MLExchange: A web-based platform enabling exchangeable machine learning workflows for scientific studies
Explore the source record for details and available documents.
Engineering topics
Publications and source records attributed to Zhao, Zhuowen.
Explore the source record for details and available documents.
The Li-S battery is a promising next-generation technology due to its high theoretical energy density (2600 Wh kg -1 ) and low active material cost. However, poor cycling stability and coulombic efficiency caused by polysulfide dissolution have proven to be major obstacles for a practical Li-S battery implementation. In this work, we develop a novel strategy to suppress polysulfide dissolution using hydrofluoroethers (HFEs) with bi-functional, amphiphlic surfactant-like design: a polar lithiophilic "head" attached to a fluorinated lithiophobic "tail." A unique solvation mechanism is proposed for these solvents whereby dissociated lithium ions are readily coordinated with lithiophilic "head" to induce self-assembly into micelle-like complex structures. Complex formation is verified experimentally by changing the additive structure and concentration using small angle X-ray scattering (SAXS). These HFE-based electrolytes are found to prevent polysulfide dissolution and to have excellent chemical compatibility with lithium metal: Li||Cu stripping/plating tests reveal high coulombic efficiency (>99.5%), modest polarization, and smooth surface morphology of the uniformly deposited lithium. Li-S cells are demonstrated with 1395 mAh g-1 initial capacity and 71.9% retention over 100 cycles at >99.5% efficiency-evidence that the micelle structure of the amphiphilic additives in HFEs can prohibit polysulfide dissolution while enabling facile Li + transport and anode passivation.
An artificial neural network is used to evaluate the effectiveness of six metrics and their combinations to assess whether slip transfers across grain boundaries in coarse-grained oligocrystalline Al foils. This approach extends the one- or two-dimensional projections formerly applied to analyze slip transfer. The accuracy of this binary classification reaches around 87% for the best single metric and around 90% when considering two or more metrics simultaneously. Here, the results suggest slip transfer mostly depends on the geometric relationship between grains. Training a double-layer network having 10 nodes per hidden layer with 40 measurements is sufficient to render the maximum accuracy.