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.

To elucidate the role of vinylene carbonate (VC) as a solvent additive in organic polar solutions for lithium-ion batteries, reductive decompositions for vinylene carbonate (VC) and ethylene carbonate (EC) molecules have been comprehensively investigated both in the gas phase and in solution by means of density functional theory calculations. The salt and solvent effects are incorporated with the clusters (EC)nLi+(VC) (n = 0-3), and further corrections that account for bulk solvent effects are added using the polarized continuum model (PCM). The electron affinities of (EC)nLi+(VC) (n = 0-3) monotonically decrease when the number of EC molecules increases; a sharp decrease of about 20.0 kcal/mol is found from n = 0 to 1 and a more gentle variation for n > 1. For (EC)nLi+(VC) (n = 1-3), the reduction of VC brings about more stable ion-pair intermediates than those due to reduction of the EC molecule by 3.1, 6.1, and 5.3 kcal/mol, respectively. This finding qualitatively agrees with the experimental fact that the reduction potential of VC in the presence of Li salt is more negative than that of EC. The calculated reduction potentials corresponding to radical anion formation are close to the experimental potentials determined with cyclic voltammetry on a gold electrode surface (-2.67, -3.19 eV on the physical scale for VC and EC respectively vs experimental values -2.96 and -2.94 eV). Regarding the decomposition mechanisms, the VC and EC moieties undergo homolytic ring opening from their respective reduction intermediates, and the energy barrier of VC is about one time higher than that of EC (e.g., 20.1 vs 8.8 kcal/mol for (EC)2Li+(VC)); both are weakly affected by the explicit solvent molecules and by a bulk solvent represented by a continuum model. Alternatively, starting from the VC-reduction intermediate, the ring opening of the EC moiety via an intramolecular electron-transfer transition state has also been located; its barrier lies between those of EC and VC (e.g., 17.2 kcal/mol for (EC)2Li+(VC)). On the basis of these results, we suggest the following explanation about the role that VC may play as additive in EC-based lithium-ion battery electrolytes; VC is initially reduced to a more stable intermediate than that from EC reduction. One possibility then is that the reduced VC decomposes to form a radical anion via a barrier of about 20 kcal/mol, which undergoes a series of reactions to give rise to more active film-forming products than those resulting from EC reduction, such as lithium divinylene dicarbonate, Li-C carbides, lithium vinylene dicarbonate, R-O-Li compound, and even oligomers with repeated vinylene and carbonate-vinylene units. Another possibility starting from the VC-reduction intermediate is that the ring opening occurs on the unreduced EC moiety instead of being on the reduced VC, via an intramolecular electron transfer transition state, the energy barrier of which is lower than that of the former, in which VC just helps the intermediate formation and is not consumed. The factors that determin...

The density functional theory (DFT) calculations have been performed for the reduction decompositions of solvents widely used in Li-ion secondary battery electrolytes, ethylene carbonate (EC), propylene carbonate (PC), dimethyl carbonates (DMC), ethyl methyl carbonate (EMC), and diethyl carbonate (DEC), including a typical electrolyte additive, vinylene carbonate (VC), at the level of B3LYP/6-311+G(2d,p), both in the gas phase and solution using the polarizable conductor calculation model. In the gas phase, the first electron reduction for the cyclic carbonates and for the linear carbonates is found to be exothermic and endothermic, respectively, while the second electron reduction is endothermic for all the compounds examined. On the contrary, in solution both first and second electron reductions are exothermic for all the compounds. Among the solvents and the additive examined, the likelihood of undergoing the first electron reduction in solution was found in the order of EC > PC > VC > DMC > EMC > DEC with EC being the most likely reduced. VC, on the other hand, is most likely to undergo the second electron reduction among the compounds, in the order of VC > EC > PC. Based on the results, the experimentally demonstrated effectiveness of VC as an excellent electrolyte additive was discussed. The bulk thermodynamic properties of two dilithium alkylene glycol dicarbonates, dilithium ethylene glycol dicarbonate (Li-EDC) and dilithium 1,2-propylene glycol dicarbonate (Li-PDC), as the major component of solid-electrolyte interface (SEI) films were also examined through molecular dynamics (MD) simulations in order to understand the stability of the SEI film. It was found that film produced from a decomposition of EC, modeled by Li-EDC, has a higher density, more cohesive energy, and less solubility to the solvent than the film produced from decomposition of PC, Li-PDC. Further, MD simulations of the interface between the decomposition compound and graphite suggested that Li-EDC has more favorable interactions with the graphite surface than Li-PDC. The difference in the SEI film stability and the behavior of Li-ion battery cycling among the solvents were discussed in terms of the molecular structures.

The decomposition of LiPF 6 and the stability of PF 5 in organic solvents, diethyl carbonate ͑DEC͒, dimethyl carbonate ͑DMC͒, ␥-butyrolactone ͑GBL͒, and ethylene carbonate ͑EC͒, have been investigated through density functional theory ͑DFT͒ calculations, in which solvent was modeled as a dielectric continuum, and also by molecular dynamics ͑MD͒ simulations which treated solvents explicitly. Both calculations showed a similar trend in which the decomposition was further promoted in more polar solvents, yet the DFT calculations predicted an endothermic decomposition, while the MD simulations indicated exothermic. This sharp contrast in the results suggests strong solute-solvent interactions, especially for PF 5 , which were not accounted for in the DFT calculations. The specific interaction between PF 5 and solvent was further investigated by DFT calculations for adduct models and also by the MD simulations for solutions. Both calculations suggest a stable formation of a PF 5 -solvent adduct in solution and its stability depends on the solvent. It was found that PF 5 is more stabilized in polar and sterically compact solvents such as EC and GBL than in less polar and bulky, linear carbonates such as DMC and DEC. The reactivity of PF 5 with organic solvents and the difference in the stability of LiPF 6 between organic and aqueous solution are also discussed.