Lithium Sulfur Superbattery Exploitating Nanotechnology
SAPIENZA INNOVAZIONE CONSORZIO
Viale Regina Elena 295 Pal Uff Complesso Regina Elena
€ 210 400
Stephen Trueman (Mr.)
Sort by EU Contribution
UNIVERSITA DEGLI STUDI DI ROMA LA SAPIENZA
€ 399 645
UNIVERSITA DEGLI STUDI GABRIELE D'ANNUNZIO DI CHIETI-PESCARA
€ 251 957
CHALMERS TEKNISKA HOEGSKOLA AB
€ 288 412
WESTFAELISCHE WILHELMS-UNIVERSITAET MUENSTER
€ 37 523,66
ZENTRUM FUR SONNENENERGIE- UND WASSERSTOFF-FORSCHUNG BADEN-WURTTEMBERG
€ 369 652
€ 243 350
CELAYA,EMPARANZA Y GALDOS INTERNACIONAL, S.A.
€ 98 780
INDUSTRY-UNIVERSITY COOPERATION FOUNDATION OF HANYANG UNIVERSITY
DEUTSCHES ZENTRUM FUER LUFT - UND RAUMFAHRT EV
€ 136 338
STENA RECYCLING INTERNATIONAL AB
€ 23 034
JOHNSON MATTHEY PLC
€ 204 181,50
ROCKWOOD LITHIUM GMBH
€ 13 693,50
KARLSRUHER INSTITUT FUER TECHNOLOGIE
€ 302 973,34
Grant agreement ID: 314282
1 September 2012
31 August 2015
€ 3 744 684,13
€ 2 579 940
SAPIENZA INNOVAZIONE CONSORZIO
Final Report Summary - LISSEN (Lithium Sulfur Superbattery Exploitating Nanotechnology)
The LISSEN project achieved a proof of concept battery, which can be scaled-up to industrial needs with:
• Improved performance (improved energy density, fast charging features and cycle life improvement through e.g. the electrode and electrolyte nano structures)
• Eco-design (recyclability and life cycle sustainability)
• Use of cost-effective electrode and electrolyte materials
More in particular, the main activities were focused on:
1) The elaboration of the specifications for rechargeable Lithium/Sulfur battery cells
2) The identification of liquid electrolyte mixtures based on the use of safe and highly stable ionic liquid, which show good ion conduction and high electrochemical stability
3) The realization of nanostructured gel polymer electrolyte membrane acting as an efficient separator in Li/S battery system
4) The development of a coupled full-cell model for Si/Sulfur battery, able to predict the discharge behavior, ion transport and solid diffusion under different operating conditions (discharge C-rates; electrodes thickness, etc.)
5) The scaled up preparation of both cathode and anode materials for larger cells realization. Scale up to 100 g. of materials preparation at kilo-scale has been proved
6) Testing of the advanced active materials for anode and cathode, as well as optimized electrolytes in full pouch cell geometry. Since the Si/C-S/C cells lack of a lithium source, a methodology to incorporate lithium into the cell and to mitigate the irreversible capacity loss of the Si/C anode is developed
7) Assembly and electrochemical tests of pouch-cells (max 0.1 Ah) using partially lithiated Si/C (anode) and partially lithiated S/C (cathode). The final cells consist of two anode and two cathode layers stacked together for a total area of 52 cm2 for the former and 48 cm2 for the latter, using liquid electrolyte mixtures
8) Realization and electrochemical tests of Z-folded pouch cells (max 0.5 Ah) with lithium metal anode and Sulphur/Carbon cathode. Z-folded pouch cells are assembled by stacking (up to 5 units) two lithium metal sheets interposed between two Sulphur-Carbon cathodes, using liquid electrolyte mixtures
9) Realization of pouch cells in a polymer configuration using the polymer electrolyte developed within the project
10) Definition of the recycling process, mainly focusing on combined hydro chemical – mechanical- thermal separation steps. The results enhance the knowledge regarding wet mechanical separation steps, the use of thermal method to promote subsequent separation steps and the use of hydrometallurgical processes to recycle important raw material
Project Context and Objectives:
LISSEN aimed at the identification and development of nanostructured electrode and electrolyte materials to promote the practical implementation of the very high-energy lithium-sulfur battery. In particular, the project was directed to the definition and test of a new, lithium metal-free battery configuration based on the use of a Si/C composite as anode and a nanostructured S/C composite as cathode. It is expected that this battery offer an energy density at least three times higher than that available from the present lithium battery technology, a comparatively long cycle life, a much lower cost (replacement of cobalt-based with a sulfur-based cathode) and a high safety degree (no use of lithium metal). All the necessary steps for reaching this goal were considered, starting from material synthesis and characterization, exploiting nanotechnology for improving rate capability and fast charging, the fabrication and test of large-scale prototypes and the completion of the life cycle by setting the conditions for the recycling process. The consortium included a number of academic laboratories, all with worldwide-recognized experience in the lithium battery field, whose task was that of defining the most appropriate electrode and electrolyte nanostructures. The project also made use of battery modeling techniques to aid in the optimization of materials as well as of electrode and cell design. Large research laboratories, having advanced and modern battery producing machineries were also involved in the preparation and test of middle size battery prototypes. Finally, chemical industries assured the necessary materials scaling-up while the fabrication and test of prototype batteries were eventually assigned to a large research laboratory; particular attention was devoted to definition and practical demonstration of its most appropriate recycling process.
More in particular, during the project duration, LISSEN developed approaches to address the following technological improvements:
1) Solubility issue: use of a selected multi-salt, ionic liquid–based electrolyte
2) Kinetic issue: use of a new morphology, homogeneous dispersion of the sulphur particles in hard carbon spherules: (HCS-S)
3) Safety issue: Replacement of the reactive Li metal anode with a Si-C composite to be utilized either pristine (for Li2S cathode) or lithiated (for HCS-C cathode)
4) Design of functional small scale Lithium sulfur full-cells utilizing advanced active materials for anode and cathode as well as optimized polymer electrolytes
5) Developing processes for cell manufacturing
6) Scaling up to large battery prototype realization
1 - Modelling activities
Modelling activities can be divided in two parts. The first part in fact was the definition of the lithium sulphur battery performance requirements while the second part was dedicated to the development of a well-founded scale-bridging model and subsequently a simulation of the performance and lifetime of sulphur-based battery cells. With this modelling and simulation approach it was possible to optimize cell chemistry and electrodes design. The model is based on elementary kinetics of the sulphur-redox-chemistry (for example: cathode side sulphur reduction, polysulfide formation, anode reactions, electrode-electrolyte interface). Another important part of the modelling and simulation approach is the multi-phases behaviour of the nano-scale composite of both electrodes, Li-S for the positive electrode and Li-Sn or Li-Si for the negative electrode.
According to the above, the part of the modelling approach is focussed on the mastering of the intimate morphology of the electrode material nanostructure (e.g. hollow carbon sphere nano composite structures). The model is based on representative micro structural features. Modelling and simulation take also into account the multi-component transport within the electrolyte (shuttle mechanism).
The technical specifications for Lithium-Sulphur battery performance requirements were defined collecting inputs from Original Equipment Manufacturers (OEM) on power/energy requirements and driving profiles. The set of parameters (capacity, power, life) for the cells were proposed considering specific battery architecture (cell arrangement), the innovative project technology and the automotive industry requirements.
Objective of the LISSEN project was the development of nano-structured cell components (anode and cathode) to be used in a high-energy scaled-up lithium/Sulphur cell prototype for automotive applications. The role of the simulation part was to analyse and model the complete system developed in the consortium considering 1) kinetic mechanisms in the electrodes and at the electrode-electrolyte interface (anodic and cathodic reactions, polysulphides formation and diffusion, etc.); 2) changes of the nano-scale composite material of both electrodes; 3) the multicomponent transport within the electrolyte. With this model it is possible to optimize electrode and cell design, as well as to gain insight into performance and lifetime of the optimized system. The micro-structural model from images of electrode materials was reproduced using the commercially available software GeoDict (Math2Market). By using that software was possible to obtain 3D images representing the particles reorganization in the electrodes and derive from the model reconstruction important chemical/physical parameters. Starting with the micro-structural model of the existing electrodes it was also possible to vary the particles and electrodes characteristics (size, density, morphology, composition, etc.) and simulate chemical/physical parameters of the new electrode's structure. The parameters were then implemented in the kinetic modelling.
In the early stage of the project the activities focused on the modelling of S/C cathode materials. As anode we model a Li metal foil. This is a configuration, which is commonly used in the research on Li-S batteries. The basic model is implemented in the software DENIS, which is a standard tool, developed for the 1D modelling of battery and fuel cell systems. It includes a description of multi-component transport of polysulphides in the liquid phase, the formation and dissolution of solid discharge products, as well as a detailed mechanism of sulphur-redox-chemistry. The software allows a detailed study of relevant processes by simulating electrochemical measurement techniques like impedance spectroscopy and discharge curves. This is an important step to ensure the exchange with the experimental partners. The parameterization of the model (transport coefficients, thermodynamic and kinetic parameters) is difficult due to the large number of unknowns, which have to be determined independently. In order to reduce the number of parameters a reduced reaction mechanism was derived. Still, the qualitative agreement to experimental data published in the literature is favourable. The reduced model was used to investigate the polysulfide shuttle, which is responsible for the capacity fading and short cycle life of Li-S batteries. The simulations give insights on the mechanism of the polysulfide shuttle and allow making qualitative predictions on the cycle life of the battery cell. At this stage the models were transferred to the software Matlab that is a widely used tool in the modelling and simulation community. This step was done in order to facilitate the cooperation between the project partners and also the development of more advanced models with a more complicated structure. Finally, the model was parameterized based on the measurement data provided by the experimental partners and the results originating from the microstructural simulations. Overall, we can report a satisfactory agreement between simulation results and the electrochemical measurements.
In the last stage of the project we also developed a model of a Si/C composite anode. The model takes into account the intercalation in both Si and graphite particles. Moreover, we model the volume change of Si particles during lithiation/delithiation, which is termed as a “breathing electrode”. Simulation results agree fairly well the experimental data provided by the experimental partners.
In a final step we combined the two electrode models of the S/C cathode and Si/C anode. This configuration called the “LISSEN cell” since it represents the final battery produced by the consortium. The model allows making predictions of the final cell based on the behaviour of the two electrodes before the actual physical cell is assembled. Therefore, the model can be used as a design and up scaling tool.
The LISSEN cell model developed was employed to do parameter variations and sensitivity analyses, e.g. variation of electrode thickness, composition, etc. The results can be used to infer manufacturing guidelines for future electrodes. Also, the model facilitates a subsequent up scaling of battery capacity. Moreover, we developed a detailed 1+1D continuum model of S/C composite cathodes, which is based on an ideal structure where all polysulphides are confined within meso- and micro-porous carbon particles. This is an important contribution since some of the most remarkable improvements in the cycle life of Li-S batteries could be achieved by nano-structuring of the electrode material. A lot of the recent work focussed on the active retention of polysulphides by the electrode host materials, mostly carbon. This approach has the advantage that it additionally ensures a good electric contact of the solid sulphur. We developed a detailed continuum model of a S/C composite cathode consisting of meso- and micro-porous particles. First, a single-particle model is derived which employs a lumped kinetic mechanism presented in the previous section. The particle model is coupled to a Li-ion battery model describing the macroscopic transport on cell level. An important assumption of our model is that dissolved polysulphides are fully confined. This corresponds to the ideal case which experimentalists are aiming for in their effort to find suitable nano-structures. The model is parameterized and validated based on experimental data from the literature. Simulations of a single particle show that for high and non-uniform sulphur loadings the specific capacity effectively decreases. Most interestingly, we identified an additional over potential caused by the transport of Li+ ions against a concentration gradient into the particle pore space. This also demonstrates the strong forces, which will drive polysulphides out of the particle if they are not perfectly retained. We investigate this issue in degradation studies, which show that the cycle life can be improved by using solvent systems with low polysulfide solubility or high salt concentration. However, simulations of the full battery cell show a detrimental effect of high salt concentrations on battery performance due to a lower ionic conductivity. Therefore, we conclude from our simulations that a well-balanced electrolyte system with a low solubility of polysulphides in combination with a sophisticated nano-structuring of the carbon host seems to be the most promising approach for an improved cycle life of Li-S batteries. The simulations demonstrate that the improved model can be utilized to study different cell designs and transport phenomena, which could not be observed in previous modelling work. Therefore, valuable insights on the operation and manufacturing of Li-S batteries with complex nano-structured cathodes can be obtained.
2 - Optimization of electrolytes
A major hurdle still hindering the practical development of the lithium-sulphur battery is the high solubility in the conventional organic--based electrolyte of the polysulphides Li2Sx (1≤x≤8) that form as intermediates during both charge and discharge processes. This high solubility results in a loss of active mass, which is reflected in a low utilization of the sulphur cathode and in a severe capacity decay upon cycling. The dissolved polysulfide anions, by migration through the electrolyte, may reach the lithium metal anode, where they react to form insoluble products on its surface; this process also negatively impacts the battery operation. Various strategies were attempted by the LISSEN partners (UNIROMA, UNICH, CHALMERS and KIT) to address the solubility issue have been explored, including the use of glassy-type electrolyte, the design of modified organic liquid electrolytes, and of polymer electrolytes.
Safety issues, such as liquid electrolyte leakages and consequent flammability, are presently seriously limiting the diffusion of the lithium battery, in particular in the electric vehicles field, which requires severe targets. Hence, solid electrolyte appeared the best choice in order to efficiently overcome this issue. Structural, spectroscopic and electrochemical study indicated the glass type, ionic conductor Li10GeP2S12 as promising electrolyte for solid-state lithium batteries. In the LISSEN project an improvement in the synthesis pathway has been developed in order to refine its structure. High conductivity values have been measured, this confirming that the Li10GeP2S12 samples are very promising materials for application as electrolyte in safer lithium batteries. The applicability of the Li10GeP2S12 has been verified by using it as solid electrolyte in a lithium/sulphur cell. The cell delivered a specific capacity of the order of 320 mAh g-1, which corresponds to a limited fraction of the overall capacity of a S-based electrode. However, under the operating conditions used for this test, the energy density delivered by the cell is about 640 Wh kg-1, evaluated on the basis of the capacity and the voltage values. This energy value, well comparable to those of the common lithium ion batteries presently offered by the market, and the very high safety content of the solid state battery configuration proposed in the project suggest that the solid electrolyte may be a valid candidate to be used for next generation lithium/sulphur batteries.
Optimization of electrolyte solution
In the LISSEN project different liquid electrolyte mixtures have been proposed and tested. The electrolytes solutions, based on PYR14TFSI ionic liquid combined with alkyl carbonates or ether solvents, with and without the addition of polysulphides, have been investigated. The IL-based, alkyl carbonate-added solutions showed satisfactory properties to be adopted as electrolyte in lithium-sulphur batteries. Good ion conduction and electrochemical stability have been demonstrated, as well as the role of the dissolved Li2S8 salt as both catholyte and buffer. Ether based solvents have been also considered as alternative to carbonate-based ones since they showed to be poorly effective for the successful utilization of the sulphur cathodes developed within the project. The optimal composition of the electrolytes was selected taking into account the compatibility with the LISSEN electrode materials and the performance of the cell. The thermal transitions of the electrolytes have been studied by DSC analyses. Comparing the ionic liquid rich solutions to the organic solvents, a much lower glass transition temperature is found for the latter. This is beneficial since it enlarges the temperature window of the liquid state and is usually also directly related to improved transport properties. The electrolyte solution containing only ionic liquid is clearly prone to crystallization, observed by the large crystallization peak following the glass transition upon heating. It is clear that in the mixed electrolyte solutions crystallization is efficiently suppressed. The addition of polysulphides to the electrolyte leads to a small increase in the glass transition temperature of the corresponding electrolyte mixtures. The ionic conductivity over a wide range of temperatures (70 to -70 °C) has been evaluated. All the samples followed non-Arrhenius behaviour. However some samples presented sudden decreases in conductivity. This change in conductivity has been attributed to a partial crystallization of the electrolyte, in agreement with the tendency for crystallization observed in the DSC experiments.
Overall, all the electrolytes showed a high conductivity around room temperature suitable for battery applications. The highest conductivity is found for the organic solvent rich samples. The addition of polysulphides is found to lead to a slight decrease in the conductivity, being this result in agreement with the slight increase in the glass transition temperature for the corresponding solutions.
Physico-chemical and electrochemical tests indicated that the ether-based electrolytes showed the best performance as neat solutions, being the addition of polysulphides beneficial for the capacity and the cyclability of Li/S cells. Galvanostatic cyclation tests have been carried out by using the selected electrolyte mixtures together with a carbon composite sulphur cathode developed in the LISSEN project. Cell performances of a standard electrolyte solution (TEGDME, 1M LiCF3SO3) have been also considered for comparison purpose. The data showed the strong improvement in the performance by the addition of polysulphides to the electrolyte solutions. This effect can be partly attributed to a buffering effect where the presence of polysulphides stabilizes the active material in the cathode.
One of the objectives of the LISSEN project was the optimization of the electrolyte materials with respect to performance, safety and adaption of synthesis procedure for up scaling. The electrolyte membrane is a key component of the LiS-battery proposed in this Project. It has to allow fast ion transport, mechanical rigidity, thermal and chemical stability and compatibility with the selected electrodes. The challenge of combining all these often-contradicting properties has been approached in the project by the development of nano-structured polymer gel-type membranes where the optimized electrolyte solution is entrapped in a polymer matrix. The work builds on two concepts previously developed by the partners based on i) electro-spun membranes and on ii) hot pressed membranes that demonstrated to be a key component of the Li/S battery. The addition of ceramic particles to the polymer matrix has also been explored in order to enhance both mechanical rigidity and potentially the retention of the electrolyte solution.
i) Electro spun polymer membranes
It has proved the possibility to realize by solution electro spinning and co-electro spinning, different electro spun mats formed by either PVdF or PEO nano fibres. The versatility of the electro spinning technology allows controlling polymer fibre morphology and fibre deposition pattern, in order to fabricate engineered nano fibrous porous membranes with a defined micro/nano-architecture, in terms of fibre size and fibre orientation. The membranes are formed by entangled fibres with nano metric diameter, which guarantee mechanical integrity joined to a very high specific surface area. It have been demonstrated that these highly porous mats are able to speed-up the swelling process by absorbing large quantities of selected electrolyte solutions at markedly lower temperatures. As a result, the entire assembly of the complete cell can be performed in dry condition and the formation of the polymeric gel electrolyte can be obtained in situ, at room temperature, by the addition of the selected electrolyte just before the cell activation step. DSC response of the gelled electro spun electrolytes showed that the melting transition of the polymer is highly influenced by the presence of the liquid component. Crystallinity of PEO (occurring around 60 °C) is strongly reduced, almost suppressed, upon gelation, giving rise to an amorphous, plasticized electrolyte system. With this respect, the nature of the electrolyte solution, with or without polysulphides, appears almost irrelevant. The main role is due to the ionic liquid (that is the major component of the liquid electrolytes), which interacts with the polymer chains, thus preventing their crystallization. Conductivity of gelled electro spun membranes was evaluated, assembling coin-type cells with stainless steel current collector electrodes where the swollen PEO membrane acts as electrolyte separator. High conductivities (σ T=25 °C = 2.8 × 10-3 S/cm) were achieved for all the gelled systems at the investigated temperatures. As expected, a temperature-activated transport mechanism is found, giving higher conductivity values at 50 °C with respect to 25 °C. All the findings above address the potentiality of the electro spun gelled electrolytes in practical lithium-sulphur batteries.
ii) Hot-pressed polymer membranes
The membranes were prepared by using poly(ethylene oxide) PEO, having a molecular mass Mw = 4M ~ 4x106 g/mol, lithium bis(trifluoromethylsulfonyl)imide LiTFSI and silicon dioxide (fumed SiO2, as filler. The membranes were prepared following a solvent-free procedure, resulting in very homogeneous, self-standing and semi-transparent membranes with geometrical surface area ranging from 30 to 50 cm2 and thickness from 100 to 150 μm. The swelling procedure, developed to realize the final gelled polymer electrolyte, has been optimized. It consists of embedding a fixed relative amount of selected liquid electrolyte solution in the membrane sample to finally reach a swelling value of ca. 250 % (computed as solution-to-membrane weights).
The different membranes were analysed by Raman spectroscopy that demonstrates the achievement of homogeneous polymer electrolytes, where the PEO is able to fully dissolve the Li-salt, regardless the presence of the inorganic silica filler. Further characterization, related to the final swelled systems, considered only the membrane without filler, being it chosen as the most easily scalable, representative sample. The ionic conductivity of the swollen membrane in the ether-based, selected solutions was determined, by using two blocking, stainless steel electrodes coin cells assembled in a controlled, argon-filled dry-box. Conductivity values from room temperature up to 70 °C have been evaluated. The polymer electrolytes showed interesting ion conductivity values and the presence of polysulfide ions seemed to affect positively the conduction properties in the membrane. Overall, the detected σ values appear suitable for battery applications in a wide temperature range.
The Li/S cell behaviour of the electrolyte systems was investigated by galvanostatic cycling tests. The coin cells used for the measurements were assembled by using the sulphur-carbon cathodes developed within LISSEN Project, with lithium foil. The tests were performed at a fixed current rate C/20, in voltage ranges of 1.5 – 3.0 V (vs. Li/Li+). The voltage profiles, shown by the Li/S cells exhibit the expected multi-plateau behaviour indicating the formation of different polysulfide ions (Li2Sx with 1 < x < 8) during discharge (reductive process) and charge (oxidative process). Capacity values achieved are of absolute interest, especially for a polymer electrolyte configuration. Stable values, levelling between 800 mAh g-1 and 900 mAh g-1 (per g of sulphur/carbon), are obtained, highlighting the beneficial effect of adding polysulphides to the electrolyte solution on the cycling performance.
3 - Optimization of electrodes
In order to select the more appropriate electrodes, both cathode and anode, that could fulfil the project goals, many different, possible candidates have been synthesized and deeply investigated. In particular, the project partners identified as cathodes, two different sulphur-loaded carbon-matrix and one type of composite material formed by mesoporous carbon coated Li2S particles, and as anodes, Si-C and Sn-C composite powder to be utilized. During the last project period, instead, the research activities have mainly been focused on Si-C and S-C composite powders, which were selected, respectively, as the most promising candidates anode and cathode for scaling-up. Moreover, fundamental, explorative research activities have been performed on Si-C, S-C and Li2S-C composite powders prepared by electro spinning.
Sulphur-loaded Mesoporous Hard-Carbon Spheres (S-MHCS) have been prepared and characterized. The hollow Mesoporous Hard-Carbon Spheres (MHCS) act both as reinforcement cage that entraps and holds the elemental sulphur and polysulphides during the charge and discharge operation of the cell, and, at the same time, as electronic conducting scaffold, which facilitates the access and release of electrons to and from the electrochemical triple point active reaction site. MHCS were prepared by a hydrothermal method using sucrose as carbon source. The electrochemical tests run on these powders showed unsatisfactory performances and further studies on these materials were discontinued.
Sulphur-loaded Porous Multi-Channel Carbon-Copper nanocomposites Electrospun Microtubes (S-PMC-Cu-EMT) have been prepared and characterized. The strategy pursued with this research line was to add to the functionality of S-MHCS those brought by metallic Cu particles. These particles, upon sulphur loading, transform in CuS, which, beside to contribute to the cell capacity, form metallic Cu and lithium polysulphides upon cycling. It was expected that the presence of nanometric Cu particles, in close proximity with both elemental sulphur and polysulphides, could contribute to the electronic conductivity of the cathode and to inhibiting polysulfide mobility due to the strong affinity between metallic Cu and sulphide anions. These innovative, functionalized powders have been prepared by electro spinning. The final sulphur content in the composite S-Cu-C powders, determined by TGA and Elemental analysis (EA) resulted to be of about 34 wt.%. The final S-Cu-C composite powders were characterized by TGA, XRD, SEM and EDX. The XRD and EDX analysis showed that the metallic Cu almost disappeared in the final product, transforming, as expected, in CuS. A presence of elemental S was also evidenced. The electrochemical characterization of cathodes prepared by utilizing these electro spun composite powders was performed by running Galvanostatic Cyclations (GC) and Cyclic Voltammetry (CV) measurements. The GC curves and the derived dQ/dE vs E curves showed the typical charge/discharge profiles and peaks associated to the electrochemical reduction and oxidation of sulphur and CuS.
Porous carbon nano fibres, to be impregnated with sulphur, were prepared by electro spinning. In order to do that, a DMF solution of the polymeric precursors (PAN:PMMA - 1:1 wt.) was prepared. The solution was electro spun in order to form fibrous mats. Porous, Multichannel, Carbon Electro spun Micro-tubes (PMCEMT) were obtained. Both the as prepared and after grinding PMCEMT were characterized by EA, TGA, SEM and TEM. EA revealed that carbon sample contain around 9 wt. % of nitrogen and some residual hydrogen and oxygen. SEM images showed the fibrous structure of the electro-spun mats before grinding. After grinding, micrometre size tubular particles were formed. TEM images revealed the porous nature of these latter carbon micro tubes, which can enhance their ability to sulphur entrapment and the electronic conductivity after sulphur loading. Finally, an apparatus to achieve under-vacuum sulphur-vapour impregnation of the porous carbon micro tubes was realized.
An innovative experimental procedure was devised to realize the very challenging task of prepare Li2S particles carbon coated. To do that it has been utilized a non-conventional co-electro spinning apparatus equipped with a concentric double needle. The final product was characterized by EA, TGA, XRD, SEM, TEM and electrochemical tests. XRD showed that amorphous carbon and Li2S formed the samples. SEM revealed the fibrous morphology of the mats and micro tubes formation after grinding. TEM made evident the porous nature of micro tubes and the entrapment of Li2S phase inside the tubular carbon matrix. The ground Li2S-C composite powders were utilized to prepare thin film electrodes. These electrodes were studied and characterized The GCs clearly showed Li2S electrochemical activity. In addition, as expected, mostly of the cycled capacity was irreversible, due to the formation of the carbon passivation layer. The electrodes showed a low reversible capacity, at relatively high rates. In conclusion, the possibility to prepare Li2S particles embedded in a carbon matrix has been demonstrated. The composite, nono-engineered powders showed electrochemical activity when cycled in a lithium cell. The low cycled capacity can be ascribed to the low Li2S/C wt. ratio and, presumably, to the low electronic conductivity of the carbon matrix.
Mesoporous Carbon Coated Li2S Composite Powders were prepared and studied. Large chunks of Li2S, prepared by Rockwood Lithium, were ball-milled a carbon black in different ratio, in order to prepare mesoporous, carbon-coated, Li2S composite powders. These powders were utilized to fabricate cathodes to characterize. The electrodes showed high initial discharge capacities, which slightly decreased after 10 cycles. The reproducibility of this system was rather difficult, with the Li2S/C 70:30 wt. ratio mixture performing better than the 50:50 one. Columbic efficiencies were quite low.
Ball-milled Sulphur-Carbon Composites were prepared and investigated. Johnson Matthey partner prepared ball-milled S-C composite powders by utilizing hard-core carbon spheres, which were subsequently impregnated with sulphur. SEM analysis showed a mixture of spherical and irregular carbon particles. The sulfuration process was carefully optimized. To conclude, ball-milled, sulphur-carbon composites were selected for further development and scaling-up activities, while the electro-spun electrodes were selected for more fundamental research studies. Three generations of, ball-milled, S/C composite powders have been prepared and characterized. The samples differed by the kind of carbon used, the milling conditions, the sulfuration process and the final sulphur loading The composite powders were utilized to prepare thin film electrodes. The S/C electrodes were characterized by GC and Electrochemical Impedance Spectroscopy (EIS). The first-generation of S/C composite samples showed a rapid decrease of the capacity and low columbic efficiency for all the samples considered, which could be attributed to the poor carbon coating of sulphur particles. The second-generation samples did show very poor performances, even if the sulphur content was higher than that of the first-generation samples. Any further study of these samples was discontinued. The third-generation samples showed a certain degree of loss of performances upon cycling. The more promising samples appeared to be SCJM2 and. Sample SCJM4 was subjected to a further thermal treatment in order to increase the degree of sulphur infiltration within the carbonaceous matrix. The sample showed an improvement of the cycled capacity, but its performances were still not satisfactory. In conclusion, Sample SCJM13 showed the best performances and was selected for the scaling-up activities.
Two generation of Si-C composite powders were prepared by ball-milling silicon nano powders mixed with carbon nano powders. The samples of the first generation were prepared by low energy ball milling, while those of the second one were prepared by high-energy. The final products were characterized by XRD, TGA, DTA and SEM analysis. XRD analysis showed the presence of only pure Si and C, with no SiOx peaks. TGA analysis confirmed a 3:1 C:Si wt. ratio. SEM analysis showed Si particles, with diameters ranging from 100 nm to 1 micrometre, well embedded with the carbon particles. These powders were utilized to prepare Si-C composite electrodes. The electrodes were tested by GC and PITT (Potentiostatic Intermittent Titration Technique). Both GC and PITT results showed that Si-C electrodes suffer from high irreversible capacity loss in the first cycles. Between all the first-generation samples, sample SiC-D showed the best performances. This sample was able to withstand extended cyclations with moderate-high reversible capacity at 0.2 mA/cm2. The second-generation samples were characterized in the same conditions as those of first generation and their performances were compared to the performances of sample SiC-B. It resulted that all the second-generation samples had poorer performances than that of SiC-B. The characterization of these samples was discontinued. All the first-generation samples were further treated by applying a high-energy ball-milling step. XRD analysis showed that the crystallite size was reduced from 67 1 nm to 25 1 nm after the treatment. Electrochemical tests run on these latter powders showed better performances in terms of cycled capacity and capacity fading. As a conclusion, it could be said that the electrochemical performances of Si-C composite sample SiC-B, belonging to the first generation materials, were the most promising one. Moreover, it was demonstrated that, by reducing the crystallite size of the powders, increased electrochemical performances could be obtained. First-generation sample SIC-B was selected for scaling-up activities. In addition, when Sol-A was utilized as electrolyte, the cell showed a high over potential, which could be attributed to the lower ionic conductivity of this solution. In conclusion, in both solutions, Si-C redox processes take place, but problems related the morphology and to the grain size not completely optimized still limit the full utilization of the active material.
Electro spun Si-C nano composites porous micro tubes were prepared in two different formulations with low Si:C and high Si:C wt. ratio, respectively. In both cases, Si particles embedded in multichannel hollow porous carbon micro tubes (Si-PMCMT) were obtained. The pyrolyzed nano-fibrous mats were manually ground and the resulting powders were characterized by TGA, SEM, TEM, XRD and Electrochemical tests. XRD analysis showed that only elemental Si and amorphous carbon were present in the composites. TGA revealed a Si content of about 15 wt.%. SEM and TEM analysis showed the nano fibrous structure of the Si-PMCMT powders with Si nanoparticles well embedded in the carbon micro tubes. The electrochemical characterization performed by GC and derivative GC showed the characteristics reversible Li-Si alloying and de-alloying processes. After an initial very high irreversible capacity, the electrodes were able to reversibly cycle approximately 400 mAhg-1 for more than 100 cycles. Derivative GC showed that the peaks related to Li-Si processes are well evident even after 100 cycles, confirming the promising electrochemical properties of electro spun, nanostructured Si-C composite powders.
Sn-C nano composite powders were prepared by utilizing the electro spinning technique. To do that, a precursor solution was prepared by dissolving polyacrylonitrile (PAN) and Polymethylmetacrylate (PMMA) in dimethylformamide (DMF) solvent. Tributylphenil-Tin was added to this solution and the resulting mixture was used to fabricate electro spun polymeric mats. The collected fibrous mats were pyrolyzed in a reducing Ar/H2 atmosphere to obtain Sn nanoparticles embedded in multi-channels hollow porous carbon micro tubes. The Sn-C final mats were ground in order to obtain a fine powder formed by nano-cylinders few hundreds nano meters wide and 1-2 micrometres long. A slurry obtained out of this powder was used to prepare thin film electrodes. GC was utilized for the electrochemical testing of the electrodes. In order to improve electrodes cyclability, the cycled capacity was limited at a maximum of 500 mAhg-1. During the first charge/discharge cycles most of the capacity was used for irreversible Li-C processes (SEI formation, etc.). As the cyclations proceed, the irreversible processes become less important and higher amounts of Li+ are reversibly exchanged. After the initial cycles, the cell is able to reversibly deliver the requested capacity of 500 mAhg-1 for 200 cycles at 1C with a Columbic efficiency close to unit. It is thus confirmed that the nanostructured carbon micro tubes can guarantee a stable Sn encapsulation and, in the cycling conditions adopted, contribute to buffer the volume variations of Li-Sn alloys during the initial ‘activation’ cycles. To conclude, the Sn-C sample studied showed promising behaviour, although, it needs further improvements aimed at reducing the particles grain size and improve the carbon coating. The Sn-C nano composite electro spun powders showed satisfactory electrochemical properties, but, the very high irreversible capacity, which dominates the first cyclations, hinders their utilization in a sulphur-based battery. For this reason further study was discontinued.
4 - Cell design
High capacity energy storage systems are needed for a large numbers of applications ranging from portable electronic devices to automotive. The project “LISSEN” focused its activities toward the development of electrochemical cells composed of Sulphur containing cathodes and high capacities anodes (Li, Si, Sn). The cell design work has been divided into three tasks aiming at: a) Designing of cell structures; b) Fabricating coin-cell laboratory prototypes; c) Developing processes for cell manufacturing. Active cathode and anode materials as well as alternative electrolytes have been provided by various partners (JM, UNIROMA1, UNICH, HANYANG). After a first period during which several materials were evaluated in terms of electrochemical performances and processability, the results indicated the Si-based material as the best anode candidate. For what concerns the cathode side the consortium agreed on the use of a Sulphur-Carbon composite (S-C) with a 50/50 weight ratio of the two components. A preliminary evaluation of the electrochemical performances of the selected materials was carried out using small lab scale T-shape and coin cells. After that single layer pouch cells having cathodes of 24 cm2 and anodes of 26 cm2 were assembled and their electrochemical performances evaluated.
Silicon negative electrodes have been produced starting from water-based slurries. Briefly, the active material (Si0.5-C0.5 provided by JM) and the binder (poly[acrylic acid]) (weight ratio 85:15) were dispersed in water and homogenized. The obtained slurries were then coated on copper foils. The layers were dried and electrodes were punched out. Thickness and loading of the electrodes were easily adjustable within a certain range considering that no particular issues concerning adhesion on the current collector were observed. Therefore very high loadings of 3-4 mg/cm2 (active material) could be achieved.
Our preliminary evaluation of the electrochemical performances showed that the Si-based electrodes could achieve gravimetric capacities of about 900 mAh/g (based on the Si-C composite mass). Tests carried out using small electrodes (1.13 cm2) showed very good electrochemical behavior in two different electrolyte systems:
1) 1M LiPF6 in ethylene carbonate (EC): dimethyl carbonate (DMC) (1:1 V/V), and
2) 1M LiTFSI + 0.25M LiNO3 in 1,3-dioxolane (Diol): dimethoxyethane (DME) (1:1 V/V).
Measurements with electrolyte #2 were carried out in view of full cell measurements where the usage of cathodes containing Sulphur impedes the utilization of electrolyte #1 (due to too high polysulphides dissolution). The characterization in half-cells showed that the Silicon negative electrodes could sustain over 70 cycles when the applied current density ranged between 50 and 500 mA/gSi-C. However decreased capacities were observed at the highest current density as expected. Of particular interest is the fact that the negative electrodes showed:
1) Comparatively low irreversible capacity loss at the 1st cycle (columbic efficiency: 0.65) and
2) Rather high columbic efficiency during the following cycles (>0.99)
which for Silicon-based electrodes represent interesting results. The galvanostatic curves did not show any particular plateau, but the potential slowly decreased and increased between the lower and upper cut-off during charge and discharge, respectively. Although the value of the irreversible capacity loss at the first cycle was found comparatively low for a Si-based electrode, still about 300 mAh/g needed to be taken into account. Therefore within the frame of the cell design work package, the ZSW also developed strategies for compensating such irreversibility. The post treatment of the anode layers with Stabilized Lithium Metal Powder (SLMP; FMC Lithium) showed in this respect very good results. Indeed it was possible to completely compensate the loss. Furthermore, pre-lithiated electrodes showed excellent electrochemical behavior especially when consider that in all cases thick electrodes having high areal capacities (3-3.5 mAh/cm2) were evaluated. It is however worth noting that after the pre-lithiation step electrodes need to be handled with care and in dry environments.
Sulphur-based cathodes were prepared starting from slurry based on an organic solvent (Anisole). Slurries were obtained by mixing the Sulphur-Carbon composite (S0.5-C0.5 provided by JM), conductive carbon and binder (Poly(ethylene oxide)) [8:2:2 ratio] in Anisole. The manufacturing of high loading sulfur cathodes was more problematic than that of the anodes. The adhesion on the current collector (Al) had to be enhanced by threating the pristine Al-foil with carbon paint before coating the active layer. Such a way made finally possible the development of cathodes with over 4 mg/cm2 loading of active material. The electrochemical performances of the Sulphur cathodes were first evaluated in half-cells (coin type). The electrolyte was 1M LiTFSI + 0.25M LiNO3 Diol:DME. Other electrolytes were also evaluated but poor compatibility with the cathodes fabricated according to the procedure described above was observed. The electrodes achieved a capacity of about 400 mAh/gS-C, which translate in about 800 mAh/g if the weight of the Sulphur only is considered. The galvanostatic plots showed two distinct plateaus at 2.25 and 2 V in full agreement with a typical galvanostatic charge/discharge process of a Sulphur electrode. In terms of cycling stability the cathodes had no issue in completing over 50 cycles with almost no loss in capacity. To demonstrate the compatibility of a sulphur/carbon composite with a silicon anode in a full cell, there was the needed of bringing available lithium into the system, at best by a scalable procedure. So far, literature does not offer any suitable procedures for this step. For the development of larger cells, an electrochemical lithiation using a sacrificial electrode could not really be an option. Thus, the ZSW developed a pre-lithiation method of the cathode side during the slurry-processing step by using the stabilized lithium metal powders (SLMP) received from FMC Lithium. For this reason, a slurry stable against metallic lithium had to be developed. In literature, one may find sources reporting about the slurry processing of polyethylene oxide/anisole mixtures (PEO/anisole). This is an interesting binder system for our approach and we tested it on a Sulphur carbon composite. The pre-lithiated cathodes showed very similar results as of those untreated opening up the way for the exploitation in full cells.
Full-cells were assembled in coin and pouch geometry. For coin cells we adopted anode of 13.5 mm in diameter and cathode of 12 mm in diameter in order to avoid lithium plating at the edge of the copper foil. A circular piece (ø=16mm) of Celgard 2325 was adopted as separator and 1M LiTFSI + 0.25M LiNO3 in 1,3-dioxolane (Diol): dimethoxyethane (DME) (1:1 V/V) served as the electrolyte. The performance of the obtained full coin-cells showed a first discharge capacity of about 425 mAh/g based on the weight of the S/C composite. The galvanostatic plots showed the classical shape for a Si-C/S-C cell. During discharge after a first steep drop, the potential slowly decreases between 2.25 and 1.5 V. At 1.5 V a rather flat plateau appears. Toward the end of the discharge the potential suddenly drops reaching the cut-off set at 1 V. The cells showed a mean potential of 1.75 V. It is actually not a surprise that the cells showed a lower potential than Li-S cells. This is because the Silicon electrode works at somehow higher potential than the Li0. The columbic efficiencies at the first cycle were always close to 0.7. During the following cycles it increased up to 0.95 although some fluctuations were observed. Although the usage of pre-lithiated electrodes was successfully exploited in full cells, the first cycle columbic efficiency never reached values close to 1. This might be explained considering some amount of polysulphides dissolution during the first charge/discharge process. The assembled coin cells were able to cycle (continuously) for 100 times at a current density of 200 mA/gS-C. As already mentioned the first discharge capacity was of 425 mAh/g with retention of about 55-60% at the hundredth cycle.
After the evaluation of the anode/electrolyte/cathode compatibility and electrochemical performances in full small coin cells, larger single layer pouch cells were manufactured. The electrodes size was the following:
- Anode: 26 cm2
- Cathode: 24 cm2
As for the coin cells the larger size of the anode avoided any eventual lithium plating on the Cu current collector at the edge of the electrode. Electrodes compositions, current collectors, electrolyte and separator were exactly the same as those used in coin cells. Necessary was the welding of the tabs, one at the cathode (Al) and one at the anode (Ni), so to have the contacts coming out from the pouch bag. Pouch cells have been built in classical two-electrode configuration but also in three-electrode configuration in which a Lithium metal stripe acted as the pseudo reference electrode. The use of a third reference electrode helped in optimizing the balancing of the cell because it allowed measuring and controlling of the single electrodes potential (Ec, Ea) as well as the full cell potential (Ec-Ea). It is worth mentioning that there were no issues in preparing larger area electrodes and that the pre-lithiation procedures could be easily applied to both the cathode and the anode. The electrochemical performances of the pouch cells were evaluated by continuous galvanostatic cycling and applying a current density of 200 mA/gS-C. The first charge capacity of the pouch cell was of around 50 mAh showing a first cycle columbic efficiency of 0.65. In the following cycles the columbic efficiency increased up to 0.9. Again as for the coin cell we believe that the electrolyte is unable to fully prevent the dissolution of polysulphides. A critical issue for the further development of the “Sulphur-cell technology” is the improvement of the ability of the electrolyte in preventing polysulfide dissolution and eventually shuttling effects. The mean cell potential was of 1.75 V. The galvanostatic curves well reflected the shape observed for the coin cells. Also for the single layer pouch cells was possible collecting up to 100 cycles with retention of about 200 mAh/gS-C.
The worked carried out aiming at the development of the “LISSEN” cell successfully showed that single layer pouch cells with electrodes size of 24 cm2 (cathode) and 26 cm2 (anode) can be assembled and cycled with good electrochemical performances. Several samples of cathode, anode and electrolytes received from the project partners were evaluated in terms of electrochemical performances. The work followed a logical plan in which the behaviors of the active materials were first evaluated in small coin half-cells followed by studies of full cells (coin type). To that aim, anodes based on Si/C composites have been successfully prepared using aqueous-based slurries and LiPAA as the binder. Very high areal loading of 3-4 mg/cm2 of active material (Si/C) have been achieved. A method for pre-lithiate the anode by using the Stabilized Lithium Metal Powder has also been developed and proven to compensate the irreversible capacity loss. Furthermore ZSW developed a preparation process for the S/C cathodes. Also for the cathode it was possible to prepare electrodes with rather high loadings (4.2 mgS-C/cm2). However, for this case it was not possible to adopt an aqueous-based slurry since the pre-lithiation method have seen the introduction of the SLMP directly during the slurry preparation step. The results obtained with the prepared electrodes showed good electrochemical performances. The scalability of the electrodes preparation and pre-lithiation steps were successfully demonstrated and single layer pouch cells were manufactured and tested. The obtained single layer pouch cells were able to run for 100 cycles even though showing some capacity decay. The main obstacles were the relatively low performance of the electrolyte but also the difficulties in obtaining high mass loading cathodes.
5 - Materials scaling up
The scale up of the materials is always a complicated step. Due to time constraints in LISSEN, scaling up was associated to the preparation of up to 500g of materials for both the cathode and anode and a polymer membrane with a homogeneous surface of an A5 size.
The best materials prepared and tested in coin cells were scaled up and used in a pouch cell configuration.
For the cathode materials, as mentioned earlier, the key limitation was the sulfuration process and the post treatment under Argon that requires specific equipment. Commercial carbons were used as a carbon support for good electronic conductivity, homogeneity and also due to availability. After studies of their electrochemical performance in half coin cells, several C-S composites were selected and scaled up. The ink processing and casting is an important step for the stability of the electrode, especially when high loadings (at least 3 mg/cm2) of sulphur are needed. Carbon black was ball milled in presence of sulphur and a heat treatment at medium temperature (160°C) was then performed.
The Si-C composites were prepared using ball milling. This process has multiple advantages: to ensure a good repartition of silicon and carbon, to break the silicon aggregates and to decrease the size of the silicon crystallites. The scaled up material had a silicon loading of 50 wt. %.
For the electrolyte, the goal was to scale up polymer electrolytes, based on a ternary mixture of poly(ethylene oxide) (PEO), a lithium salt and an inorganic filler. The final target was to realize polymer electrolyte films with an area of 20x20 cm and a maximum thickness of 100 µm. The recipe for the polymer electrolyte was provided by UNIROMA and KIT conducted the scaling up. The polymer electrolytes were prepared in a dry room (with a relative humidity below 1%), using the hot-press technique.
The first attempt to scale up the original recipe of UNIROMA resulted in brittle and very inhomogeneous films, which was attributed to the intrinsic properties of the used polymer. Therefore, KIT tried several approaches, including addition of solvents and ionic liquids, to obtain homogeneous films with the diameter and thickness matching the set requirements. KIT succeeded in preparing homogeneous polymer films with the required diameter and thickness, however, just by using an organic solvent. The results of this investigation were presented in the review meeting in month 30. As the aim of the task was to scale up the initial recipe by hot-press technique without the use of solvents it was agreed during the meeting to slightly modify the original recipe. For further investigations PEO with a lower average molecular weight was used first without inorganic filler in order to determine if the filler had a negative impact on the up scaled membranes.
After the change to lower molecular weight PEO KIT was able to obtain polymer electrolytes with and without inorganic filler that were matching the requirements in terms of film area and thickness. As it was also planned to swell the polymer electrolyte film with ionic liquid based solutions prepared by UNIROMA KIT was conducting swelling experiments using the standard electrolyte for lithium-sulphur batteries. These experiments revealed that the polymer electrolyte films were partly dissolving in the liquid resulting in gel-like substances without mechanical stability. In order to overcome this issue a UV-initiator was added to the system and the films were UV-cross-linked after the hot-pressing step. Following this procedure KIT succeeded in the preparation of polymer electrolyte films having the required area and thickness, which did not show any dissolution in the liquid electrolyte.
In summary, it was possible to scale up the initial recipe of UNIROMA by changing from high molecular weight to low molecular weight PEO. KIT was able to prepare polymer electrolyte films with and without inorganic filler with an area of at least 20x20 cm and a thickness below 100 µm. By cross-linking the polymer films using a UV-initiator the films could be swollen with liquid electrolyte without losing their mechanical integrity.
6- Cells scaling up
The work carried out within the frame of the cell scaling up has been divided into two parts, considering the two different cell chemistries involved. The “LISSEN” project aimed indeed at:
1) Design and realization of pouch full cells (max 0.1 Ah) using partially lithiated Silicon/carbon anode and Sulphur/carbon cathode;
2) Realization of Z-folded pouch cells (max 0.5 Ah) using Sulphur/carbon composite as the cathode and lithium metal as the anode.
For the realization of the negative and positive electrodes, the Si-C and S-C composites were provided by JM, which took care about the scaling up of the synthesis. For the case of cells using lithium metal as the negative electrode, Rockwood Lithium kindly provided foils of the alkali metal.
Design and realization of pouch full cells (max 0.1 Ah) using partially lithiated Silicon/carbon anode and Sulphur/carbon cathode
The basis for the full cell design was a minimum active area of 24 cm2 with a maximum total electrode area of around 48 cm2 for max 0.1 Ah targeted capacity. Considering that a source of lithium is missing in the Sulphur/Silicon cell, one of the biggest challenges was to provide enough lithium ions so that the cell could operate. Therefore, to demonstrate the compatibility of a sulphur/carbon composite with a silicon anode, ZSW developed procedures for the pre-lithiation of both the cathode and the anode. A description of the two processes is given in the chapter “Cell Design”. All cathode layers have been manufactured in a glove box where H2O and O2 level were kept constantly below 0.1 ppm. Both cathodes and anodes have been prepared at the ZSW using small lab coaters and are meant to be single side coated. Wet thickness was set at 250 μm for the cathode and 120 μm for the anode. Pouch cells have been assembled in a glove box because of the high reactivity of the lithiated electrode materials. The multi layers pouch cells consisted of two cathodes and two anodes stacked together. The electrodes were winded into Celgard 2325 separator. Before assembling the cells, electrodes were welded with their tabs. Aluminium tabs are used for cathodes and nickel ones for anodes. The total anode area in the cell was of 52 cm2, whereas ca. 48 cm2 was that of the cathode. The multi layers stack was finally placed in a pouch bag and filled with electrolyte (1M LiTFSI, 0.25 M LiNO3 in a mixture of 1,2-dimethoxyethane and 1,3-dioxolane (1/1 V%/V%)). We here note that the addition of the electrolyte leads to an increase of the cell temperature. Such operation must therefore be performed very carefully. The cells filled with the electrolyte were finally sealed and let resting for 48h. Their electrochemical performances were then evaluated showing that over 250 galvanostatic charge/discharge cycles could be sustained even if a drop of the capacity over cycling was observed.
Realization of Z-folded pouch cells (max 0.5 Ah) using Sulphur/carbon composite as the cathode and lithium metal as the anode.
The objective was to assemble Z-folded pouch cells using lithium metal as anode and Sulphur/carbon composite material as cathode. Z-folded pouch cells were assembled by stacking (up to 5 units) two lithium metal sheets interposed between two Sulphur-Carbon cathodes. The expected capacity of the final cells was of max 0.5 Ah. The Sulphur composite cathodes and the liquid electrolyte were prepared by ZSW and the final pouch cells built by KIT, which also provided the lithium metal. The composition of the Sulphur-based cathodes as well as the process used for their fabrication is reported in the “Cell Design” chapter. KIT fabricated the Z-folded pouch cells using dry-room facilities with relative humidity below 1%. Each prototype cell was assembled by hand rolling of the cathode and anode layers in between the separator (Celgard 2325) layers. After hand rolling, the “tails” of the Al current collectors were welded together on a bigger piece of Al foil, using ZSW facilities. For the negative electrode, Ni foil was attached to the tails of lithium metal. The obtained roll was inserted into an empty pouch bag. The electrolyte (1M LiTFSI 0.25M LiNO3 in DME:DOL (1:1, v/v)) was inserted into the cell using a blister prior to vacuum-sealing the cell. After sealing the blister was opened. The cell was left for 24 h rest.
After the rest, the cell impedance was measured. The results showed that the impedance of all cells was very high resulting in a very low capacity during the subsequent cycling. The impedance measurements on small symmetrical Li/separator/Li and cathode/separator/cathode cells revealed that the main resistance was originated from the cathode material. This was attributed to the cathode preparation sequence. Another possible reason for high impedance was the lack of electrolyte in the Z-folded cell. Therefore, an excess of electrolyte was added to the cell. However, the impedance was still very high and thus no satisfactory cycling results could be obtained. Due to the limited time within the task, further investigations were not conducted.
7 - Recyclability and Life cycle sustainability
The LISSEN battery with its new and unique chemistry will give the recycling industry novel challenges to solve. Today´s recycling of Li-ion batteries is adopted for the recovery of valuable metals such as cobalt. For LISSEN batteries, recycling will most likely be motivated by sustainability requirements and the recovery of lithium. From a value point of view, the most interesting metals are the copper and aluminium foils supporting the anode and cathode materials, as well as the aluminium “pouch”. The reuse of expensive organic solvents and additives in the electrolyte could also be of interest. From a resource point of view, recycling of lithium is important if these batteries should be used in large volumes. The low metallic content in the battery indicates that it will be difficult to fulfil the Battery Directives demand of 50% recyclability only recovering copper, aluminium and lithium. To fulfil the Battery Directive, a more progressive process is needed, recycling also the carbon and/or silicon and sulphur. This is an important conclusion and shows that the recycling requirements in the Battery Directive are not adapted to certain modern forms of batteries where the element value as well as the environmental benefit from recycling of these elements can be low.
Recycling of lithium sulphur cells possesses a number of health and safety challenges. The LISSEN batteries differ from today´s Li-ion batteries in the use of lithium sulphide as cathode material. In air, lithium sulphide hydrolyses and reacts to form hydrogen sulphide and lithium hydroxide. This is a problem both in manufacturing and recycling of the batteries. The presence of lithium sulphide has also implications when looking at different ways to recycle these batteries. For example, the sulphur can either form sulphur dioxide or hydrogen sulphide in a thermal process, depending on the oxygen level or form sulphide ions or hydrogen sulphide in a water solution, depending on pH.
Since no “real” LISSEN batteries have been available, commercially batteries have been used for the tests. A LISSEN battery has been simulated in the hydro chemical experiment by adding sulphide ions. The thermal and mechanical tests have not included the presence of sulphur. This aspect has, however, been included in the proposed process solutions.
The hydro chemical experiments demonstrate that lithium can be removed from the washing water using a strong acid ion exchange resin while the anions can be recovered either using a weak base ion exchanger or by solvent extraction with a lipophilic triamine. It was shown that the lithium could be recovered by treatment of the ion exchange resin with aqueous acid. The presence of sulphide ions in the leach solution can modify the leaching behaviour of some metals and needs to be considered designing a wet chemical treatment process. But the presence of sulphides is not an insurmountable obstacle to a hydrometallurgical process. Valuable organics such as fluorine containing anions and other components of ionic liquids could be recycled, it has been shown that the lipophilic anions used in lithium ion cells can be absorbed onto ion exchange resin thus paving the way for a recycling process for these anions.
Based on the hydro chemical test at Chalmers and the mechanical/thermal tests at Stena, two different process routes are suggested:
This process scheme is based on a solution with shredding under water combined with sink-and-float. First, the batteries are discharged and dismantled to give only battery cells with minimum energy content. The electronics, cables and metallic covers will be sent directly to recycling. The battery cells are then shredded under water. An alternative way is to shred the batteries in an inert atmosphere (N2, CO2 or argon) and dispose them directly into a water bath to reduce the risk of fire. The water will contain soluble organic compounds, H2S(aq) and Li+-ions. In order to keep the H2S in the solution (as HS- and S2- -ions) it is important that the pH will be maintained above nine. The shredded cells are passed through a sink-and-float and sieving device. The solid material is divided into three streams; a light fraction mainly consisting of plastic separators, a large, heavy, fraction with copper and aluminium electrode supports plus aluminium from the casing, and finally a fine fraction made of insoluble anode materials, carbon black and PVDF. The plastics can be reused as a raw material or used as a fuel while the aluminium/copper mix is used as a raw material in secondary aluminium industry, where copper is commonly needed as an additive. The fine fraction is filtered out by means of a filter press and sent to a limekiln as a fuel. The water continues to a solvent extraction step where the lithium is separated with selective organic extractants or ion exchange resins. After stripping, the lithium ions are precipitated as lithium carbonate or what the market demands. This method gives an opportunity to separate aluminium, copper and polymers from the black mass and soluble materials in the battery, e.g. lithium sulphide and organic carbonates. Lithium and valuable organics can further be extracted from the water. It is important that pH always is above 9 to avoid formation of hydrogen sulphide. Good ventilation is required due to the presence of the organic carbonates but also as a back up, if hydrogen sulphide accidentally is formed.
After discharging and dismantling (same as Process 1 above), the battery cells are heated at a temperature of ca 450 oC. During the pyrolysis process, the organic content of the electrolyte, separator and binder will decompose to a mixture of small organic compounds (mainly methane, ethane and propane), hydrogen fluoride, carbon monoxide, some carbon dioxide and water. The sulphur may leave the system as elemental sulphur or as hydrogen sulphide. By adding oxygen to the process, more combustion like conditions can be achieved, and the gaseous product will contain more carbon dioxide, sulphur dioxide and water. After pyrolysis, the battery cells can be handled without any further risk for fire and are shredded in a standard knife shredder. The shredded material are sieved and separated into a fine fraction and coarse fraction. The coarse fraction will mainly consist of the aluminium and copper supports and the aluminium from the casing, and are sent to a secondary aluminium smelter for further processing. The fine fraction is leached with water to dissolve the lithium present in the battery. After leaching, there will be a residue of insoluble material, mainly carbon, that will be filtered of by a filter press and send to a lime kiln as a fuel. The lithium are precipitated as lithium carbonate or what the market demands. The off gas from the pyrolysis unit will be treated in a flue gas cleaning facility.
Thus, from a technical point of view there are probably no difficulties to handle a LISSEN-battery even if a process most likely will be more expensive due to the extra concern that has to be put on treating its content of sulphur in a good way. However, from an economical point of view, it is questionable if a LISSEN-battery will cover its treatment cost due to the low value materials it consist of and therefore a fee is most likely needed if the batteries are to be recycled, at a level in relation to the directive, in a safe and environmental responsible way.
Life Cycle Assessment
The life cycle sustainability assessment has been made to analyse and assess the potential for the LISSEN battery to contribute to sustainable development. The assessment is focused on environmental issues. Environmental impacts of two LISSEN cell prototypes have been assessed: one with a Li metal anode and one with a Si/C anode. Their environmental performances have been compared to those of an NMC cell and a diesel fuel producing the same mechanical energy. The lifetime of the LISSEN cells has been assumed to be 5000 cycles, and the lifetime of the NMC cell 7500 cycles. Losses of electric energy during the use phase are allocated to the cell and losses of chemical energy during the use of diesel fuel are allocated to the diesel.
Three weighting methods have been used to assess the overall environmental performance: The EPS v2015d (4), the Ecoindicator99 (5), and the Recipe (6) methods. The Ecoindicator99 and the Recipe methods are similar, and assess common attitudes towards environmental impacts. The EPS 2015d method includes the same types of impacts but valuation is based on monetary values of changes imposed on environmental and human state indicators. The EPS 2015d also differs in that it has a longer time perspective on mineral resources.
The results from the overall comparison of the four technologies indicate as expected, that the diesel alternative is the least favourable. Also somewhat expected is that the NMC alternative is less favourable in a long-term sustainability perspective (EPS 2015d) than in a conventional environmental perspective (EI99 and Recipe). Much of the overall impacts from the different battery alternatives come from electricity production and from other types of energy production, when manufacturing different materials. If the energy used changes to more green types, the overall impact will decrease. The present damage cost, as estimated by the EPS 2015d method, is 22 € per 1000 kWh, or 0.022 € per kWh. This means that if the external environmental damage costs would be added to the price of the use of a LISSEN battery, it would be a marginal change and not preventing its use. The LISSEN battery concept’s potential for good environmental sustainability performance, is partly due to no use of scarce resources. Al, C, Si and S are abundant materials. From a sustainability point of view, the Li metal anode concept for the LISSEN battery is to prefer compared to the Si/C anode, mainly because of not having to use copper in the anodes. Besides giving an overall lower impact value, this also means that the choice of recycling technology is more flexible. There are in particular two ways of improving the sustainability results: one is the use of more “green” energy in the material production and electricity production; another is an extended battery life.
LISSEN is providing advances to a number of scientific challenges for new lithium based battery cell technologies and their performance. The successful resolution of these will lead to breakthroughs in automotive lithium-sulphur batteries for electric vehicles and thus to the development of a sustainable mobility and quality of life.
The research and innovation proposed by the LISSEN project will contribute to the strengthening of the competitiveness of the European industry of automotive batteries and in global markets through the development of innovative technologies, architectures and chemistries; moreover, through the research on Lithium recycling, LISSEN will further strengthen the European competitiveness also in the area of recycling.
Current and near-term (i.e. lithium ion) battery technology development is one of the key factors on the Mobility Electrification and the large scale production of these automotive batteries and reducing their costs is, in fact, critical for market entry and acceptance of Electric Vehicles. In order to achieve a break-even cost with internal combustion engines, battery costs must be reduced from the current estimated range of 675 € to 500 € per kilowatt-hour at high volume production (order of 100 k units) down to 350 to 275 €/kWh by 2020. In the present scenario for Li-ion technology, raw materials and processing account for approximately 1/3 of total battery costs (Powertrain 2020, Li-ion Battery Value Chain Analysis, Roland Berger Strategy Consultants, Aachen, March 2011).
LISSEN addresses further reduction costs driven not only by the higher energy density of the Lithium-Sulphur technology, but also due to the change in electrode chemistry. Considering material cost breakdown, the cathode accounts for 35-40% in Li-ion cells so we can expect a significant cost reduction for the Li-S technology.
Moreover, it should be noted, that the proposed materials are safe, non-toxic and environmentally benign and as a consequence derived processes, such as electrode preparation and cell assembly are very "working-condition-friendly", i.e. gaseous emission-free, toxic solvent free, etc. There is no doubt, that replacement of standard lithium-ion batteries by lithium metal free Li-S lithium batteries will have a long term influence on the improvement of safety, working conditions, environment and thus eventually on quality of life.
Main Dissemination and Exploitation Activities
The overall aim of the dissemination and exploitation activities was to ensure that the project results are disseminated and that the results arising from the LISSEN project are exploited. Specifically, the objectives of the dissemination and exploitation activities were to:
• Promote the dissemination of project results
• Organize dissemination management for the consortium
• Support the partners in identifying project exploitable results
LISSEN partners agreed that the exploitation of project results is of a common interest. Therefore, dissemination and exploitation activities were planned, as much as possible, at project level and not at individual level.
To raise the public awareness of the objectives and the research applied in LISSEN, the Consortium followed several paths including conference organization, public project presentation and typical academic publication activities. To such purpose, the dissemination activities concentrated primarily on:
• Presentations at conferences and workshops both at national and international level
• Publications in scientific peer reviewed journals (both national and international) and magazines
• Setting up and maintenance of the LISSEN web site
• Organization of the LISSEN final event
All the above-mentioned activities were targeted primarily at the International Scientific Community and at potential business partners.
During the project duration the following activities were carried out:
• 5 publications in peer reviewed journals, plus 4 prepared and submitted/accepted
• 16 presentations and poster presentations at national and international conferences and workshops
• LISSEN final open event, which took the form of an international workshop “Towards next generation lithium-ion batteries” organized in Ulm on 27 July 2015; during the workshop, attended by about 50 people, 6 invited speakers made presentations on issues related to batteries for Electric Vehicles, while 6 presentations on project main results were given by project partners
The exploitation of the project results can be of interest for LISSEN industrial partners as well as for third parties. In order to streamline exploitation activities, partners identified the following eleven key exploitable results:
1) Theoretical concepts applied to cell modeling
2) Knowledge about functioning and degradation of cells
3) Optimized electrolyte solutions
4) Optimization of the electrode materials
5) Optimized cell design
6) Process for production of nanostructured Lithium sulphide composites for lithium sulphur batteries (patent filed)
7) S-C cathode
8) Si-C composite anode
9) Li2S-C cathode
10) Proof of concept large cells
11) Recycling Process
For each of the above-mentioned key exploitable results, partners already identified possible future paths for exploitation such as further research activities, patenting and licensing activities and direct use by industrial partners.
List of Websites:
The website (www.lissen.eu) is operative since mid-November 2012; it presents the project and its partners to the wider public and gives detailed, tailored information on e.g. project progress, activities and results.
During the project duration, the LISSEN web site has been visited by 5,290 people, 82% of the people visiting the website are new visitors. Apart from participating countries, persons have visited the site from 52 countries, including Europe (almost all EU countries, Russia), Asia (India, China, Japan), North America (USA and Canada) and South America (Brazil).
Grant agreement ID: 314282
1 September 2012
31 August 2015
€ 3 744 684,13
€ 2 579 940
SAPIENZA INNOVAZIONE CONSORZIO
Deliverables not available
Grant agreement ID: 314282
1 September 2012
31 August 2015
€ 3 744 684,13
€ 2 579 940
SAPIENZA INNOVAZIONE CONSORZIO
Grant agreement ID: 314282
1 September 2012
31 August 2015
€ 3 744 684,13
€ 2 579 940
SAPIENZA INNOVAZIONE CONSORZIO