IMPROVING ELECTROLYTES FOR LITHIUM-ION AND LITHIUM OXYGEN BATTERIES

There is an ever increasing demand for fossil fuels. Lithium ion bat teries (LIBs) can effectively reduce the production of greenhouse gases and lessen the ne ed for fossil fuels. LIBs also have great potential in electric vehicle appli cations as an alternative to petroleum modes of transportation. Understanding the chemical reacti ons between the electrolyte and electrodes in LIBs is very crucial in developi ng batteries which can work over a wide temperature range and also give a wide potential window. The Solid Electrolyte Interface (SEI), formed by the reduction of solvent molecules on the surface of electrodes, is an important component of LIBs. The SEI is very essential to the performance of LIBs. One electron reduction pathway products of olvent molecules was investigated using lithium-naphthalenide. Methylene thylene carbonate, a high temperature additive has been synthesized and its performance has been tested at 60 C. Lithium-Oxygen batteries have an energy density ten times grea ter than that of LIBs. However, lithium-oxygen batteries have rechargability problems associate d with them. The most common electrolyte used in this type of batteries is L iPF6 in carbonate or ether based solvents. LiPF 6 inherently decreases electrolyte stability, since LiPF 6 can undergo thermal dissociation into PF 5 and LiF. PF5 being a strong Lewis acid, can react with electron rich species. The thermal decomposition reac tions of LiPF6 based electrolytes are studied in detail with regard to LIBs. The com prehensive study has been conducted on the thermal degradation of several electrolyte sys tem in the presence of Li 2O2.


Chapter-1 Introduction
Battery is an electrochemical devise which converts chemical energy to electric energy. The reduction-oxidation (red-ox) processes that occur in this electrochemical devise are responsible for the conversion of chemical energy to electrical energy. The major components of battery are: 1. Anode is the active material of the battery which gives off electrons to external circuit and gets oxidized.
2. Cathode is the active material that takes the electron from the external circuit and gets reduced.
3. Electrolyte helps in transport of ions from anode to cathode and vice versa during the red-ox process. The electrolyte should be electrically nonconductive but conducts the ions between the electrodes. 4. Separator prevents the direct electrical contact between anode and cathode.
The separator should be permeable to electrolyte.
An ideal combination of anode and a cathode is that which can give high voltage and also high energy density. Lithium is the most electropositive metal (-3.04V Vs SHE) and lightest metal (6.94 g/mole) makes it ideal choice as anode material 1 .
Thus the lithium batteries (primary) have very high energy densities. The theoretical specific capacity of lithium metal is 3,884m.A.h.g -1 . The rechargeability of such a battery is limited due to dendrite formation which on further cycling leads to the shorting of the cell. These safety issues lead to the development of rechargeable lithium ion batteries. Replacing lithium metal with intercalation compounds for lithium ions solves many problems arising due to primary lithium batteries. Lithium-ion batteries are based on such interaction compounds 2 .

Design of Lithium ion battery:
With the demand for fossil fuels ever increasing the need for alternative energies is getting lot of attention in the scientific community. Lithium ion batteries can provide an alternative solution to petroleum based transportation 3 . A typical rechargeable lithium-ion battery (LIB) consists of a Lithium transition metal oxide cathode and a graphite anode. The lithium transition metal oxide is coated on an aluminum foil which act as current collector and the graphite is coated on copper foil which acts as current collector. The anode and cathode are separated by a thin polyolefin film (polyethylene or polypropylene) separator. The electrolyte used in LIB's is usually a lithium salt (lithium hexafluorophosphate) dissolved in a blend of organic carbonates and esters.  Figure 1 shows the typical construction and discharge process of the lithium-ion battery. Lithium ions are removed from the cathode metal oxide lattice and intercalated in to the anode during the charging process. The transition metal undergoes an increase in oxidation state during this process and the reverse process occurs during discharge of the LIB. The LIB is constructed in discharge state so, these batteries have to be charged before any useful electric work can be extracted out of them 1

Interface in Li-ion batteries
One of the main reasons behind the cycleability of the present lithium ion battery is the solid electrolyte interface formed during the initial charging of the LIB. The lithiation process, which is insertion of lithium ions in to the graphene structure, occurs at a potential of ~0.2V vs. Li. So, during the charging process the potential of the anode is taken very close to the potential of lithium metal. This potential provides a highly reductive environment for the decomposition of electrolyte.  [6][7][8] Path A represents a two electron reduction of cyclic carbonates, path B and C represents one electron reduction of cyclic carbonate.

Additives for Lithium-ion batteries:
A major requirement for the lithium ion batteries for electric vehicles is long cycle life and also wide temperature window. The SEI formed by the standard electrolyte formulation ( 1M LiPF 6 in 1:1:1 EC:DMC:DEC) is not stable at high temperatures (above 60 o C). the ustable SEI can lead to parasitic reaction and accelerate the electrolyte degrardation. This leads to capacity fade in lithium ion batteries. To suppress this capacity fade and improve the cycle life, additives which can form a better SEI are used. Chapter 3 of this thesis deals with one of the additive that is developed by our research group which improved the high temperature performance of LIB.

Lithium/oxygen battery:
Most of the modern electronic devices have lithium-ion battery. This technology is being used in Hybrid electric vehicle to decrease the demand for fossil fuels 10 . The energy density of lithium ion batteries is dictated by the active material in the electrodes of these batteries. Li/O 2 battery has more energy density than the conventional Li-ion battery 10,11 . The theoretical energy density of Li/O 2 battery is around 11680Wh/kg, but the current state of rechargeable lithium /O 2 battery is limited to only few cycles 1 . The first rechargeable Li/O 2 battery was invented by

Introduction
Organic Carbonates are the major solvents used in lithium-ion batteries. The reversibility of the present Lithium-ion battery is mainly attributed to the electrolyte system used in these batteries 1 . Solid electrolyte interface (SEI) is formed by reduction of organic carbonates on the surface of the anode in lithium-ion batteries.
The SEI has the properties of the electrolyte which permits only Li + migration into and out of the electrodes but prevents electron tunneling through it 2 . The growth of the SEI at the expense of the electrolyte at high temperatures leads to the capacity fade in lithium-ion batteries. Understanding the reactions that transform these organic carbonates to SEI is essential in developing better lithium-ion batteries. A two electron pathway for the reduction of ethylene carbonate was proposed initially 3 , which was later challenged by Aurbach et. al. 1987. They proposed one electron pathway for the reduction of cyclic carbonates and later extended it to linear carbonates 5 . The major products of one electron reduction are semi carbonates and gases 4,5 . Other than semi carbonates the SEI also consists of lithium oxalates and alkoxides when cycled with carbon based anodes 6 .
The reaction mixture is stirred over night at room temperature. A dark green color solution of Lithium-naphthalenide is obtained.

NMR analysis of precipitates
The molecular structures of the precipitates formed in the reaction between lithium naphthalenide and various carbonates are analyzed via a combination of 1 H, 13 C NMR and FTIR. The 1 H and 13 C NMR spectra of the precipitates show spectral features similar to that of their parent carbonates but have a different chemical shift values. Figure 1 shows the 1 H NMR of the precipitates formed by reaction between Lithium naphtahalenide with EC. The 1 H NMR shows a singlet around 3.51 ppm which corresponds to methylene protons. Figure 2 contains the corresponding 13 C NMR spectrum which contains a singlet at 62.5 ppm and another singlet around 168.3 ppm.
The singlet around 62.5 ppm is due to methylene carbon (O-CH 2 -CH 2 -O-) and the singlet around 168.3 due to carbonyl carbon. Figure 3 shows the 1 H NMR of the precipitate formed by the reaction between Lithium naphthalenide and PC. 1 H NMR shows a doublet around 1ppm due to -CH 3 protons coupled to a single proton and multiplet around 3.3ppm which is due to -CH 2 type protons. There is another multiplet around 3.7 ppm which is due to the asymmetric proton. This type of splitting pattern is similar to that of PC. Figure 4 shows the 13  and -C(O)-type carbons respectively. Figure 5 shows the 1 H NMR of the precipitate formed in reaction between Lithium naphthalenide and DEC which shows a triplet at 1.12 ppm which is due to CH 3 protons coupled with CH 2 type protons and a quartet at 3.5 ppm due to CH 2 protons coupled with CH 3 protons. Figure Table 1 shows all the chemical shifts of the precipitates formed in the reactions.

GC-MS analysis of Gases
Since gas evolution was observed during the reduction reaction of carbonates with lithium naphthalenide, the gasses evolved during reaction were analyzed by GC-MS.
Upon reaction of lithium naphthalenide with EC the reaction mixture evolves gas which matches to the NIST library spectrum of ethylene. Reaction of propylene during the reaction of Li naphthalenide with DMC and DEC have not been fully characterized due to interference of Nitrogen and Oxygen gases in GCMS analysis.

Conclusions:
A model compound Li napthalenide has been used to simulate one electron reduction             However, the generation of poly(alkyl carbonate) on the graphite surface is typically cited as the primary source of performance enhancements. In this manuscript, we report a novel anode SEI film forming additive, methylene ethylene carbonate (MEC) [10,11]. MEC is readily synthesized via a mercury catalyzed cyclization reaction [12]. images were taken on a JEOL 5900 scanning electron microscope. Fourier transfer infrared spectroscopy (FTIR) was conducted on a Thermo Scientific Nicolet iS10

Incorporation of MEC into lithium ion cells results in a signifi
Spectrometer with an attenuated total reflection (ATR) accessory. The spectra were acquired with a resolution of 4 cm−1 and a total of 128 scans.

Results and discussion
Electrochemical testing on full cells

Surface analysis
The surfaces of three sets of electrodes were analyzed by X-ray photo electron spectroscopy (XPS), scanning electron microscopy (SEM), and infra-red spectroscopy with attenuated total reflectance (IR-ATR). The electrode samples were: fresh uncycled electrodes, electrodes after five formation cycles, and electrodes from cells with 30 cycles that were further stored at 60 o C for one week to simulate accelerated aging. The cells were dismantled at a full state of charge in an argon glove box and the anodes and cathodes were extracted. The electrodes were washed with DMC and dried under vacuum.

XPS analysis
The electrodes were analyzed by XPS.

FTIR analysis
The IR spectra of both the anodes and cathodes are dominated by peaks from PVDF binder at 1400, 1170, 1070, 877, and 840 cm −1 (Fig. 4).         during the reversible charging and discharging of the cell. [10][11][12][13] Initial investigations used electrolytes similar to those used in lithium ion batteries, LiPF 6 in carbonates. 1,2,9 However, carbonate solvents have been reported to be unstable to oxidation in Li-O 2 cells. [10][11][12] Other studies have investigated the use of glymes as solvents for Li-O 2 batteries but problems with the solvents have still been encountered. 3,11 In addition, the stability of the lithium salts used in the electrolyte has also been questioned. One reactive species which is generated during cycling, lithium superoxide (LiO 2 ), has recently been reported to react with lithium bis(oxalato borate) (LiBOB). 13 In order to develop a and Raman spectroscopy. XPS spectra were acquired on PHI5500 system using Al Kα radiation (hν = 1486.6 eV) under ultra high vaccum. The C-H peak (285 eV) is used as a reference peak for final adjustment of energy spectra. The spectra obtained were analyzed by Multipak (6.1A) software and fitted with XPS peak software (version 4.1). The Raman spectroscopy is performed on a Bruker optics FT-Raman microscope using a 785 nm excitation wavelength and 50x microscope objective. The Raman shifts were collected in a range of 3500-150cm -1 .

Results and Discussion
The reactivity of a LiPF 6 in a 1  (Figure 1). There were no observable changes in the 1 H NMR spectra (Figure 2). Continued storage of the electrolyte for two weeks results in only small changes to the spectra supporting a slow increase in the concentration of LiF and O=PF 2 OLi but no changes to the carbonate solvents were observed. Extended storage in carbonate based solvents at 55 o C (two weeks) results in the generation of OPF(OLi) 2 evidenced by a doublet in 19 F spectra ~60 ppm (920 Hz) and a doublet in 31 P spectra -10 ppm (920 Hz) consistent with further thermal decomposition of the LiPF 6 /carbonate electrolyte as previously reported. 15 Additional storage experiments were conducted with different ratios of The reaction of other salts including LiBOB, LiBF 4 and LiTFSI with Li 2 O 2 were investigated in the presence of PC and DME. The reactions were monitored by 11 B, experiments suggest that there is no degradation of LiTFSI in DME upon storage for an additional two weeks at 85 o C (Figure 4).

Surface analysis
The         investigations of the mechanism of the thermochromic transition, there are still questions relating to how the structure of the polythiophene (regioregularity, 3,4 substitution pattern, 5 molecular weight, 6 impurities, 7 precipitation method, 8,9 etc.) affects the mechanism. 10 Although most polythiophenes have a simple two-phase thermochromic transition, mesophase formation is observed in some poly (3alkylthiophene)s with long side chains. 11,12 In an effort to better understand the thermochromic properties of polythiophenes with different types of substitution patterns, we have prepared poly(3-docosoxy-4methylthiophene) (PDMT). On heating thin films of PDMT, the color of the polymer changes from purple to yellow. When the samples are rapidly quenched, a red-orange mesophase is generated. On heating a second time, the mesophase changes color irreversibly. This two-step thermochromism was investigated by reflection and fluorescence spectroscopies and differential scanning calorimetry (DSC).

Materials
All reagents were used without further purification. Reagents and solvents were  Optics S2000 instrument using a cylindrical fiber optics probe containing one source fiber and seven collecting fibers. The spectra were taken in reference to white background and dark background between 400 and 800 nm using a tungsten halogen lamp. Variable-temperature fluorescence spectra are recorded on Ocean Optics S2000 instrument with a blue LED light source (λ = 470 nm). The polymer films for reflection and fluorescence spectroscopy were prepared by drip-coating the polymer solution in THF onto a piece of white paper and evaporating the solvent with a heat gun. The mesophase was generated by heating the paper containing the polymer film and rapidly cooling with a cold metal plate. The polymer film was placed on aluminum block containing the thermometer.
The aluminum block was placed on a hot plate and heated at 2 o C/min. NMR spectra of monomer were recorded on a Bruker 300-MHz NMR spectrometer using CDCl3 as solvent. Molecular weight of the polymer was estimated by size exclusion chromatography in THF.

Results and discussion
Copper (I) chloride-catalyzed Ullmann coupling of 3-bromo -4-methylthiophene with 1-docosanol was used to prepare 3-docosoxy-4-methylthiophene. Oxidative polymerization of 3-docosoxy-4-methylthiophene was afforded by addition of FeCl 3 in methylene chloride followed by precipitation in methanol. PDMT was isolated as a deep violet powder. Thin films of PDMT are violet at room temperature and bright yellow at high temperature. The rapidly quenched film exhibits a red-orange color.
The red-orange color is indefinitely stable at room temperature. On heating the rapidly quenched films, the red-orange color first changes to violet and then changes to yellow at high temperature. Related thermochromic responses have been observed for poly(3alkylthiophene)s with very long linear alkyl side chains (>C22) and is consistent with the formation of a mesophase on quenching. 11,12 Variable-temperature reflectance spectra (Fig. 1) Where T 1 and T 2 are the transition temperatures, ∆T 1 and ∆T 2 are the width of the transitions, a 1 and a 2 are the reflection changes through the transition, and R 0 is the baseline reflection. The transition temperatures as determined from reflectance data at 600 nm are summarized in Table 1. The reflection spectra of the mesophase are similar to the annealed ground-state phase, but the two low-temperature phases are quite different from the high-temperature phase. Related variable temperature fluorescence experiments were conducted on PDMT films containing the mesophase (Fig. 2). The fluorescence spectrum (excited at 470 nm) of the mesophase contains a weak, broad emission with a maximum centered at ~ 650 nm. On heating, the emission intensity decreases at intermediate temperatures (60-110 o C), followed by an increase in intensity at high temperature (>120 o C). The emission band of the high temperature phase has peaks at 540 and 565 nm, with much greater intensity than the emission band of the mesophase. The temperature-dependent intensity of the emission spectra of a PDMT film containing the mesophase at 650 nm is provided in Figure 3.
The transition temperatures as determined from fitting the emission data to eq 1 are similar to the transition temperatures determined from the reflection data ( Table   1).The thermal behavior of the polymer was investigated by DSC. The DSC thermogram of the PDMT in the mesophase (Fig. 4) was conducted starting at -10 o C and ramped at 10 o C/min to 140 o C. The first heating cycle contains three peaks. The first peak at ~38 o C corresponds to a melt transition of the side chain, the second peak at ~48 o C corresponds to the loss of the mesophase, and a high-temperature peak at ~111 o C corresponds to the reversible thermochromic transition associated with mainchain melting. 11,12 Unlike the alkyl analogues, PDMT does not show an unusual peak shape in the central transition, suggesting that the melting/recrystallization proposed for the poly(3-alkylthiophene)s does not occur in PDMT. After a slow cooling cycle to anneal to the polymer, the second heating cycle contains only two peaks. The absence of one transition is consistent with a loss of the mesophase on slow cooling. The two transitions are consistent with side chain melting and main chain melting. The DSC, reflection spectra, and emission spectra indicate the presence of a partially disordered mesophase of PDMT that is structurally and electronically different from the ordered annealed low-temperature phase and the high-temperature disordered phase. The mesophase is generated via rapid cooling of the high-temperature phase. All three techniques are in agreement regarding the transition temperatures: the lower transition from the mesophase to the annealed state occurs at ~40 o C, whereas the transition to the hightemperature phase occurs at ~120 o C. The different phases of PDMT have differing degrees of π-π stacking and extended π conjugation. The annealed lowtemperature phase has the most extended p conjugation as evidenced by the longest wavelength reflection spectrum and the greatest π-π stacking as evidenced by the weak fluorescence intensity. 2 The hightemperature phase has little extended conjugation and weak π-π stacking interactions as evidenced by the shortest wavelength reflection spectra and greatest intensity fluorescence spectra. The mesophase has an intermediate extent of conjugation and π-π stacking. The spectral and thermal properties of PDMT are very similar to those previously reported for poly(3-docosylthiophene), suggesting similar mechanisms of thermochromism for poly(3-alkylthiophene)s and poly (3alkoxy-4-methylthiophene)s. 11,12 A mechanism for the phase transitions is provided in Figure 5.
Uncertainties are ±3 o C for the spectroscopy measurements and ±1 o C for the DSC measurements.