No abstract

The solubility of lithium salts in dimethyl carbonate ͑DMC͒ found in solid electrolyte interface ͑SEI͒ films was determined. The salt-DMC solutions evaporated, and the salts were transferred into water for ion conductivity measurements. The salts examined included lithium carbonate ͑Li 2 CO 3 ͒, lithium oxalate ͓͑LiCO 2 ͒ 2 ͔, lithium fluoride ͑LiF͒, lithium hydroxide ͑LiOH͒, lithium methyl carbonate ͑LiOCO 2 CH 3 ͒, and lithium ethyl carbonate ͑LiOCO 2 C 2 H 5 ͒. The salt molarity in DMC ranged from 9.6 ϫ 10 −4 mol L −1 ͑LiOCO 2 CH 3 ͒ to 9 ϫ 10 −5 mol L −1 ͑Li 2 CO 3 ͒ in the order of LiOCO 2 CH 3 Ͼ LiOCO 2 C 2 H 5 Ͼ LiOH Ͼ LiF Ͼ ͑LiCO 2 ͒ 2 Ͼ Li 2 CO 3. X-ray photoelectron spectroscopy measurements on SEI films on the surface of the negative electrode taken from a commercial battery after soaking in DMC for 1 h suggested that the films can dissolve. Separately, the heat of dissolution of the salts was calculated from computer simulations for the same salts, including lithium oxide ͑Li 2 O͒, lithium methoxide ͑LiOCH 3 ͒, and dilithium ethylene glycol dicarbonate ͓͑CH 2 OCO 2 Li͒ 2 :LiEDC͔ in both DMC and ethylene carbonate ͑EC͒. The results from the computer simulations suggested that the order in which the salt was likely to dissolve in both DMC and EC was LiEDC Ͼ LiOCO 2 CH 3 Ͼ LiOH Ͼ LiOCO 2 C 2 H 5 Ͼ LiOCH 3 Ͼ LiF Ͼ ͑LiCO 2 ͒ 2 Ͼ Li 2 CO 3 Ͼ Li 2 O. This order agreed with the experiment in DMC within the experimental error. Both experiment and computer simulations showed that the organic salts are more likely to dissolve in DMC than the inorganic salts. The calculations also predicted that the salts dissolve more likely in EC than in DMC in general. Moreover, the results from the study were used to discuss the capacity fading mechanism during the storage of lithium-ion batteries.

Density functional theory (DFT) is used to reveal that the polycrystalline Young’s modulus (E) of graphite triples as it is lithiated to LiC6 . This behavior is captured in a linear relationship between E and lithium concentration suitable for continuum-scale models aimed at predicting diffusion-induced deformation in battery electrode materials. Alternatively, Poisson’s ratio is concentration-independent. Charge-transfer analyses suggest simultaneous weakening of carbon–carbon bonds within graphite basal planes and strengthening of interlayer bonding during lithiation. The variation in bond strength is shown to be responsible for the differences between elasticity tensor components, Cij , of lithium–graphite intercalation (Li-GIC) phases. Strain accumulation during Li intercalation and deintercalation is examined with a core–shell model of a Li-GIC particle assuming two coexisting phases. The requisite force equilibrium uses different Young’s moduli computed with DFT. Lithium-poor phases develop tensile strains, whereas Li-rich phases develop compressive strains. Results from the core–shell model suggest that elastic strain should be defined relative to the newest phase that forms during lithiation of graphite, and Li concentration-dependent mechanical properties should be considered in continuum level models.

Lithium insertion and removal in lithium ion battery electrodes can result in diffusion induced stresses (DISs) which may cause fracture and decrepitation of electrodes. Many lithium ion electrode materials undergo formation of two or more phases during lithium insertion or removal. In this work, we mathematically investigate the DISs in phase transforming electrodes using a core-shell structural model. We examine the concentration jumps at phase boundaries that result in stress discontinuities, which in turn can cause cracking. The influence of the mechanical properties of the two phases on stress evolution, stress discontinuity, and strain energy are clarified. The trends obtained with the model may be used to help tune electrode materials with appropriate interfacial and bulk properties so as to increase the durability of battery electrodes.

Direct observation of the particles of active material as they react in situ is necessary to build an understanding of the interaction mechanisms present when large volume-change active materials are employed in composite electrodes. In situ optical observations of a-Si 0.64 Sn 0.36 composite electrodes show dramatic interparticle movement; particles from below the electrode surface are pushed up, particles above the electrode surface shift downward, particles crash and overlap, particles become inactive, and some particles move so far that they become detached from the electrode.

No abstract

A review of recent literature on Si:C composite and nanocomposite electrode materials is first presented emphasizing that most authors do not compare the experimental specific capacity of the composite with that expected based on the phases present. We provide such a comparison and suggest that much of the apparent confusion in the literature, when taken as a whole, can be understood if nanocomposites prepared by "aggressive" methods like high energy milling and high temperature heat-treatment contain significant amounts of amorphous or nanocrystalline SiC. In order to help resolve the confusion, samples of Si 1−x C x were prepared by high-energy mechanical milling for 0.25 Ͻ x Ͻ 0.5 and by combinatorial co-sputtering for 0 Ͻ x Ͻ 0.8. X-ray diffraction shows the mechanically milled samples to be a mixture of nanocrystalline SiC and Si. Electrochemical studies of the mechanically milled samples show that the attained specific capacity can be described accurately assuming that the Si is active and can reversibly react with 3.75 Li atoms per Si atom ͑Li 15 Si 4 ͒, while the SiC is inactive. The co-sputtered samples are amorphous or extremely nanostructured for all x. For 0 Ͻ x Ͻ 0.5, the specific capacity decreases with increasing x, from about 3580 mAh/g at x = 0, to about 1000 mAh/g at x = 0.5, presumably due to the formation of inactive regions of a-SiC. The capacity of the co-sputtered samples does not reach small values at x = 0.5, unlike the ballmilled samples, because there are presumably some regions of a-Si and a-C among the inactive a-SiC regions due to the high quenching rate of the sputtering process. Commercially relevant compositions are identified.

In situ atomic force microscopy measurements of patterned amorphous Sn-Co-C sputtered films reacting with Li in an electrochemical cell have been made. Prismatic-shaped patches of Sn0.34Co0.19normalC0.47 were found to undergo reversible volume expansion of 175±5% [(Vfinal−Vinitial)∕Vinitial] without fracture. The a-Sn0.34Co0.19normalC0.47 was found to have a reversible specific capacity of 700±10mAh∕g when cycled vs a Li metal negative electrode.