Viscosity Prediction of Lubricants by a General Feed-Forward Neural Network.

Modern industrial lubricants are often blended with an assortment of chemical additives to improve the performance of the base stock. Machine learning-based predictive models allow fast and veracious derivation of material properties and facilitate novel and innovative material designs. In this study, we outline the design and training process of a general feed-forward artificial neural network that accurately predicts the dynamic viscosity of oil-based lubricant formulations. The network hyperparameters are systematically optimized by Bayesian optimization, and strongly correlated/collinear features are trimmed from the model. By harnessing domain knowledge in the selection of features, the quantitative structure-property relationship model is built with a relatively simple feature set and is versatile in predicting the dynamic viscosity of lubricant oils with and without enhancement by viscosity modifiers (VMs). Moreover, partial dependency, local-interpretable model-agnostic explanations, and Shapley values consistently show that the eccentricity index, Crippen MR, and Petitjean number are important predictors of viscosity. All in all, the neural model is reasonably accurate in predicting the dynamic viscosity of lubricant solvents and VM-enhanced lubricants with an R2 of 0.980 and 0.963, respectively.

Fast water desalination by carbon-doped boron nitride monolayer: transport assisted by water clustering at pores.

The well-being of the ever-escalating world population hinges largely upon the adequacy of clean, fresh water. Desalination is one of the most promising approaches in such an endeavor. Using molecular dynamics simulations, we take a close look at nanoporous hexagonal boron nitride nanosheets as desalination membranes, and study how C dopants affect their performance. The calculations predict that the desalination performance of C-doped BN membranes compares favorably to that of MoS2 membranes: the water flux through the 0% (0CB-0CN), 25% (3CB-0CN), 75% (3CB-6CN), and 100% C terminated BN membrane (6CB-6CN) is 29.9, 47.5, 95.3, and 81.5 molecules ns-1 per pore, respectively, and there is a strong correlation between the water flux and the axial diffusion coefficient. Through our study of the effect of C content on the desalination performance, it is found that more clustering of water molecules at membrane pores due to a smaller hydration free energy and pore energy barrier assists water transport through the pores, and allows a greater water flux.

Out-of-plane structural flexibility of phosphorene.

Phosphorene has been rediscovered recently, establishing itself as one of the most promising two-dimensional group-V elemental monolayers with direct band gap, high carrier mobility, and anisotropic electronic properties. In this paper, surface buckling and its effect on its electronic properties are investigated by using molecular dynamics simulations together with density functional theory calculations. We find that phosphorene shows superior structural flexibility along the armchair direction allowing it to have large curvatures. The semiconducting and direct band gap nature are retained with buckling along the armchair direction; the band gap decreases and transforms to an indirect band gap with buckling along the zigzag direction. The structural flexibility and electronic robustness along the armchair direction facilitate the fabrication of devices with complex shapes, such as folded phosphorene and phosphorene nano-scrolls, thereby offering new possibilities for the application of phosphorene in flexible electronics and optoelectronics.

Robust magnetic domains in fluorinated ReS22monolayer.

Transition metal dichalcogenides are layered materials that are typically bound together by van der Waals forces. An exception in the family is ReS2; the geometrical distortion of its lattice structure due to Peierls distortion dimerizes Re atoms to form zigzag chains, thus decoupling the layers electronically and vibrationally. This in turn reduces the layer dependence of its physical and chemical properties. In order to tailor the properties of ReS2 monolayer, an alternative way by fluorination above and between Re chains in the lattice is investigated. The results of density functional theory calculations show the site-dependent electronic properties of fluorinated ReS2: (i) F atoms above the Re chains induce metallic mid-gap states which are ferromagnetically-coupled within the Re chains, and antiferromagnetically-coupled between chains; (ii) F atoms between the Re chains induce semiconducting mid-gap states which are non-magnetic. Unlike other states observed, the mid-gap states associated with top sites above the Re chains are generally insusceptible to the external electric field. The electron localization and negative Laplacian plots show that not only bonds between F and S atoms are ionic in character, the Re chains are also coupled ionically. The emergence of these robust metallic mid-gap states in the localized domains suggest that electrons could conduct along the Re chains by hopping. The electron conduction in such functionalized ReS2 is anisotropic in nature, and hence could be applied in spintronic devices, such as spin-transfer torque and spin-wave logic devices.

Re-ordering chaotic carbon: origins and application of textured carbon.

Formation of nanocrystals with preferred orientation within the amorphous carbon matrix has attracted lots of theoretical and experimental attentions recently. Interesting properties of this films, easy fabrication methods and practical problems associated with the growth of other carbon nanomaterials such as carbon nanotubes (CNTs) and graphene gives this new class of carbon nanostructure a potential to be considered as a replacement for some applications such as thermal management at nanoscale and interconnects. In this short review paper, the fabrication techniques and associated formation mechanisms of these nanostructured films have been discussed. Besides, electrical and thermal properties of these nanostructured films have been compared with CNTs and graphene.