Effects of Annealing on the Performance of Silicon Thin Film Anodes

MISSING IMAGE

Material Information

Title:
Effects of Annealing on the Performance of Silicon Thin Film Anodes
Physical Description:
1 online resource (52 p.)
Language:
english
Creator:
Mohan, Rohit
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Materials Science and Engineering
Committee Chair:
JONES,KEVIN S
Committee Co-Chair:
SIGMUND,WOLFGANG MICHAEL
Committee Members:
DEMPERE,LUISA AMELIA

Subjects

Subjects / Keywords:
annealing -- anode -- native-oxide -- silicon -- stainless-steel -- thin-films
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre:
Materials Science and Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
Lithium ion batteries are used to power portable electronic devices and hybrid electric vehicles, and hold great potential for future energy storage. The current technology uses a graphite anode, the capacity of which is 372 mAh/g. Silicon has a theoretical capacity of 4200 mAh/g, which is why it has been a topic of interest for many years. However, the insertion of lithium in silicon during discharge results in a nearly 400 percent volume expansion, resulting in pulverization of the silicon, and loss in the gravimetric capacity after a few charge-discharge cycles. The use of nanostructures and thin films have mitigated this problem to a certain extent, although a lot more still remains to be done. In this work thin films of silicon deposited on stainless steel have been annealed from 475 to 600 degrees C, intervals of 25 degrees C for 90 minutes each. Although expected, no silicide formation was observed. However, annealing did improve the gravimetric capacity and capacity retention in pouch cells assembled using these anodes and Lithium metal as the cathode. The best capacity and retention was seen in case of the specimen annealed at 525 degrees C. The increase in capacity may be related to a reaction between the native oxide at the interface and the deposited silicon resulting in improved film adhesion.
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 Rohit Mohan.
Thesis:
Thesis (M.S.)--University of Florida, 2014.
Local:
Adviser: JONES,KEVIN S.
Local:
Co-adviser: SIGMUND,WOLFGANG MICHAEL.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2014
System ID:
UFE0046757:00001


This item is only available as the following downloads:


Full Text

PAGE 1

EFFECTS OF ANNEALING ON THE PERFORMANCE OF SILICON THIN FILM ANODES By ROHIT MOHAN A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2014

PAGE 2

2014 Rohit Mohan

PAGE 3

To Mom, Dad. For everything.

PAGE 4

4 ACKNOWLEDGMENTS At the outset, I would like to thank Ethan Kennon, Caleb Barrett, Joe Ault, Tom Martin, Mithun Nair and Ryan Murray for making my time in the SWAMP lab memorable Nick Vito has my most sincere thanks for being the re as a mentor and helping me lea rn the fundamentals of batteries Aaron Lind and Henry Aldridge have my deepest gratitude for h elping me out with the technicalities of various lab equipment and also for all the good times in the lab I also thank Dr. Shadi Al khateeb for many inter esting discussions on battery technology. I would like to acknowledge the help from Dr. Nancy Ruzycki for access to the optical microscope I also take this opportunity to thank my parents for all their support, putting up with me for all these years, and standing by my every decision. Last, but not the least, I thank my advisor Dr. Kevin Jones for giving me this opportunity to explore the nuances of batteries, and hence complete this work.

PAGE 5

5 TABLE OF CONTENTS Page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ............................. 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 Lithium Ion Batteries ................................ ................................ ............................... 13 Anode Materials for Lithium Ion Batteries ................................ ............................... 15 Objectives and Overview ................................ ................................ ........................ 17 2 EXPERIMENTAL METHODS ................................ ................................ ................. 19 Plasma Enhanced Chemical Vapor Deposition (PECVD) ................................ ....... 19 Electrode Mass Calculation ................................ ................................ .................... 20 Annealing ................................ ................................ ................................ ................ 20 Pouch Cell Assembly ................................ ................................ .............................. 21 Electrochemical Characterization ................................ ................................ ........... 23 Galvanostatic Cycling ................................ ................................ ....................... 23 Cyclic Voltammetry ................................ ................................ ........................... 24 Material Characterization ................................ ................................ ........................ 24 Optical Microscopy ................................ ................................ ........................... 25 Focused Ion Beam (FIB) ................................ ................................ .................. 25 Cross sectional Transmission Electron Microscopy (XTEM) ............................ 27 Energy dispersive X ray Spectroscopy (EDS or EDX) ................................ ..... 28 3 ANNEALING AND ITS EFFECTS ON THE PERFORMANCE OF THE CELL ....... 33 Overview of Experiments ................................ ................................ ........................ 33 Annealing ................................ ................................ ................................ ......... 33 Electrochemical Characterization ................................ ................................ ..... 33 Results ................................ ................................ ................................ .................... 33 Electrochemical Characterization ................................ ................................ ..... 33 Galvanostatic Cycling ................................ ................................ ................ 34 Cyclic Voltammetry ................................ ................................ .................... 35 Material Characterization ................................ ................................ .................. 35 Discussion ................................ ................................ ................................ .............. 35

PAGE 6

6 Concluding Remarks ................................ ................................ ............................... 39 4 CONCLUSIONS ................................ ................................ ................................ ..... 47 Summary ................................ ................................ ................................ ................ 47 Future Work ................................ ................................ ................................ ............ 47 LIST OF REFERENCES ................................ ................................ ............................... 50 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 52

PAGE 7

7 LIST OF TABLES Table page 3 1 Gravimetric discharge capacities for all cells for select cycles. ........................... 46

PAGE 8

8 LIST OF FIGURES Figure page 2 1 Schematic of a PECVD system. ................................ ................................ ......... 29 2 2 Microsoft Excel plot used for electrode mass calculation ................................ ... 29 2 3 Pouch cells ................................ ................................ ................................ ........ 30 2 4 Schematic representation of the basic optical microscope. ................................ 31 2 5 Schematic of th e FIB ................................ ................................ ......................... 32 3 1 Results of galvanostatic cycling of the as deposited electrode. .......................... 40 3 2 Result of annealing at 475C for, compared to the as deposited electrode. ....... 41 3 3 Comparison of cycling performance of the as deposited and 475, 525 and 600C annealed electrodes ................................ ................................ ................ 41 3 4 Comparison of cycling performance of as deposited and all annealed electrodes ................................ ................................ ................................ ........... 42 3 5 Galvanostatic cycling test results for as deposited, optimum anneal and ion implanted specimens ................................ ................................ .......................... 42 3 6 Trend in gravimetric capacity of the 5 0 th cycle with annealing temperature. ....... 43 3 7 Cyclic voltammograms of the 1 st 2 nd and 20 th cycle s ................................ ........ 43 3 8 XTEM image of the as deposited specimen. ................................ ...................... 44 3 9 XTEM image of the specimen annealed at 525C for 90 minutes. ..................... 44 3 10 ED S scans ................................ ................................ ................................ .......... 45 3 11 Micrograph of the cycled electrode for the specimen ann e aled at 600C. ......... 45

PAGE 9

9 LIST OF ABBREVIATIONS CV Cyclic Voltammetry DMC Dimethyl Carbonate EC Ethylene Carbonate EDS Energy dispersive X ray Spectroscopy EV Electric Vehicle FIB Focused Ion Beam GHG Greenhouse Gas HEV Hybrid Electric Vehicle PECVD Plasma Enhanced Chemical Vapor Deposition PHEV Plug in Hybrid Electric Vehicle PP Polypropylene PVA Polyvinyl Alcohol RTA Rapid Thermal Annealing SAD Selected Area Diffraction SEI Solid Electrolyte Interphase SEM Scanning Electron Microscopy SHE Standard Hydrogen Electrode SS Stainless Steel STEM Scanning Transmission Electron Microscopy TEM Transmission Electron Microscopy XTEM Cross sectional Transmission Electron Microscopy

PAGE 10

10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science EFFECTS OF ANNEALING ON THE PERFORMA N CE OF SILICON THIN FILM ANODES By Rohit Mohan May 2014 Chair: Kevin S. Jones Major: Materials Science and Engineering Lithium ion batteries are used to power portable electronic devices and hybrid electric vehicles and hold great potential for future energy storage The current technology uses a graphi te anode, the capacity of which is 372 mAhg 1 Silicon has a theoretical capacity of 4200 mAhg 1 which is why it has been a topic of interest for many years However, the insertion of lithium in silicon during discharge results in a nearly 400% volume exp ansion resulting in pulverization of the silicon, and loss in the gravimetric capacity after a few charge discharge cycles. The use nanostructures and thin films have mitigated this problem to a certain extent, although a lot more still remains to be done. In this work thin films of silicon deposited on stainless steel hav e been annealed from 4 75 to 600 C, in intervals of 25 C, for 90 minutes each. Although expected, no silicide formation was observed However, ann ealing did improve the gravimetric capacity and the capacity retention in pouch cells assembled using these anodes and Li thium metal as the cathode The best capacity and retention was seen in case of the specimen annealed at 525 C. The increase in capacit y may be related to a reaction

PAGE 11

11 between the native oxide at the interface and the deposited silicon resulting in improved film adhesion.

PAGE 12

12 CHAPTER 1 INTRODUCTION Mankind has been heavily dependent on fossil fuels for more than two and a half centuries for its progress and development. The ease of availability, low cost, stable nature and high energy densities have allowed extensive use of fossil fuels over the years. It is common knowledge that these fuels have formed on the earth over eons and that their rese rves are finite. The continued use of such fuels has led to an increase in the propor tion of greenhouse gases (GHGs), especially carbon dioxide (CO 2 ) in the sea levels rising, it is important to develop and use sustainable, alternative means of producing energy. These global changes are bound to relocate populations, shuffle production of food and industry and may spa rk conflicts for food and water [1] The per capita consumption of electrical energy is often seen as a marker for economic development. As developing economies of the world develop, their consumption of electricity increases. The course of natural evolution has man improving the quality of h is life prompting an increase in electricity consumption [2] Such increase in consumption coupled with the depleting reser ves of fossil fuels drives the need to look for other alternative means to produce electrical energy. A vast quantity of fossil fuels are burnt in in internal combustion engines in vehicles, trains and aircraft. To reduce the consumption of fuels, it beco mes necessary to realize and put into practice the use of hybrid electric vehicles (HEVs), plug in hybr id electric vehicle (PHEVs) and fully electric vehicles (EVs). The development of lithium ion batteries holds a lot of promise for such vehicles Automob ile giants like Toyota and General Motors are engaged in intense research to improve their HEVs, the Prius and

PAGE 13

13 the Chev rolet Volt, respectively [3] Though electricity is a relatively clean form of energy, a vast majority of electricity is produced by the combu stion of fossil fuels or nuclear fuels in power plants. Solar and wind energy are now harnessed to produce electricity. Thus producing electric energy from clean sources and storing them efficiently is essential [1] The success of such vehicles largely depends on having an efficient means of storing electric energy. Batteries are used to supply energy to portable electronic devices like laptop computers, cellular phones, cameras, and also in TV remote controls, calculators, and other hand held devic es, radio sets, toothbrushes, pacemakers, central locking systems in cars etc. [4] The development of battery technology has made it poss ible to employ them in electric vehicles It is evident that batteries will play a very crucial role in storing electrical energy in the future. The inception and development of various battery types, and chemistries is outlined by Dell (2000), including zinc manganese dioxide primary and secondary cells lead acid batteries etc. [4] Lithium Ion B atteries Lithium (Li), with atomic number 3 is the most electropositive element in the periodic table. Being the lightest metal, it is a good choice for high energy density storage systems. The Li/Li + redox system ha s an electr ical potential of 3.040 V with respect to the standard hydrogen electrode (SHE) [5] The fir st lithium metal batteries wer e used to power watches, calculators or implantable medical devices, mainly due to their variable discharge rate and high energy density. Exxon developed a lithium intercalation battery using TiS 2 due to its favorable, layered structure. But these batteri es had an explosion hazard due to the uneven formation of lithium dendrites when used as an electrode in an electrochemical cell with each subsequent charge

PAGE 14

14 discharge cycle [6] The problem of dendrite formation was tackled by using lithium aluminum (Li Al) alloy electrod es [7] But these elec trodes had poor cycle life, and lasted only for a few cycles due to very high volume changes during the charge discharge cycles [6] The switch to the sulfide did improve safety of the cells, but reduced the voltage derivable from the cell. A breakthrough in obtaining a high voltage with Li metal as one of the electrodes was seen with Goodenough et al. introducing LiCoO 2 as the electrode material in 1981 The deeper Fermi level of oxides makes them the material of choice as compared to sulfides for such electrodes. The layere d structure of LiCoO 2 was found to be good for Li + intercalation and deintercalation [8] Sony Corporation introduced the first commercial rechargeable (or rocking chair) lithium ion cell, consisting of LiCoO 2 and graphite (C) in 1991 The use of a lithium intercalation compound serves to do away with the explosion hazards associated with the use of lithium metal [4, 6] The g eneral reaction chemistry for a lithium ion intercalation electrode is given as follows: Full cell reaction: LiCoO 2 + C 6 CoO 2 +LiC 6 (1 1) Anode: C 6 + x Li + + x e Li x C 6 (1 2) Cathode: Li (1 x) CoO 2 CoO 2 + xLi + + xe (1 3) These reactions are reversible. During discharge, oxidation occurs at the anode and reduction, at the cathode. The oxidation results in the liberation of electrons that move through the external circuit, and lithium ions move through t he electrolyte and intercalate the cobalt oxide to f orm LiCoO 2 [5] During charging, the reverse reaction takes place and the lithium intercalates into the carbon anode. Now, the reduction

PAGE 15

15 happens at the anode and oxidation occurs at the cathode. Thus, these cells are known [5] For a cell to function effectively, t here should be no electrical or physical contact between the anode and the cathode. However, there should be an effective medium that allows ionic conduction betwee n the two electrodes. Thus ionic conduction and electronic isolation is achieved by using an electrolyte, which may be inorganic or organic. Also, to prevent shorting between the anode and the cathode, a separator is used which is usually a polymer that a llows the flow of ions [9] Anode Materials for Lithium Ion Batteries Carbon based anodes are still widely used in commercial lithium ion battery systems The theoretical capacity of batteries with these carbon ba sed anodes is 372 mAhg 1 The LiC 6 phase corresponds to this theoretical capacity. Thus, 6 carbon atoms are required to intercalate one lithium atom. The need for battery systems with greater capacity has driven research towards using materials like silico n, germanium, tin or alloys of the se elements as anode materials. Of these, silicon has the highest theoretical capacity of 4200 mAhg 1 for the Li 4.4 Si phase, nearly an order of magnitude higher than those exhibited by carbon (graphite) based anode systems This corresponds to the Li 22 Si 5 phase which effectively packs in 4.4 lithium atoms per atom of silicon No system is yet to match up to this theoretical value, though systems with capacities exceeding that of the carbon based anode have been realized, a t l east at the laboratory scale [10] One of the major issues with silicon is the ~ 300 to 400 % volume expansion that occurs during cycling due to the intercalation of Li in silicon [11] Charging and discharging the cell is accompanied by delithiation and lithiation. This means the silicon expands to roughly

PAGE 16

16 400% its initial volume and contracts every time the cell discharges and charges. This causes the film to crack, and ultimately delaminate from the substrate. Pal et al. have developed a model to predict cracking and delamination of sili con thin film anodes deposited on a copper substrate [12] Various other geometries in silicon such as nanowires, nanopillars, etc have been employed at an attempt to improve the capacity retention of cells with these silicon anodes by preventing them from delaminating from the substrate material Silicon nanowires have shown promising results even at higher cycling rates of 1 C [13] Nano pillars of silicon, grown via reactive ion etching on silicon substrates have also been employed by Green et al., which also shows promising results with good capacity up to 50 cycles [14] Ion beam mixing was also employed which showed that a dose of 1x 10 16 cm 2 shows very good cycling stability and capacity retention, even up to 100 cycles. Many tin alloys with various morphologies are being studied for use as an anode in the lithium ion battery. Silicon tin alloy electrodes of varying compositions deposited by magnetron sputtering were studied by Hatchard and Dahn, and these show reversible capacities of up to 3500 mAhg 1 for maximum Si content and just under 2 000 mAhg 1 for the maximum Sn content alloy. T he alloys were of the type Si 1 x Sn x where 0 < x < 0.45 [15] By optimization of the bulk oxygen content, controlled surface oxidation and low temperature (200 C) annealing, Abe l et al. were successful in developing a nanostructured thin silicon film anode that had a capacity of ~2200 mAhg 1 with improved cycling stability upon coupling with lithium metal as the counter electrode [16] Germanium (Ge) is another material that has sparked interest in the area of electrode materials. It has a the oretical capacity of 1600 mAhg 1 for the Li 4.4 Ge phase.

PAGE 17

17 Ion beam mixed germanium electrodes have met with success in retaining near theoretical capacity of ~1500 mAhg 1 as demonstrated by Rudawski et al [11] In this work, a morphous silicon was deposited on stainless steel shim and annealed at various temper atures, ranging from 475 to 600 C to determine their us e as potential anodes for lithium in batteries These anodes were then subject to electrochemical and material characterization tests to determine the effects of annealing and changing the annealing temperature on the electrochemical performance of the pou ch cells into which these anodes were incorporated. Objectives and Overview While bulk systems are used primarily in cars and other heavy duty applications, thin films find uses in niche applications like medical devices like pacemakers and hearing aids, where a small size is of essence. Thin film technology could find use in the smartphone technology as well, given the trend to have thinner devices with longer battery life. The formation of a silicide is expected to improve the capacity as compared to the as received material by annealing While different methods like ion beam mixing and nano structures like nanowires and nanopillars have be en used to bring about better capacity retention in silicon anodes have been studied extensively, annealing the as re ceived films of silicon is relatively unexplored. Chen et al have studied the annealing of silicon films deposited on a copper substrate. T here is no focus on the details of what goes on at the interface, or the mention of any silicide being formed at all A capacity of 3134 mAhg 1 was achieved, but cycling was performed at a C/40 rate [17] The main objectives of this s tudy are to determine the effect of annealing and to explore the effect

PAGE 18

18 of changing annealing temperatures on the performance of the anode in these thin film batteries

PAGE 19

19 CHAPTER 2 EXPERIMENTAL METHODS This chapter discusses the various processes, equipment and experiments used during the course of this work. It outlines PECVD, electrode mass calculation, annealing, pouch cell assembly, electrochemical and material characterization procedures used Plasma Enhanced Chemical Vapor Deposition (PECVD) PECVD is a widely us ed technique to deposit films on various substrates at relatively lower temperatures. Low temperatures allow the use of a wider range of substrates PECVD also has the advantage of better coverage. It is similar to CVD in most ways, except that the dissociation, ionization and excitation reactions are brought about by the plasma instead of high temperatures Plasma is a partially ionized gas. PECVD is done in a vacuum chambe r, at low pressures of about 1 T orr. The vacuum chamber contains tw o parallel plates connected to a voltage DC power supply. A high voltage source, such as a charge capacitor is used to produce an arc momentarily to start the plasma. The plasma glows on account of various excitation and relaxation processes that take plac e in the mix. The arc ionizes the gas, and accelerates the positive ions to the cathode and the negative ions to the anode. The ions being heavy, release a cloud of electrons from the cathode. The secondary electrons released can sustain the plasma if a su fficiently high voltage is applied. A schematic of a PE CV D system is shown in figure 2 1 The applied voltage, the gas flow rate, separation between the electrodes, etc. affect film deposition rate The stainless ste ) w as washed with water, ethanol, acetone and isopropyl alcohol in order to remove any surface impurities. An RF

PAGE 20

20 frequency of 13.56 MHz at 30W was used to deposit the film with the substrate maintained at 300 C. the chamber pressure was maintained at 750 mTor r. Silane was supplied for deposition at approximately 46 standard cubic centimeters per minute (sccm). The carrier gases were argon and nitrogen, supplied at ~450 and ~2250 sccm respectively. The 350 nm amorphous silicon films were deposited in 37 minutes Electrode Mass Calculation Three hundred and fifty nanometers of amorphous s ilicon was deposited on the stainless shim substrate by PECVD at 300 C the thickness of which was measured using XTEM. Specimens of different sizes and shapes were cut to measur e areal mass density. Each of the samples was weighed and their area measured using Image J to analyze optical micrographs of the cut specimens. Based on the density of silicon for such an amorphous film from literature, the weight of silicon per unit area of the SS substrate was determined. This data was plotted in Microsoft Excel as shown in figure 2 2 and used to obtain the mass of the silicon, and hence determine the amount of current required or galvanostatic tests for a given mass of silicon film on stainless steel. Annealing Annealing involves heating a sample and then allowing it to cool. In this case, the heating is expected to bring about the formation of a layer of iron silicide between the stainless steel substrate and the deposited silicon thi n film. Annealing Procedure and Equipment Pieces of the anode were cut from the sheet of silicon dep osited on 25 micron thick stainless steel shim (TBI, USA) and weighed. The furnace (Lindberg) was set to the required temperature using a Lindberg controller and allowed to achieve a steady state at the set temperature. The flow of argon was introduced at a rate of 7.5 Nl/min

PAGE 21

21 The specimen was then placed on a quartz boat and introduced into the tube furnace using a glass rod. Each s ample was annealed for 90 minutes. After removing from the furnace the samples were allowed to cool for at least 1 hour before placing them back into their holders. The annealed electrodes were used to prepare pouch cells as described in the next section after a minimum of 12 hours following removal from the furnace The correctness of the set temperature was checked using a Cole Parmer Digi Sense scanning thermometer No difference in electrode mass was noticed before and after annealing. Annealing temp eratures ranged from 475 C to 600 C, at intervals of 25 C Pouch Cell Assembly The pouch cell is made from aluminum laminate sheets, polypropylene polyvinyl alcohol strips, stainless steel contacts, anode, cathode, electrolyte, and a separator. The aluminu m laminate sheets form the pouch and also double up as an environmental of the PP PVA strips, placed on each edge of the laminate piece. Another laminate piece of the same dimensions has one edge with the PP PVA strip. The strips are held in place by heating the strips using a soldering iron at intervals of about one quarter inch. Two strips of the same stainless shim are placed on this laminate, and held in place on t he laminate pieces by heating the strips at points at us ing a soldering iron. Figure 2 3 A shows a half assembled pouch cell The SS strips are placed so as to allow electrodes to be placed one over the other. The pouch is then sealed using an impulse seal er at this edge This is the first heat seal. A small piece of the separator polypropylene Celgard model C480 (Celg ard USA, Inc.) is placed between the electrodes to prevent any direct physical contact between the

PAGE 22

22 electrodes The separator is held in place by a small piece of the Kapton HN polyimide tape. Next, the anode, (silicon on SS) is placed under one of the contacts, with the silicon facing down on the separator. The anode, weighed and in this case, annealed, is fixed in p lace using a piece of the Kapton tape. The rest of the assemb ly is done in an inert, argon atmosphere in the glove box An MBraun Unil ab glove box was used to complete the remaining assembly of the pouch cells. An inert atmosphere was maintained in the gl ove box by using argon gas. It was ensured that the water vapor level in the glove box was less than 1 ppm. This is because both lithium and the electrolyte are sensitive to the presence of water and oxygen. The half assembled pouch cells were placed in th e antechamber, and the antechamber of the glove box was evacuated, followed by filling with argon. This process was repeated 3 to 4 times. Lithium metal pieces, marginally larger than the anode were cut from a strip of lithium metal (Aldrich) The passivat ion layer was scraped off, and the piece was then fixed to the other contact and secured in place using a piece of the Kapton tape. Next, the electrolyte, a solution of 1M LiPF 6 in 1:1 DM C: EC solvent was introduced into the cell, after two more sides wer e heat sealed multiple times to prevent any leaks. After introducing sufficient electrolyte into the pouch making sure that the electrodes were flooded by the electrolyte the last side was sealed, and any excess solvent on the sides of the p ouch were wip ed using KimWipes. The finished cells were then taken out of the glove box for electrochemical characterization. A complete p o uch cell is shown in figure 2 3 B Certain pouch cells were deprocessed for analysis using optical microscopy The deprocessing w as done in the argon filled glove box. To deprocess the cells, a small

PAGE 23

23 potion was cut, and the electrolyte was drained into the liquid waste s container Thereafter, the sides that do not contain the electrical contacts were cut. The cell was propped open a nd the anode was retrieved from under the contact and the Kapton tape that was used to hold it in place. The electrode is washed with a few drops of DMC to dissolve any remaining electrolyte and allowed to dry. Any excess DMC was charged into the liquid w astes container. The electrode is held over a filter paper covering the wastes container, so that the electrode does not fall into it. The electrode is then stored in a sample holder. Next, the piece of lithium metal is retrieved and discarded in the vial used exclusively for disposal of used lithium The remainder of the cell is put into the interim tra sh receptacle in the glove box. Electrochemical Characterization Galvanostatic cycling and cyclic voltammetry tests were performed on an Arbin BT2000 batter y tester. The testing equipment has 16 channels in the main unit and 8 in an auxiliary unit, thus enabling the testing of 24 batteries at a time, with the provision to test both pouch and coin cells. The testing equipment was controlled using the MITSPro software, provided by Arbin. These tests were used in evaluating the electrochemical performance of the electrodes. Data from these tests wer e obtained and analyzed using Microsoft Excel. Before connection to the leads of the equipment, the cells were pack ed between two glass microscope slides, and clamped together with binder clips to ensure good packing and to prevent any movement of the electrodes within the cell. Galvanostatic Cycling Galvanostatic testing involves maintaining a constant current across the electrochemical cell. The cell is allowed to discharge to a set m inimum voltage, and then charge to a set maximum voltage at the given current, determined by the mass of

PAGE 24

24 the electr ode and the C rate to be used. In case of any problem that prevents the cell from charging or discharging to the set voltage, a time limit is specified for which the particular step would go on. Once this time has elapsed, the next step starts. All cells (annealed, as well as the as received) were subject to a galvanostatic te st between 1.5 V and 0.01 V. The lower limit of 0.01 V is to ensure that the cell starts cycling. The C rate used was C/1 for 100 cycles for all the cells throughout the course of this work. To ensure that the cells cycled properly, all cells w ere cycled f or 2 cycles at a rate of C/10 before cycling at a rate of C/1 for 100 cycles. Cyclic Voltammetry The cyclic voltammetry (CV) technique involves applying a constant step voltage to the cell and measuring the current response. Any peaks in the current indica te that a reaction has taken place at that particular voltage A shift in the peaks thus indicates a change in the reaction that occurs in the cell. The CV tests also give some information regarding the formation of the solid electrolyte interphase (SEI) layer. The cells were cycled at 0.414 mVs 1 for 20 cycles. The voltage limits were set between 1.5 and 0.01 V. Material Characterization Optical microscopy was used to observe the films after cycling. XTEM and EDS (EDX) were used in order to characterize the as deposited and 525 O C annealed sample. Only this annealed sample was chosen, as it gave the best electroche mical performance Sample preparation for the XTEM and EDS was done using a FIB apparatus.

PAGE 25

25 Optical Microscopy Optical microscopy, as the name suggests, makes use of light to magnify and resolve features of the object of interest to the naked eye. It works on the principle of is expressed as : (2 1) Here, used to illuminate the sample angle of collection used in a medium A schematic of the ba sic setup of an optical m icroscope is shown in figure 2 4 A microscope uses a system of lenses to provide a range of magnification. The denominator in equation 2 1 is called as the numerical aperture and is often approximated to unity. For a wavelength of ~550 nm, the resolution is approximately 300 nm, which is a good resolution for an optical microscope [18] Images of the cycled electrodes were ob tained at 200x using a L eco DMIL LED inverted optical microscope. A calibration slide where one division was 0.01 mm was used to calibrate the scale of the images obtained at a given magnification. Focused Ion Beam (FIB) The FIB is a popular technique to prepare samples for analysis in the TEM. The F IB uses a heavy ion like Ga + to mill the specimen to the required level of electron transparency suited for use in the TEM It is just like a scanning electron microscope; the only difference is that it uses positive ions instead of electrons. The use of a positive ion will damage the specimen, because of which ions are not used to image specimens. To prevent artifacts in the sample, a protective layer of carbon is deposited on the

PAGE 26

26 specimen. An organometallic precursor gas containing platinum is used along with the Ga+ ions to deposit platinum over the area of interest This acts as mask for further milling. The sample is then milled on opposite sides of this mask. After a thickness of about one micrometer is reached, the s pecimen is cut free from the base and one side. Now, the sample is further made thinner for ex situ transfer to a TEM sample grid using or the transfer to a grid in situ is done using a micromanipulator. In the in situ method, the micromanipulator is atta ched to the sample using platinum, and the other end is then milled, to attach the sample to the micromanipulator. This sample is then placed on the grid and the platinum used to attach the sample to the probe is removed by ablation to free the sample fro m the micromanipulator. The sample is then further thinned to the required thickness (under 100 nm) by milling the opposite sides of the sample in small increments and at lower accelerating voltages. The grid is then taken out of the chamber and is ready f or analysis in a TEM [19] The sample preparation by using the FIB is outlined in figure 2 5. Sample preparation was done on a FEI dual beam Strata DB235 FIB at MAIC. This FIB stage can be tilted between 10 and 56 O and the elect ron and ion beams are separated by an angle of 52 using a 300 pA aperture. Milling was done at an accelerating voltage of 30 kV, with the currents ranging between 100 and 5000 pA. To prevent too much of the protective platinum layer from damage, a small off axis tilt of about 4 was applied to the sample from either side. Cleaning was done using a 300 pA aperture at 7 kV.

PAGE 27

27 Cro ss sectional Transmission Electron Microscopy (XTEM) Transmission Electron Microscopy (TEM) is a characterization tool known for its versatility and ability to structurally and chemically characterize materials. The TEM operates on the basis of complex dynamical scattering theory. In a transmission electron microscope, electrons are directed on to thin specimens by magnetic fields. These magnetic fields allow very precise control over the path of the electrons. These sion electron microscope. In TEM, electrons that go through the sample are analyzed unlike SEM, where the electrons that are reflected back from the surface of the sample are collected and analyzed. The electrons travel through the sample, and hence detail s about the structure, latt ice parameter and composition can be obtained. Sample preparation for the TEM is done using the so that the electron may pass through the spe cimen and reveal the desired information. Since the FIB can damage the specimen and render the very purpose of obtaining micrographs useless, a thick coat of ca rbon is coated on the specimens using an SPI Module carbon coater and control using the pulsing technique In the bright field imaging mode, the electron beam incident on the sample is parallel to the optic axis of the instrument and it illuminates the area of interest on the sample through the objective aperture. The cross sectional view allows ob servation of different layers and measurement of the thicknesses of those layers. The high resolution mode uses the direct and one diffracted beam to from the image. TEM can also provide diffraction patterns of materials, and this is done by weakening the intermediate lens of the TEM. Placing a suitable aperture on the back focal plane of the objective lens produces a SAD pattern. An important thing to remember about TEM is

PAGE 28

28 that the image seen is a 2 D projection of a 3 D specimen. Also, the TEM sample is n ot representative of the whole sample, as a very small portion of the sample is actually used in obtaining the micrographs [18, 20] XTEM was performed on a JEOL 2010F TEM operating at 200 kV. Bright field i mages of the sp ecimens were obtained at 10, 20, 50 and 100 kx. Energy dispersive X ray Spectroscopy (EDS or EDX) EDS is a powerful tool for chemical characterization of a specific part of the sample. X rays are generated when a focused probe of electrons is made inciden t on the sample. These X rays are like fingerprints for the atom, and depend on electronic transitions in the atom that occur as the outer electrons jump in to fill inner shell vacancies For instance, a transition from the L shell to the K shell electron can occupy a given quantum state in an atom and there is a possi bility to get slig htly different binding energies for different transitions. S uch transitions are identified with a number after the shell Greek letter label. In EDS, the energy of the X ray emitted after the electron beam hits the sample is measured. The X rays are detected using the detector, and analyzed using suitable computer software to identify chemical species present in the are a of interest in the specimen [18, 20] EDS was performed using the detector system a ttached to the JEOL 2010F TEM. Line scan data at the interface were obtained using the Inca software on an attached computer. Data was suitably obtained in a Microsoft Word document.

PAGE 29

29 Figure 2 1. Schematic of a PECVD system. Figure 2 2 Microsoft Excel plot used for e lectrode mass calculation y = 53.953x 0.04 R = 0.9997 0 1 2 3 4 5 6 7 8 9 10 0 0.05 0.1 0.15 0.2 Area (cm 2 ) Mass (g) Series1 Linear (Series1) Cathode Anode Plasma Gas Inlet To vacuum Film

PAGE 30

30 Figure 2 3 Pouch cells. A ) In progress. B) C ompleted. Images by Rohit Mohan

PAGE 31

31 Figure 2 4 Schematic representation of the basic optical microscope Eye piece Objective lens Specimen Condenser lens Light source

PAGE 32

32 Figure 2 5. Schematic of the FIB. A) Forming trenches on two sides. B) Attaching the micromanipulator. C) Specimen lift off.

PAGE 33

33 CHAPTER 3 ANNEALING AND ITS EFFECTS ON THE PERFORMANCE OF THE CELL Overview of Experiments Annealing The as received films were deposited on the 304 stainless steel substrates using PECVD. T he electrodes were cut, weighed, annealed and used as anodes in the batteries as described in Chapter 2. The annealing is expected to form a silicide at the interface between the SS shim and the deposited layer. The results of the annealing were examined by cross sectional TEM to determine if any silicide was formed EDS was used to identify the c hemical species present in the annealed samples in comparison with the as received samples. Electrochemical Characterization The as deposited and annealed samples were tested for 100 cycles galvanostati cally, at a C/1 rate of 4200 mAh g 1 This corresponds to the Li 22 Si 5 phase. For the Li 15 Si 4 phase, this current corresponds to roughly 1.2 C. However, to ensure that all cells started cycling properly, all the cells were cycled for two cycles at a ra te of C/10 before cycling for 100 cycles at C/1. The cyclic voltammetry tests were performed at a voltage step rate of 0.414 mVs 1 for 20 cycles. Both the tests were performed within the v oltage range of 0.01 to 1.5 V. T he details of these tests and related equipment are described in Chapter 2, under electrochemical characterization Results Electrochemical Characterization Results of the galvanostatic testing and cyclic voltammetry tests are presented in this section.

PAGE 34

34 Galvanostatic Cycling The galvanostatic test results for the as deposited electrode is shown in figure 3 1 The cell shows high capacity of 2910 mAhg 1 for the first cycle, but this drops to 1608 mAhg 1 for the second cycle. The fade in capacity is very fast, and the cell practically out 400 mAhg 1 At the end of 100 cycles, it shows a poor capacity of 228.6 mAhg 1 Annealing at 475 C also does not show any impressive capacity retention but there is some improvement, as shown in figure 3 2 The discharge capacities for the 1st, 2nd an d 20th cycles are 1513 mAhg 1 1261 mAhg 1 and 1248 mAhg 1 respectively. This shows low initial capacity, but better retention as compared to the as deposited sample. It shows nearly 800 mAhg 1 in the 50 th cycle and 414 mAhg 1 for the 100 th cycle. However, annealing at 500 C shows a marked improvement over the as deposited electrode. This one exhibits capacities 2296 mAhg 1 for the 1 st and 1725 mAhg 1 for the 2 nd cycle. It also exhibits 1656 mAhg 1 for the 20 th and 1334 mAhg 1 for the 50 th cycle. The electrode annealed at 525 C has only marginally better capacity than the sample annealed at 500 C It exhibits a slightly delayed start, the capacity for the 1 st cycle being only 373 mAhg 1 It shows 1644 mAhg 1 and 1811 mAhg 1 in the 3 rd and 6 th cycl e respectively. This also shows a low capacity fade and has a capacity of 1627 mAhg 1 at the 50 th cycle. It manages to retain nearly 90% of its maximum capacity. However, annealing at 550, 575 and 600 C show capacities of 1468 mAhg 1 1259 mAhg 1 and 1326 mAhg 1 for the 20 th and 1261 mAhg 1 1053 mAhg 1 and 964 mAhg 1 for the 50 th cycles respectively. The results of galvanostatic testing of the lowest, optimum and highest temperature annealed s pecimens are shown in figure 3 3 Figure 3 4 presents the cyclin g of all annealed electrode cells and the as deposited electrode cell.

PAGE 35

35 These results are compared to the ion beam mixed samples, data for which was obtained from previous exp eriments, as shown in figure 3 5 Only the 10 16 cm 2 ion implanted electrode seems to have a superior capacity retention. A dose of 10 15 cm 2 results in a loss of capacity in the cell after about 50 60 cycles, whereas annealing gives a capacity retention for a 100 cycles, albeit lower than the ion implanted samples. The discharge capacities of select cells at certain cycles are provided in tabl e 3 1. The capacities for all annealed specimens are in figure 3 6. Cyclic Voltammetry The CV t ests for the as deposited 525 C annealed samples are shown in figure 3 7. These voltammograms suggest that there is no difference in the reaction that occurs at the electrodes for all the samples. The peaks do not show any significant shift. The reduction peak at ~0.6 V is indicative of the SEI layer formation, and is seen for the first cycle only. In these figures, the 1 st 2 nd and 20 th cycles are plotted. Materi al Characterization XTEM micrographs for the as deposited and 525 C anneal are shown in figure 3 8 and 3 9 respectively It is clear that no silicide is actually formed. EDS scans are shown in figure 3 10 There is no significant difference between the two The layer seen in the XTEM image of the annealed sample is the result of having cut the specimen across at an angle to the surface normal, and not perpendicular to it. Optical microscopy of the cycled electrodes reveals crack formation as expected. Figur e 3 11 is the optical micrograph of the specimen annealed at 600 C Discussion Though this system of silicon on SS has not been studied in depth with reference to silicide formation, there are many articles in literature that have studied the formation

PAGE 36

36 of the silicide when a thin film of iron is deposited on a silicon wafer and then annealed [21, 22] The paper by S. S. Lau and others was used as a guide to decide the annealing time and temperatures [22] T his paper suggests annealing times less than 10 minutes to form a thin silicide layer not more than 50 nm thick. However, to account for the presence of a native oxide on the substrate, an annealing time of 90 minutes was chosen. Many publications have reported that silicon is the diffusing species and that the silicide formation is time dependent [21 23] As seen from the XTEM image s in figure s 3 8 and 3 9 no silicide layer seems to have formed in case of the 525 C annealed sample. This is evident by comparison with the XTEM images in the paper by Baldwin and Ivey, where t he silicide formation is clearly visible However this paper investigates the formation of the silicide on a n etched silicon wafer after a 40 nm layer of iron was deposited on it [21] It is evident from the cyclic voltammograms that there is no change in the reaction that is occurring in electrodes. The reduction peak at ~ 0.6 V corresponds to the SEI layer, and is absent for the 2 nd and subsequent cycles This is clearly evident from f igure 3 7. Silicon is the only species that intercalates the lithium, expands to nearly 4 times its volume and shrinks again. Thus, any silicon that is involved in the silicide formation does not participate in the reaction. This means that the cells were charged and discharged at a rate higher than C/1. This means the actual capacities of these cells are better than those reported in the preceding subsection. But no silicide formation is actually observed as seen from the XTEM images and EDS scans; which means the formation of silicide is not the reason for the improved performance, as hypothesized. In fact, the formation of a silicide has been reported to adversely impact

PAGE 37

37 the capacity of a cell and deteriorate it [24] Silicide not forming is possibly due to the layer of native oxi de (containing iron and chromium oxides) present on the steel surface [25] It is also possible that the format ion of SiO 2 takes place. It is worthwhile to note here, t h at other studies involving the deposition of iron on silicon do not contain the oxide. In most studies, the silicon is sputtered or etched to remove the native SiO 2 before depositing iron. In this c ase, however, no etching or sputtering was done prior to silicon deposition. This makes it impossible to rul e out the presence of the oxide There is improvement in adhesion between the SS substrate and the deposited silicon as a result of annealing. As expected, cracking is observed in the deposited silicon films on cycling [11, 26, 27] The cycled electrodes were deprocessed as outlined in the chapter on experimental methods. Optical micrographs were obtained for all the annealed electrodes. N o reasonable inference on a trend in the capacity in relation to the island sizes can be drawn without the u se of SEM for imaging the cycled anodes. However, in the case of t he 600 C annealed electrodes (figure 3 11) large island sizes are observed and these exhibited correspondingly lower gravimetric discharge capacities. This trend has been reported for anod es by Li et al. where thinner films of a Si were found to show higher gravimetric capacities and also smaller island sizes [27] Size dependent fracture studies have also been performed by Liu et al. for silicon nanoparticles, where smaller nanoparticles exhibited finer cracks, and less breaking [28] Almost no silicon was left on the substrate, which means almost all of it was delaminated by the 100 th cycle for the as deposited samples. This confirms poor adhesion of the deposited film with the SS substrate. All annealed anodes performed better than t he as deposited anode for over 50 cycles, evident from figure s 3 4 and 3 6

PAGE 38

38 As stated earlier, the surface of the stainless steel substrate was not etched to remove any native passivation oxides Recently, it was shown that a very thin layer about 3 5 mm thick, of iron and chromium oxides is present on the surface of stainless steel by aberration corrected STEM. The presence of chromium oxide close to the matrix of the steel and a depletion of chromium close to the oxide/gas interface was observed [25] The study of the oxidation of steel by Langevoort et al. reports the enrichment of the surface of the steel with chromium and its preferential oxi dation at temperatures above 377C and the absence of an iron oxide layer [29] These prove useful to gain an understanding of what might actually be happening at the steel silicon interface. Also, s ilicon has a stable oxide, which forms readily compared to iron or chromium oxide, evident from the Ellingham diagram. The following is possibly what happens at the interface as the temperature is increased, the silicon atoms reduce the iron, by taking up oxygen fro m iron oxide. This continues to happen till all the oxygen from the iron oxide is used up. As the free energy of formation of silic on dioxide is lower than that of chromium oxide, silicon now starts to take up oxygen from the chromium oxide. A study on the oxide layers on stainless steel for bonding with glass for hermetic seals in feed through connectors have shown that the formation of SiO 2 affects adhesion in the case of metal adhesion on glass This study reports that the formation of SiO 2 as small islands at the interface results in better adhesion between the chromium and silicon oxides and that a more continuous layer of SiO 2 results in poorer adhesion [30, 31] Thus it is possible that for the thin films studied in this experiment, lower annealing temperatures resulted in some SiO 2 formation at the interface which improved the film adhesion and thus the cycling b ehavior. However further annealing

PAGE 39

39 may have resulted in the formation of a continuous oxide, which may reduce the adhesion leading to a maximum in the effect iveness of annealing on battery performance. Further studies of the oxide at the interface are need ed to determine if there is merit to this hypothesis. Another explanation for loss in the gravimetric capacity of the cells for annealing temperatures above 525C is the possible formation of a silicide, thereby reducing the amount of silicon actually available for reaction with lithium. A very small amount of silicide may be forme d, less than 10 nm thick for the higher temperature anneals. Copper silicide formation in amorphous silicon films 40 0 nm in thickness deposited on copper has been reported for higher temperatures (600 C ) by He et al. A similar trend of decreasing capacity is seen with increase in the annealing temperature and silicide formation [24, 32] Also, in creasing the thickness of the film to greater than 350 nm may not help, because earlier experiments with 700 nm and 1500 nm films have proved 350 nm to be an optimal thickness. Even ion beam mixing in such 700 nm films as electrodes have shown poor performance. These t hicker films are prone to forming larger cracks, and delaminate faster than thinner films, resulting in faster capacity fade. Thinner films have proven to minimize the probability of cracking and delay subsequent failure of the electrode [27, 32] Finally, all cells were tested at a high cycling rate of C/1. Cycling the cells at more lower rates that correspond to those in actual devices will show better improvement in capacity and retention. Concluding Remarks Thin films of amorphous silicon were deposited on stainless steel substrates, and subsequently tested as potential battery anodes. The annealing resulted in an increase

PAGE 40

40 in the capacity retention upon cycling, with the best improvements seen for the sample annealed at 525 C. This is possibly because of the formation of some SiO 2 at the silicon substrate interface, which improved the adhesion at the interface and hence the cycling performance. Further annealing possibly resulted in the formation of a continuo us layer of SiO 2 which reduced the adhesion, and in turn the cycling performance. Also, no silicide formation was observed. Cyclic voltammograms confirm the reaction of only silicon in the cells. Figure 3 1. Results of galvanostatic cycling of the as deposited electrode

PAGE 41

41 Figure 3 2 Result of annealing at 475 C for compared to the as deposited electrode. Figure 3 3. Comparison of cycling performance of the as deposited and 475, 525 and 600 C annealed electrodes

PAGE 42

42 Figure 3 4. Comparison of cycling performance of as deposited and all annealed electrodes Figure 3 5 Galvanostatic cycling test results for as deposited, optimum anneal and ion implanted specimens

PAGE 43

43 Figure 3 6 Trend in gravimetric capacity of the 50 th cycle with anneal ing temperature. Figure 3 7 Cyclic voltammograms of the 1 st 2 nd and 20 th cycles. A ) A s deposited B ) 525 C annealed electrodes 310

PAGE 44

44 Figure 3 8 XTEM image of the as deposited specimen. Figure 3 9 XTEM image of the specimen annealed at 525 C for 90 minutes

PAGE 45

45 Fi gure 3 10. EDS scans. A) A s deposited specimen. B ) 525 C annealed specimens. Figure 3 11. Micrograph of the cycled electrode for the specimen annealed at 600 C Darker areas correspond to the silicon and the lighter parts to the SS substrate.

PAGE 46

46 Table 3 1. Gravimetric discharge capacities for all cells for select cycles. Cell Gravimetric discharge capacity (mAhg 1 ) Cycle 1 Cycle 2 Cycle 20 Cycle 50 Cycle 100 As deposited 2910.49 1378.28 399.99 309.51 228.70 475 C anneal 1573.19 1261.41 1248.58 797.27 414.01 500 C anneal 2268.84 1725.44 1656.00 1556.99 1007.65 525 C anneal 372.67 132.54 1761.53 1627.22 1212.21 550 C anneal 1495.49 1251.00 1468.51 1261.88 323.01 575 C anneal 1346.88 1166.77 1259.24 1053.80 582.54 600 C anneal 1284.43 1080.11 1326.71 964.32 529.56 1e15 ion implanted 1044.94 634.10 2320.44 1812.41 535.18 1e16 ion implanted 3266.32 2713.47 2533.30 2366.38 1715.64

PAGE 47

47 CHAPTER 4 CONCLUSIONS Summary In summary, electrodes annealed for 90 minutes over a range of temperatures were assembled in pouch cells and electrochemically tested using galvanostatic and cyclic voltammetry methods While annealing was expected to form a silicide and aid the gravimetr ic capacity and its retention, no silicide was formed. This was confirmed by XTEM and EDS scans at the SS silicon interface. A more thorough literature survey has shown that thick layers of silicide are detrimental to cell performance. How ever, annealing did help, and an impressive capacity of ~1550 mAhg 1 was observed for the electrodes annealed at 525 C in spite of the high cycling rate, unlike the as deposited electrodes, which lost almost all capacity after about 20 cycles. Cyclic volta mmograms confirmed no changes in reaction at any of the anneali ng temperatures. Better adhesion between the deposited silicon films and the substrate is what seems to bring about this improvement in performance. A possible explanation for what is possibly going on at the interface is also proposed. Also, varying the annealing temperature has shown a trend with the maximum improvement in capacity exhibited for electrodes annealed at 525 C Future Work While it is now established that annealing has significan tly improved the performance of these cells, it will be interesting to study the effect of annealing time, at a given temperature I t is possible that the same set of results may arise from annealing for shorter periods at higher temperatures or longer per iods at lower temperatures. This will help establish a suitable schedule for annealing depending o n the thermal budget. Also, in situ optical/electron microscopy will help in understanding such systems better.

PAGE 48

48 Variations in cycling rate will also give valu able insights into the effects of annealing on cell performance Analysis of SEM images of cycled electrode will help to build a cor relation between the size of the islands on the cracked electrodes and annealing temperature and/or annealing time. Further investigations of the interface via aberration corrected STEM will also offer profound insights into annealing and its effects at the interface. To understand exactly what happens at the interface, chemical characterization of the interface, which detects the oxidation state of chromium or silicon, needs t o be done, using EELS. It will be interesting to study annealing of silicon films deposited on a foil of iron, instead of SS, and see if silicide formation occurs. However, the added possibility of oxidat ion of iron during storage can pose a problem. It would, in that case, be a good idea to store such specimens in an in ert chamber or in the glove box Another interesting approach would be to try and use the inverted system where iron is deposited on ultrathin wafers of silicon, and then annealed and used as the anode. Such an inverted system, as stated earlier in this work, has been the subject of intens e research for a very long time Silicon holds a lot of promise as an anode material. The use of t hin films may be restricted to use in small medical devices, and probably small portable electronics. But, the simple method of depositing a film and heating it to a favorable temperature bring s about quite an impressive improvement in the capacity of the cell. This is a relatively inexpensive method compared to synthesizing nanostructures or using ion beam mixing seen from ion beam mixing, this method can prove to be useful where capacities in the

PAGE 49

49 range of those exhibited by these annealed anodes are acceptable from the viewpoint of reducing costs. RTA at higher temperatures can possibly replace the longer annealing time which could allow this system to be commercially intro duced.

PAGE 50

50 LIST OF REFERENCES [1] P. Poizot, F. Dolhem, Energy & Environmental Science, 4 (2011) 2003 2019. [2] M.S. Alam, A. Roychowdhury, K.K. Islam, A.M.Z. Huq, Energy, 23 (1998) 791 801. [3] E.J. Cairns, P. Albertus, Batteries for Electric and Hybrid Electric Vehicles, in: J.M. Prausnitz, M.F. Doherty, M.A. Segalman (Eds.) Annual Review of Chemical and Biomolecular Engineering, Vol 1, 2010, pp. 299 320. [4] R.M. Dell, Solid State Ionics, 134 (2000) 139 158. [5] H.A. Kiehne, Electrical engineerin g and electronics 60, (2003) xx, 515 p. [6] J.M. Tarascon, M. Armand, Nature, 414 (2001) 359 367. [7] B.M.L. Rao, R.W. Francis, H.A. Christopher, Journal of the Electrochemical Society, 124 (1977) 1490 1492. [8] K. Mizushima, P.C. Jones, P.J. Wiseman, J.B. Goodenough, Solid State Ionics, 3 4 (1981) 171 174. [9] C.M. Hayner, X. Zhao, H.H. Kung, Annual Review of Chemical and Biomolecular Engineering, Vol 3, 3 (2012) 445 471. [10] J.P. Maranchi, A.F. Hepp, P.N. Kumta, Electrochemical and Solid State Letters, 6 (2003) A198 A201. [11] N.G. Rudawski, B.R. Yates, M.R. Holzworth, K.S. Jones, R.G. Elliman, A.A. Volinsky, Journal of Power Sources, 223 (2013) 336 340. [12] S. Pal, S.S. Damle, S.H. Patel, M.K. Datta, P.N. Kumta, S. Maiti, Journal of Power Sources, 246 ( 2014) 149 159. [13] C.K. Chan, H. Peng, G. Liu, K. McIlwrath, X.F. Zhang, R.A. Huggins, Y. Cui, Nature Nanotechnology, 3 (2008) 31 35. [14] M. Green, E. Fielder, B. Scrosati, M. Wachtler, J. Serra Moreno, Electrochemical and Solid State Letters, 6 (2003) A 75 A79. [15] T.D. Hatchard, J.R. Dahn, Journal of the Electrochemical Society, 151 (2004) A1628 A1635. [16] P.R. Abel, Y. M. Lin, H. Celio, A. Heller, C.B. Mullins, Acs Nano, 6 (2012) 2506 2516. [17] L.B. Chen, J.Y. Xie, H.C. Yu, T.H. Wang, Journal of Applied Electrochemistry, 39 (2009) 1157 1162.

PAGE 51

51 [18] D.B. Williams, C.B. Carter, SpringerLink (Online service), in, Springer,, New York, 2009, pp. 1 online resource (lxii, 760 p., (I761 715)). [19] L.A. Giannuzzi, F.A. Stevie, Micron, 30 (1999) 197 204. [20 ] B. Fultz, J.M. Howe, Transmission electron microscopy and diffractometry of materials, Springer, Berlin ; New York, 2001. [21] N.R. Baldwin, D.G. Ivey, Journal of Materials Science, 31 (1996) 31 37. [22] S.S. Lau, J.S.Y. Feng, J.O. Olowolafe, M.A. Nicole t, Thin Solid Films, 25 (1975) 415 422. [23] D.G. Ivey, D. Wang, Canadian Journal of Physics, 70 (1992) 860 865. [24] J. H. Cho, X. Li, S.T. Picraux, Journal of Power Sources, 205 (2012) 467 473. [25] E. Hamada, K. Yamada, M. Nagoshi, N. Makiishi, K. Sato, T. Ishii, K. Fukuda, S. Ishikawa, T. Ujiro, Corrosion Science, 52 (2010) 3851 3854. [26] J.P. Maranchi, A.F. Hepp, A.G. Evans, N.T. Nuhfer, P.N. Kumta, Journal of the Electrochemical Society, 153 (2006) A1246 A1253. [27] J. Li, A.K. Dozier, Y. Li, F. Yang, Y. T. Cheng, Journal of the Electrochemical Society, 158 (2011) A689 A694. [28] X.H. Liu, L. Zhong, S. Huang, S.X. Mao, T. Zhu, J.Y. Huang, Acs Nano, 6 (2012) 1522 1531. [29] J.C. Langevoort, I. Sutherland, L.J. Hanekamp, P.J. Gellings, Applied Surfa ce Science, 28 (1987) 167 179. [30] D.F. Susan, J.A. Van Den Avyle, S.L. Monroe, N.R. Sorensen, B.B. McKenzie, J.E. Christensen, J.R. Michael, C.A. Walker, Oxidation of Metals, 73 (2010) 311 335. [31] F.H. Stott, G.J. Gabriel, F.I. Wei, G.C. Wood, Werkstof fe Und Korrosion Materials and Corrosion, 38 (1987) 521 531. [32] Y. He, Y. Wang, X. Yu, H. Li, X. Huang, Journal of the Electrochemical Society, 159 (2012) A2076 A2081.

PAGE 52

52 BIOGRAPHICAL SKETCH Rohit Mohan was born in Mu mbai in January 1990 He was always intrigued by the things around him as a mischievous, inquisitive and restless 10 year old. The inquisitive young boy went on to obtain a bachelor t echnology from the Institute of Chemi cal Technology, Mumbai, in May 2012. Thirsty for more knowledge and wanting to contribute to science, he began broadening his horizons at the University of Florida, Gainesville. He spends his spare time in photography, reading listening to music and tr ave ll ing He graduated with a Master of Science in materials s cie nce and e ngineering in May 2014