Search

A reliable heating device coupled with a FTIR gas analyzer has been tailored with the aim of evaluating the role of state-of-the-art lithium-ion battery components and environmental conditions on thermal and toxic hazards. Here, we demonstrate its effectiveness in accurately assessing the role of fully charged 0.6 Ah pouch cells confinement, electrolyte composition and separator coating on heat release and toxic gas generation-related risks. The fire safety international standards developed by the ISO TC92 SC3 subcommittee were used to determine the asphyxiant and irritant gases toxicity. Cells tighting confinement proves to be a very efficient way to diminish and delay (from 180 to 245 °C) the thermal runaway phenomenon occurrence and relating toxic gas release. Vinylene carbonate as electrolyte additive is able to shift (+20 °C) the onset temperature, while substitution of 1/3 M LiPF6 by LiFSI does not modify the thermal behavior, nor the toxic risks. The coating of a tri-layer separator influences the irritant gas toxicity related risk, by decreasing fluorinated components release. This study highlights that some improvements regarding LIB safety can be achieved through appropriate component selection and cells integration design at a module/pack level.

Here two chemical reduction pathways to synthesize the vinylene carbonate (VC) and poly(VC) reduction products are investigated, with the precise aim of further deciphering the lithium-ion battery solid electrolyte interphase (SEI) layer composition and the associated reduction mechanisms. The liquid synthesis pathway offers the opportunity of varying the concentration of Li-4,4′-Di-tert-butylbiphenyl reducing agent, whereas the dry synthesis pathway by ball milling allows to solve issues related to solvent-induced side reactions and washing procedure. As a result, the two syntheses do not unveil the same reduction mechanisms, favouring either carboxylate or carbonate salts as the major end product. The latter pathway is very efficient in terms of providing SEI-layers products resulting in well-defined IR spectra and comparisons with simulated spectra enable us to obtain IR fingerprints of the Li di-vinylene di-carbonate (LDVD) compound. Taken together the synthesis procedures provide information on conditions favouring radical polymerization and further poly(VC) reduction into Li2CO3 and polyacetylene. Overall, this chemical simulation of SEI-layers formation assists in a proper characterization of the SEI-layers created on graphite surfaces by their IR spectra showing that Li2CO3, LDVD and poly(VC) are all present in different proportions dependent on the VC content in the electrolyte.

International audience; The risk of thermal runaway is, for Li-ion batteries, a critical issue for large-scale applications. This results in manufacturers and researchers placing great emphasis on minimizing the heat generation and thereby mitigating safety-related risks through the search for suitable materials or additives. To this end an in-depth stepwise investigation has been undertaken to provide a better understanding of the exothermic processes that take place at the negative electrode/electrolyte interface as well as an increased visibility of the role of the state-of-the-art electrode binders, additives and lithium salt by means of the classical DSC technique.A reliable experimental set up helped quantify the beneficial or harmful contribution of binder polymers to the exothermic behavior of the CMC/SBR containing graphite electrode film in contact with 1 M LiPF6 in EC:DMC:EMC (1:1:1 v/v/v) electrolyte. Further, the role of the VC, FEC and VEC electrolyte additives (2 wt.%) in reinforcing the protective SEI layer towards thermally induced electrolyte reduction is discussed in the light of infrared spectroscopy and transmission electron microscopy analyzes results.Moreover, after a preliminary corrosion study of LiPF6/LiFSI mixtures, we showed that the 0.66/0.33 molar composition can be used in commercial NMC-based LiBs with a positive effect on the thermal runaway.

International audience; The effect of the replacement of the carbonyl oxygen in VEC additive by =C(CN)2 in the analogous dicyano ketene vinyl ethylene acetal (DCKVEA) on the electrochemical reduction profile is significant. Yet, the additives were proven, through IR spectroscopy supported by DFT computations, by applying EELS techniques and performing synthesis of a reduction product, to reduce in a similar way. Interestingly, the reduction-induced capacities were found to be quite different and can be explained either by the different properties of the SEI, from lithium carbonate and its malononitrile homologue, or by thedifferent abilities of the two additives to solvate Li+.

In recent years, greener transportation has become of major interest to limit air pollution and global warming. For this purpose, Li-ion batteries (LIBs) are considered as the most promising power source for electric vehicles (EV) and hybrid electric vehicles (HEV) due to their high energy density. The use of high-energy multi-cell battery packs imposes ever more stringent requirements on LIBs in term of safety and long-term cyclability. The formation of an effective SEI passivation layer at the negative electrode / electrolyte interface was found to be of paramount importance in order to enable LIB long-life cycling and controlling the threshold for thermal runaway. This is why SEI-forming additives have been used in the electrolyte to reinforce these protective properties, with the most common additives being vinylene carbonate (VC) (1), fluoroethylene carbonate (FEC) (2), and vinyl ethylene carbonate (VEC) (3). To our knowledge, the modification of EC and PC on the carbonyl group, rather than on the alkylene bridge as for VC, VEC, or FEC, has not previously been attempted.(4,5) It is known that the =C(CN)2 group is an “oxygen equivalent” being even slightly more electronegative than O itself and extending considerably the conjugation. Hence, we hypothesized that modified EC or PC, with C=C(CN)2 replacing the carbonyl groups, could lead to a more facile reduction at higher potentials and a stable SEI. The dicyanoketene propylene (and ethylene) acetal, DCKPA (DCKEA), have both been synthesized according to a simple procedure.(6) The reduction process has been investigated for DCKPA by means of GC/MS and IR analysis and the efficiency as a SEI-reinforcing additive demonstrated by the analysis of the soluble products using liquid GC/MS. The cycling tests using a pouch cell configuration at both 20 and 45°C were realized with only 0.5 wt.% of additive in the electrolyte and resulted in higher capacity retention. Moreover, a post-mortem analysis by DSC revealed an improvement in term of safety due to an improved lithiated graphite/electrolyte interface. [1] H.-H. Lee, Y.-Y. Wang, C.-C. Wan, M.-H. Yang, H.-C. Wu, D.-T. Shieh, Journal of Applied Electrochemistry. 35 (2005) 615–623. [2] I.A. Profatilova, S.-S. Kim, N.-S. Choi, Electrochimica Acta. 54 (2009) 4445–4450. [3] Y. Hu, W. Kong, H. Li, X. Huang, L. Chen, Electrochemistry Communications. 6 (2004) 126–131. [4] C. Forestier, P. Jankowski, L. Coser, G. Gachot, L. Sannier, P. Johansson, et al., Journal of Power Sources. 303 (2016) 1–9. [5] Pending patent [6] W.J. Middleton, V.A. Engelhardt, Journal of the American Chemical Society. 80 (1958) 2788–2795. Figure 1: Results of cycling tests at 45°C without and with 0.5 wt.% of DCKPA using a pouch cell configuration, and the post mortem DSC / liquid GC/MS analyses. Figure 1

Gel polymer electrolyte (GPE) membranes based on two polymers, a copolymer of poly(vinylidene fluoride–hexafluoropropylene) (PVdF–HFP), and the poly(ethylene oxide) (PEO), together with a plasticizer, the dibutylphtalate (DBP), were elaborated in two ways. Firstly, the polymers and the plasticizer were mixed together to obtain a single membrane. Secondly, a bi-layer separator membrane was made by adjunction, through lamination, of a DBP plasticized PVdF–HFP film and a homemade DBP–PEO thin film. We report here a protocol based on a simple galvanostatic polarization of Li/GPE/Li symmetric cells as a way to rapidly screen new viable membranes. Such a procedure enables to quickly discriminate separators by leading experiments that do no not exceed 1 week compared to hundreds of days needed with classical batteries. The validity of such an approach was confirmed by investigating the performances of the membranes in Li/GPE/Li4Ti5O12 flat battery configuration. Besides, through this study we also highlighted the role of the macroscopic PEO–PVdF interface toward dendrite of bi-layered separator, while a single blended membrane does not seem to be suitable for a practical use. Post-mortem pictures and SEM investigation have confirmed this result.

We report on experimental and theoretical studies of dendritic growth in Li/polymer/Li symmetric cells. Potential evolution with time, impedance and in situ microscopy experiments enable to characterise the onset and evolution of dendrites. In particular we observe that dendrites may burn when a high enough current goes through them, a thermo-fusible effect predicted in a previous paper and confirmed by SEM experiments. We present a calculation that gives a quantitative description of this effect: our results enable to understand a series of experimental data published in the literature concerning impedance variations observed while cycling lithium-polymer cells.

A new effective Gel Polymer Electrolyte membrane based on two polymers, the polyethylene oxide (PEO), a poly(vinylidene fluoride-hexafluoropropylene) (PVdF-HFP) copolymer and a plasticizer, the dibutylphtalate (DBP), was realized. This separator membrane was made by adjunction, through lamination, of an industrially made DBP/PVdF-HFP film and a homemade DBP/PEO thin film. Once the plasticizer was removed and the separator gelled by the electrolyte, the PEO enables the formation of a good interface with the lithium while the PVdF-HFP film brings the mechanical strength to the membrane. The electrochemical behavior of lithium batteries based on this bi-layer separator was investigated versus temperature, cycling potential and cycling rate. Owing to the promising results obtained with laboratory cells, a 1 Ah prototype was successfully assembled, and its cycling and rate performances were reported.

Polymer dispersed liquid crystals (PDLC) were obtained by polymerization-induced phase separation of the nematic E7 liquid crystal in a reactive diepoxide–diamine thermosetting matrix prepared by polycondensation (in two thermal steps). The evolution of some important electrooptical characteristics (transmittance in the off-state, threshold voltage, maximum transmittance of the on-state, transmittance at maximum applied voltage) were studied at three different reaction conversions of the polymer matrix beyond the gel point. The electrooptical curves greatly depended on the reaction conversion, especially beyond the gel point of the polymer matrix. It was found that the transmission in both the off-state and the on-state decreased with the extent of cure. In addition, threshold and saturation voltages increased with the reaction conversion. The electrooptical curve showed unusual behavior at approximately 60% of reaction conversion (i.e., near the gel conversion). At higher conversions, the expected normal mode was recovered. We discussed these electrooptical characteristics in the light of the very significant evolution of some properties of the crosslinked polymer matrix (glass-transition temperatures, concentration of crosslink points, molar mass, and weight fraction of the residual uncrosslinked oligomers), whereas the droplet morphology did not exhibit dramatic changes. © 2004 Wiley Periodicals, Inc. J Appl Polym Sci 92: 2621–2628, 2004

High voltage spinel oxides with composition LiMn 2 − x M x O 4 (M, a transition metal element) have remarkable properties such as high potential, high energy density and high rate capability. We believe that these positive electrode materials could replace the widespread commercial layered nickel cobalt oxides in some applications. The present assessment highlights electrochemical performance of optimized LiNi 0.5 Mn 1.5 O 4 and substituted counterparts, all having a spinel structure (cubic close-packed oxygen array) similar to the relative LiMn 2 O 4 . To fully emphasize the benefit from high potential spinel oxides, tests have been performed versus lithium metal, Li 4 Ti 5 O 12 and graphite, using various electrode loadings (0.3–4.5 mAh cm −2 ) and cycling rates (from C/20 to 60C rate). Steady capacity retention (130–140 mAh g −1 for nearly 500 cycles) and flat voltage (4.7 V vs. Li + /Li) have been obtained at C/5 rate at room temperature. Effect of cycling at high temperature has been shown to be less critical than for LiMn 2 O 4 . High voltage spinel oxides still sustain 100 mAh g −1 and over after 400 cycles at 55 °C at 1C rate. Rate capability is also excellent, with only 4% loss of capacity when comparing C/8 and 8C rates (thin electrodes).

Although widely used, the most promising Li-based technologies still suffer from a lack of suitable electrodes. There is therefore a need to seek new materials concepts to satisfy the increasing demands for energy storage worldwide. Here we report on a new layered electrode material, Cu(2.33)V(4)O(11), which shows a sustainable reversible capacity of 270 mA h g(-1) at a voltage of about 2.7 V, and electrochemically reacts with Li in an unusual and spectacular way. The reaction entails a reversible Li-driven displacement process leading to the growth and disappearance of Cu dendrites with a concomitant reversible decomposition and recrystallization of the initial electrode material. We show from structural considerations that the uniqueness of Cu(2.33)V(4)O(11) is rooted in the peculiar flexibility of the stacked [V(4)O(11)](n) layers, which is due to the presence of pivot oxygen atoms. Fully reversible displacement reactions could provide a new direction for developing an alternative class of higher energy density Li storage electrodes.

In Li-ion batteries, a layer called Solid Electrolyte Interphase (SEI) forms at the interface between electrodes and electrolyte. The protective role of this layer is of paramount importance for aging and safety issues, as it prevents any further destruction of electrode or electrolyte and thus ensures the good cycle life. In the past two decades, many research efforts have been dedicated to SEI layer characterization at the interface between the negative electrode and the electrolyte, but so far the relationship between its composition/structure and protective property remains unclear. In this study, we propose a direct and practical method to monitor the evolution of this interphase property during cycle life. The principle of this method is to analyze the chemical evolution of the electrolyte upon cycling through liquid GC/MS. The technique gives access to the presence of ethylene glycol bis-(alkyl carbonate), hereafter ethylene bis-carbonate (ROCO2CH2CH2OCO2R, with R = alkyl, reported as 11 1 or alkyl dicarbonates1–3). These soluble compounds are formed at the very beginning of battery cycle life; linear carbonate such as DMC, DEC or EMC gives a lithium alkoxide (ROLi) after electrochemical reduction. This compound can be partly dissolved in electrolyte and initiate chemical degradation of electrolyte through nucleophilic attack, thus yielding the ethylene bis-carbonate compounds. Their formation is responsible for the early change in electrolyte composition, and emphasizes the degradation by forming oligomers4. Ethylene bis-carbonate formation can be one of the reasons for battery performance decay, since it reduces the ionic conductivity of electrolytes. A thorough study on the generation profile of ethylene bis-carbonates shows that it depends on potential and on temperature. With our in-depth knowledge on the influence of the chemical nature of linear carbonates, we clearly show that the soluble compounds are probes for SEI protective property evolution. From this study, we discern several major advantages of this approach using liquid GC/MS: a) Easy and practical to apply, it does not require any sample washing. b) Quantification of the electrolyte degradation and estimation of the amount of electrolyte consumed for lithium alkoxide generation. So far, direct study of SEI composition and morphology through XPS, TOF-SIMS and TEM could give information on the SEI thickness, but would not give any information on the amount of electrolyte that was consumed to form each component. c) Indicator of SEI dissolution, especially when more soluble compounds are detected while electrolyte reduction is not possible. Monitoring the ethylene bis-carbonate generation profile can give information on SEI layer modifications upon first charge/discharge as well as aging. d) Probe for SEI-layer protective quality, especially the ethylene bis-carbonate compound. We successfully tested the efficiency of SEI-reinforcing additives such as VC, FEC, VEC and 1,3-PS and this technique is currently applied to test other new additives on Li and Na-ion batteries. This technique can be applied to all energy storage systems using carbonate-based electrolyte. Moreover, it gives complementary information to the direct analysis of SEI layer through XPS, TOF-SIMS or FTIR. Combining the two of them helps to better understand the relationship between morphology/composition and protective property. 1. Gachot, G. et al. Deciphering the multi-step degradation mechanisms of carbonate-based electrolyte in Li batteries. J. Power Sources 178,409–421 (2008). 2. Sasaki, T., Abe, T., Iriyama, Y., Inaba, M. & Ogumi, Z. Formation mechanism of alkyl dicarbonates in Li-ion cells. J. Power Sources 150,208–215 (2005). 3. Sasaki, T. et al. Effect of an Alkyl Dicarbonate on Li-Ion Cell Performance. J. Electrochem. Soc. 152,A1963–A1968 (2005). 4. Gachot, G. et al. Gas Chromatography/Mass Spectrometry As a Suitable Tool for the Li-Ion Battery Electrolyte Degradation Mechanisms Study. Anal. Chem. 83, 478–485 (2011).

not Available.

Gel polymer electrolyte ~GPE! membranes based on two polymers, the polyethylene oxide ~PEO! and a copolymer of polyvinylidene fluoride-hexafluoropropylene ~PVdF-HFP!, and a plasticizer, the dibutylphthalate ~DBP!, were elaborated in two ways. First, the polymers and the plasticizer were mixed together to obtain a single membrane. Second, a bilayer separator membrane was made by adjunction, through lamination, of a DBP plasticized PVdF-HFP film and a homemade DBP-PEO thin film. The physicochemical properties of the gels were analyzed. AC impedance spectroscopy was carried out on symmetric Li/GPE/Li cells using either the single layer or bilayer membrane as a function of aging ~isothermal at 20 and 70°C!, temperature ~240 to 70°C!, and finally, galvanostatic cell polarization. Both GPE membranes exhibit high ionic conductivities, but the most spectacular result was the measured decrease in the interface resistance, indicative of a deep modification of the interface Li/GPE when the cells were polarized. Aside from having a good interface with the Li metal electrode, such membranes were also shown to form good interfaces with the cathode because assembled Li/GPE/Li4Ti5O12 flat cells were able to sustain, at room temperature, more than

We report live observations of lithium polymer batteries upon cycling within the microscope antechamber by means of a scanning electron microscope, equipped with a transfer system that avoids sample air exposure. The well-established direct correlation between current density and dendrite formation is confirmed, and the formation of a mossy or dendritic interface was evidenced to be at the origin of the delaminating between the substrate, where the lithium is plated, and the polymer. These experiments enable a better understanding of the dendrite formation mechanism, like the important finding that they grow tipwise as well as sidewise. © 2002 The Electrochemical Society. All rights reserved.

Polymer dispersed liquid crystal (PDLC) films can be switched electrically from a light scattering off-state to a highly transparent on-state. Mixtures including different amounts of a high molecular weight copolymer and the eutectic nematic low molecular weight liquid crystal E7 were prepared in acetone which acts as a common organic solvent. PDLC films were obtained by a solvent induced phase separation (SIPS) initiated by evaporation of the solvent. The optical transmission properties of the obtained PDLC films were investigated as a function of composition, film thickness, and amplitude of the applied AC voltage. The electro-optical curves depend on the liquid crystal content and are highly reproducible. The particular properties of the copolymer allow to obtain a strong physical polymer network which in turn leads to well defined films compared to results obtained for other linear polymer/E7 systems.