and mor phologies of TiO 2 flakes. The surface morphology of F TiO 2 and the thickness are shown in Figure 5 3(a) ( d ). The lateral dimension of the F TiO 2 was found to be in the order of 20 3 (Figure 5 3(a)) and the thickness of these flakes is approximately 4 0 nm (Figure 5 3 (c)). The apparent thin sheet morphology was explored and the aspect ratio of the lateral dimension to thickness ranged from 250:1 to 500:1. The flakes were further treated by calcination at 400 C in air for 2 hrs. Aggregation was not app arent when comparing the CF TiO 2 (Figure 5 3(b)) with the F TiO 2 (Figure 5 3(a)). Moreover, the thickness of the CF TiO 2 did not change by the heat treatment, as shown in Figure 5 3(d). A comparison of the measured particle size distribution of F TiO 2 and C F TiO 2 by laser diffraction is shown in Figure 5 4. With the differential volume distribution of flakes that moved in turbulence and passed the laser beam, the maximum diameter was measured through the average random orientation of the flakes. The lateral dimension of flakes were estimated by dispersing both flakes in deionized water with liquid modules and were fairly close to the results taken from the SEM images (Figure 5 3(a) and (b)). The D10, D50, D90, mean and standard deviation values for the partic le size distributions are shown in Table 5 1. F TiO 2 has a broad size distribution spanning from 1 50 for the F TiO 2 and CF TiO 2 are of 39.1 and 23.5 m, respectively. It is evident that some larger flakes broke or cracked during dehydrat ion and crystallization. The crystalline structure of the flakes was further investigated under HR TEM. The images show that the F TiO 2 and CF TiO 2 were comprised of crystalline platelets

PAGE 105

105 of about 5 and 8 nm in diameter (Figure 5 5(a)(b)), respectively. T he interference lattice fringes can be seen in the TEM images and has a separation distance of 0.35 nm, corresponding to the interplanar spacing of the (101) planes for anatase [122] Random orientation of individual grains over both F TiO 2 and CF TiO 2 samples is observed from the concentric diffraction rings in the selected area diffraction mode and consistent with the anatase (101), (004), (200), (105) diffraction planes (the insets of Figure 5 5(a) and (b)). On a closer inspection, an amorphous layer can be seen surrounding the smaller crystallites in the F TiO 2 sample. An overall dense structure was seen in the F TiO 2 as shown in Figure 5 5(c). After the calcination process, pores with an average size of 7 nm were developed, as shown in Figure 5 5(d), due to local rearrangement caused by the growth of crystal grains and by removing the residual solvent. Consequently, it is considered that the CF TiO 2 are polycrystalline consisting of fine nano grains and nano size pores. In order to investigate the change of s urface area and por osity after the calcination treatments the nitrogen absorption isotherms were used H igher specific surface areas of anode materials can increase the electrode and electrolyte contact area. Compared to a commercial TiO 2 nanoparticles, the surface area of F TiO 2 was 15 times higher (in Table 5 2), while the surface area of the CF TiO 2 was still 7.5 times higher than that of TiO 2 nanoparticles. The BET measurements also show that the pore size of the calcined flakes is about 7.2nm, which is consistent with t he TEM observation. Figure 5 6 shows the typical UV visible diffuse reflectance spectra in the wavelength range of 250 500 nm for the F TiO 2 and CF TiO 2 The sharp decrease in the diffuse reflectance in the UV region is attributed to the fundamental light absorption of

PAGE 106

106 the TiO 2 flakes. The spectral profile for CF TiO 2 was apparently blue shifted compared with that of the F TiO 2 The general relation between the absorption coefficient and the band gap energy is given by the equation: (5 2) W here m is an index depending on the nature of the electron transitions, is the absorption coefficient, h is the Planck constant, is the frequency of electromagnetic radiation, and Eg is band gap energy of the semiconductor. The estimated band gap energies of the as prepared and calcined flakes are 3.25 and 3.33 eV, respect ively. Both values are higher than the band gap energies of nanoparticles (3.2 eV). The increase of the band gap is most likely due to the quantum confinement effect [122] 5. 3.2 Electrochemical P roperties 5.3.2.1 Charge d ischarge s tudies The electrochemical properties of the TiO 2 powder, F TiO 2 and CF TiO 2 were investigated using lithium half cells with a constant current 0.1 mA/cm 2 The first discharge /charge profiles between 3.5 and 1.0 V are shown in Figure 5 7(a)(b) As Figure 5 7(a) shows, there is a plateau around 1.75 V for TiO 2 nanoparticles and CF TiO 2 that corresponds to the occurrence of the two phase equilibrium of Lithium poor (Li 0.01 TiO 2 ) and Lithium rich (Li 0.6 TiO 2 ) domains [123] However, there is no voltage plateau observed in the F TiO 2 The amorphous phase of the F TiO 2 seen by TEM a nd the smaller grain size [99] may be the contributing factors to the sloping voltage curve. The lithium interca lation/deintercalation reaction in this region can be written as ( 5 3 )

PAGE 107

107 The amount of lithium insertion depend s on the crystallinity, morphol ogy as well as the microstructure. Generally, increasing the specific surface area or decreasing the particle size for anatase TiO 2 particle s can increase the discharge capacity [100] In our work, the flakes show a similar trend. The capacity, grain size and specific surface area for F TiO 2 CF TiO 2 and nanoparticles ar e (238 mAh /g,5 nm 309 m 2 /g ), ( 213 mAh/g, 8 nm 151 m 2 /g ), and (170 mAh/g 50 nm 20 m 2 /g ) respectively Although the F TiO 2 ha ve the largest 1 st cycle discharge capacity the first cycle charge capacity is very small, as shown in Figure 5 7(b). Apparently the lithiation reaction was not reversible for the F TiO 2 Although the surface area and the grain size are key factor s to determine the electrochemical properties, the amorphous phase (Figure 5 2, Figure 5 5(a) ) residual organic precursors (Figure 5 1) a nd denser microstructure (Figure 5 5(c) ) lead to the poor electrochemical performance. The supercritical drying process during the preparation of the F TiO 2 can induce a partially crystallized anatase phase and keep the largest surface area, but the enha nced electrochemical performance can be achieved C post synthetic calcination process. For the anatase TiO 2 nano particles the discharge/charge capacity is about 170 mAh/g corresponding to Li 0.5 TiO 2 that is identical to the conventional anatase TiO 2 materials. The C F T iO 2 exhibit a 1 st cycle discharge capacity exceeding 200 mAh/g and subsequent discharge/charge capacities of 180 mAh/g. Interesting observations can be made through comparing the discharge and charge curves in Figure 5 7: The length of the biphasic region (plateau region) corresponds to the number of lithium ions inserted into the bulk lattice [100] Therefore, the capacity in the plateau region should decrease for smaller grain sizes, as observed by others [100] Although the grain size of the CF

PAGE 108

108 TiO 2 (8 nm) is much smaller than that of the nanopowders (50 nm), t he capacity of the plateau region is similar to that of nanoparticles. Ob viously, the CF TiO 2 still keep some bulk characteristics O n the other hand, the capacity of the sloping region for CF TiO 2 is larger than that in the TiO 2 nano particles, thus implies that Li ions are further inserted into the surface layer [101] Upon the first charging, a sloping region can be clearly observed at the beginning of the charge as well Wagemaker et al have shown that anatase TiO 2 nanoparticle s can completely convert to the Li 1 TiO 2 phase if the particle size is smaller than 7 nm as shown in Figure 5 8(a) [9 9] The grain size s of the flakes are between 5 8 nm which suggests the new Li 1 TiO 2 phase may form during the lith i ation process and lead to a larger capacity compared to that of the nanoparticles ( 50 nm) In theory, the bulk Li 0.5 TiO 2 and LiTiO 2 phase ha ve distinct diffraction peaks and the extra peaks should be detected by the ex situ XRD. However, it is difficult to differentiate the overlapping peaks of nano sized Li 0.5 TiO 2 and LiTiO 2 because the serious peak broadening for the nano sized materials. Th e XRD patterns of Li 0.5 TiO 2 and LiTiO 2 were simulated using the Crystal Diffraction software with 5 nm particle size, shown in Figure 5 8 (b) .The major peaks of Li 0.5 TiO 2 and LiTiO 2 overlap significantly. In practice, the peak broadening from the laboratory instrument also has to be considered. Therefore, the Li 1 TiO 2 phase cannot be identified clearly by conventional laboratory XRD. High resolution neutron diffraction experiments are necessary to identify the detailed structural change. According to Wagemaker et al[4], the neutron diffraction has identified the new LiTiO 2 phase in 7 nm TiO 2 nanoparticles. 5. 3.2.2 Rate apability performance To study the rate capability of TiO 2 nanomaterials electrodes were made with similar loading of active materials in each cell (about 5 mg active material/cm 2 ). For

PAGE 109

109 each material, the cells were discharged to 1 V at constant rates (C/20) and charged to 3.5 V at various current rates (C/20, C/10, C/5, C/2, C and 2C). The measured 1 st cycle charge capacities with different cur rent density were shown in Figure 5 9 Clearly, CF TiO 2 show s better rate performances than TiO 2 nano particles The charge capacity of the CF TiO 2 tested at 2C still retains 48.44 % of its capacity, compared to the capacity at C/20. On the other hand, the charge capacity of TiO 2 nanoparticles tested at 2C decreased to 26.81 %, compared to the capacity at C/20 N ano size effects of electrochemical energy storage systems can be divided into two categories ; [117] The first is the true size effects, which involve changes of actual materials, including the change of structure and phase stability, electronic band gap and intrinsic diffusivity etc The second is the tr ivial size effects, and the result of trivial size effect is largely relied on the surface to volume ratio the shape and the pores of the nanomaterials. TiO 2 is a n intrinsically low electronic conductive material. The quantum confinement effects show th at the band gap will increase for decreased particle size, confirmed in our UV Vis diffuse reflectance measurement. The band gap of nanoparticles, F TiO 2 and CF TiO 2 are 3.20 eV, 3.25 eV and 3.33 eV respectively. This implies that the intrinsic electronic conductivity of the calcined TiO 2 flakes will decrease. Moreover, Wagemaker et al [124, 125] used static 7 Li NMR T2 relaxation measurements to detect the Li ion mobility of micro and nanosize anatase TiO 2 The resu lt shows that the lithium diffusion coefficient for nanomaterials is lower than that for the microsized materials. Nevertheless, our electrochemical data clearly show that the CF TiO 2 have superior electrochemical properties.

PAGE 110

110 For the flakes, the trivial s ize effects should be the key factors that contribute to the excellent rate capability. The specific surface area of the CF TiO 2 is 7.5 times than the nanoparticles. The large surface area can lead to larger electrode/electrolyte contact area that can impr ove the ionic conductivity. At the same time, nanostructured materials can supply the short Lithium ions and electron transportation distance even though the nanomaterials have intrinsically lower electronic and ionic conductivity. Considering that the dif fusion length is scaled with time by l t 2 the Li diffusion time may be much lower for 8 nm grains in the CF TiO 2 Moreover, the l ower specific current density of flakes can be contributed by the larger surface area of nanostructured electrodes and the exc ellent rate performance can be achieved by the l ower specific current density [117] 5. 3.2.3 Cycling performance The cycling performance of CF TiO 2 and TiO 2 nano particles at a constant current density (C/20) was shown in Figure 5 10 TiO 2 nanoparticles show a fast fading with the increasing of cycl e number while there is virtually no fading for the CF TiO 2 The F TiO 2 have poor cycling performance due to the residual solvent or amorphous phase present. CF TiO 2 with nano grains and nano porous structure show an excellent cycling performance with high capacity. This makes it become a excellent anode candidate for Li ion batteries. T he porous structure is ex pected to supply the space to accommodate the volume change and preventing the undesired fractures or electronic disconnects of the electrode materials during the de lithiation and lithiation process [107] On the other hand, the lack of a two phase boundary in nanosize < 8 nm TiO 2 is another possible reason for the improved cycling performance. Previous work by neutron diffraction shows two different lithiated mechanisms for the bulk and nano size anatase

PAGE 111

111 materials [99, 126] For bulk materials, the Li titanate crystal phase will grow at th e cost of the crystal anatase phase and induce extra strain during the li thia tion process. However, for nano sized particles (7 nm), either the anatase or the Li titanate phase is detected during the lithiation process This indicates that in extremely sma ll particles, the two phases do not coexist in the same particle during the intersion process as shown in Figure 5 11 The unique nano grains and nano porous structure in CF TiO 2 and the possible elimination of the phase boundary during lithiation lead to the excellent cycling performance. The advantage of the porous structure can be tarnished by the low volumetric energy density. However in the case of flakes, it is possible to improve the volumetric energy density by packing the sheets of flakes aligned in a horizontal orientation. This methodology is in development and will be reported elsewhere. 5 4 Summary In this work, titanium dioxide flakes were fabricated by a simple spreading method This proce s s is eas y and scalability is high The low surface t ension hydrocarbon and titanium n butoxide were used as the precursors and continuously added into the solution on the surface of flowing water. A supercritical drying process was performed to remove the residual water. The size of the synthesized flake is about 40nm in the lateral dimension and was composed of 5nm nanograins. After a 400 C calcination process, the flakes transformed in to a pure anatase phase that is free of any impurity. In this phase, the flakes are characterized to have 8 nm grains and the 7nm pores. The crystal structure, surface area, pore size, electronic properties and electrochemical properties are compared for flakes and nanoparticles T he electrochemical and structural properties of the flakes are significantly influenced by

PAGE 112

112 subse quent heat treatment process The CF TiO 2 exhibit s larger reversible charge/discharge capacity, better rate capability and excellent cycling stability The nano sized grains (8 nm) may lead to the absence of two phase in single particles during lithiation, improving the cycling performance significantly A larger interface between electrod e and electrolyte, the short diffusion path of lithium ions and electron and lower specific current density of in the CF TiO 2 lead to a much better rate capability

PAGE 113

113 Ta ble 5 1 Particle diameter statistics for F TiO 2 and CF TiO 2 D10(m) D50(m ) D90( m) Median(m) StandardDeviation(m) F TiO 2 5.2 39.1 81.6 20.8 58.9 CF TiO 2 5.1 23.5 50.1 16.4 18.8 Table 5 2 Surface area and pore size of TiO 2 particles, F TiO 2 and CF TiO 2 Specific surface area (m 2 /g) Specific pore volume (cm 3 /g ) Average pore diameter (nm) F TiO 2 309 CF TiO 2 151 0.511 7.2 Particles 20

PAGE 114

114 Figure 5 1 TG/DTA measurements of F TiO 2 Figure 5 2 X ray diffraction patterns of TiO 2 samples.

PAGE 115

115 (a) (b) (c) (d) Figure 5 3 FESEM images of TiO 2 samples. (a) F TiO 2 (b) CF TiO 2 (c) edge view of F TiO 2 (d) edge view of CF TiO 2

PAGE 116

116 Figure 5 4 Particle size distribution for F TiO 2 and CF TiO 2 by laser diffract ion

PAGE 117

117 (a) F TiO 2 (b) CF TiO 2 (c) F TiO 2 (d) CF TiO 2 Figure 5 5 HR TEM images of TiO 2 samples (the SAD pattern as ins et). (a) F TiO 2 (b) CF TiO 2 The diffraction rings are indexed as (1) 101 (2) 004 (3) 200 (4) 105 for anatase. (c) F TiO 2 (d) CF TiO 2

PAGE 118

118 Figure 5 6 UV visible diffuse reflectance spectra of F TiO 2 and CF TiO 2

PAGE 119

119 (a) (b) Figure 5 7 (a) First disc harge curves of TiO 2 samples (b)First charge curves of TiO 2 samples

PAGE 120

120 (a) (b) Figure 5 8 (a)Neutron diffraction patterns of maximum lithiated LixTiO 2 for different particle sizes [30] (b) The XRD patterns of Li 0 5 TiO 2 LiTiO 2 simulated by Crystal Diffractio n

PAGE 121

121 Figure 5 9 Rate capability comparison of CF TiO 2 and TiO 2 particles Figure 5 10 The variation of specific discharge capacity with respect to the cycle number of CF TiO 2 and particles

PAGE 122

122 Figure 5 11 Lithiation mechanism of bulk and nanosized m at erial [ 30]

PAGE 123

123 CHAPTER 6 A STUDYOF S URFACE C HARACTERISTISTICS IN LINI X MN 2 X O 4 HIGH VOLTAGE SPINEL MATER IALS FO R L ITHIUM I ON B ATTERIES 6.1 Introduction Capacity fading do es not create from single source the fading mechanism was contributed by interaction of different processes Since the processes happen at same timescales the understand ing of fading m e chanism s is limit The major factors induce the capacity fading are summarized as follows [12, 65, 1 27, 128] 6.1.1 Inactive M aterials Conventional cathode electrode consisted of active material conductive additive and polymer binder to enhance the electronic conductivity and binding the particles together. The corrosion of metal, the reaction of binde r and carbon additives with electrolyte during charge/discharge process may induce capacity fade. 6.1.2 Structural C hanges D uring C ycling (Jahn Teller D istortion) If an extra lithium ion is introduce d into LiMn 2 O 4 lattices, the valence of Mn will decrease a nd t he injected electrons will occupy Mn e g orbitals The enhanced interaction of Mn e g and O 2p orbitals create the instability of the lattice. In order to decrease the repulsion force, the lattice expands and the cubic phase transfer to tetragonal struct ure [9] The capacity loss on cycling at 3 V is thought to be mainly a result of Jahn Teller distortion, which causes a 16% increase in the c / a ratio of the unit cell during extraction and insertion of lithium from or into this spinel. 6.1.3 Structural C hanges o f D elithiated L ithium M anganese Spinel ; With the delithiatian process an ordered superstructure L i 0.5 Mn 2 O 4 [129] will form with the double hexagonal phase [130] and lead to the structure change during

PAGE 124

124 charge/discharge process. The phase transformation should be avoided in order to keep the stable structure. 6.1.4 Chemical D issolution R eaction ; At high tempe rature, the metal dissolution reaction in electrolyte l eads to serious capacity fading The weight loss of manganese dissolution leads to the decreasing of capacity Moreover the impedance increases due to the compound formed by the side reaction such as MnF 2 and MnCO 3 Finall y the Mn 3+ will move to anode and included in the SEI Manganese ions can be dissolved by two different mechanisms. At low operating voltage, Mn 3+ will disproportionate into Mn 3+ and Mn 2+ the basic chemical con be expressed as follows [131] : 2Mn 3+ Mn 4+ + Mn 2+ The Mn 2+ ions will be dissolved in to the electrolyte and lost the weight of a ctive mate r ial At high voltage, The hydrolysis reaction of the LiPF 6 electrolyte leads the formation of LiPF 5 and F 2 If any water present, the active material will be dissolved by the formed HF [132] 6.1.5 Oxidation o f E lectrolyte C omponents A nd S urface F ilm F ormation The increase of the impedance will induce the power loss and the high temperature and high operation voltage will accelerate the reaction. T he oxidation of electrolyte components and the carbonate film formed at surface induce the surface films. Some gaseous also formed by the side reaction of the electrolyte [134] This work aims to fabricate the high voltage LiMn 2 x Ni x O 4 and explore the intrinsic surface stability in the high voltage LiMn 2 x Ni x O 4 spinel type cathode. Compare

PAGE 125

125 to LiMn 2 O 4 cathode, previous research indicate the oxidation of electrolyte components and the carbonate film formed at surface are the major sources inducing capacity fading of LiMn 2 x Ni x O 4 cathode. However, detailed information p ertaining to the electrode/electrolyte interface are still not properly investigated. A solid electrolyte interface (SEI) typically forms at the cathode and is said to originate from the decomposition products of the electrolyte at very high (>4.4V) voltag es. While the SEI layer can protect the electrode if stable, it also creates a barrier for lithium diffusion during electrochemical cycling and results in an increase in cell impedance and decreases the cycling efficiency. Several groups have reported on t he characterization of the SEI layer using composite electrodes containing conductive agents and polymeric binders and show that it mainly consists of lithium alkyl carbonates (ROCO 2 Li), phosphates, LiF, and Li 2 C O 3 stemming from the electrolyte components [135 146] However, these results are typically not a direct representation of the surface stability in the cathode material, rather an overall composite study, as the conductive agents and polymer binders tend to react with the electrolyte at high voltages and accelerates the formation of the SEI layer. Thin film electrodes offer the ability to investigate the intrinsic activity in the cathode surface, free of po lymeric binders and conductive agents, while the controlled geometry enables more accurate measurements of ionic diffusion coefficients. To adequately assess the intrinsic surface stability this work aims to compare both conventional powder composites and thin film electrodes after extended cycling.

PAGE 126

126 6.2 Experiment Method 6.2.1 Synthesis of P owders The sol solutions were prepared from the stoichiometric mixtures of Li(CH 3 COO)2H 2 O (Fisher), Ni(CH 3 COO) 2 4H 2 O (Fisher), and Mn(CH 3 COO) 2 4H 2 O (Fisher) in distill ed water. Then the solution was added drop wise to a solution of citric acid which was used as a chelating agent pH value was adjusted to 6.5 by adding. ammonium hydroxide A gel was obtain ed by heated at 75 C overnight The gel were calcined at 450 C for 10 hours to remove the carbonate in air and calcined at 800 C for 10 hours. 6.2.2 Synthesis o f Thin Films LiNi 0.5 Mn 1.5 O 4 target was synthesized by a solid state reaction. MnO 2 NiO and LiOH precursors were mixed by 3d ball mill er for 5 h r and were col d pressed into a pellet and calcined in air at 750C for 24 h. Then, the pellet was grinded and re fabricate by static press machine and calcined at 900C for 2 h The mole rat i o of Lithium and transition was controlled at 1.3 in order to avoid the evapor ation of lithium during the deposition process The LiNi 0.5 Mn 1.5 O 4 thin films were deposited on stainless steels (SS304) or SiO 2 /Si. In the deposition process KrF excimer laser with 248 nm was used. Laser fluence with 500 mJ and repetition rate with 5Hz was used in the process S ubstrate temperature was controlled at 600 and in the oxygen partial pressure at 300mTorr for 40minutes and calcined with extra 20minutes in oxygen atmosphere. 6.2.3 Characterization X ray diffraction (XRD) data were collected using a Philips APD 3720 diffractometer using a Cu K radiation so urce operated at 45 k V and 20 m A.The scan

PAGE 127

127 speed was 3 sec/step with a step size of 0.02 degrees two theta. SEM Images were collected using a field emission scanning electron microscopy (FEG SEM, JEOL JSM 6335F) X ray photoelectron spectroscopy data were co llected using a PHI 3056 spectrometer equipped with a hemispherical detector, 54.7 degrees off normal, with a Mg anode source operated at 15kV. The energy of the spectra was calibrated by the binding energy of the hydrocarbon C1s (C H) at 284.6 eV, which c orresponds mainly to the carbon black in the composite electrode and adventitious carbon. The data was analyzed using the software CasaXPS and all peaks were fit using a Shirley type background. Powder samples and thin Film samples were disassembled in an Ar filled glove box, washed with acetonitrile three times, then transferred to ORNL in a custom made transfer case designed to prevent atmospheric gases from reacting with the samples. The samples were subsequently transferred to XPS, from the glove box, without exposure to air 6.2.4 Electrochemistry The electrochemical properties of power electrodes were measured using 2016 coin cells consisting of metallic lithium as an anode and an electrolyte comprised of 1M 1M LiPF 6 in ethylene carbonate (EC)/dimethyl carbonate (DMC) with 1:1 volume ratio Celgard type C480 (CelgardInc.USA) was used as separator. Cathode films were prepared from slurries of LiNi 0.5 Mn 1.5 O 4 powders with 7.5% polyvinylidene fluoride (PVDF) and 7.5% acetylene carbon black dissolved in N me thyl 2 pyrrolidinone (NMP). The mixtures were cast onto Al foil using a doctor blade and dried at 100 C overnight. Electrode discs were cut into cathodes containing 2 3 mg of active materials before

PAGE 128

128 storing them in an argon filled glove box (H 2 O level<2 pp m). The coin cells were assembled in an argon filled glove box and tested using an Arbin BT2000 instrument in galvanostatic mode. The tests were conducted between 5V and 3.5V at a constant current rate of C/20.The The electrochemical properties of cathode thin film was measured using Swagelok type cell consisting of metallic lithium as an anode, an electrolyte comprised of 1M LiPF 6 in a 1:1 volume fraction of ethylene carbonate (EC)/dimethyl carbonate (DMC) and Celgard type C480 separator (CelgardInc.USA). 6.3 Results a nd Discussion 6.3.1 Char acter ization SEM images of both the powder and thin film samples are shown in Figure 6 1 Formation of the spinel structure at 600C using the PLD method shows a dense yet relatively faceted crystal morphology with well defined grains ranging between 100 200 nm. The grains exhibited an octahedral type morphology reflecting the cubic spinel structure. Compared to the PLD method, the sol gel synthesis produced 100 nm particles with a pseudo polyhedral morphology when calci ned at 800C. Figure 6 2 (a) shows the XRD patterns for the NiMn spinel thin film deposited on a stainless steel at 600C and the sol gel synthesized powder. All diffraction peaks from the thin film electrode can be indexed to the spinel structure with an F d 3m space group and the SS substrate. A (111) textured structure is seen for the thin film, as evidenced by the XRD pattern, resulting in larger relative intensities for the (111)/(311) and (111)/(222) peaks compared to that of the synthesized Ni Mn spinel powder material. Rietveld refinement on the powder material similarly shows the disordered Fd 3m space group with no NiO impurities as shown in Figure 6 2 (b) [147] The refinement results show the lattice

PAGE 129

129 parameter of 8.1804, similar to that of previously published data [13] For the disordered (Fd 3m) structure both the Mn 4+ and Ni 2+ ions occupy the o ctahedral 16(d) sites 6.3.2 Electrochemical M easurement The electrochemical properties of the NiMn spinel powder and thin film were investigated in lithium half cells. The charge/discharge profiles between 3.6 and 5.0V are shown in Figure 6 3 (a) and (c). The NiMn spinel powder and thin film electrodes have a reversible capacity of ~130 mAh/g and ~125 mAh/g, respectively. In addition, both cells have two distinct voltage plateaus around 4.7V, which corresponds to the Ni 2+ Ni 4+ redox couples. A relatively sm aller plateau can be found in the 4.0V region for the thin film material during the first charge cycle corresponding to the Mn 3+ Mn 4+ redox couple. The high vacuum deposition process of the PLD system may cause oxygen loss and can lower the average oxidati on state of Mn and introduce more Mn 3+ surface ions. The lower deposition temperature associated with the thin film process may also be attributed to the lower reversible capacity compared to the powder electrodes. To study the rate capability of the NiM n spinel electrodes, half cells were charged to 5.0V at a constant current density (C/20) and discharged to 3.6V at various current densities. The measured discharge capacities with the various current densities are shown in Figure 6 3 ( b) and (d ). When the r ate increases from C/20 to 2C for the thin film electrodes only a relatively small capacity loss (~15%) is shown. In contrast, capacity loss the composite electrode is ~40%. The rate capability performance for the thin film electrode in the absence of car bon black and binder indicates the intrinsically fast Li diffusion in the NiMn spinel material due to the 3D Li ion diffusion pathway. Cycling performance tested at number at a constant current density (C/20) shows little capacity fading after 5 cycles fo r both composite and thin film electrodes as shown in Figure 6 4

PAGE 130

130 6.3.3 X ray P hotoelectron S pectroscopy X ray photoelectron spectroscopy was used to examine surface features of the thin film and powder electrode materials after electrochemical cycling. A ll data was collected by Dr. Carroll in Oak Ridge National laboratory. The thin film batteries were disassembled, in the fully discharged state, in an argon filled glove box and transferred to the XPS using a load lock system to eliminate exposure to air. Figure 6 5 shows the spectra observed for the C1s, Ni2p, and F 2p regions for pristine and cycled samples. The thin film electrodes offer the ability to probe the surface of the cathode material more accurately than that of the composite powder electrodes. This is critically important when trying to determine failure mechanisms associated with the cathode material. For example, the thin film electrodes show no significant formation of decomposed electrolyte such as the formation of carbonate species after el ectrochemical cycling, evident in the C1s data (Fig 6 5 d ). This suggests that even at the higher voltages (5V) during charging, the NiMn spinel material is intrinsically stable and does not contribute to the decomposition of the electrolyte. In contrast, t he composite electrode shows several peaks, which are associated with the carbon ad ditive and PVD F binder (Figure 6 5 a ). The C1s spectra can be de convoluted into 4 peaks. The peak at ~284.6 eV (C C, C H) corresponds mainly to carbon black while the peaks at 285 eV ( CH 2 ), and 290.5 eV (CF 2 ) correspond to the PVD F binder. [148] A peak at 287 eV (C O) corresponds to the EC/DMC electrolyte. The powder comp osite electrodes shown a gradual increase in the C O peak with increasing cycle numbers corresponding to the decomposition of the ethylene carbonate and dimethylene carbonate (EC/DMC) electrolyte solution. In addition, the O1s spectra for the composite ele ctrodes show the gradual formation of a C O (~533 eV) species after cycling

PAGE 131

131 consistent with the C1s spectra for the decomposition of the EC/DMC electrolyte during cycling as shown in Figure 6 6 (a). In contrast, no C O formation is seen for the thin film el ectrode after extended cycling as shown in Fi g ure 6 6 (b). Changes can also be seen in the Ni2p region where a significant loss of spectral features, particularly a loss in intensity and peak shape, are present for the composite electrode (figure 6 6b). Thi s can be attributed to a large signal from the carbon black and PV D F binder and a relatively lower stoichiometry when compared to Mn in the spinel structure, resulting in a lower percentage of surface exposed Ni. Loss of spectral intensity detracts from ad equately assessing the effect cycling on the surface of the NiMn spinel material. The Ni2p region scans for the thin film, however, have a much higher signal to noise ratio than those of the composite electrode due to the elimination of carbon black and PV D F binder (Fig 6 6e). The Ni2p region shows a slight loss in spectral resolution and broadening of the Ni2p 3/2 peak upon extended cycling. It is hypothesized that this may be contributed to a higher nickel valence (Ni 3+ /Ni 4+ ) and suggests some irreversibil ity of the surface Ni species, which contributes little to the total capacity. De convolution of the Mn 2p 3/2 spectra for the thin film and composite electrodes reveals two peaks, a minor peak at 641.2 eV (Mn 3+ ) and major peak at 642.5 eV (Mn 4+ ) shows Mn 3+ appear similar to previous reports by Shaju et al. [149, 150] The F1s and P2p region scans offer the most contrast between the thin film and composite electrodes for the formation of the SEI layer. It has been rep orted for composite electrodes, LiPF 6 will react and form LiF and PF 5 as byproducts, which further reacts with water forming POF 3 and HF on the surface of the particles. [128] This trend is consistent with our composite electrode s, which show the presence of

PAGE 132

132 phosphates after electrochemical cycling, shown in F1s spectra (figure 6 5 c ) However, in the thin film XPS analysis, no phosphate formation is shown and only a small amount (<5%) of LiF after 20 cycles, which suggests minimal electrolyte decomposition on the surface of the electrode after extended cycling. 6.4 Summary The results from this analysis implies that the surface of the NiMn spinel material is not directly contributing to the SEI formation mechanism; rather the SEI formation seen in the composite electrodes is a result of the incor poration of carbon black and PVD F. These results may also explain the enhancement seen in the rate testing for the thin film electrodes relative to the composite electrode, which enables fa sters Li mobility without the formation of an SEI la yer and without the inactive PVD F binder.

PAGE 133

133 Figure 6 1 SEM images of thin film (a,b) and powder (c,d) NiMn spinel electrode materials.

PAGE 134

134 (a) (b) Figure 6 2 (a) XRD patterns for thin film and powder composite NiMn spinel electrodes (b)The refinement pattern of the composite powder NiMn

PAGE 135

135 Figure 6 3 Charge/discharge profiles and rate testing for NiMn spinel powder (a,b) and thin film (c,d) electrodes. Figure 6 4 Discharge capacity performance as a function of cycle number for thin film and powder electrodes.

PAGE 136

136 Figure 6 5 XPS data collected using a Mg anode on pristine and post electrochemically cycled powder composite (a c) and (d f) thin film NiMn spinel electrodes.

PAGE 137

137 (a) (b) (c) (d) Figure 6 6 XPS plots of the Mn2pand O1s region scans for t he composite (a,c) and thin film (b,d) electrode s

PAGE 138

138 CHAPTER 7 SOLID ELECTROLYTE DE POSITION BY PLD In order to fabricate all solid state thin film batteries, the solid electrolyte must be prepared following the cathode electrolyte deposition. The LVSO t hin films were deposited using the PLD system with different oxygen pressure. The amorphous and dense film can be prepared in the low oxygen pressure. Although the XPS results show the existence of oxygen vacancy, the EDS and LEAP analysis show exact stoic hiometry and uniform distribution of each element. 7.1 Introduction An suitable electrolyte material should follow some requirements : high ionic conductivity, no chemical reaction between electrodes and electrolyte and high electrochemical decomposition vo ltage [151] Moreover, with the aim of examin ation the reaction of interface by TEM in situ or ex situ the solid electrolyte should have dense structure and keep stable in the electron beam(TEM) and ion beam(FIB). Several methods were developed to prepare solid electrolyte, including solid state method [152] thermal evaporation [153] sputter method [154] and PLD method [155 158] I n order to fabricate th e all solid thin film batteries the PLD deposi tion process was selected as the deposition method in this research because it is easier to fabricate dense film and it is possible to fabricate the all solid state thin film batteries only by pulsed laser deposition method. However, it is difficult to fabricate the solid electrolyte films by PLD technique : Li 2 O SiO 2 V 2 O 5 (LVSO) [157 161] and Li 2 O P 2 O 5 +N 2 (LiPON) [40, 41, 153, 156, 162 164] are the widely used electrolyte can be prepared by PLD technique In this research, Li 2 O SiO 2 V 2 O 5 (LVSO) was deposited by PLD process..

PAGE 139

139 7.2 Experiment Method 7.2.1 Thin F il ms P reparation The Li 2 O SiO 2 V 2 O 5 ( LVSO ) target was brought from Plasmaterials. Energy of the laser system and frequency were set at 300mJ, 5Hz .The films were deposited at room temperature and 5m t orr 200mtorr oxygen pressure Stainless Steel 304 and SiO 2 /S i substrate were used as substrate. 7.2.2 Characterization X ray diffraction (XRD) data were collected using a Philips APD 3720 k V and 20 m A. The scan speed was 3 sec/step with a step size of 0.02 degrees two theta .X ray diffraction (XRD) data of LVSO target was collected using Philips XPert MRD because of the thickness issue. Images were collected using a field emission scanning electron microscopy (FEG SEM, JEOL JSM 6335F) The energy dispersive X ray spectroscopy( EDS) system of JEOL SEM 6335 was also used to identify the ratio of V:Si element. X ray photoelectron spectroscopy (XPS) Perkin Elmer PHI 5100 System was performed using ray source to investigate the valence number of each element. The incident X ray beam was 45 off normal to the sample while the X ray photoelectron detector was normal to the sample. Binding energies of the photoelectron are co rrelated to the C 1s peak at 285 eV. A pass energy of 22.36 eV and 30 scans per region were taking with a step size of 0.100 eV Local electrode atom probe (LEAP) facility at Alabama University, Tuscaloosa was used to get 3D compositional images: LEAP Mode l is LEAP 3000XSi the laser wavelength and pulse Energy used is 532 nm : 0.5 nJ separately. The experiment was performed by Dr. Santhanaqopalan in my group.

PAGE 140

140 Local electrode atom probe is a unique analytical microscopy with highest spatial resolution pro ducing 3D compositional images. A sharp specimen is used for obtaining such 3D compositional images. A high voltage is applied to the apex of the sharp specimen to create electric fields of the order of 10 10 V/m or even higher. Such a high field can strip off the electrons of the atoms sitting at the tip of the specimen by the phenomena called field ionization. The ionized atoms are pulled off from the surface and accelerated by the electric field towards the counter electrode [165] This process of pulling off of ionized atoms is called field evaporation. To produce such high e lectric fields for field evaporation the specimen tip radius of curvature must be about 100 nm or less and the voltage must be around kV. The field evaporated ions are detected by a large area 2D position sensitive detector to give the X Y position of atom s. The tip apex is about 100 nm across and the detector is 100 mm across and hence, the image of atoms on the surface is projected onto the detector with a projection magnification of about one million times. At this magnification, interatomic distances of 0.2 nm are 0.2 mm on the detector, which is relatively easy to resolve. More importantly with this technique the ionization voltage is controlled in a fashion to evaporate one atom at a time, so that the sequence of evaporating events would give informati on about the z position. 7.3 Results A nd Discussion Figure 7 1 shows the XRD pattern of the L 3.4 V 0.6 S 0.4 O 4 (0.6(Li 4 SiO 4 ) 0.4(Li 3 VO 4 ))target. Although there are some small peaks the major peaks belong to crystalline Li 4 SiO 4 and Li 3 VO 4 p hases that means the target is suitable for subsequent deposition process.

PAGE 141

141 Figure 7 2 shows the XRD pattern of the LVSO film deposited in different pressure range, all films are amorphous that are ident ical with the previous literature. Figure 7 3 shows the SEM pictures of LVSO film deposited in different pressure range. When the film deposited at low pressure(5 20m T orr), the dense and smooth f ilm will form as show in Figure7 3 (a ) However, the film is discontinuous when the deposition pressure increase to 200mtorr as shown in Figure 7 3 (b).The thickness of the electrolyte film deposited on low pressure is about 600 800nm Figure 7 4 shows the EDS of LVSO thin films deposited on stainless steel deposited at 5 m T orr, Fe, Cr, Ni, V and Si were investigated by EDS, the signals of Fe, Cr, Ni are belong to substr a te, the V : Si ratio is 3:2 that is identical with the theoretical values. The LEAP technique reveals the distribution of Li, V, Si and O element in the LVSO film as shown in Figure 7 5 The differ ent color means the different amount of the element. Except the edge, all element show similar color which represent the uniform element distribution. XPS results of LVSO thin films were shown in Figure7 6 The binding energy of the Li 1 s and V2p 3/2 for LV SO film is 55.1 eV and 516.94 eV that is similar with the standard for Li 1+ and V 5+ separately. The binding energy of S i2p as shown in Figure 7 6 (c) is smaller than the standard for Si 4+ about 1.8eV that suggest oxygen deficient in the LVSO film because of the low oxygen pressure deposition condition (5m T orr). In order to examine the effects of solid electrolyte after anode deposition process, the prepared solid electrolyte was put in vacuum chamber and heated up to sccm the th in film will crack after the process as show in Figure 7 7 (a)(b) which sets the limitation of the subsequent deposition temperature of the anode electrode.

PAGE 142

142 7.4 Summary LVSO solid electrolyte was prepared by PLD method successfully. The low oxygen depositi on condition will form the dense amorphous film that is suitable for manufacture of nano batteries. Different analysis methods were performed to investigated the properties of LVSO film, the XPS show some oxygen vacancy exhibit in the thin film because of the low oxygen deposition pressure, the EDS of the film show the exact stoich i ometry compared the theoretical value. The LEAP analysis shows the excellent uniformity of each element. b

PAGE 143

143 Figure 7 1 XRD pattern of LVSO target Figure 7 2 XRD pa ttern of LVSO thin film deposited at different pressure

PAGE 144

144 (a) (b) Figure 7 3 SEM of LVSO thin films deposited on different pressure (a)5mtorr(b)200mtorr Figure 7 4 EDS of LVSO thin films/stainless steel deposited on 5 mtorr

PAGE 145

145 Figure 7 5 The element distribution of LVSO investigated by LEAP

PAGE 146

146 (a) (b) Figure 7 6 XPS results of LVSO thin films(a)Li(b)V(c)Si

PAGE 147

147 (c) Figure 7 6 Continued (a) (b) Figure 7 7 SEM pictures of LVSO after 650 C

PAGE 148

148 CHAPTER 8 ELECTROCHEMICAL PROP ERTIES OF TIO 2 AND LI 4 TI 5 O 12 THIN FILMS AND ALL SOLID S TATE THIN FILM BATTE RIES PREPARED BY PULSED LASERDEPOSITION 8. 1 Introduction The all solid state thin film battery is a promising micro powder source that can be fabr icated by a typical thin film deposition process [158, 167] Different thin film cathodes, such as LiCoO 2, LiMn 2 O 4, LiNi 0.5 Mn 1.5 O 4 [168 ] V 2 O 5 [169] and various solid elec trolytes including LVSO [170] LIPON [171] and Li 2 SO 4 [172] were fabricated successfully by various techniques. Compared to cathodes and soli d electrolyte materials, it is more difficult to choose a suitable anode as there are differences in thermal expansion coefficients along with layer diffusion during the final layer deposition synthesis process Several materials were selected as anode can didates for fabrication of all solid state thin film battery, such as Li metal, Li 4 Ti 5 O 12 [173] a S i [174] SnO [33] Si Sn [175] Si 0.7 V 0.3 [176] and Ni 3 Sn 2 [177] a Si and alloy anode has relatively higher capacity but the large volume difference between the lithiated and host induce losses of electrical contact between active material and current collector lead to the poor cycling performance. The initi al irreversible capacity of SnO anode also inhibits the availability Li 4 Ti 5 O 12 has been recognized as a zero strain anode with excellent cyclability [23] Titanium oxide is also a promising anode electrode material for Li ion batteries due to its safety against overcharging and stable voltage plateau at 1.78V. TiO 2 and Li 4 Ti 5 O 12 thin film s alread y deposited successfully by various deposition methods, such as PECVD [178] sputtering [179, 180] sol gel [181] atomic layer deposition [182] and pulsed laser deposition [183 185] The opt ical electrical properties, surface characterization and deposition parameter were explored detail in previous research [185, 186] However, there are limited literature s that discuss the

PAGE 149

149 electrochemical properties of Li 4 Ti 5 O 12 and TiO 2 thin films, this chapter uses a PLD method to fabricate TiO 2 and Li 4 Ti 5 O 12 thin films on different substrate and temperature. The phase, structure, morphology were examined by X ray diffraction (XRD), and scanning electron microscopy (SEM). The valence of Titanium and surface environment was investigated by X ray photoelectron spectroscopy(XPS).The relationships between crystal structure, surface morphology and electrochemical properties of thin films were examined. The crystalline TiO 2 thin film can be fabricated at low temperature that make it is a powerful anode candidate for all solid state thin film batteries fabrication. Finally, the all solid state thin film battery with TiO 2 / LiNi 0.5 Mn 1.5 O 4 / L 3.4 V 0.6 S 0.4 O 4 /SiO 2 /Si stack was fabr icated successfully by PLD process for the first time 8.2 Experimental M ethods 8.2.1 Preparation o f Li 4 Ti 5 O 12 M aterials Li 4 Ti 5 O 12 powder was synthesized by a typical s ol gel method Li(CH 3 COO)2H 2 O and tetrabutyl titanate [Ti(OC 4 H 9 ) 4 ] with suitable s toich iometric amounts were dissolved in alcohol Then, the citric acid with half amount of mole ratio compared to titanium was also dissolved in alcohol and drop wise to precursor solution. The solution was dried at 80 C for 6 h to form a gel and calcined at 700 C in air for 6 h and then calcined at 850 C for 24 h to get Li 4 Ti 5 O 12 powder. Li 4 Ti 5 O 12 target was synthesized by a solid state reaction. The starting powders were TiO 2 and LiOH with proper molar ratio The mixture was stirred by 3 D ball mill er for 5 h and then calcined in oxygen at 850C for 20 h r After calcination, the powder was ground again and remade to a pellet .Finally, in order to densify the target the target was calinced in oxygen at 900C for 2 h r

PAGE 150

150 The deposition process was performed by a 248nm KrF excimer laser. The energy of the laser system and frequency were set to 5 00mJ, 5Hz The films were deposited between room temperature and 800 and the oxygen pressure was set to 300mTorr for 40minutes on Au and SiO 2 substrate 8.2.2 Preparation o f TiO 2 M aterials TiO 2 target was also synthesized by a solid state reaction and starting materials anatase (50nm, Fisher) powder was grinded and press ed into a pellet by cold press stainless steel and SiO 2 /Si after the pressure lower than 10 5 Torr. The deposition process was performed by a 248nm KrF excimer laser. The energy of th e laser system and frequency were set to 480 mJ, 5Hz The films were deposited between room temperature and 850 and the oxygen pressure was set to 50 mTorr for 40 minutes 8.2.3 Preparation o f A ll S olid S tate T hin F ilm B attery First, the LiNi 0.5 Mn 1.5 O 4 thi n films were deposited on stainless steels substrate (SS). The deposition process was performed by a 248nm KrF excimer laser. The energy of the laser system and frequency were set to 480 mJ, 5Hz The films were deposited between room temperature and 600 a nd the oxygen pressure was set to 300 mTorr for 40 minutes After the deposition process, the samples were put into chamber with extra 20minutes in oxygen atmosphere. Second, the solid electrolyte LVSO was deposited. Laser fluence and repetition rate were controlled at 300mJ and at 5Hz. Film depositions were carried out at room temperature and in the oxygen partial pressure at 5mTorr for 30min. Finally, the anode TiO 2 and in the oxygen partial pressure at 50mTorr for 40minutes.

PAGE 151

151 8.2.4 Characterization X ray powder diffraction of all samples were taken using XRD Philips APD 3720 using a Cu Kalpha radiation source operated at 45kV and 20m A. The scan speed was 3 sec/step and a step size of 0.02 degrees in two theta. Field emission scanning electron microscopy (FEG SEM, JEOL JSM 6335F) was performed to investigate the thickness and morphologies. Images were collected with an accelerating vol tage of 15kV. X ray photoelectron spectroscopy data were collected using a PHI 3056 spectrometer equipped with a hemispherical detector, 54.7 degrees off normal, with a Mg source in Oak Ridge National laboratory. 8.2.5 Electrochemical T esting The electroc hemical properties of powder and thin films were measured using coin cell 2016 and swagelok consisting of metallic lithium as an anode and an an electrolyte composed of 1M 1M LiPF 6 in ethylene carbonate (EC)/dimethyl carbonate (DMC) with 1:1 volume ratio Celgard model C480 separator (Celgard Inc.USA) was used as separator. The swagelok cells w ere assembled in an argon filled glove box and tested using an Arbin BT2000 instrument in galvanostatic mode. The tests were conducted between 3.5V and 1V (TiO 2 ) and 2.5V and 1V(Li 4 Ti 5 O 12 )at a constant current rate of C/4, C/2, C, 2C, 4C and 8C. For Cycling Performance test, the tests were conducted between 3.5V and 1V at a constant current rate of C/4. 8.3 Results a nd D iscussion 8.3.1 Materials C haracterization a nd E l ectrochemical P roperties O f Li 4 Ti 5 O 12 Figure 8 1 shows the XRD diffraction patterns for the Li 4 Ti 5 O 12 powder synthesized by sol gel method and the Li 4 Ti 5 O 12 target synthesized by solid state reaction. Although small amount anatase TiO 2 phase appears, most peaks can be

PAGE 152

152 identified as the spinel structure with an Fd 3m space group, the average of cubic lattice constants of the samples are 8.381 8.373 separately. Figure 8 2 shows the XRD patterns of the Li 4 Ti 5 O 12 thin film deposited on SiO 2 /Si substrate f rom room temperature to 850 only thin films deposited higher than 700 C show the spinel structure T he pure spinel phase can be formed at relative lower temperature compared to chemical sol gel method. A (111) textured structure is seen for the Li 4 Ti 5 O 12 thin film deposited on 70 0 C However, the anatase TiO 2 phase appears for the sample deposited on 800 C High temperature deposition leads to the evaporation of the Li ion in vacuum and induce s the TiO 2 impurity Figure 8 3 shows SEM photos of Li 4 Ti 5 O 12 thin film deposited on SiO 2 /Si substrate from 500 800 C (a d)and on gold substrate on 700C(e f), The thin films deposited at 500 600 C are c o mposed of agglomerated particles. The films deposited at 700 800 C (b,d) show solid and flat morphology with grain size between 50 and 100nm Figure 8 4(g h) depicts the microstructure of the Li 4 Ti 5 O 12 film deposited on gold meters to a few micrometers. The high magnification image shown in Figure 8 3(f) also indicat es a unique architecture containing smaller grain size <100 nm. The electrochemical properties of the powder and thin film spinel Li 4 Ti 5 O 12 were investigated in Swagelok cells. The charge/discharge profiles between 2.5 and 1V are shown in Figure 8 4(a) and (b). The powder and thin film electrodes have a reversible capacity of ~147 mAh/g and ~163 mAh/g, respectively. In addition, both cells have one distinct voltage plateaus around 1.55V. However, the sloping region of Li 4 Ti 5 O 12 thin film extend, the smal l grain size of the sample [187] or the shape of grain, size distribution

PAGE 153

153 and surface environment [188] lead to the difference of the voltage profile With the decreasing of the particle size, the increa sing of the elastic energy penalty lead the decreasing of the mixing Gibbs free energy at the two phase boundary that extend the sloping region The Li 4 Ti 5 O 12 thin film with pure spinel phase can only be deposited at high deposition temperature which lim its its application as the anode in all solid state batteries The differences in thermal expansion coefficients along with layer diffusion during the final anode deposition synthesis process is the fact that prevents Li 4 Ti 5 O 12 thin film from being used fo r all solid state thin film batter despite its intrinsic excellent electrochemical performance 8.3.2 Materials C haracterization o f TiO 2 M aterials Figure 8 5 shows the XRD pattern for the TiO 2 target. The peaks can be identified as a rutile structure wi th P42/mnm space group. The TiO 2 thin films were fabricated on stainless steels and SiO 2 /Si separately. Only films deposited on stainless steels can be used for electrochemical measurements. The thickness of the t hin films was estimated by the samples depo sited on the SiO 2 /Si substrates and the samples also can be used for evaluat ing the properties when the TiO 2 anode deposited on amorphous solid electrolyte. Crystalline struct ure changes of the TiO 2 thin films deposited between 25 C and 850 C and different substrate were also monitored by XRD as sho wn in Figure 8 6 and Figure 8 7 When the films are deposited at the temperature lower 450 C on SiO 2 /Si substrate, the film is of an amorphous structure. As the substrate temperature increases to 450 C, the texture (101) anatase phase with I41/amd space group is observed. When the substrate temperature increases to 650C, the (004) peak increases rapidly.

PAGE 154

154 At 850C deposition conditi on, the mixed anatase and rutile phases were formed. The average crystal grain size can be calculated by the Scherrer equation. (8 1) Where d is the calculated grain size, is the wavelength of X ray (Cu K 1.54), B is the full width at half maximum intensity, and B is the Bragg diffraction angle. The grain size of anatase phase is determined to be 10.3, 32.9, 57.3nm separately for 450, e. Compared to the TiO 2 thin films deposited on SiO 2 substrate, the thin films deposited on SS show similar phase transformation trend. When the temperature increases, the anatase phase will transf orm to rutile phase. However, the transf ormation occurs a t relatively lower temperature as shown in Figure 8 7 .The pure anatase phase with space group I41/amd and preferred se. I n opposition, the unknown second phase was investigated at 34 degree. It may be the phase of inter diffusion between SS and TiO 2 thin film Figure 8 8 shows the FESEM images of the films deposite d at different substrate temperatures. At room temperature, a kinetic energy to form a dense film. The surface morphologies of films deposited was observed. However, the pores appear between fine grains and t he pores will shrink form and the pore size is about 50nm.From the cross sectional view of the film grown

PAGE 155

155 on the SiO 2 substrate as shown in Figure 8 8 (e), the film thickne ss is estimated to be about 150 nm The X ray photoemission spectrum of the TiO 2 film /SS deposited at different temperature were shown in Figure 8 9(a)(b). The Ti 2p 3/2 and 2p 1/2 are located at 459.78 icate the presence of Ti 4+ oxidation states. The Ti 2p 3/2 and 2p1/2 are located at 458.48 and 464.18eV for thin the low oxygen at high temperature process. Surface OH r atio can be also estimated by the O 1s signal, two components are shown in the typical O 1s XPS spectrum as shown in Figure 5(b). One represents the oxygen element in titania lattice (530.3 eV) and the other corresponding to the surface hydroxyl species (5 32.3 eV). It should be noted that the thin film samples has higher surface hydroxyl concentration (14.4%) than commercial P25(less than 10%) that imply the thin films has a larger surface area. 8.3.3 Charge D ischarge S tudies o f TiO 2 T hin F ilms The el ectrochemical properties of the TiO 2 thin films were investigated using lithium half cells with a constant current 0.02mA/cm 2 In order to measure the electrochemical properties, the stainless steel was used as a substrate and current collector. The first discharge/charge profiles between 3.5 and 1.0V are shown in Figure 8 10(a)(b). The specific capacity of all films were estimated by theoretical density of anatase TiO 2 (3.8cm 3 /g) and the thickness was examined by SEM. The lithium intercalation/deintercalati on reaction in this region can be written as: (8 2 )

PAGE 156

156 The amount of lithium insertion depends on the crystallinity, morphology as well as the microstructure. In our work, the discharge capacity and grain size for thin films largest capa phases of anatase and rutile. It seems the nannosize rutile and anatase phase contribute extra capacity. Wagemaker et al [99] have shown that anatase TiO 2 nanoparticles can completely convert to the Li 1 TiO 2 phase if the particle size is smaller than 7nm. Although the grain size of thin film is larger than 7nm, some LiTiO 2 phase should form during the Li insertion proce ss. Nevertheless, the nanosiezd rutile TiO 2 can also supply the extra capacity [106, 189 192] although only small amount of lithium can insert in to bulk rutile. As Figure 8 10(a) shows, there is an obvious plateau around 1.75V for TiO 2 occurrence of the two phase equilibrium of Lithium poor (Li 0.01 TiO 2 ) and Lithium rich (Li 0.6 TiO 2 ) domains [123] However, there is only very small voltage plateau observed for the TiO 2 be th e contributing factor to the sloping voltage curve [106] The length of the plateau region also changed with different deposition temperature that c orresponds to the ratio With the temperature increasing, the ratio of anatase phase decrease and the plateau shrink. Moreover, the thin film samples shows longer slopin g length that also shows that reducing the crystalline size reduces the miscibility gap of two phase reactions up lithiation [22] With the decreasing of the particle size, the increasing of the elastic

PAGE 157

157 energy penalty lead the decreasing of the mixing Gibbs free energy at the two phase boundary that extend the sloping region Moreover, Li ions inser tion into the surface layer of the nanosized thin film may be an other reason contributing to the sloping region [101] 8.3.4 Rate C apability S tudies o f TiO 2 T hin F ilms Without the addi tion of the carbon black and PVDF, the thin film el ectrode is a n ideal sample to explore the intrinsic properties of TiO 2 material itself. To study the rate capability of TiO 2 electrodes, the cells were discharged to 1V at various current rates (C/4, C/2, C, 2C, C, 4C and 8C). The measured discharge capaci ties with different current density were shown in Figure 8 and 6 The discharge capacity of the TiO 2 thin film tested at C still retains 75.89% of its capacity, co mpared to the capacity at C/4. On the other hand, the discharge capacity of TiO 2 thin film deposited at The true size effects, which involve changes of actual materials and the trivial size effects which depend on the surface to volume ratio and the shape and the pores of the nanomaterials both affect the performance of electrochemical properties [100] For example, anatase and rutile TiO 2 materials show different Li insertion properties and different diffusion coefficient. Wagemaker et al [99, 124] use static 7 Li NMR T2 relaxation measurements to probe the Li ion mobility of micro and nanosize anatase TiO 2 The result shows that the lithium diffusion coefficient for nanomaterials is lower than that for the microsized materials. For r utile TiO 2, it owns one dimensional diffusion channels, the lower Li+ diffusion coefficien t along ab plane limits the transportation of the lithium ions Malik et al [22] prove that for the decreasing of particle

PAGE 158

158 size with the 1D diffusion channel, the diffusion constant will increase The lith ium ion diffusion pathway will be blocked if static defects exist in bulk material. Because there are limit defects for smaller particles, lithium ions can move with high rate without obstruction has relatively larger grain size of The chemical data clearly show that the TiO 2 Besides the structure effects, the trivial size effects should be the key factors that contribute to the excellent rate capability. From the SEM pictures and the grain sizes extracted from XRD pattern, the porous structure of TiO 2 thin film deposited at 350 conductivity. At the same time, nanostructured materials can supply the short Lithium ions and electron transportation distance the Li diffusion time may be much lower for porous anantase thin film. Moreover, the larger surface area of nanostructured thin film electrodes that suggested from XPS analysis also can signific ant diminish the specific current density of the cell The high capacity can maintain even testing at higher rate with t he low specific current density 8.3.5 Cycling P erformance o f TiO 2 T hin F ilms The cycling performance of TiO 2 thin films deposited on d ifferent temperature at a constant current density (C/4) was shown in Figure 8 12. As we can seen, the TiO 2 is virtually no fading for the thin films deposited at This film have poor cycling performance may due to the dense structure and possibly serious inter diffusion

PAGE 159

159 between stainless steel and TiO 2 thin film. TiO 2 thin film with nano grains, nano porous structure show an excellent c ycling performance with high capacity. This makes it a promising anode material for all solid state Li ion batteries. The porous structure can supply the space to accommodate the volume change and avoid ing the during the de lithiation and lithiation proces s. 8.3.6 Characterization o f A ll S olid S tate T hin F ilm B attery The cross section of TiO 2 / LiNi 0.5 Mn 1.5 O 4 / L 3.4 V 0.6 S 0.4 O 4 /SiO 2 /Si stack was fabricated using focus ion beam by Dr. Santhanaqopalan in my group. The cross section of all solid state thin film b attery was examined by SEM as shown in Figure 8 13.The thickness of TiO 2 anode, LVSO solid electrolyte and LiNi 0.5 Mn 1.5 O 4 cathode are 150nm, 900nm and 300nm respectively. The total thickness of the stack is about 1.5um that is suitable for In situ TEM in vestigation. Moreover, there is no crack or holes even after the anode TiO 2 deposition. The TEM images and selected area diffraction pattern are shown in Figure 8 1 4 confirming the random orientation of cathode and anode thin films as observed in XRD resul ts. In spite of the FIB processing and the e beam observation the solid electrolyte is amorphous as shown by the diffraction pattern. The interfaces between the layers are found to be sharp and well defined. 8 4 Summary In this work, Li 4 Ti 5 O 12 and TiO 2 th in films were fabricated by PLD method at different temperature. The relationships between crystal structure, surface morphology and electrochemical properties of thin films were examined. The Li 4 Ti 5 O 12 thin film with pure spinel phase only can be deposit ed at high deposition temperature that limits its application.

PAGE 160

160 The TiO 2 exhibits the largest capacity coupled with excellent cycling performance. The smaller grain size of the rutile and anat ase phases in the thin films most likely enables the formation of the new LiTiO 2 phase during the lith i ation process. However, enhanced rate performance can only be achieved if the thin film is deposited at 350 The porous le ads to a larger electrode/electrolyte contact area and shorter solid state path lengths for both Li ion and electron transport that results in the better rate capability The cycling performance was also significantly improved by the p orous structure of th e thin films deposited at this lower temperature. Finally, the all solid state thin film battery with TiO 2 / LiNi 0.5 Mn 1.5 O 4 / L 3.4 V 0.6 S 0.4 O 4 /SiO 2 /Si stack was fabricated successfully only by PLD process.

PAGE 161

161 Figure 8 1 XRD pattern of Li 4 Ti 5 O 12 powde r synthesized by sol gel method and solid state method Figure 8 2 XRD pattern of Li 4 Ti 5 O 12 thin film deposited on SiO 2 /Si

PAGE 162

162 (a) (b) (c) (d) (e) (f) Figure 8 3 FESEM images of Li 4 Ti 5 O 12 thin film deposited at different temperatures(a)500C(b)600C (c)700C (d)800C (e) 700C on Au (f) 700C on Au

PAGE 163

163 (a) (b) Figure 8 4 Charge/Discharge curves of Li 4 Ti 5 O 12 (a) powder and(b)thin film

PAGE 164

164 Figure 8 5 X ray diffraction patterns of the TiO 2 target. Figure 8 6 X ray diffraction patterns of TiO 2 thin films deposited onto a SiO 2 /Si substrate

PAGE 165

165 Figure 8 7 X ray diffraction patterns of TiO 2 thin films deposited onto a stainless steel substrate

PAGE 166

166 (a) (b) (c) (d) (e) ( f) Figure 8 8 FESEM images of TiO2 thin films deposited at different temperature (a)

PAGE 167

167 (a) (b) Figure 8 9 XPS plots of the Ti2p(a) and O1s(b) region scans for the TiO 2 thin films

PAGE 168

168 (a) (b) Figure 8 10 (a) First discharge curves of TiO 2 thin films (b)First charge curves of TiO 2 thin films

PAGE 169

169 Figure 8 11 Rate capability comparison of TiO 2 thin films deposited at different temperatures.

PAGE 170

170 (a) ( b) (c) (d) (e) (f) Figure 8 12 Cycling Performance of TiO 2 thi n films deposited at different temperatures.

PAGE 171

171 Figure 8 13 Cross sectional SEM image of the all solid state battery prepared by PLD and subsequent FIB processing. Figure 8 14 TEM image of the all solid state battery prepared by PLD and diffrac tion pattern of all the three layers

PAGE 172

172 CHAPTER 9 CONCLUSIONS AND SUGG ESTIONS FOR FUTURE W ORK 9. 1 Conclusion The main goal of this study,as delineated in the introduction, is to develop strategies to improve the electrochemical performance of electro des for Li ion batteries. First, i n order to improve the rate capability of the high voltage spinel material a series of bi doped spinel oxides of LiNi x Cu y Mn 2 x y O 4 (0
PAGE 173

173 significantly A larger contact area between electrode and electrolyte, short diffusion path way for both Li ion and electron move ment and lower specific current density of in the CF TiO 2 lead to a much be tter rate capability Third, the various thin film electrodes and electrolyte were prepared by PLD process, including LiNi 0.5 Mn 1.5 O 4 cathode, LVSO solid electrolyte, LI 4 Ti 5 O 12 and TiO 2 anode s The thin film electrodes offer the ability to probe the surfac e of the material without the need of a conductive agent and polymer binder typically used in composite electrodes .The surface stability of the high voltage Li ion cathode LiMn 2 x Ni x O 4 (x= 0.5, 0.45) were compared by comparing thin film and powder compo site electrodes after cycling using X ray photoelectron spectroscopy. The results from this analysis implies that the surface of the NiMn spinel material is not directly contributing to the SEI formation mechanism; rather the SEI formation seen in the comp osite electrodes is a result of the incorporation of carbon black and PVD F. LVSO solid electrolyte was also prepared by PLD method successfully. The low oxygen deposition condition will form the dense amorphous Although some oxygen vacancy exhibit in the thin film because of the low oxygen deposition pressure, LVSO thin film with exact stoich i ometry and high uniformity can be fabricated Li 4 Ti 5 O 12 and TiO 2 thin films were also fabricated by PLD method at different temperature. The relationships between cry stal structure, surface morphology and electrochemical properties of thin films were examined. The Li 4 Ti 5 O 12 thin film with pure spinel phase only can be deposited at high deposition temperature that limits its application.

PAGE 174

174 The TiO 2 thin film deposited at 3 50 show high capacity coupled with excellent cycling performance. The smaller grain size in the thin films most likely enables the formation of the new LiTiO 2 phase during the lith i ation process. The porous leads to a larg er electrode/electrolyte contact area results in the better rate capability Finally, the all solid state thin film battery with TiO 2 / LiNi 0.5 Mn 1.5 O 4 / L 3.4 V 0.6 S 0.4 O 4 /SiO 2 /Si stack was fabricated successfully only by PLD process 9.2 Suggestions f or Future W ork First, the refinement works of LiNi x Cu y Mn 2 x y O 4 (0
PAGE 175

175 amount of PVDF(20%) must be added to fabricate a well adhered electrode on Al foil. (2)In the electrode making process, the flakes may be partially destroyed due to ball milling The authors suggest a graphene paper flakes comp osite and further improve the electrochemical properties by which the addition of PVDF can be diminished and the flakey shape of TiO 2 can be kept. Moreover, it is also possible to improve the volumetric energy density by packing the sheets of flakes aligne d in a horizontal orientation by this method. Third, scanning probe microscopy (SPM) approach, referred to as electrochemical strain microscopy (ESM) is an excellent tool to explore Li ion transport in thin films The thin films electrodes deposited at dif ferent conditions and different lithiation level were used to explore the Li diffusivity in nano scale. Moreover, the in situ nano scale observation of chemical and structural changes across the interfaces of the all state thin film batteries will also pe rformed by TEM

PAGE 176

176 LIST OF REFERENCES [1] Y. Wang, G. Cao, Advanced Materials, 20 (2008) 2251 2269. [2] M. Winter, R.J. Brodd, Chemical Reviews, 104 (2004) 4245 4270. [3] M.S. Whittingham, Chemical Reviews, 104 (2004) 4271 430 2. [4] J.M. Tarascon, M. Armand, Nature, 414 (2001) 359 367. [5] C. Rosenkranz, EVS 20 Plug in Hybrid Workshop. Nov. 15, Monaco., (2003). [6] K. Mizushima, P.C. Jones, P.J. Wiseman, J.B. Goodenough, Materials Research Bulletin, 15 (1980) 783 789. [7] G.G. Amatucci, J.M. Tarascon, L.C. Klein, Journal of The Electrochemical Society, 143 (1996) 1114 1123. [8] A. Van der Ven, M.K. Aydinol, G. Ceder, G. Kresse, J. Hafner, Physical Review B, 58 (1998) 2975 2987. [9] M.M. Thackeray, W.I.F. D avid, P.G. Bruce, J.B. Goodenough, Materials Research Bulletin, 18 (1983) 461 472. [10] D. Guyomard, J.M. Tarascon, Solid State Ionics, 69 (1994) 222 237. [11] J.M. Tarascon, F. Coowar, G. Amatuci, F.K. Shokoohi, D.G. Guyomard, Journal of P ower Sources, 54 (1995) 103 108. [12] A.D. Pasquier, A. Blyr, P. Courjal, D. Larcher, G. Amatucci, B. Gerand, J.M. Tarascon, Journal of The Electrochemical Society, 146 (1999) 428 436. [13] T. Ohzuku, S. Takeda, M. Iwanaga, Journal of Power Sourc es, 81 82 (1999) 90 94. [14] C. Sigala, D. Guyomard, A. Verbaere, Y. Piffard, M. Tournoux, Solid State Ionics, 81 (1995) 167 170. [15] M.A. Arillo, G. Cuello, M.L. L ez, P. Mart, C. Pico, M.L. Veiga, Solid State Sciences, 7 (2005) 25 32. [16] H. Fang, Z. Wang, B. Zhang, X. Li, G. Li, Electrochemistry Communications, 9 (2007) 1077 1082.

PAGE 177

177 [17] A.K. Padhi, K.S. Nanjundaswamy, J.B. Goodenough, Journal of The Electrochemical Society, 144 (1997) 1188 1194. [18] A. Yama da, S.C. Chung, K. Hinokuma, Journal of The Electrochemical Society, 148 (2001) A224 A229. [19] A.S. Andersson, J.O. Thomas, B. Kalska, L. Haggstrom, Electrochemical and Solid State Letters, 3 (2000) 66 68. [20] S. Yang, P.Y. Zavalij, M. Stanley Whittingham, Electrochemistry Communications, 3 (2001) 505 508. [21] S. Y. Chung, J.T. Bloking, Y. M. Chiang, Nat Mater, 1 (2002) 123 128. [22] R. Malik, D. Burch, M. Bazant, G. Ceder, Nano Letters, 10 4123 4127. [23] T. Ohzuku, A. U eda, N. Yamamoto, Journal of The Electrochemical Society, 142 (1995) 1431 1435. [24] V. Subramanian, A. Karki, K.I. Gnanasekar, F.P. Eddy, B. Rambabu, Journal of Power Sources, 159 (2006) 186 192. [25] A.R. Armstrong, G. Armstrong, J. C anales, P.G. Bruce, Angewandte Chemie International Edition, 43 (2004) 2286 2288. [26] A.R. Armstrong, G. Armstrong, J. Canales, P.G. Bruce, Journal of Power Sources, 146 (2005) 501 506. [27] M. Wagemaker, A.P.M. Kentgens, F.M. Mulder, Na ture, 418 (2002) 397 399. [28] W. Yong gang, Z. Xiao gang, Electrochimica Acta, 49 (2004) 1957 1962. [29] H. Huang, W.K. Zhang, X.P. Gan, C. Wang, L. Zhang, Materials Letters, 61 (2007) 296 299. [30] M. Wagemaker, W.J.H. Borghols, F.M. Mulder, Jo urnal of the American Chemical Society, 129 (2007) 4323 4327. [31] Y.S. Hu, L. Kienle, Y.G. Guo, J. Maier, Advanced Materials, 18 (2006) 1421 1426. [32] S. Bruno, Electrochimica Acta, 45 (2000) 2461 2466.

PAGE 178

178 [33] Y. Idota, T. Kubota, A. Mats ufuji, Y. Maekawa, T. Miyasaka, Science, 276 (1997) 1395 1397. [34] T. Brousse, R. Retoux, U. Herterich, D.M. Schleich, Journal of The Electrochemical Society, 145 (1998) 1 4. [35] J. Yang, M. Winter, J.O. Besenhard, Solid State Ionics, 90 (1996) 281 287. [36] C.K. Chan, H. Peng, G. Liu, K. McIlwrath, X.F. Zhang, R.A. Huggins, Y. Cui, Nat Nano, 3 (2008) 31 35. [37] C.K. Chan, X.F. Zhang, Y. Cui, Nano Letters, 8 (2007) 307 309. [38] H. Xia, H.L. Wang, W. Xiao, M.O. Lai, L. Lu, I nternational Journal of Surface Science and Engineering, 3 (2009) 23 43. [39] M. Duclot, J. L. Souquet, Journal of Power Sources, 97 9 8 (2001) 610 615. [40] J.B. Bates, N.J. Dudney, G.R. Gruzalski, R.A. Zuhr, A. Choudhury, C.F. Luck, J.D. Robertson, Solid State Ionics, 53 5 6, Part 1 (1992) 647 654. [41] J.B. Bates, N.J. Dudney, G.R. Gruzalski, R.A. Zuhr, A. Choudhury, C.F. Luck, J.D. Robertson, Journal of Power Sources, 43 (1993) 103 110. [42] S. Zhao, Z. Fu, Q. Qin, Thin Solid F ilms, 415 (2002) 108 113. [43] A. Khorassani, A.R. West, Journal of Solid State Chemistry, 53 (1984) 369 375. [44] H. Ohtsuka, J. i. Yamaki, Japanese Journal of Applied Physics, 28. [45] S. Zhao, Q. Qin, Journal of Power Sources, 122 (2003) 174 1 80. [46] M.N. Rahaman Ceramic Processing (2007) CRC Press. [47] M. Ebelmen, Ann. Chimie Phys. 16 (1846) 129. [48] C.J.a.S. Brinker, G. W., Sol Gel Science, Academic Press, New York, (1990). [49] C.J. Brinker, A.J. Hurd, P.R. Schunk, G.C. Frye C.S. Ashley, Journal of Non Crystalline Solids, 147??48 (1992) 424 436.

PAGE 179

179 [50] G.W. Scherer, J. Am. Ceram. Soc., 73 (1990). [51] F.a.C. Breech, L Appl. Spect., 16 (1962) 59. [52] H.M.S.a.A.F. Tuner, Appl. Opt., 4 (1965) 147. [53] D. Dijkkamp, T. Venkatesan, X.D. Wu, S.A. Shaheen, N. Jisrawi, Y.H. Min Lee, W.L. McLean, M. Croft, Applied Physics Letters, 51 (1987) 619 621. [54] M.N.R. Ashfold, F. Claeyssens, G.M. Fuge, S.J. Henley, Chemical Society Reviews, 33 (2004) 23 31. [5 5] G.K.H. Douglas B. Chrisey, 1994., Pulsed laser deposition of thin films, J.Wiley, New York: (1994). [56] M.D. Shirk, Journal of laser applications, 10 (1998) 18. [57] N. Bloembergen, Proceedings of LAMP, (1992) 3 8. [58] P.P. Pronko, S.K. Du tta, D. Du, R.K. Singh, Journal of Applied Physics, 78 (1995) 6233 6240. [59] S.M.M.a.V.P. Veiko, Springer, Berlin, Heidelberg, (1994). [60] J.H. Song, Y.H. Jeong, Solid State Communications, 125 (2003) 563 566. [61] W. F. Hu, X. j. Zhao, W. Pen g, T. S. Wang, W. Liu, L. Li, C. Lei, Y. f. Chen, L. Li, Journal of Crystal Growth, 231 (2001) 493 497. [62] S. Metev, K. Meteva, Applied Surface Science, 43 (1989) 402 408. [63] Y.H. Rho, K. Kanamura, Journal of Solid State Chemistry, 177 (2004 ) 2094 2100. [64] C.J. Wen, B.A. Boukamp, R.A. Huggins, W. Weppner, Journal of The Electrochemical Society, 126 (1979) 2258 2266. [65] Y. Shin, A. Manthiram, Electrochemical and Solid State Letters, 5 (2002) A55 A58. [66] H. Fan g, L. Li, G. Li, Journal of Power Sources, 167 (2007) 223 227.

PAGE 180

180 [67] J.C. Arrebola, A. Caballero, M. Cruz, L. Hernn, J. Morales, E.R. Castelln, Advanced Functional Materials, 16 (2006) 1904 1912. [68] J.C. Arrebola, A. Caballero, L. Hernan, J. Morales, Electrochemical and Solid State Letters, 8 (2005) A641 A645. [69] M. Kunduraci, J.F. Al Sharab, G.G. Amatucci, Chemistry of Materials, 18 (2006) 3585 3592. [70] Y. Ein Eli, J.T. Vaughey, M.M. Thackeray, S. Mukerjee, X.Q. Yang, J. McBreen, Journal of The Electrochemical Society, 146 (1999) 908 913. [71] Y. Ein Eli, J.W.F. Howard, S.H. Lu, S. Mukerjee, J. McBreen, J.T. Vaughey, M.M. Thackeray, Journal of The Electrochemical Society, 145 (1998) 1238 1244. [72 ] Y. Ein Eli, J.W.F. Howard, Journal of The Electrochemical Society, 144 (1997) L205 L207. [73] Y. Ein Eli, S.H. Lu, M.A. Rzeznik, S. Mukerjee, X.Q. Yang, J. McBreen, Journal of The Electrochemical Society, 145 (1998) 3383 3386. [74] J. Mo lenda, J. Marzec, K. Swierczek, W. Ojczyk, M. Ziemnicki, M. Molenda, M. Drozdek, R. Dziembaj, Solid State Ionics, 171 (2004) 215 227. [75] A. Van der Ven, G. Ceder, M. Asta, P.D. Tepesch, Physical Review B, 64 (2001) 184307. [76] K. Kang, Y.S. Meng, J. Breger, C.P. Grey, G. Ceder, Science, 311 (2006) 977 980. [77] T.A. Arunkumar, A. Manthiram, Electrochimica Acta, 50 (2005) 5568 5572. [78] D. Li, A. Ito, K. Kobayakawa, H. Noguchi, Y. Sato, Journal of Power Sources, 161 (20 06) 1241 1246. [79] A. Ito, D. Li, Y. Lee, K. Kobayakawa, Y. Sato, Journal of Power Sources, 185 (2008) 1429 1433. [80] M. Aklalouch, J.M. Amarilla, R.M. Rojas, I. Saadoune, J.M. Rojo, Journal of Power Sources, 185 (2008) 501 511. [81] G Kresse, D. Joubert, Phys. Rev. B, 59 (1999) 1758 1775.

PAGE 181

181 [82] G. Kresse, J. Furthmuller, Comput. Mater. Sci., 6 (1996) 15 50. [83] G. Kresse, J. Hafner, Phys. Rev. B, 49 (1994) 14251 14269. [84] G. Kresse, J. Furthmuller, Physical Review B, 54 (1996) 11 169 11186. [85] J.P. Perdew, K. Burke, Y. Wang, Phys. Rev. B, 54 (1996) 16533 16539. [86] M. Cococcioni, S. de Gironcoli, Phys. Rev. B, 71 (2005) 16. [87] F. Zhou, M. Cococcioni, C.A. Marianetti, D. Morgan, G. Ceder, Phys. Rev. B, 70 (2004). [8 8] B. Xu, S. Meng, J. Power Sources, (2010). [89] A.I. Liechtenstein, V.I. Anisimov, J. Zaanen, Phys. Rev. B, 52 (1995) R5467 R5470. [90] B.N. Figgis, M.A. Hitchman, Ligand field theory and its applications, Wiley VCH, New York, 2000. [9 1] J.M. Lloris, B. Len, C. Prez Vicente, J.L. Tirado, M. Womes, J. Olivier Fourcade, J.C. Jumas, Journal of Solid State Electrochemistry, 8 (2004) 521 525. [92] T. Ohzuku, K. Ariyoshi, S. Takeda, Y. Sakai, Electrochimica Acta, 46 (200 1) 2327 2336. [93] G.T. K. Fey, C. Z. Lu, T.P. Kumar, Journal of Power Sources, 115 (2003) 332 345. [94] C.J. Wen, B.A. Boukamp, R.A. Huggins, W. Weppner, Journal of The Electrochemical Society, 126 (1979) 2258 2266. [95] H. Xia, Y.S. Meng, L. Lu, G. Ceder, Journal of The Electrochemical Society, 154 (2007) A737 A743. [96] R.D. Shannon, Acta Crystallogr. Sect. A, 32 (1976) 751 767. [97] J.H. Kim, S.T. Myung, C.S. Yoon, S.G. Kang, Y.K. Sun, Chemistry of Materials, 16 (2004) 906 914.

PAGE 182

182 [98] G. Armstrong, A. Armstrong, P. Bruce, P. Reale, B. Scrosati, Advanced Materials, 18 (2006) 2597 2600. [99] M. Wagemaker, W.J.H. Borghols, F.M. Mulder, Journal of the American Chemical Society, 129 (2007) 4323 4327. [100] C. Jiang, M. Wei, Z. Qi, T. Kudo, I. Honma, H. Zhou, Journal of Power Sources, 166 (2007) 239 243. [101] G. Sudant, E. Baudrin, D. Larcher, J. M. Tarascon, Journal of Materials Chemistry, 15 (2005) 1263 1269. [102] V. Subraman ian, A. Karki, K.I. Gnanasekar, F.P. Eddy, B. Rambabu, Journal of Power Sources, 159 (2006) 186 192. [103] A.R. Armstrong, G. Armstrong, J. Canales, P.G. Bruce, Journal of Power Sources, 146 (2005) 501 506. [104] A.R. Armstrong, G. Armstrong, J. Canales, P.G. Bruce, Angewandte Chemie International Edition, 43 (2004) 2286 2288. [105] A.R. Armstrong, G. Armstrong, J. Canales, R. Garca, P.G. Bruce, Advanced Materials, 17 (2005) 862 865. [106] Y.S. Hu, L. Kienle, Y.G. Guo, J. Maier, Advanced Materials, 18 (2006) 1 421 1426. [107] M.V. Reddy, R. Jose, T.H. Teng, B.V.R. Chowdari, S. Ramakrishna, Electrochimica Acta, 55 3109 3117. [108] S. J. Bao, Q. L. Bao, C. M. Li, D. Zhi Li, Electrochemistry Communications, 9 (2007) 1233 1238. [109] B. L. He, B. Dong, H. L. Li, Electrochemistry Communications, 9 (2007) 425 430. [110] J. Xu, C. Jia, B. Cao, W.F. Zhang, Electrochimica Acta, 52 (2007) 8044 8047. [111] G.F. Ortiz, I. Hanzu, T. Djenizian, P. Lavela, J.L. Tirado, P. Knauth, Chemistry of Materials, 21 (2008) 63 67.

PAGE 183

183 [112] L. Kavan, J. Rathousk, M. Gr zel, V. Shklover, A. Zukal, Microporous and Mesoporous Materials, 44 45 (2001) 653 659. [113] P. Kubiak, J. Geserick, N. H ing, M. Wohlfahrt Me hrens, Journal of Power Sources, 175 (2008) 510 516. [114] M. C. Tsai, J. C. Chang, H. S. Sheu, H. T. Chiu, C. Y. Lee, Chemistry of Materials, 21 (2009) 499 505. [115] S. W. Kim, T.H. Ha n, J. Kim, H. Gwon, H. S. Moon, S. W. Kang, S.O. Kim, K. Kang, ACS Nano, 3 (2009) 1085 1090. [116] P.G. Bruce, B. Scrosati, J. M. Tarascon, Angewandte Chemie International Edition, 47 (2008) 2930 2946. [117] C. Jiang, E. Hosono, H. Z hou, Nano Today, 1 (2006) 28 33. [118] Y. G. Guo, J. S. Hu, L. J. Wan, Advanced Materials, 20 (2008) 2878 2887. [119] A. Matsuda, T. Matoda, T. Kogure, K. Tadanaga, T. Minami, M. Tatsumisago, Chemistry of Materials, 17 (2005) 749 757. [120] T. Sasaki, S Nakano, S. Yamauchi, M. Watanabe, Chemistry of Materials, 9 (1997) 602 608. [121] S. Tedeschi, N. Stevens, D. Cepeda, Y. Y. Lee, K. Powers, M. Ranade, H. El Shall, Powder Technology, 191 (2009) 188 193. [122] T. Sasaki, M. Watanabe, The Journal of Physical Chemistry B, 101 (1997) 10159 10161. [123] M. Wagemaker, A.P.M. Kentgens, F.M. Mulder, Nature, 418 (2002) 397 399. [124] M. Wagemaker, W.J.H. Borghols, E.R.H. van Eck, A.P.M. Kentgens, G.J. Kearley, F .M. Mulder, Chemistry A European Journal, 13 (2007) 2023 2028. [125] M. Wagemaker, R. van de Krol, A.P.M. Kentgens, A.A. van Well, F.M. Mulder, Journal of the American Chemical Society, 123 (2001) 11454 11461.

PAGE 184

184 [126] W. Dr eyer, J. Jamnik, C. Guhlke, R. Huth, J. Moskon, M. Gaberscek, Nat Mater, 9 448 453. [127] J. Vetter, P. Nov k, M.R. Wagner, C. Veit, K.C. M ller, J.O. Besenhard, M. Winter, M. Wohlfahrt Mehrens, C. Vogler, A. Hammouche, Journal of Power Sources, 147 (2005) 269 281. [128] K. Edstrm, T. Gustafsson, J.O. Thomas, Electrochim. Acta, 50 (2004) 397 403. [129] M.R. Palac n, G.G. Amatucci, M. Anne, Y. Chabre, L. Seguin, P. Strobel, J .M. Tarascon, G. Vaugha n, Journal of Power Sources, 81??2 (1999) 627 631. [130] Y. Gao, J.R. Dahn, Solid State Ionics, 84 (1996) 33 40. [131] R.J. Gummow, A. de Kock, M.M. Thackeray, Solid State Ionics, 69 (1994) 59 67. [132] D.H. Jang, S.M. Oh, Journal of The Electrochemical Society, 144 (1997) 3342 3348. [133] M. Broussely, S. Herreyre, P. Biensan, P. Kasztejna, K. Nechev, R.J. Staniewicz, Journal of Power Sources, 97??8 (2001) 13 21. [134] R. Imhof, P. Novak, Journal of The Electr ochemical Society, 146 (1999) 1702 1706. [135] D. Aurbach, I. Weissman, A. Schechter, H. Cohen, Langmuir, 12 (1996) 3991 4007. [136] D. Ostrovsky, F. Ronci, B. Scrosati, P. Jacobsson, J. Power Sources, 94 (2001) 183. [137] A. Schechter, D. Aurbach, H. Cohen, Langmuir, 15 (1999) 3334 3342. [138] L. Dahron, R. Dedryvre, H. Martinez, M. Mntrier, C. Denage, C. Delmas, D. Gonbeau, Chem. Mater., 20 (2007) 583 590. [139] L.M. Bull, B. Bussemer, T. Anupold, A. Reinho ld, A. Samoson, J. Sauer, A.K. Cheetham, R. Dupree, J. Am. Chem. Soc., 122 (2000) 4948.

PAGE 185

185 [140] R. Dedryvre, S. Laruelle, S. Grugeon, P. Poizot, D. Gonbeau, J.M. Tarascon, Chem. Mater., 16 (2004) 1056 1061. [141] R. Dedryvre, H. Mart inez, S. Leroy, D. Lemordant, F. Bonhomme, P. Biensan, D. Gonbeau, J. Power Sources, 174 (2007) 462 468. [142] A.P. Grosvenor, M.C. Biesinger, R.S.C. Smart, N.S. McIntyre, Surf. Sci., 600 (2006) 1771 1779. [143] S. Leroy, H. Martine z, R. Dedryvre, D. Lemordant, D. Gonbeau, Appl. Surf. Sci., 253 (2007) 4895 4905. [144] K.M. Shaju, G.V. Subba Rao, B.V.R. Chowdari, Electrochim. Acta, 48 (2003) 1505 1514. [145] N. Sharma, K.M. Shaju, G.V.S. Rao, B.V.R. Chowdari, J. Power Sources, 139 (2005) 250 260. [146] C. Wu, F. Wu, L. Chen, X. Huang, Solid State Ionics, 152 153 (2002) 335 339. [147] M. C. Yang, B. Xu, J. H. Cheng, C. J. Pan, B. J. Hwang, Y.S. Meng, Chem. Mater., 23 (2011) 2832 2841 [148] N. Yabuuchi, Y. C. Lu, A.N. Mansour, S. Chen, Y. Shao Horn, J. Electrochem. Soc., 158 (2011) A192 A200. [149] K.M. Shaju, G.V. Subba Rao, B.V.R. Chowdari, Solid State Ionics, 152 153 (2002) 69 81. [150] K.M. Shaju, G.V. Subba Rao, B.V.R. Chowdari, Electrochim. Acta, 48 (2002) 145 151. [151] H. Xia, H.L. Wang, W. Xiao, M.O. Lai, L. Lu, International Journal of Surface Science and Engineering, 3 (2009) 23 43. [152] N. Kamaya, K. Homma, Y. Yamakawa, M. Hirayama R. Kanno, M. Yonemura, T. Kamiyama, Y. Kato, S. Hama, K. Kawamoto, A. Mitsui, Nat Mater, 10 682 686. [153] F. Vereda, R.B. Goldner, T.E. Haas, P. Zerigian, Electrochemical and Solid State Letters, 5 (2002) A239 A241.

PAGE 186

186 [154 ] E.J. Jeon, Y.W. Shin, S.C. Nam, W.I. Cho, Y.S. Yoon, Journal of The Electrochemical Society, 148 (2001) A318 A322. [155] S. i. Furusawa, A. Kamiyama, T. Tsurui, Solid State Ionics, 179 (2008) 536 542. [156] S. Zhao, Z. Fu, Q. Qin, T hin Solid Films, 415 (2002) 108 113. [157] J. Kawamura, N. Kuwata, K. Toribami, N. Sata, O. Kamishima, T. Hattori, Solid State Ionics, 175 (2004) 273 276. [158] N. Kuwata, R. Kumar, K. Toribami, T. Suzuki, T. Hattori, J. Kawamura, Sol id State Ionics, 177 (2006) 2827 2832. [159] Y. Tao, D. Yi, J. Li, Solid State Ionics, 179 (2008) 2396 2398. [160] C. Yada, Y. Iriyama, T. Abe, K. Kikuchi, Z. Ogumi, Journal of The Electrochemical Society, 153 (2006) A1148 A1153. [161] S. Zhao Q. Qin, Journal of Power Sources, 122 (2003) 174 180. [162] D. Nancy J, Journal of Power Sources, 89 (2000) 176 179. [163] X. Yu, J.B. Bates, J.G.E. Jellison, F.X. Hart, Journal of The Electrochemical Society, 144 (1997) 524 532. [164] J.B Bates, N.J. Dudney, B. Neudecker, A. Ueda, C.D. Evans, Solid State Ionics, 135 (2000) 33 45. [165] T.F. Kelly, M.K. Miller, Review of Scientific Instruments, 78 (2007) 031101 031120. [166] D.J.L.a.T.F. Kelly, Microscopy and Analysi s 20 (2006) 59. [167] D. N.J, Materials Science and Engineering: B, 116 (2005) 245 249. [168] H. Xia, S.B. Tang, L. Lu, Y.S. Meng, G. Ceder, Electrochimica Acta, 52 (2007) 2822 2828. [169] J.M. McGraw, C.S. Bahn, P.A. Parilla, J.D. Perkins, D.W Readey, D.S. Ginley, Electrochimica Acta, 45 (1999) 187 196.

PAGE 187

187 [170] N. Kuwata, J. Kawamura, K. Toribami, T. Hattori, N. Sata, Electrochemistry Communications, 6 (2004) 417 421. [171] Y. Xiaohua, J.B. Bates, G. E. Jellison, Jr., F. X. Hart, Journal of The Electrochemical Society, 144 (1997) 524 532. [172] N. Kuwata, N. Iwagami, J. Kawamura, Solid State Ionics, 180 (2009) 644 648. [173] Y.H. Rho, K. Kanamura, M. Fujisaki, J. i. Hamagami, S. i. Suda, T. Umegaki, Solid State Ionics, 151 (2002) 151 157. [174] H. Xia, S. Tang, L. Lu, Materials Research Bulletin, 42 (2007) 1301 1309. [175] L.Y. Beaulieu, K.C. Hewitt, R.L. Turner, A. Bonakdarpour, A.A. Abdo, L. Christensen, K.W. Eberman, L.J. Kra use, J.R. Dahn, Journal of The Electrochemical Society, 150 (2003) A149 A156. [176] S. J. Lee, H. K. Baik, S. M. Lee, Electrochemistry Communications, 5 (2003) 32 35. [177] Y. L. Kim, H. Y. Lee, S. W. Jang, S. J. Lee, H. K. Baik, Y. S Yoon, Y. S. Park, S. M. Lee, Solid State Ionics, 160 (2003) 235 240. [178] W.G. Lee, S.I. Woo, J.C. Kim, S.H. Choi, K.H. Oh, Thin Solid Films, 237 (1994) 105 111. [1 79] P. L l, M. Huppertz, D. Mergel, Thin Solid Films, 251 (1994) 7 2 79. [180] S. Tanemura, L. Miao, P. Jin, K. Kaneko, A. Terai, N. Nabatova Gabain, Applied Surface Science, 212 213 (2003) 654 660. [181] C. Garzella, E. Comini, E. Tempesti, C. Frigeri, G. Sberveglieri, Sensors and Actuators B: Chemi cal, 68 (2000) 189 196. [182] V. Pore, A. Rahtu, M. Leskel, M. Ritala, T. Sajavaara, J. Keinonen, Chemical Vapor Deposition, 10 (2004) 143 148. [183] J. H. Kim, S. Lee, H. S. Im, Applied Surface Science, 151 (1999) 6 16. [184] N. Inoue, H. Yu asa, M. Okoshi, Applied Surface Science, 197 198 (2002) 393 397.

PAGE 188

188 [185] F. Cao, G. Oskam, P.C. Searson, J.M. Stipkala, T.A. Heimer, F. Farzad, G.J. Meyer, The Journal of Physical Chemistry, 99 (1995) 11974 11980. [186] J. Domaradzki, D. Kaczmarek, E.L. Prociow, A. Borkowska, D. Schmeisser, G. Beuckert, Thin Solid Films, 513 (2006) 269 274. [187] G. Kobayashi, S. i. Nishimura, M. S. Park, R. Kanno, M. Yashima, T. Ida, A. Yamada, Advanced Functional Materials, 19 (2009) 395 403. [188] A. Van der Ven, M. Wagemaker, Electrochemistry Communications, 11 (2009) 881 884. [189] P. Kubiak, M. Pfanzelt, J. Geserick, U. H rmann, N. H sing, U. Kaiser, M. Wohlfahrt Mehrens, Journal of Power Sources, 194 (2009) 1099 1104. [190] E. Baudrin, S. Cassaignon, M. Koelsch, J.P. Jolivet, L. Dupont, J.M. Tarascon, Electrochemistry Communications, 9 (2007) 337 342. [191] W.J.H. Borghols, M. Wagemaker, U. Lafont, E.M. Kelder, F.M. Mulder, Chemis try of Materials, 20 (2008) 2949 2955. [192] D. Dambournet, I. Belharouak, K. Amine, Chemistry of Materials, 22 (2009) 1173 1179. [193] M.V. Koudriachova, N.M. Harrison, S.W. de Leeuw, Solid State Ionics, 157 (2003) 35 38.

PAGE 189

189 BIOGRAPHICAL SKETCH Ming Che Yang was born in Taipei of northern Taiwan. He did his undergraduate studies at the Applied Chemistry Department at National Chiao Tung University, and got his master degree in m aterials s cience and e ngineering at Na tional Tsing Hua University Hsinchu, Taiwan. After a three month military training, he worked for four and half years as an assistant researcher in National Nano Device Labor a tory, Hsinchu, Taiwan. He focus on the research about High K, Low K ferroelectri c materials and develop a novel method to fabricate Si Nanograss by simple hydrogen etching method, this structure was used as a field emission source, a antireflection layer, a superhydrophobic surface and a Mold for polymer nanotip. He also got the paten t in Taiwan and Korea for this technique. He published eight papers in peer reviewed journals and presented his results in five conference s when he worked in Taiwan. H e continued his education in pursuit of a Doctor of Philosophy degree in the Departmen t of Ma terials Science and Engineering at University of Florida, Gainesville. He worked with Dr. Shirley Meng in UF since 2008 on developing and optimizing novel electrode s of Li Ion Batteries Now his research interests ar e high voltage spinel materials nanosized anode materials and all solid solid state thin film Lithium ion batteries. By using the all solid state thin film batteries, the intrinsic properties of materials can be explore d especially for the interface between the electrode and electrolyte He published four papers in peer reviewed journals and gave 6 presentations in conference when he stay ed in Dr.