Unveiling the Effects of Laser Structuring on Thick Electrodes Using an Electrochemical Model

The lithium ion battery (LIB) electrolyte is a delicate system susceptible to a manifold of influences both from the outside and inside. Thermal stress and overcharging of the LIB can result in unwanted side reactions such as rapid electrolyte degradation and the so-called thermal runaway, respectively. Here, the degradation of the conducting salt, namely lithium hexafluorophosphate (LiPF6 )poses many risks to the LIB and the environment. Hydrolysis of LiPF6 leads to the formation of hydrofluoric acid (HF) and POF3 which can react with the organic compounds within the electrolyte to form highly toxic organofluorophosphates (OFPs). The formation of OFPs in LIBs due to thermal stress has been covered in literature. In this study the electrochemically induced formation of OFPs is discussed. Electrolytes containing fluoroethylene carbonate (FEC) as a film-forming additive indicated the presence of OFPs forming at operation voltages of 4.8 V in self-assembled LIB coin cells. Cycling of the coin cells with cut-off voltages of 4.8 V gave rise to 15 non-acidic and two acidic OFPs after 10 charge/discharge steps. The quantity of FEC had an impact on the amount of OFPs formed which gives reason to incorporate FEC in the formation pathways proposed in this study. The formation pathway of OFPs through EC-polymerization proposed in literature is evaluated and an alternative mechanism with FEC as the carbonyl-carbon donor is presented. Structure elucidation and separation of the formed OFPs is achieved by utilization of a) hydrophilic interaction liquid chromatography (HILIC) for acidic OFPs and b) reversed-phase (RP) chromatography for non-acidic OFPs hyphenated to a high-resolution ion trap time-of-flight mass spectrometer (IT-TOF-MS). Based on the findings in this study and due to the highly toxic nature of OFPs further investigation of the formation of OFPs in FEC-containing electrolytes is necessary. Further, the impact of the operation voltages needs to be assessed to ensure for safe and efficient LIBs.

Solid polymer electrolytes for bipolar lithium ion batteries requiring electrochemical stability of 4.5 V vs. Li/Li+ are presented. Thus, imidazolium-containing poly(ionic liquid) (PIL) networks were prepared by crosslinking UV-photopolymerization in an in situ approach (i.e., to allow preparation directly on the electrodes used). The crosslinks in the network improve the mechanical stability of the samples, as indicated by the free-standing nature of the materials and temperature-dependent rheology measurements. The averaged mesh size calculated from rheologoical measurements varied between 1.66 nm with 10 mol% crosslinker and 4.35 nm without crosslinker. The chemical structure of the ionic liquid (IL) monomers in the network was varied to achieve the highest possible ionic conductivity. The systematic variation in three series with a number of new IL monomers offers a direct comparison of samples obtained under comparable conditions. The ionic conductivity of generation II and III PIL networks was improved by three orders of magnitude, to the range of 7.1 × 10−6 S·cm−1 at 20 °C and 2.3 × 10−4 S·cm−1 at 80 °C, compared to known poly(vinylimidazolium·TFSI) materials (generation I). The transition from linear homopolymers to networks reduces the ionic conductivity by about one order of magnitude, but allows free-standing films instead of sticky materials. The PIL networks have a much higher voltage stability than PEO with the same amount and type of conducting salt, lithium bis(trifluoromethane sulfonyl)imide (LiTFSI). GII-PIL networks are electrochemically stable up to a potential of 4.7 V vs. Li/Li+, which is crucial for a potential application as a solid electrolyte. Cycling (cyclovoltammetry and lithium plating-stripping) experiments revealed that it is possible to conduct lithium ions through the GII-polymer networks at low currents. We concluded that the synthesized PIL networks represent suitable candidates for solid-state electrolytes in lithium ion batteries or solid-state batteries.

The use of laser-based processes in battery manufacturing is becoming more and more common [1]. The majority of the research and applications in this field focus on pulsed nanosecond lasers, primarily due to a lower capital cost. There are potential applications for shorter pulse length lasers in battery manufacture due to the change in ablation regime, from a thermal to a cold ablation process that results in significantly less damage to the structure of the batteryIn this research, a femtosecond laser has been used to cut and structure graphite anodes for coin cell batteries. The lasers fundamental emission wavelength is in the infra-red region at 1029nm and it has a maximum average power of 40W. The same laser is also capable of operating at the second (515nm) and third (343nm) harmonics, with a maximum average power of 20W and 10W respectively. This allowed for a direct comparison of how modifying the wavelength of the laser affects the structuring of the anode material, in this instance a graphite coating with an SBR-PVDF binder, with a specific focus on any amorphisation of the crystalline structure which may occur during the structuring process. A further comparison of optimised processes at each wavelength, with respect to time, are presented.A reduction in amorphization at all wavelengths was observed for both cutting and structuring graphitic material when compared to longer pulse length lasers. These findings could allow the use of lasers in creating 3D structured electrodes without the loss in performance of the final cell that is observed when using nanosecond lasers [3]. It has been shown that the creation of structures in the electrode material of a lithium ion battery can improve a cells stability under higher C-rates due to an improvement in ion diffusion kinetics by increasing the surface area of the active material [4]. As well as this, structuring can aid in the wetting of electrolytes into active materials, this is especially prevalent when looking into thick film electrodes such as those over 50µm [5].It has been shown that the specific capacity of a cell with a thick-film NMC electrode drops with increasing thickness, even for C rates as low as C/5 [1]. Laser structuring helps alleviate the problem by increasing the surface area of the active material and while this is done by removing some of said material, it can still lead to an increase in the effective capacity of a constructed cell. While some studies have investigated the machining of high aspect ratio micro-channels, the effect of wavelength on laser structuring of anodes with ultrafast lasers has not been thoroughly investigated until now.With more information on how various laser based structuring techniques effects the electrode material further improvements can be made not only to the individual performance of cells but the manufacturing process of lithium ion batteries in general.[1] W. Pfleging, “A review of laser electrode processing for development and manufacturing of lithium-ion batteries,” Nanophotonics, vol. 7, no. 3, pp. 549–573, Feb. 2018, doi: 10.1515/nanoph-2017-0044.[2] T. Jansen, M. W. Kandula, S. Hartwig, L. Hoffmann, W. Haselrieder, and K. Dilger, “Influence of Laser-Generated Cutting Edges on the Electrical Performance of Large Lithium-Ion Pouch Cells,” Batteries, vol. 5, no. 4, p. 73, Dec. 2019, doi: 10.3390/batteries5040073.[3] M. Mangang, H. J. Seifert, and W. Pfleging, “Influence of laser pulse duration on the electrochemical performance of laser structured LiFePO 4 composite electrodes,” J. Power Sources, vol. 304, pp. 24–32, 2016, doi: 10.1016/j.jpowsour.2015.10.086.[4] P. Smyrek, Y. Zheng, H. J. Seifert, and W. Pfleging, “Post-mortem characterization of fs laser-generated micro-pillars in Li(Ni 1/3 Mn 1/3 Co 1/3 )O 2 electrodes by laser-induced breakdown spectroscopy,” 2016, no. March 2016, p. 97361C, doi: 10.1117/12.2210815.[5] W. Pfleging and J. Pröll, “A new approach for rapid electrolyte wetting in tape cast electrodes for lithium-ion batteries,” J. Mater. Chem. A, vol. 2, no. 36, pp. 14918–14926, 2014, doi: 10.1039/C4TA02353F.

To meet the growing performance demands for personal electronics and electric vehicles the energy density of lithium-ion batteries can be increased by incorporating thicker electrodes. We present thick “honeycomb” electrodes based on patterned, vertically aligned carbon nanotubes (CNTs) on Cu foils. Thick electrodes are created by Si deposition on >100 μm tall honeycomb patterned CNTs. Si-CNT electrodes are cycled in half-cells, demonstrating electronic connection between the Si and Cu foil via the aligned CNTs. For ~4.7 mAh cm−2 capacity the honeycomb patterning improves capacity retention (78%) over 30 cycles compared to non-patterned electrodes (58%). We attribute this improvement to the honeycomb’s ability to accommodate Si expansion, thereby reducing cracking that causes active material loss and solid electrolyte interphase instability, and to provide pathways for Li-ion transport into the electrode. The Si-CNT electrode capacity is further increased to 20 mAh cm−2 by increasing the Si loading. Finally, a fluoroethylene carbonate containing electrolyte is used to increase cell lifetime. Here, the honeycomb electrodes have a higher areal (~10.2 mAh cm−2) and retained (65%) capacity over 180 cycles, and exhibit superior rate performance to their non-patterned counterparts. Our work demonstrates the role of patterning in enabling aligned CNTs as a robust template for thick battery electrodes.

For widespread adoption of electric vehicles, lithium-ion batteries need to achieve energy densities of around 350 Wh/kg, cost less than $100/Wh, and be able to charge quickly within 15 minutes. The production of most inactive materials within modern Li-ion cells (e.g. separator, current collector, packaging) have already undergone tremendous optimization but many opportunities remain for improving the performance of the electrodes and reducing the cost and time of wetting, formation and aging processes. By increasing the thickness of electrodes clear opportunities exist for enhancing the energy densities of conventional cell materials and reducing cost. For example, doubling the thickness of electrodes in full cells from 50 μm to 100 μm can increase the energy density of the cell by about 16% and reduce the cost of the cell by 50% (from $249/kWh to $172/kWh)1. These benefits arise from increasing the ratio of electrode active material to inactive material within cells; however, increased thickness is often accompanied by reduced rate performance and increasingly difficult electrolyte wetting and formation processes due to the challenge of infiltrating the electrode pore structure with electrolyte and displacing gases during formation2. One emerging electrode processing technology involves the laser-ablation of small micro pores or channels into Li-ion battery electrodes to enhance their electrolyte wetting and rate performance. Pulsed lasers are increasingly being used in advanced manufacturing processes due to their ability to weld, sinter, or remove materials quickly and with very high precision. When applied to high energy density thick electrodes, micro-structuring by laser patterning has been shown to improve the rate capabilities of cells while maintaining high capacity retention3. These laser ablated pores or channels provide low tortuosity paths for lithium ions to reach deeper regions in the thick electrodes more quickly than before, limiting concentration gradients and thus reducing unwanted lithium plating at elevated rates. Despite these benefits, the utilization of ultrafast lasing ablation systems as an in-line technique to selectively remove patterned regions of active material from lithium-ion battery electrodes during production has not yet been widely adopted by the industry. In this work, we identify and manufacture laser-ablated 3D electrode architectures for enhanced Li+ transport and electrochemical performance with the hope of inspiring confidence in the benefits, practicality and cost effectiveness of the laser patterning technique. Preliminary modelling work at NREL has explored the limitations of planar electrodes for fast charge performance and the importance of microscale features for improving ion-transport4-8. With the correct pore/channel geometry, the energy density of ablated electrodes can be significantly improved compared to traditional architectures. We have combined simulation models and experimental data to optimize the 3-D laser patterning of our electrodes. Modelling elucidated the electrodes baseline transport limitations and quickly identified the most promising electrode architectures for favorable cost, energy- and power-density. Next, we used our benchtop ultrafast femtosecond laser ablation system to pattern our thick (ca. 100 µm) electrodes. After processing, the electrodes were inspected using cross-section scanning electron microscopy (SEM) so that ablation parameters such as laser wavelength, power and repetition rate could be optimized and distributions of feature sizes could be identified giving insight into the limitations of the laser system. We observed that the anode and cathode materials responded significantly different to the laser ablation process and that this must be considered during electrode preparation. Next, scanning electron microscopy with energy dispersive spectroscopy (SEM-EDS) and x-ray diffraction (XRD) were used to probe for local chemical or structural changes in the electrodes’ with patterning to determine if the remaining active materials experienced permanent laser-induced damages. Once successful achievement of the target electrode architecture was confirmed, the benefits of the achieved 3-D patterning were investigated by cycling the electrodes in full cell coin cells and comparing their performance against standard cells with unmodified electrodes (benchmark). Cycling data clearly shows improved rate performance and capacity retention derived from the patterning of the thick graphite and NMC electrodes. Direct structuring of the electrodes has improved their ion transport and reduced lithium concentration gradients (cell polarization) resulting in high specific energy- and high power-density. While clear evidence of rapid lithium plating was observed in the standard benchmark cells at rates exceeding 1C, this has been largely mitigated in the laser ablated electrodes. Lastly, we probed the wetting of these electrodes with electrochemical impedance spectroscopy to map in real time the improved rate and extent of wetting achieved through the adoption of our 3-D laser patterns. Achieving shorter wetting times will correspond to reductions in cell manufacturing costs and is expected to present a significant market advantage. Figure 1

Introduction An internal resistance in an electrode for lithium ion batteries (LIB) is consisted of three resistances, which are an electric resistance, ionic resistance and charge-transfer resistance. Since a length of Li-ion conductive pathways decreases with decreasing electrode thickness, the internal resistance can be reduced by decreasing the electrode thickness [1]. Therefore, decreasing the electrode thickness is one of the effective methods to improve the power of LIB. On the other hand, making the thin electrodes by the same procedure as the thick electrodes probably causes a change in the electrode structure and increase of the internal resistance. However, it has not been clarified. In this study, the LIB electrodes with various thicknesses were examined to clarify the influence of decreasing the electrode thickness on its structure and the internal resistance. Half cell tests and scanning electron microscopy with focused ion beam (FIB-SEM) analyses were carried out to evaluate electrochemical and structural properties of the electrodes. Results and Discussion Fig. 1 shows the electrode thickness dependence on the three resistances: the ionic resistance (Rion), charge-transfer resistance (Rct), and total internal resistance (R).The resistances are normalized by the weight of active materials. At around 15 μm of the electrode thickness, R showed the minimum. At less than 15 μm of the electrode thickness, R increased due to the Rct increase. Fig. 2 shows 3D reconstructed images of active material phase, acetylene black (AB) and binder phase, and pore phase in the thin (9 μm) and thick (25 μm) electrodes . The porosity of the thin electrode was larger than that of the thick electrode. In addition, the ABs and binder in the thin electrode were dispersed less uniformly than those in the thick electrode. The large porosity and localization of the ABs and binder could produce a high contact resistance among active materials. As a result, the reduced reactive surface area increased the Rct in the electrode. In our presentation, we will discuss the influence of decreasing the electrode thickness on .the Rct more detail with other analysis results, and the method for improving Rct. Acknowledgment This work was supported by New Energy and Industrial Technology Development Organization (NEDO) of Japan. Reference [1] C. Fongy et al. , J. Electrochem. Soc , 157, A885 (2010). Figure 1

Electrochemical characteristics and energy densities of lithium-ion batteries using mesoporous silicon and graphite as anodes

Recent development of ionic liquid-based electrolytes in lithium-ion batteries

Modeling the effect of electrode thickness on the performance of lithium-ion batteries with experimental validation

Phenyl tris-2-methoxydiethoxy silane as an additive to PC-based electrolytes for lithium-ion batteries

Lithium ion batteries have been investigated extensively due to their widespread applications in portable electronic devices and electrical/hybrid vehicles. However, significant challenges still exist for an extended calendar life at a wide temperature range. Due to the intrinsic drawbacks of the commonly used LiPF6-carbonate electrolyte systems, such as insufficient thermal stability at elevated temperature and unavoidable HF contaminants, much effort has been paid to exploring novel additives. It is well known that the introduction of additives into electrolyte systems is one of the most effective and economic approaches to improve performance of lithium ion batteries. In this dissertation, a novel class of borate compounds has been successfully synthesized and screened as additives for electrolyte of lithium ion battery. Nuclear magnetic resonance (NMR) spectroscopy was utilized to characterize compounds by dissolving additives into deuterated solvents. The cycling performance of these novel additives and other commercialized additives was compared by adding them into 1M LiPF6 EC/EMC electrolyte, LiNi0.5Mn1.5O4/Graphite cells are cycled under both room temperature and elevated temperature up to 4.8V. Electrochemical impedance spectroscopy (EIS) and Linear sweep voltammetry (LSV) were used to investigate electrochemical activity of additives. The investigation of the interrelationship of cycling performance, additive structure, and electrode surface film structure has been conducted by scanning electron microscopy (SEM), transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS) and FT-IR instruments. SEM and TEM images showed that novel additives can form uniform solid electrolyte interface (SEI) and cathode solid electrolyte interface (CEI). XPS and FT-IR spectra were acquired to analyze main components of SEI and CEI, and they are beneficial for further understanding how addition of additives changed surface chemistry of electrodes. The surface reactions of both additives and electrolytes with the graphitic anode and lithium nickel manganese oxide cathode of lithium ion batteries have been speculated. New additives can not only form more uniform SEI on surface of anode, but also beneficial for forming uniform CEI on surface of cathode.

Structural regulation chemistry of lithium-ion solvation in nonflammable phosphate-based electrolytes for high interfacial compatibility with graphite anode

The pursuit of higher energy density in lithium-ion batteries has led to the development of new cathode materials that can operate at elevated voltages and provide increased specific capacities. [1-2] One such class of materials is the nickel-rich layered oxide cathodes known as LiNixMnyCozO2 (NMC). These cathodes offer high specific capacities and demonstrate electrochemical stability at high cutoff voltages, making them promising candidates for advanced lithium-ion batteries. [3-4] However, NMC cathodes face a significant challenge in the form of capacity fading during cycling. This issue primarily arises from the voltage instability of the conventional carbonate-based electrolytes used in lithium-ion batteries, which are typically designed for the 4-V lithium-ion chemistry. [5-6] As a result, researchers and scientists have directed extensive efforts toward developing new and improved electrolyte systems that can withstand the high voltage requirements of NMC cathodes and other high-energy applications. [7-8]Designing and synthesizing new molecules for use as electrolyte solvents in lithium-ion batteries is indeed a complex and challenging task. Researchers often focus on optimizing one specific property, which can lead to the development of molecules that excel in that aspect while potentially overlooking other crucial features that are necessary for stable battery cycling. For example, the development of α-fluorinated sulfones as electrolyte solvents was driven by their exceptional anodic stability. [9] This stability is achieved by introducing strong electron-withdrawing trifluoromethyl groups directly attached to the sulfonyl group. However, while this electron-withdrawing effect enhances anodic stability, it can also significantly increase the reduction potential of α-fluorinated sulfones. This heightened reduction potential renders them unstable when used with graphite anodes, highlighting the delicate balance required in designing electrolyte solvents that perform well across various aspects of battery operation.In this presentation, we embraced the concept of a "golden middle way" when designing and synthesizing new electrolyte solvents. The recently developed β-fluorinated sulfone, TFPMS, doesn't claim to be the best in any single property, but it strikes a balance across various key characteristics. This equilibrium has proven to be highly effective in ensuring the long-term stability of high-voltage graphite||NMC622 full cells. Positioned at the β site of the sulfone molecule, the strong electron-withdrawing trifluoromethyl group renders β-fluorinated sulfone sufficiently stable against the high-voltage NMC622 cathode, even if it possesses a slightly lower oxidation potential compared to its α-fluorinated counterparts. Furthermore, its reduction potential is lower than that of α-fluorinated sulfone, making it considerably more stable when paired with the graphite anode. While it may not possess the same high lithium solvating power as the typical sulfone (EMS), the lithium solvating capacity of β-fluorinated sulfone falls somewhere in between, mitigating transition metal dissolution and deposition in the graphite||NMC622 full cell that utilizes a β-fluorinated sulfone-based electrolyte. As a result, the full cell equipped with TFPMS-based electrolyte demonstrates superior cycling performance, with a significantly higher average capacity than cells using regular sulfone or α-fluorinated sulfone-based electrolytes. In summary, our approach exemplifies the successful application of the "golden middle way" in designing and synthesizing new electrolyte solvents.Reference: Santhanam, R.; Rambabu, B., Power Sources 2010, 195, 5442.Lin, F.; Markus, I. M.; Nordlund, D.; Weng, T.-C.; Asta, M. D.; Xin, H. L.; Doeff, M. M., Commun. 2014, 5, 3529.Li, W.; Song, B.; Manthiram, A., Chemical Society Reviews 2017, 46 (10), 3006.Xu, J.; Lin, F.; Doeff, M. M.; Tong, W., Journal of Materials Chemistry A 2017, 5 (3), 874.Xia, J.; Petibon, R.; Xiong, D.; Ma, L.; Dahn, J., Journal of Power Sources 2016, 328, 124.Chen, S.; Wen, K.; Fan, J.; Bando, Y.; Golberg, D., Journal of Materials Chemistry A 2018, 6 (25), 11631.Zheng, J.; Lochala, J. A.; Kwok, A.; Deng, Z. D.; Xiao, J., Advanced Science 2017, 4 (8), 1700032.Yamada, Y.; Wang, J.; Ko, S.; Watanabe, E.; Yamada, A., Nature Energy 2019, 4 (4), 269.Su, C.-C.; He, M.; Redfern, P. C.; Curtiss, L. A.; Shkrob, I. A.; Zhang, Z., Environ. Sci. 2017, 10 (4), 900.

For Gas Chromatography Chemical Ionization Mass Spectrometry (GC-CI-MS) method development, a standard lithium-ion battery (LIB) electrolyte was thermally aged at 95 °C for a faster generation of decomposition products.

Lithium-ion batteries (LiBs) have been successfully utilized as power sources in various applications ubiquitous in our daily lives, ranging from portable electronic devices to electric vehicles. To meet the increasing energy demand, however, development of LiBs with a higher energy density is inevitable. Using silicon (Si) as an anode is promising due to its high theoretical capacity (approximately 4200 mAhg-1), appropriate operating voltage (~0.4 V vs. Li/Li+), abundance, and environmentally benign nature.1 Despite the aforementioned advantages, several challenges remain before the commercial utilization of Si anodes, namely the low conductivity of its readily-formed surface oxide (SiO2), a large volume change (up to 400 %) during lithiation/delithiation, and the instability of Si/electrolyte interface (SiEI).2-3 As the large volume change leads to fracture and continuous evolution of SiEI, achieving a stable SiEI is a key to developing successful LiBs with Si anodes. Water, which is known to react with LiPF6 salt in carbonate-based electrolytes to generate corrosive hydrofluoric acid (HF),4 is a source for the electrode/electrolyte interface destabilization. Thus, it is essential to understand how the presence of even trace amounts of water in the system affects the interfacial chemistry of the LiB electrodes. While the performance of LiBs with graphite or lithium titanate negative electrodes have been previously studied with excess water in the electrolyte,5-6 information on the effect of water concentration in the electrolyte on Si anodes, and the SiEI in particular, is limited. In this work, we evaluated the electrochemical performance of Si anodes in a standard electrolyte (1.2 M LiPF6 in EC:EMC (3:7 wt%)) with varying concentration of water (10 - 1,000 ppm). Subsequent analyses on the cycled Si anodes were performed utilizing a variety of surface and bulk characterization techniques. The surface chemistry (e.g., composition and evolution) of SiEI was analyzed using Fourier-transform infrared spectroscopy (FTIR), Raman spectroscopy, time-of-flight secondary ion mass spectrometry (TOF-SIMS), and X-ray photoelectron spectroscopy (XPS). Surface morphologies of cycled Si anodes were measured with tapping-mode atomic force microscopy (AFM), and the surface resistivities were mapped in contact-mode with scanning spreading resistance microscopy (SSRM). The information obtained is utilized to understand water concentration effects on the properties, reactivity, and evolution of the SiEI and the electrochemical cell performance. References Feng, K.; Li, M.; Liu, W.; Kashkooli, A. G.; Xiao, X.; Cai, M.; Chen, Z., Silicon-Based Anodes for Lithium-Ion Batteries: From Fundamentals to Practical Applications. Small 2018, 14 (8), 1702737.Chae, S.; Ko, M.; Kim, K.; Ahn, K.; Cho, J., Confronting Issues of the Practical Implementation of Si Anode in High-Energy Lithium-Ion Batteries. Joule 2017, 1 (1), 47-60.Philippe, B.; Dedryvere, R.; Allouche, J.; Lindgren, F.; Gorgoi, M.; Rensmo, H.; Gonbeau, D.; Edstrom, K., Nanosilicon Electrodes for Lithium-Ion Batteries: Interfacial Mechanisms Studied by Hard and Soft X-ray Photoelectron Spectroscopy. Chem. Mater. 2012, 24 (6), 1107-1115.Xu, K., Nonaqueous Liquid Electrolytes for Lithium-Based Rechargeable Batteries. Chem. Rev. 2004, 104 (10), 4303-4417.Burns, J. C.; Sinha, N. N.; Jain, G.; Ye, H.; VanElzen, C. M.; Scott, E.; Xiao, A.; Lamanna, W. M.; Dahn, J. R., The impact of intentionally added water to the electrolyte of Li-ion cells. I. Cells with graphite negative electrodes. J. Electrochem. Soc. 2013, 160 (11), A2281-A2287.Burns, J. C.; Sinha, N. N.; Jain, G.; Ye, H.; Van Elzen, C. M.; Scott, E.; Xiao, A.; Lamanna, W. M.; Dahn, J. R., The impact of intentionally added water to the electrolyte of Li-ion cells: II. Cells with lithium titanate negative electrodes. J. Electrochem. Soc. 2014, 161 (3), A247-A255.