THE EFFECT OF ADDITIVES AND SURFACE MODIFYING AGENTS ON THE SOLID ELECTROLYTE INTERFACE IN LITHIUM ION BATTERIES

Lithium ion batteries have become the most widely used and state of the art energy storage device in the current century, due to its high volumetric and gravimetric capacity, which makes portable electronic devices a possibility. The new and improved battery material have further increased the energy density of lithium ion batteries which made it possible to be used in high power applications such as electric and hybrid electric vehicles, aerospace applications as well as grid energy storage. Many problems still exist with the use of Li-ion batteries, such as safety, cost, material availability, environmental impact of used material, and etc. With the gain of the popularity of these devices, the need for even higher energy density keep on arising. Researchers have constantly been working on understanding the reactions taking place in the batteries as well as developing materials to improve the inherent problems of these systems. One major concern is the inability of li-ion batteries to work in a wide temperature window. Batteries tend to freeze at low temperatures during winter. Although this does not destroy the battery, it makes it harder for use and reduces the reliability. Also, at higher temperatures, due to the use of organic solvents, batteries have the tendency to catch fire. One approach to overcome this problem is to use a solvent like propylene carbonate, which has a working temperature range from -42 to 240 C. But propylene carbonate tends to co-intercalate into graphite anode and reduce in-between the graphite layers leading it to exfoliate and destroy the cell. In this study alkali metal ions were used as additives to understand the effect of them on the solid electrolyte interphase, that can improve the overall cell performance and use of propylene carbonate as a co-solvent. Another major problem is the low gravimetric capacity of the commercially used graphite anode. Many scientists have been working on using high energy density material such as silicon, which has ten times the capacity of graphite, as an anode material. But silicon has a propensity to change its volume up to ~300% in the charge discharge cycle which leads to cracking of the silicon particles and continuous consumption of the electrolyte. Binders play a large role in the performance of silicon-based electrodes. The binder not only helps in the direct mechanical properties of the electrode, but also participate in the solid electrolyte interphase (SEI) formation. Most studies conducted by the research community show the use of polymers as binders. This study shows that the use of single molecules with similar functional groups, can still act as good binders as they react on the surface of silicon to form a better SEI. One study shows that citric acid, which is a tri carboxylic acid, when used as a binder gives similar performance to polyacrylic acid, which is a polycarboxylic acid binder. When looking at the surface, lithium citrate was observed on the SEI. This gave an insight to functionalize the surface of silicon nanoparticles with citric acid, in order to form an artificial SEI containing lithium citrate. These surface modified nanoparticles have shown better performance with the conventional binders compared to unmodified particles consistent with the findings in the previous study. Many natural polymeric gums such as chitosan, guar gum, xantham gum, pine resins etc, have been studied as binders. This led us into using a single molecular natural glue, casein, a cheap, alternative binder material for silicon based electrodes. This has shown better performance compared to the conventional PVDF binder.


Background
With the increase of greenhouse gas emission from industry and motorized vehicles, governments around the world took the initiative to fuel greener energy research 1 .
Preceded by numerous inventions carried out by researchers such as Goodenough 3 , and Yazami 4 , the first Li ion battery was commercialized by Sony Co. in 1990s.
Although battery research at the beginning was very slow, technological advancement in the field of electronics led to a boom in popularity 2 . Soon the Li ion batteries became the power source of all mobile consumer electronic devices, such as laptops, cell phones, cameras etc. Battery manufactures have invented new ways to increase the energy density of materials and develop better cell architecture as well as improve safety aspects of lithium ion batteries to be used in consumer electronics.
Continuous research and development are still required with the rise of battery usage in electric vehicles, use in medical implants, grid storage etc 5 . In the first few charging cycles, as the solvated li ions approach the anode, they reduce on the surface due to the working potential forming a passivation layer called the solid electrolyte interphase (SEI). The SEI is a li ion conducting but electron insulating layer. This acts as a protective layer on the electrode to minimize the electrolyte from further decomposing on the surface 6 .

How a Li ion Battery works
Optimization of the SEI formation is important because the consumption of the electrolyte will lead to faster capacity fade.

Review of the problem
Continuous improvement of Li ion batteries is required due to cost of material, need for higher energy density, improved working voltage range, temperature range, lifetime and safety. Both anode and cathode material development is needed for use in high power applications. Electrolyte and additive development is required to address the other challenges. This work confronts these challenges by 1. Investigating additives to work on a wider temperature range 2. Investigating binders and surface functionalization to improve the performance of high energy density material, Silicon Interest in Li ion batteries (LIB) has significantly increased over the last two decades due to the widespread use of consumer electronics such as cellular phones, laptops, and cameras. There is also interest in the use of LIB for high power and wide temperature range applications such as electric vehicles [1], electric torpedoes [2], as well as military and aerospace applications [3]. While many materials such as lithium (Li), graphite, silicon (Si) and tin (Sn) have been recently investigated as anodes for LIB, graphite is the most widely used anode material in commercial LIB [4]. This is due to a low flat working potential vs. Li, and good cycling stability which provides long operational lifetimes [5] [6]. Graphite also offers good electronic conductivity, stable capacity, and low cost [7][8] Lithium is intercalated into the graphite structure as LiC6 with a specific capacity of 372 mAh/g. LIB electrolytes consist of a highly dissociating salt dissolved in a mixture of linear and cyclic carbonate solvents. The most frequently used cyclic carbonates are ethylene carbonate (EC) and propylene carbonate (PC). EC is considered an essential component in the electrolyte due to the importance of EC in formation of a stable solid electrolyte interphase (SEI), allowing reversible lithium intercalation/deintercalation for graphite electrodes [9][10][11].
However, EC has a relatively high melting point (36.4 [12,13]. Another approach is to replace EC with PC which has a broader operating temperature window of -48.8 o C to 242.0 o C [14] [15].
Unfortunately, PC co-intercalates into graphite during cycling resulting in graphite exfoliation and destruction of the anode [16].
One effective method to prevent graphite exfoliation with PC electrolytes is to generate a stable SEI during the first cycle. A more stable SEI can be generated via incorporation of electrolyte additives or electrolytes with high salt concentrations [17].
While electrolytes containing high salt concentrations have received significant recent interest, high concentrations of lithium salts typically reduce the conductivity of the electrolyte. Thus, most commercial LIB electrolytes have salt concentrations between 1.0 and 1.2 M [18][19][20]. The SEI formed on the graphite anode consumes lithium irreversibly from the electrolyte resulting in a decrease in the first cycle coulombic efficiency. Therefore, formation of a stable and uniform surface film with minimum electrolyte reduction is critical for LIB [21] [22]. Low concentrations of SEI forming electrolyte additives have been investigated to generate a stable SEI [23]. The additives are typically preferentially reduced before the carbonate solvents and lead to improved cycle life [24]. Recent reports have described the use of many different additives including sulfur containing species, unsaturated compounds, and other alkali metal cations [16] [25][23] [26]. Investigations of electrolytes containing both sodium and lithium salts, show that lithium intercalates into graphite in the presence of sodium without sodium intercalation or deposition [27]. A recent study has revealed that cesium cations can enhance the performance of lithium ion batteries containing PC based electrolytes [28][29][30] The role of different alkali metal ions as additives for graphite electrodes in PC based electrolytes has been systematically investigated via electrochemical and ex-situ been investigated with 1M LiPF6 in EC: PC: EMC (5:2:3) with and without added lithium, sodium, potassium and cesium acetates. The coulombic efficiency improves with an increase in cation radius of the acetate salt. Ex situ surface analysis was conducted via a combination of TEM, IR-ATR, XPS and ICP-MS to develop a better understanding of SEI formation, structure, and function for the SEI generated in the presence of different alkali metal ions.

Material
Battery

Electrochemical Measurements
All electrochemical measurements were performed in triplicate (< 1% variation) on 2032 coin cells using an Arbin BT2000 battery cycler. The cells consist of a graphite working electrode, lithium foil counter electrode, one 19 mm and one 15  To investigate the role of cesium and potassium ions in the anode SEI, cells containing Cs-Ac and K-Ac as additives were cycled at C/20 for a single cycle and carefully disassembled. The electrode was carefully rinsed twice using 500 L portions of the carbonate solvent mixture (EC: PC: EMC 5:2:3) to remove residual electrolyte. The electrode was further rinsed with three 500 L portions of DMC to remove residual EC. These electrodes were dried for 10 minutes under vacuum in the antechamber of an argon filled glovebox and then used as an electrode for a new cell with the STD electrolyte formulation without added Cs-Ac or K-Ac. The cell was cycled at C/20 for one cycle and the voltage profiles were compared.

2.4.
Results and Discussion:

Electrochemical Properties
The voltage profiles for the first cycle for each cell are provided in Figure 1. A plateau at ~0.7 V is observed for cells cycled with standard, Li-Ac and Na-Ac electrolytes. The voltage profiles after 10 cycles are provided in figure 1(b) and contain no high potential (> 0.4 V) shoulders consistent with no further PC reduction, and the reversible capacity and efficiency are good for all cells. The discharge capacity and coulombic efficiency are provided in Figure 1c and d, respectively. As the ionic radius of the alkali metal increases, the first cycle efficiency increases as depicted in Figure   1. The first cycle efficiency for cells containing the STD, Li-Ac and Na-Ac electrolytes is around 25%. The first cycle efficiency is increased to 40-50% for K-Ac and is further increased above 80% for Cs-Ac. The first cycle discharge capacity is similar for all electrolytes, suggesting that while there are significant differences in the quantity of electrolyte reduced, graphite damage and exfoliation is minimal during the first cycle. Cells cycled with the Cs-Ac and K-Ac electrolytes have higher specific capacity after 10 cycles than cells cycled with Na-Ac, Li-Ac, or STD electrolytes.
Cells containing Na-Ac, Li-Ac and STD electrolytes have relatively similar efficiency and capacity retention and more capacity fade over the first 10 cycles (~20 %). While all the cells have reasonable efficiency after 6 cycles, the electrodes extracted from cells cycled with the STD, Li-Ac or Na-Ac electrolytes have signs of considerable damage as evidenced by very poor contact with the current collector. Alternatively, the electrodes extracted from cells cycled with the Cs-Ac or K-Ac electrolytes adhere well to the current collector after 10 cycles. Oddly, these results differ from previous investigations where Na was beneficial while K had a negative impact on the graphite anodes [33]. with the different electrolytes are also provided in Figure 2. The spectra are similar for all electrolytes investigated, suggesting that the organic components of the SEI are similar for all electrolytes investigated.

X-ray photoelectron spectroscopy
The C 1s, F 1s, and O1s XPS spectra of graphite electrodes extracted from cells after 1 and 10 cycles with different electrolytes are provided in Figure 3. The XPS spectra of electrodes cycled with all of the different electrolytes are similar. The C 1s spectra contain a peak at 290 eV characteristic of the CO3 present in Li2CO3 and lithium alkyl carbonates [39]. In addition, the C1s spectra contain peaks characteristic of C-O (286.5 eV), C-H (285 eV), and graphite (284 eV). The primary difference in the C 1s XPS spectra for the different electrolytes is the intensity of the graphite peak at 284 eV. The relative intensity of the graphite peak is greatest for the Cs-Ac and K-Ac electrolytes, suggesting that the SEI is thinner for these electrolytes since the SEI is thinner than the depth of penetration of XPS, ~ 5 nm [40]. The O 1s spectra of electrodes cycled with all of the electrolytes are dominated by a broad peak centered at ~533 eV consistent with the presence of a mixture of C-O and C=O containing species characteristic of lithium acetate, lithium alkyl carbonates and Li2CO3 [41]. The F1s spectra of electrodes cycled with the different electrolytes contain two peaks characteristic of LiF at 685 eV and LixPFyOz at 687 eV. After the first cycle, the electrodes cycled with most of the electrolytes have similar intensity for the LiF and LixPFyOz peaks. However, the F 1s peak for LiF has much greater intensity for the electrode cycled with the Cs-Ac electrolyte. The F 1s spectra of the electrodes after 10 cycles also contain the same two peaks, but the LiF peak intensity is greater for the electrodes cycled with the STD and Li-Ac electrolytes. The P2p spectra of the cycled electrodes contain weak peaks between 134 and 137 eV (not shown) characteristic of LixPFyOz and residual LiPF6 [42]. The electrode cycled with the Cs-Ac electrolyte also contains very weak Cs 3d peaks ( Figure 4)

Transmission Electron Microscopy (TEM)
TEM images (figure 6) of the electrodes cycled with different electrolytes reveal significant differences in the thickness of the anode SEI and are in agreement with the XPS sputtering data. Electrodes cycled with the STD, Li-Ac and Na-Ac electrolytes have a very thick film on the electrode. The SEI is amorphous and discontinuous consistent with significant deposits of electrolyte reduction products. The cells cycled with electrolyte containing added K-Ac and Cs-Ac have a thin grainy SEI which has a similar morphology to the LiF rich SEI generated on graphite electrodes with concentrated electrolytes [42]. The gains are ~10 nm diameter and densely packed on the graphite surface. The TEM and XPS data suggest the presence of a LiF rich SEI composed of LiF nano-particles. The presence of Cs + or K + appears to initiate the formation of a thin LiF nano-particle passivation layer. The larger cations may initiate LiF nanoparticle aggregation by impeding the kinetics of Li ion intercalation into the graphite or providing a site for initial CsF generation acting a seed for LiF aggregation.

Inductive couple plasma -Mass Spectrometry (ICP-MS)
To further investigate the presence of cesium or potassium in the SEI, elemental analysis of the extracted electrodes was conducted by ICP-MS. The concentrations of Cs and K for the cycled electrodes are provided in Table 2. The concentrations of Cs and K are 100 times greater when cells are cycled with Cs-Ac or K-Ac, respectively, consistent with incorporation of the Cs or K into the SEI during formation cycling, as previously observed by XPS data (Figure 4).

Electrochemical Cycling of Harvested Electrodes
In an effort to confirm the generation of a stable SEI for cells cycled with Cs-Ac or K-Ac electrolytes, cells were prepared with the STD, K-Ac, and Cs-Ac electrolytes.
Electrodes were extracted from cells after one cycle and the pre-cycled electrodes were used to prepare new cells containing STD electrolyte (no added Cs-Ac or K-Ac).

Conclusion
A systematic investigation was carried out on the incorporation of alkali metal acetates as additives for PC based electrolytes for graphite anodes in lithium-ion batteries. The electrochemical performance and ex-situ surface analysis of cells containing the STD electrolyte and the Li-Ac electrolyte are nearly identical suggesting that the effect of the acetate anion on the electrochemical performance and SEI structure is small. Upon increasing the ionic radius of the alkali metal ions, a significant improvement in the performance was observed along with significant decrease in PC reduction. A combination of ex-situ surface analysis via XPS, IR-ATR, and TEM suggests that the organic SEI components are similar for all electrolytes, but higher concentrations of electrolytes. A morphological difference of the surface films was observed for the different electrolytes. Electrodes cycled with the STD, Li-Ac, and Na-Ac electrolytes are thick and inhomogeneous while electrodes cycled with the K-Ac and Cs-Ac electrolytes are thin and grainy. The larger cations may initiate LiF nano-particle aggregation and facilitate the generation of a thin, homogeneous, and robust LiF rich SEI.

2.6.
Li-Ion Battery Anodes, J. Electrochem. Soc Table 1. Elemental composition of cells cycled a) first cycle and b) 10 cycles Table 2. ICP-MS data for digested graphite electrodes. Data given in milligrams of metal in an electrode.

Introduction
The rapid development of portable electronic devices such as smartphones, tablets, This layer is also expected to undergo an electrochemical reduction process, resulting in a citric acid-derived SEI that protects the electrode from the electrolyte reduction in a manner similar to that observed for PAA. (23) In addition, citric acid is inexpensive, which is important for practical use.
We have investigated the electrochemical cycling performance of Si nanoparticles utilizing citric acid as a surface-modifying agent and binder. Several other carboxylic acids have been investigated along with the effect of PVDF and PAA binders on SEI formation on silicon nanoparticle anodes. The changes to the surface of the Si nanoparticles has been investigated via infrared spectroscopy with attenuated total reflection (IR-ATR) and X-ray photoelectron spectroscopy (XPS). The use of citric acid as a SEI modifier and binder provides improved capacity retention and a more stable SEI.

Experimental Section
Silicon nanoparticles (Alfa Aesar), super C (Timcal), and carboxylic acids (Acros) with a weight ratio of 50:25:25 (23) were thoroughly mixed with a mortar and pestle for 1 h using N-methyl-2-pyrrolidone (NMP) as a solvent. The slurry was then transferred to a vial containing a magnetic stirring bar and stirred for 3 h. The wellmixed slurry was spread on a copper foil and dried in a convection oven at 60 °C. Si electrodes with a PVDF (Mw = 600 000, MTI) and PAA (Mw = 450 000, Sigma-Aldrich) binder were also prepared in the same way for reference. The electrodes were punched into 14-mm-diameter disks and dried in a vacuum oven at 110 °C overnight.
The dry electrodes were not calendared. The thickness of the electrode laminates was ∼15 μm (excluding copper foil), and the total material loading was ∼1.2 mg/cm2 (0.6 mg/cm2 for Si).
Battery-grade solvents, salts, and additives were received from BASF. Coin cells

Electrochemical Cycling Behavior
The The superior performance of Si-CA compared to that of Si-PVDF suggests that CA plays an important role in enhancing the cycling performance of the Si electrode.
Because CA is small molecule, the mechanical and adhesion strength, which are considered to be critical properties for binders, are poor compared to those of the longchain PAA binder. However, the comparable cycling performance of Si-CA and Si-PAA suggests that the primary factor contributing to the performance enhancement is the reactivity of the CA, which modifies the surface of the silicon similarly to that reported previously for PAA and CMC. (23) The small molecular size of CA allows better uniform coverage of the surface of Si via the strong interactions of a large number of carboxylic and hydroxyl groups. Therefore, ex situ surface analysis has been conducted via a combination of ATR IR and XPS to investigate the SEI on fresh and cycled Si-CA anodes.

Ex Situ Characterization of the SEI Formed on Si-CA Electrodes
The IR spectra for fresh and cycled Si electrodes with PVDF and citric acid are and lithium carbonate is slightly enhanced but remains relatively small compared to peaks for lithium citrate. The lithium citrate derived from citric acid likely forms a uniform surface film on the Si nanoparticles, which inhibits electrolyte decomposition.
The C 1s, F 1s, and O 1s XPS spectra of the Si-PVDF, Si-PAA, and Si-CA electrodes are provided in Figure 5. The elemental atomic concentrations are provided in Figure   6. XPS provides information on the top layer of the SEI, ∼10 nm depth, whereas ATR IR provides information to a greater depth of ∼600 nm. The fresh Si-PVDF electrode contains peaks at 291 and 688 eV in the C 1s and F 1s spectra, respectively, characteristic of the CF2 group of the PVDF binder. The peak at 286.5 eV in the C 1s spectrum is from the CH2 of PVDF, and the peak at 284 eV corresponds to super C.
The broad peak at 532. 5

Conclusions
Several carboxylic acids were investigated as surface-modifying agents and binders The authors declare no competing financial interest.

Abstract
Silicon electrodes are of interest to the lithium ion battery industry due to high gravimetric capacity (~3580 mAh/g), natural abundance, and low toxicity. However, the process of alloying and dealloying during cell cycling, causes the silicon particles to undergo a dramatic volume change of approximately 280% which leads to electrolyte consumption, pulverization of the electrode, and poor cycling. In this study, the formation of an ex-situ artificial SEI on the silicon nanoparticles with citric acid has been investigated. Citric acid (CA) which was previously used as a binder for silicon electrodes was used to modify the surface of the nanoparticles to generate an artificial SEI, which could inhibit electrolyte decomposition on the surface of the silicon nanoparticles. The results suggest improved capacity retention of ~60% after 50 cycles for the surface modified silicon electrodes compared to 45% with the surface unmodified electrode. Similar improvements in capacity retention are observed upon citric acid surface modification for silicon graphite composite/ LiCoO2 cells.

Introduction
Lithium ion batteries have been widely used in the portable electronic device market for over two decades due to high energy density, good rate capability, and long cycle life 1-3 . Graphite is the most frequently used commercial anode material. Although graphite has good performance, low cost and high capacity retention, the relatively modest storage capacity (~370 mAh/g) has driven investigations of alternative anode materials 4 . Different anode materials with greater storage capacity have been investigated including lithium metal, tin, silicon and other metal alloys. While lithium metal anodes are very appealing, dendrite formation after long term cycling results in significant safety concerns [5][6][7] . Therefore, the use of lithium alloying compounds has been intensively investigated over the last decade. Silicon is the most attractive alloying anode material due to its high theoretical capacity. Lithiation of silicon results in the formation of alloys such as Li15Si4 with a theoretical capacity of 3580 mAh/g 8,9 .
In addition, silicon has a high volumetric capacity of 9786 mAh/ cm 3 . 10 Silicon is abundant and has low toxicity which makes it a good candidate for an anode material for commercial batteries. Silicon also exhibits a discharge voltage of ~0.4 V vs Li/Li + which allows it to maintain an open circuit potential which avoids lithium plating 9,11-13 . While theoretically interesting, there are numerous factors that make silicon electrode use difficult. Some of the important factors include the volume variation during the lithiation and delithiation process of 280% which leads to pulverization of the electrode, instability of the SEI due to the volume variation, and damage to the electrode laminate 14,15 . Numerous strategies have been undertaken to solve these stress induced problems that affect the electrochemical properties of the electrode including the use of nanoparticles, which limit the stress induced damage of from large volume changes 16,17 , containing the capacity of silicon to 1200 mAh/g 18 , the use of electrolyte additives for better SEI formation 19 , novel binders that can accommodate the stress or modify the surface of the silicon particles 20,21 and structure modification of the electrode materials 22 .
This investigation is focused on improving the interfacial properties of the silicon electrode by using surface modification. Citric acid (CA) has been previously reported as a binder for silicon-based anodes 23 . The use of citric acid binder provided comparable electrochemical performance to silicon anodes prepared with poly(acrylic acid) (PAA), one of the better binders reported for silicon anodes 24,25 . Ex-situ surface analysis of the citric acid based anodes indicates that the citric acid decomposes on the surface of the electrode to form lithium citrate which functions as a pre-formed SEI.
Our previous manuscript investigated the use of citric acid and a binder for silicon anodes. 25 In this manuscript, we investigate citric acid as a pre-treating agent to modify the surface of the silicon particles, which therein could generate an artificial SEI on the silicon particles. The use of citric acid as a surface modifying agent also allows the use of citric acid modified silicon in silicon/graphite composite electrodes.

Surface modification of Silicon nanoparticles with citric acid
Surface modification was carried out through sonication and subsequent stirring of

Preparation of silicon and silicon graphite composite electrode laminates
Laminates were prepared with both fresh silicon nanoparticles (Si-np) and modified

Results and discussion
The surface modification was carried out as described in the experimental section and

Surface analysis of the electrodes
In an effort to develop a better understanding of the source of performance improvement of the citric acid modified silicon particles, ex-situ surface analysis of the electrodes before and after cycling has been conducted. XPS and IR-ATR spectra of the fresh and cycled electrodes were obtained without exposure to air.
The IR-ATR spectra of the Si-np/Gr composite electrodes and the M-Si-np/Gr composite electrodes before and after cycling are depicted in Figure 6.

Conclusion
Surface modification of silicon nanoparticles with citric acid, a small molecule tricarboxylic acid, has been investigated to generate a more stable SEI and reduce study has shown improved performance for the surface modified silicon-based electrodes due to the formation of the pre-SEI. The surface modified Si-np/Gr composite/lithium cells have much better capacity retention, 60%, than the unmodified Si-np/Gr composite||lithium cells, 45%, after 50 cycles. The surface modified Si-np/Gr composite||LiCoO2 cells also have much better capacity retention, 78%, compared to the unmodified Si-np/Gr composite||LiCoO2 cells. The results suggest that surface modification of the silicon nano-particles with citric acid can significantly improve the performance of silicon/graphite composite electrodes.         Tables   Table 1 -Elemental concentrations obtained by XPS for fresh and cycled (20 cycles) Si-np/GR and M-Si-np/Gr electrodes.

Manuscript in Progress for Journal of Power Sources
High capacity electrode material has been studied over the years due to the constant demand for high energy density applications in li ion batteries. Among others, silicon has been considered a more promising candidate for its high theoretical capacity of ~ 3600mAh/g, abundant resources, low cost and low toxicity. However, silicon-based electrodes face rapid degradation due to the extensive volume variation (~300%) in the charge discharge process. Binders used in the electrode fabrication plays a crucial role in these high-performance electrodes since it can reduce the mechanical fracture in the cycling process.
Utilization of polymeric material as binders to hold the active material has been the most common approach used in the li ion battery electrode preparation. Recent studies carried out by Nguyen et. al have shown the use of small molecular carboxylic acids as binders which showed an improvement in the cell performance for silicon electrodes. In this study we introduce a cheap and environmentally friendly alternative that could be used as a non-polymeric binder for silicon electrodes. Further, the electrode preparation can be done in water medium which reduces the introduction of toxic organic solvents such as NMP to the environment. Casein is a milk protein found in bovine milk rich in amine groups and carboxylic acid groups which can form bonds with the silanol groups in silicon. A comparative study conducted between PVDF and Casein as binders have shown that when casein was used as binder, it shows better performance compared to PVDF. It has 30% higher capacity retention compared to PVDF with a 1300 mAh/g capacity after 200 cycles. Surface morphology and Solid

Introduction
The demand for high energy density material for lithium ion battery electrodes increase by the day due to the boom in the electric and hybrid electric vehicle market.
The current electric vehicle battery has a rather low driving range per single charge and has a high price tag associated with the production making it unaffordable to the wide population. Utilization of a high volumetric energy density electrode can give a high driving range with a similar sized unit cell which could drop the cost and make it affordable 1 . Currently, most commercial batteries use graphite as the anode material which was introduced in 1991. Graphite has good cycle life, low cost, low flat working potential vs Li, good electronic conductivity and good overall performance 2,3 .
Advancement in technology have made it possible to almost reach the theoretical capacity of graphite (~372 mAh/g), which is a specific energy of ~150 Wh/kg. This is still far too less for the energy demand of the electric vehicles and power grid energy storage applications 4 . Many alternative anode materials have been studied over the years and silicon has been the most promising candidate with a theoretical capacity of ~3600 mAh/g in its alloyed form (Li15 Si4) 4,5 which is almost 10 times higher than graphite. At higher temperatures the Li22Si4 alloying state can be reached which can give up to ~4200 mAh/g capacity [6][7][8] , making it the best of all other known anode material. This is due to silicon being able to alloy with about four lithium atoms whereas, six carbon atoms are needed to bond with one lithium atom. The half reactions for silicon and graphite are as follows 9 .
LixSi → x Li + + Si + x e -------(1) LiC6 → Li + + 6 C + e ------- (2) Silicon has many other advantages apart from the higher volumetric and gravimetric energy density. The working potential (0.5 V vs Li/Li + ) reduces the safety concerns of lithium deposition and dendrite formations. It is the second most abundant element on earth crust making it a cheap alternative and it is environmentally friendly 10 However, the practical implementation of silicon as an anode material has been challenging. The alloying/ dealloying of silicon with lithium experiences a volume variation of ~300%. This expansion induces stress on the particles which leads to instability, loss of contact between active material and the current collector which leads to pulverization of the electrode. This in turn leads to low capacity stability.
Secondly, the continuous formation of the SEI contributes to the fast capacity fade. In a conventional cell, the first cycle produces an SEI which acts as a barrier to stop the electrolyte from being further consumed. But due to the effects of pulverization, silicon exposes fresh surfaces throughout the cycling process which leads to consumption of lithium leading to capacity fade 6,11,12 .
Investigating binders is one of the various methods used to improve cycling stability of silicon-based electrodes. A common approach used by many researchers is to work with polymeric binders because of the polymeric structures ability to hold the silicon particles with each other as well as the current collector more stably. But in the case of silicon as well as most other high capacity material, the structure of the binder also play a significant role in the SEI formation in the charge and discharge process. In a study by Nguyen et. al. it was shown that small molecules, that has the ability of forming a functionally and morphologically more stable SEI can act as a good binder.
In the study a small molecular tricarboxylic acid, citric acid when used as a binder has outperformed the properties of PVDF and showed comparable performance to polyacrylic acid binder 13,14 .
Many naturally occurring adhesives such as xanthan gum 15 , guar gum 16 , chitosan 17 , cellulose, gelatin etc. has been studied in lithium ion batteries as binders. All these structures are known biopolymers. In this study another alternative single molecular structure, casein was used as a binder. Casein is a naturally occurring milk protein, which can easily be extracted from bovine milk in a very cost-effective method and can be used as a natural adhesive. Instead of harmful chemicals, water can be used for processing casein as a binder. Electrochemical studies were conducted to compare the performance of electrodes made with casein in multiple methods vs the common PVDF binder and the SEI was characterized using XPS, IR and SEM.

Cell construction and Cycling Procedure
All Infrared spectroscopy with attenuated total reflectance (IR-ATR) (Bruker Tensor 27 with LaDTG detector) was conducted inside a high purity N2-filled glovebox to prevent the reactions of samples with O2 and moisture. All spectra were collected with 512 scans at a spectral resolution of 4 cm −1 .
The surface morphology and elemental distribution of fresh and cycled electrodes was examined by scanning electron microscopy, coupled with EDS (JEOL 5900).

Results and Discussion
Four types of electrodes have been prepared and compared in this study. In each case 25% binder was used. First one is a conventional silicon electrode prepared with the most commonly used polymeric binder PVDF, shown in figure 2(d). In the same way a casein-based electrode was prepared with a similar slurry consistency. However, it was observed that when using a small amount of solvent, (solvent : solid = ~1 v/v) casein tend to foam producing air bubbles which cannot be removed using conventional methods such as vacuum suction or sonication. Electrodes were coated with the air bubbles with a higher thickness using a doctor blade. The punched electrode is shown in figure 2(a). A second method was then utilized to reduce the foamy nature of the electrode slurry. With the addition of more solvent (2.5 ml was further increased to 5 ml), the foaming could be reduced. Also, sonication with vacuum suction further reduced the formation of these large bubbles giving a smoother electrode as shown in figure 2(b). Furthermore, an electrode was prepared with a combination of casein and PVDF as the binders where no foaming was observed.

Electrochemical performance
Cell cycling was carried out in 2032 type cells according to the procedure described in the experimental section. The specific capacity of half cells is depicted in figure 3(a).
A rapid capacity fade is observed in both the electrodes containing PVDF. After 50 cycles cells the PVDF electrode retains a capacity of ~1000 mAh/g. The electrode with a mixture of PVDF and casein has a much lower first cycle capacity, but the capacity retention is slightly better than the PVDF electrodes after the first 80 cycles.
The casein electrode with the textured surface retains a capacity of ~1550 mAh/g after 50 cycles, which is significantly better than PVDF electrodes. The casein electrodes with excess water retains more than 1200 mAh/g capacity after 200 cycles, which is a significant improvement compared to all other electrodes. The first cycle efficiency is lowest for the electrode containing casein + PVDF consistent with the low first cycle capacity, while the casein electrode prepared with excess water had the highest first cycle efficiency.
An equivalent circuit diagram for a typical Nyquist plot for Li/ele/Si electrode is shown in figure 3(e). The semicircle, which is in the high to moderate frequency range is usually attributed to the impedance caused due to the SEI and charge transfer. The semicircle in the low frequency range is attributed to Li-ion diffusion 18 .

Surface Analysis
Ex-situ surface characterization was performed on fresh and cycled electrodes to understand the varying difference in the cycling performance. X-ray photoelectron spectroscopy reveal a few noteworthy differences in the fresh electrodes as shown in The FTIR-ATR spectra of the fresh and cycled electrodes are provided in Figure 7.

Understanding surface morphology
Surface analysis has shown that the change in the processing method used when utilizing casein as a binder not only improves the overall performance of the electrode but also alters the surface species and the distribution of the binder. Therefore scanning electron microscopy studies were conducted. As shown on figure 8

Conclusion
Casein was investigated as potential binder for silicon-based electrodes. showing the viability of using casein as a binder for silicon-based anodes. Further studies need to be conducted to improve the processing method to further smoothen the surface that could show even better performance.  Metal cell parts, glassware and plastic vials, plastic spatulas and pipette were washed with DI water and acetone and dried overnight in a vacuum oven before being introduced into the argon or nitrogen filled glove boxes. All the work required for the projects were carried out in a nitrogen or argon filled glovebox due to air and moisture sensitivity of the materials used in cell preparation.

Electrode Coating Procedures and Equipment
Mortar and pestle as well as an Ultra turax tube dispenser were used for electrode slurry preparation. A doctor blade coupled with an MSK-AFA-iii Automatic thin film coater from MTI corp. was used for electrode laminate preparation. Depending on the type of material used, and the particle size the methods used in slurry preparation differ. Calendaring of electrodes were carried out using MSK-HRP MR100DC high precision rolling press from MTI.

Cell Construction, Cycling, and Deconstruction
Coin cells type 2032 were constructed in an argon glove box to avoid exposure to air and moisture. Cells contain an exoskeleton of a can and a cap with a gasket to help create an air tight seal. Half cells consist of an electrode that needs to be analyzed with a lithium counter electrode and full cells will consist of graphite or silicon electrode as the negative electrode and lithium metal oxide electrode as the positive electrode. Two polyolefin discs each 15mm and 19mm, or two 19mm with a 16mm or 17mm glass fiber disc sandwiched between them were used as the separator. Each cell consists of exactly 100 of the electrolyte solution. The separators were sandwiched between the electrodes and the cell are sealed using an MTI crimper. Cell construction is subject to change based on the needs of the experiment but doesn't change drastically.
All cells were cycled on a battery cycler, at a given current rate based on the study conducted, and will undergo a charge, taper, a rest, a discharge, a rest, and will repeat as many times as necessary. The current rates are calculated in accordance with the theoretical capacity of the limiting electrode. If the capacity of the limiting electrode is X mAh, 1 C will represent X mA. Cells are cycled at different rates, and at different voltage ranges depending on what is needed for the experiment.
Cells were deconstructed in an argon filled glove box to isolate the anode and cathode for surface analysis as needed.

Electrode Preparation for Surface Analysis
All electrodes that will be used in surface analysis will be washed three times with dimethyl carbonate (DMC) or the electrolyte solvent mix without the salts and additives and dried overnight in an argon filled glove box antechamber. These samples are extremely air sensitive and thereby will be transferred to instrument used for analysis using Argon filled vials or vacuum sealed holders.

Battery Cyclers
Cell cycling is carried out in Arbin BT2000 battery cyclers coupled with temperaturecontrolled ovens.

Electrochemical Impedance Spectroscopy (EIS)
Princeton and Biologic VSP instruments were used for electrochemical impedance spectroscopic studies. This information will provide details about the cell impedance at various stages of the SEI formation and cell cycling. This instrument have also been used for cyclic voltammetry measurements which can be used to understand voltage ranges and reduction potentials.

Argon and Nitrogen Glove Boxes
Since LIB are sensitive to moisture, and oxygen even at very low levels, and argon glove box is needed to build and disassemble cells. The moisture content of the glove box should be kept below 0.1 ppm. In addition to cell building, the glove boxes are used for other moisture and air sensitive experiments such as electrolyte additive synthesis, electrolyte degradation studies and electrode slurry preparation.

Inductive Coupled Plasma Mass Spectrometry (ICP-MS)
ICP-MS analysis of the digested cycled electrodes was conducted using iCAP Q ICP-MS in Inbre Laboratory, Department of Pharmacy. ICP-MS is a very sensitive technique that can be used to quantify trace amounts of metals on the SEI. When metal containing additives are used, it is important to analyze the SEI to verify the presence of the added metal.

Scanning Electron Microscope (SEM)
SEM Joel 5900, was used to characterize the surface morphology, of the electrode and the solid electrolyte interphase at different stat of charge and cycling conditions. EDX is used to understand the elemental distribution on the SEI.

Transmission Electron Microscopy (TEM)
TEM imaging was conducted using a JEOL JEM-2100F TEM (Perbody, MA). Is used to characterize the solid electrolyte interphase thickness.

X-ray Photoelectron Spectroscopy (XPS)
XPS, Fisher Thermo Scientific K-Alpha using Al-Kα radiation source (hυ = 1486 eV), was used. XPS help reveal bonding information on the surface of the electrodes, giving insight on the functional groups and oxidation states of the SEI components. It can also be used for depth profiling of the electrode where the surface is etched with argon to analyze the chemical composition and oxidation states at deeper levels. It can also be used for area mapping of elements on the electrode surface. The TGA (TA Instruments Q5000) was used with nitrogen purge gas to quantify the extent of surface modification. Platinum pans were used for the analysis

Vacuum and Convection ovens
Fisher Scientific Isotemp vacuum oven-282A was used for drying cell parts, vials and powders. Jeio Tech TC-ME wide temperature chamber was used to carry out analysis at negative temperatures and higher temperatures. Fisher Scientific Isotemp incubators were used to maintain temperatures for cell cycling from 10 to 60 0 C. Thermo Scientific Precision convection ovens are used for all other drying purposes.