No abstract

No abstract

No abstract

Li-S cells have a low voltage (∼ 2.1 V), but their potentially high energy density (200-500 Wh/kg) makes them a promising system for next generation batteries. To obtain high energy densities on cell level, the weight fraction and load of the active material should be as high as possible, while inactive material is reduced to a minimum. Conventionally, sulfur slurry cathodes with an aluminum current collector are used. However, binder-free CNT-coated carbon structures are a promising method of achieving higher loads and higher ratios of active material. Using a specially designed test cell it was demonstrated that sulfur cathodes without a metal current collector can deliver enough power to meet the requirements of consumer electronics at simultaneously high capacities of up to 600 mAh g −1 for the entire electrode and current collector. A literature study compared various equivalent circuits used for Li-S electrochemical impedance spectroscopy (EIS), and enabled the selection of the most suitable one for the system used here. EIS measurements during charge and discharge delivered vital information about the specific resistances of the sulfur cathodes with a carbon current collector.Li-S cells deliver high theoretical capacities of 1672 mAh g −1 at a relatively low average discharge voltage of ∼ 2.1 V, thus providing energy densities of around 200-500 Wh kg −1 on cell level. In most publications, electrodes with sulfur loadings below 2 mg cm −2 are reported, 1 leading to active mass ratios clearly below 40% for the entire electrode and current collector. We believe that this ratio can be improved in some applications with smaller cell size (e.g. smartphones) by replacing the polymer binder and the metal current collector with a carbon current collector in a binder free sulfur cathode.Recent publications also show a growing interest in binder-free electrodes: Elazari et al. demonstrated good sulfur utilizations between 50-60% with a microporous activated carbon fiber cloth and a sulfur load of 6.5 mg cm −2 . 2 Zhou et al. prepared a flexible CNTbased membrane (CNT-S: 2-3 mg cm −2 ) and obtained high capacities of around 700 mAh g −1 S at very high currents of 6 A g −1 . 3 The high rate performance of these CNT-based electrodes is confirmed by Su et al. who achieved around 900 mAh g −1 S at 4C. 4 Kim et al. examined the effects of high temperature conditions on cell capacity, rate capability and cycle durability of a vertically-aligned CNT electrode synthesized on a Ni substrate by a CVD process. 5 With a comparable electrode acceptable capacities of around 700 mAh g −1 S were reported at sulfur loads of 7.1 mg cm −2 and 90 wt% sulfur in the electrode. 6 Higher sulfur loads of up to 20 mg cm −1 with sulfur utilizations around 50% can be obtained with CNT-coated carbon fiber structures .7The electrode we applied also consists of CNTs coated on a carbon fiber structure by a CVD process with subsequent melt infiltration of sulfur. Our belief is that that the optimization of such a cathode can allow active material rat...

Li-S cells can have high gravimetric energy densities above 300 Wh kg −1 when the electrodes and cell components are optimized. Low electrolyte/sulfur mass ratios or more generallly, the relative amount of electrolyte in a Li-S cell have an especially high impact on the achievable gravimetric energy density. A negative side effect of low electrolyte/sulfur ratios are low cycle numbers due to electrolyte decomposition and the possibility that electrolyte becomes inaccessible at the lithium metal anode when the lithium becomes more and more porous during cycling. Electrode thickness measurements were performed during cycling for various cell chemistries such as lithium-sulfur (Li-S) with different cathode sulfur loadings and porosities, lithium-hard carbon (Li-HC), lithium-silicon (Li-Si), prelithiated HC-sulfur (LiHC-S), prelithiated Si-sulfur (LiSi-S) and Li-ion. The thickness measurements provided information about mechanical stress and irreversible thickness changes. The thickness measurements also helped to explain different electrolyte decomposition behavior and they can be used to discuss the impact of thickness changes on gas analysis. The electrolyte decomposition of the Li-S standard electrolyte based on lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) in dimethoxyethane (DME): dioxolane (DIOX) with LiNO 3 was examined by online mass spectrometry (MS) within Li-S, Li-HC, Li-Si, LiSi-S and LiHC-S cells. Several electrolyte decomposition products were verified by post-mortem gas chromatography (GC) and gas chromatography-mass spectrometry (GC-MS) of Li-S cells with cycled electrolyte. Li-S prototypes demonstrate that high gravimetric energy densities of 300 Wh/kg and above can be obtained. Possible strategies to optimize future sulfur cells are the reduction of the weight amount of passive components e.g. by decreasing the thickness of separator and current collectors and/or the application of thin perforated current collectors. Additionally with lithium metal being conductive, the copper current collector can be removed completely providing high energy densities despite low sulfur loaded (1-3 mg cm −2 ) cathodes. By utilizing a copper current collector high sulfur load cathodes (>5 mg cm −2 ) are required to compensate for the copper's passive weight.1 Thick, high load sulfur cathodes reduce the electrode and separator coating length in a cell and therefore save costs. However high load sulfur cathodes also have the drawback that high transported capacities during cycling stress the lithium metal anode and increase the chance of lithium induced shorts 2 (next to the costs of the copper). Despite all this, the electrolyte is a major weight source in Li-S cells even if the electrolyte/sulfur weight ratio (E/S) is low. With the Li-S standard electrolyte based on lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) in dimethoxyethane (DME): dioxolane (DOL) with LiNO 3 additive, the lowest obtainable E/S ratio is likely >3:1. This is due to the high porosity of the sulfur cathode (usually ∼60-80%) which has ...

Lithium-sulfur cells are potential high gravimetric energy density cells with 200-500 Wh/kg. Unfortunately typical used ether based electrolytes are not stable towards Li and degrade during cycling next to a decreasing active Li mass. Typical Li-S publications therefore work with an excess of electrolyte and Li combined with relatively low sulfur loads allowing high cycle numbers at the cost of low practical relevance. High cell energy densities can only be obtained with a low electrolyte/sulfur ratio (ml/g) next to high sulfur utilization, fraction and load. Unfortunately low amounts of electrolyte typically decrease the sulfur utilization negatively and most sulfur cathodes need electrolyte/sulfur ratios between 5:1 and 10:1 to deliver maximum sulfur utilization. Unfortunately these ratios are too high to surpass the energy density of commercialized Li-Ion technology. We therefore suggest that the reserach focus should be switched away from maximizing only the sulfur utilization to optimization of the whole electrode capacity incl. the electrolytes weight. But even with such electrodes the obtainable cycle number will be low because of the electrolyte degradation. Since no superior, stable electrolytes exist up to now a good strategy might be the use of (lithiated) Si anodes instead of Li metal anodes. Si anodes face severe volume changes during cycling but are at least not completely converted like Li metal anodes. Thus, the electrolyte degradation might be slowlier with Si anodes and allow higher cycle numbers and also higher safety compared to Li metal anodes. In the talk we would like to give a short overview about state of the art Li-S research, discuss the impact of the electrolyte/sulfur ratio on cell energy density, show some S8-LixSiy full cell results with very high active material loads (Si > 2mg/cm²!) and conlude with safety test results of small Li-S multi layer pouch cells.

It is well known that silicon presents one of the most important anode materials for the improvement of lithium-ion cells in terms of energy density. Indeed, silicon’s high theoretical specific capacity to lithium (more than 3800 mAh/g at room temperature), environmental friendliness, low potential compared to lithium and material abundance turns silicon to a strong candidate for the replacement of carbon-based anodes [1]. However, one of the main drawbacks of silicon’s application to the lithium-ion technology is its poor electrochemical cycling stability over several galvanostatic cycles, mainly due to silicon’s huge volume change (around 300%) during lithiation and delithiation that lead to high internal mechanical stress [2]. Many alternatives have been proposed that alleviate this mechanical expansion through the use of nanostructured silicon sometimes combined with carbon-based materials [3]. However, most of proposed solutions are technically and financially demanding at least for production. In this presentation, we investigate the physical and the electrochemical properties of micro-grain structured silicon deposited on special copper with nodular grains. Silicon films with high loadings (0.8-1 mg/cm2) have been deposited with the aid of DC sputtering technique on top of copper foil with nodular grains. Half-cells with silicon as anode and Lithium metal as counter electrode were prepared. The gravimetric specific capacity was estimated by measuring the weight of silicon. Scanning Electron Analysis (SEM) has been performed in order to investigate the physical properties of the deposited material. Figure 1 demonstrates the SEM picture of the micro-grain structure of silicon when deposited on the copper foil with nodular grains. It is shown that silicon is organized in columnar form with grains of an average size of 1-2 μm. By KaptonTM tape tests, the adhesion of this material on the copper foil was found to be excellent. The equivalent silicon thickness that has been grown is estimated to be more than 4 μm (loading 0.8-1 mg/cm2). Figure 2 shows the electrochemical behavior of such half-cells when cycled with current of 0.6 mAh/cm2 (C/3 rate). Concerning the specific capacity, we observe a slight decrease from 2200 mAh/g to 1800 mAh/g and after the 5th cycle, an increase up to the initial value. It is also mentioned that the silicon anode achieves high specific areal capacity up to 2 mAh/cm2. According to McDowell at al. [4], amorphous silicon spheres with diameter of around 1 μm do not fracture upon lithiation and therefore anodes fabricated with such material demonstrate stable capacity over galvanostatic cycling. Thus, our experimental results further support that the DC sputtered amorphous spheres provide excellent capacity retention and good stability. Acknowledgement This work is supported by European Regional Development Fund and National Funds/German-Greek Bilateral R&D Cooperation Initiative/2013-2015. References [1] C. John Wen, Robert A. Huggins, Journal of Solid State Chemistry, 37, 271-278 (1981) [2] Candace K. Chan, Hailin Peng, Gao Liu, Kevin McIlwrath, Xiao Feng Zhang, Robert A. Huggins and Yi Cui, Nature Nanotechnology 3, 31 - 35 (2008) [3] Tae Hoon Hwang, Yong Min Lee, Byung-Seon Kong, Jin-Seok Seo, and Jang Wook Choi Nano Letters 12, 802-807 (2012) [4] Matthew T. McDowell, Seok Woo Lee, Justin T. Harris, Brian A. Korgel, Chongmin Wang, William D. Nix, and Yi Cui Nano Letters 13, 758-764 (2013) Figure 1

Under the German-Greek bilateral R&D cooperation initiative 2013-2015, the Safe High Energy Lithium-ion Cells for Electric Vehicles” – SHELION project aimed to investigate both the anode and the cathode materials with a clear objective to manufacture cells with a specific energy density of more than 150 Wh/kg with adequate cycling charging/discharging performance, complying with industrial manufacturing practices, safety standards and environmental regulations. Towards this direction, this project is focused on the lithium-ion technology and most precisely with the Silicon (Si)-Lithium Iron Phosphate (LFP) electrochemical system that due to silicon should provide excellent energy density and due to LFP excellent safety properties and cost advantage [1]. Silicon as anode in Li-ion technology is considered as a promising material that will eventually replace graphite mainly thanks to its high specific capacity that may attain more 4200 mAh/g (theoretical value) compared to graphite’s specific capacity (372 mAh/g, theoretical value) [2,3]. However, silicon undergoes mechanical stress that is induced during lithiation and delithiation, which leads to poor life cycle [4]. During this project, silicon deposited by DC sputtering was investigated and after multiple substrates used, anodes with high specific capacity of more than 2000 mAh/g and 2.5 mAh/cm2 were manufactured [5,6]. Silicon’s capacity was found to be stable for more than 50 galvanostatic cycles. Concerning the cathode, slurry based LFP was developed and achieved specific capacity of more than 3.5 mAh/cm2 and 180 mAh/g. The electrochemical characterization of these materials and their combinations with various electrolytes was performed in small cells. Based on these results, 820 mAh pouch cells were designed, developed and manufactured (demonstrators). Galvanostatic cycling were performed and an overcharge safety test according to IEC 62660-2. The demonstrator cell with a capacity of nominal 820 mAh was put into a sealed stainless steel can, equipped with an internal camera to observe the cell during the test. Additionally the can was linked to a mass spectroscope, a gas chromatograph and a FTIR device to measure released gas online. Temperature sensors on the demonstrator pouch cell enabled a monitoring of the cell temperature during the test. The demonstrator was charged with C/2 until the state of charge (SOC) was measured to be 200% and therefore fulfilled the guidelines of IEC 62660-2 successfully. A thermal runaway was enforced when the demonstrator was charged with 4C and had a SOC of 300%. The temperature at thermal runaway was measured to be ~340 °C. This contribution will show a distribution of released gas components during the overcharge safety test and provide information about the harmfulness of these substances. To our knowledge, this is the first time that safety tests were conducted to Si/LFP cells. This work was supported by the European Regional Development Fund and National Funds/German-Greek Bilateral R&D Cooperation Initiative/2013-2015. References [1] G.B.Cho, M.G.Song, S.H.Bae, J.K.Kim, Y.J.Choi, H.J.Ahn, K.K.Cho, K.W.Kim, J. Power Sources 189 (2009) 738-742 [2] H.Kim, E.-J.Lee, Y.-K. Sun, Mater. Today 17 (2014) 285-297 [3] B.Scorsati, J.Garche, J. Power Sources 195 (2010) 2419-2430 [4] H.Yang, F.Fan, W.Liang, X.Guo, T.Zhu, S.Zhang, J. Mech. Phys. Solids 70 (2014) 349-361 [5] F. Farmakis, M. Hagen, P. Fanz, A. Kovacs, S. Schiestel, P. Selinis, and N. Georgoulas, 277th ECS Meeting (Chicago, May 2015) [6] F.Farmakis, C.Esmasides, P.Fanz, M.Hagen, N.Georgoulas, J. Power Sources 293 (2015) 301-305 Figure 1