No abstract

Ion conduction is of prime importance for solid-state reactions in ionic systems, and for devices such as high-temperature batteries and fuel cells, chemical filters and sensors. Ionic conductivity in solid electrolytes can be improved by dissolving appropriate impurities into the structure or by introducing interfaces that cause the redistribution of ions in the space-charge regions. Heterojunctions in two-phase systems should be particularly efficient at improving ionic conduction, and a qualitatively different conductivity behaviour is expected when interface spacing is comparable to or smaller than the width of the space-charge regions in comparatively large crystals. Here we report the preparation, by molecular-beam epitaxy, of defined heterolayered films composed of CaF2 and BaF2 that exhibit ionic conductivity (parallel to the interfaces) increasing proportionally with interface density--for interfacial spacing greater than 50 nanometres. The results are in excellent agreement with semi-infinite space-charge calculations, assuming a redistribution of fluoride ions at the interfaces. If the spacing is reduced further, the boundary zones overlap and the predicted mesoscopic size effect is observed. At this point, the single layers lose their individuality and an artificial ionically conducting material with anomalous transport properties is generated. Our results should lead to fundamental insight into ionic contact processes and to tailored ionic conductors of potential relevance for medium-temperature applications.

Twenty seven LiCoO 2 /graphite wound prismatic cells containing a variety of electrolyte additives as well as high or low surface area LiCoO 2 were studied during high temperature storage using an automated storage system. The same cells had been previously studied using high precision coulometry. Cells were initially cycled to measure the capacity, charged and then stored for one month at either 40 or 60 • C, then cycled again to measure the reversible and irreversible capacity loss. The process was then repeated. During storage, the open circuit potential was automatically measured every 6 hours. The mechanisms responsible for the voltage drop which occurred during storage and the capacity loss after storage were analysed using a Li inventory model. The voltage drop during storage is caused primarily by parasitic reactions (electrolyte oxidation, transition metal dissolution, etc.) that insert Li into the positive electrode, because the potential of the Li x C 6 electrode is virtually constant on the stage-2/stage-1 plateau even if its Li content changes due to solid electrolyte interface (SEI) growth. The experimental results show that the combination of the electrolyte additive, vinylene carbonate, and low surface area LiCoO 2 minimizes the voltage drop and capacity loss during storage presumably by reducing the amount of electrolyte oxidation occurring at the positive electrode. The same cells had charge endpoint capacity slippages that were closest to 0.00%/cycle during cycling tests monitored with high precision coulometry. Storage experiments, in concert with precision coulometry, allow a clear picture of the effect of additives to be determined.Lithium-ion batteries are now being used in electrified vehicles. The cycle and calendar life requirements in vehicular applications are far more demanding than in computer and phone applications. Therefore it is utmost importance to understand cell degradation mechanisms and to use new electrode materials, electrolytes and electrolyte additives to minimize degradation.Capacity loss in Li-ion batteries occurs during storage and cycling. 1-5 There are many possible undesired or parasitic processes, such as dissolution of transition metals from charged positive electrodes, corrosion of current collectors, electrolyte oxidation at the positive electrode, electrolyte reduction at the negative electrode leading to SEI growth, etc. that lead to capacity loss. Capacity retention and storage life of Li-ion cells are critically dependent on the stability of the passivation layers that form on both electrodes. Control of the electrode/electrolyte interfaces is therefore key to obtain Li-ion cells with long lifetimes.It was suggested by Broussely et al. 2 that lithium consumption at the negative electrode affected cell capacity during storage at high temperature. They also concluded that electrolyte oxidation at the positive electrode resulted in additional losses during storage at high voltage.Electrolyte additives, such as vinylene carbonate (VC), are known to improve cycl...

No abstract

Methods and criteria for assessing the commercial viability of Si-based materials are discussed and demonstrated with the 3M V6 alloy and 60 nm nano Si powder. These materials are firstly evaluated through the cycling of neat electrodes containing only alloy and binder to characterize the capacity, first cycle efficiency, binder compatibility, and microstructure stability of the material. The alloy displays higher first cycle efficiency, lower fade, and a more stable amorphous microstructure compared to the nano Si, which displays a variable microstructure with a rate dependent presence of crystalline Li 15 Si 4 . The materials are then evaluated in graphite-containing composite electrodes having high areal capacities (> 2 mAh/cm 2 ). In a well designed composite electrode including carbon nanotubes, 3M V6 material was found to cycle with little fade and high coulombic efficiency (∼99.8%) while maintaining a stable dQ/dV. A composite electrode of equivalent volumetric capacity with nano Si powder shows similar capacity retention over 50 cycles but an unacceptably low coulombic efficiency (∼99.2%). High precision coulometry and calorimetry results show surface area as the dominant factor in levels of parasitic reactions with Si based materials.

Si 1Ϫx Sn x samples for 0 Ͻ x Ͻ 0.5 were prepared by magnetron sputtering using a combinatorial materials science approach. The room-temperature resistivity and X-ray diffraction ͑XRD͒ patterns of the samples were used to select materials having both an amorphous structure and good conductivity for further study. The reaction of lithium with amorphous Si 0.66 Sn 0.34 was then studied by electrochemical methods and by in situ XRD. The electrode material apparently remains amorphous throughout all portions of the charge and discharge profile, in the range 0 Ͻ x Ͻ 4.4 in Li x Si 0.66 Sn 0.34. No crystalline phases are formed, unlike the situation when lithium reacts with tin. Using the Debye scattering formalism, we show that the XRD patterns of the a-Si 0.66 Sn 0.34 starting material and a-Li 4.4 Si 0.66 Sn 0.34 can be explained by the same local atomic arrangements as found in crystalline Si and Li 4.4 Si or Li 4.4 Sn, respectively. In fact, the in situ XRD patterns of a-Li x Si 0.66 Sn 0.34 , for any x, can be well approximated by a linear combination of the patterns for x ϭ 0 and x ϭ 4.4. This suggests that predominantly only two local environments for Si and Sn are found at any value of x in a-Li x Si 0.66 Sn 0.44. However, based on differential capacity vs. potential results for Li/a-Si 0.66 Sn 0.34 there is no evidence for two-phase regions during the charge and discharge profile. Thus, the two local environments must appear at random throughout the particles. We speculate that the charge-discharge hysteresis in the voltage-capacity profile of Li/ a-Li x Si 0.66 Sn 0.34 cells is caused by the energy dissipated during the changes in the local atomic environment around the host atoms.

LiCoO 2 /graphite and LiCoO 2 /Li 4 Ti 5 O 12 wound prismatic cells were examined with and without electrolyte additives using the high precision charger at Dalhousie University. The additives tested were vinylene carbonate, trimethoxyboroxine, and lithium ͑bis͒ trifluoromethanesulfonimide. The voltage curves, charge and discharge end point positions, fade, and coulombic efficiency were compared to gain an understanding of the effects of the electrolyte additives on the cells. Long term cycling data ͑capacity loss over 750 cycles͒ was compared with predicted lifetime measurements based on high precision coulometry. Design of experiments was used in order to help interpret the results from the 20 groups of cells tested.Lithium-ion batteries ͑LIBs͒ used in electrified vehicles and for grid energy applications require larger capacities, longer cycle life, and longer calendar life than previously required in portable electronics ͑e.g., laptops and cell phone͒ applications. Electrolyte additives have been extensively studied and are used to improve the lifetime of LIBs. [1][2][3][4][5][6][7][8] One of the most commonly used electrolyte additives is vinylene carbonate ͑VC͒. 2-5 Aurbach et al. 2 studied the impact of VC and found that its reduction at a graphite negative electrode takes place before the reduction of ethylene carbonate, which forms a flexible and cohesive polymeric surface species that acts as a more stable solid electrolyte interface ͑SEI͒. The authors also believed that this type of unique surface reaction may be occurring on the positive electrode, stabilizing its SEI as well. Ota et al. 3 examined the improved SEI formed on graphite with the addition of VC. Another paper from the same group showed the beneficial impact of VC at higher temperature and attributed it to increased Li + ion mobility and stated that the addition of VC has a large impact on the negative electrode but could benefit the positive electrode as well. 4 Other less studied additives include trimethoxyboroxine ͑TMOBX͒ and LiN͑CF 3 SO 2 ͒ 2 ͑called HQ-115 here͒. TMOBX is made of a ͑BO͒ 3 ring with methyl groups attached to the boron atoms. Mao et al. 6 showed that the presence of ͑BO͒ 3 rings dissolved in electrolyte reduced the capacity loss during cycling of 18,650-size cells ͑LiCoO 2 /graphite and LiMn 2 O 4 /graphite͒ and that the additive with the methyl groups attached to the ring proved more effective. Small concentrations ͑Ͻ 1%͒ were added to see this benefit. 6 HQ-115 is a Li-salt ideal for organic electrolyte-based lithium batteries. HQ-115 has better thermal stability than LiPF 6 resulting from the strong covalent bonding nature of the negative ion. 7,8 It is believed by the authors that HQ-115 is used as an electrolyte additive in prismatic and pouch-type Li-ion cells to limit gas generation during operation.A major problem for researchers is the inability to conveniently determine the cycle life and calendar life of lithium-ion batteries under actual duty cycles that will be used in the field. For example, a pure elect...

In recent papers, our laboratory has reported new cathode materials based on the Li͓Ni x Co 1Ϫ2x Mn x ]O 2 series. These materials show good performance and appear to be much less reactive with electrolyte at high temperatures than LiCoO 2 at the same potential; however, the tap density and resulting electrode density of samples reported previously is lower than required for many industrial applications in Li-ion batteries. This paper focuses on changes to the synthesis procedure to increase the tap and pellet densities. Denser samples ͑pellet density between 3.7 and 4.1 g/cm 3 ͒ can be obtained using an improved coprecipitation method, followed by heating at 1100°C. Different cell performance is obtained depending on the cooling conditions and on the stoichiometry, x, in Li͓Ni x Co 1Ϫ2x Mn x ]O 2 . The best capacity retention for cycling to 4.4 V is obtained for Li͓Ni 0.25 Co 0.5 Mn 0.25 ]O 2 heated to 1100°C followed by slow cooling in oxygen. The volumetric energy density of electrodes ͑not including the current collector͒ of this material is 2110 Wh/L compared to 2270 Wh/L for optimized LiCoO 2 cycled under the same conditions ͑40 mA/g, 2.5 to 4.4 V͒. The volumetric energy density of Li͓Ni 0.25 Co 0.5 Mn 0.25 ]O 2 would exceed that of LiCoO 2 if the irreversible capacity ͑about 20 mAh/g͒ could be eliminated, and that will be the focus of future work.