Shaohua Chen

Recent experimental discovery of the ultrahigh ductility of the body-centered cubic (BCC) Nb45Ta25Ti15Hf15 refractory multiprincipal element alloy (RMPEA) exhibits a clear distinction from the MoNbTaW RMPEA with low ductility. To unravel the physical origins of this remarkable ductility, this study employs first-principles calculations to investigate the factors governing intrinsic ductility of the RMPEAs. The energy calculation results indicate a stronger tendency for double cross-slip, dislocation multiplication, and twinning deformation of the Nb45Ta25Ti15Hf15 RMPEA. Integrated with fracture criteria, it is further confirmed that multiple dislocations and twins can emit from the crack tip in this alloy. Analysis of the electron structure reveals that the compositional substitution induces the charge delocalization, bonding metallization and a higher and smoother density of states curve near the Fermi level of the Nb45Ta25Ti15Hf15 RMPEA. Furthermore, increasing Nb content generally raises the calculated fault and surface energies, while favorable plasticity tendencies persist across the investigated BCC configurations. Nb addition can increase the energetic barrier to fault formation and dislocation nucleation without suppressing the intrinsic propensity for crack-tip plasticity over a broad composition range. These findings can provide energetic and electronic insights into the intrinsic ductility of the BCC RMPEAs containing group IV elements.

To address the limited generalization and weak physical consistency of purely data-driven surrogate models with limited CFD samples, a physics-informed stacking ensemble surrogate model (PISM) is developed by exploiting the complementary strengths of random forest (RF) and extreme gradient boosting (XG). Physical features, including opening ratio and aspect ratio, are incorporated as inputs, while equivalent slip length replaces pressure drop as the supervised target. For parameter optimization of spanwise-grooved drag-reducing microstructures, a framework integrating CFD simulations, surrogate modeling, and a genetic algorithm is established. Groove width and depth are treated as design variables, and a pressure-drop dataset is generated through Latin hypercube sampling and gas–liquid two-phase flow simulations. Comparisons with RF, XG, and other models show that PISM achieves a coefficient of determination of 99.39% for pressure-drop prediction under limited-sample conditions, with higher accuracy and stability and lower prediction bias in the low-pressure-drop region. CFD validation further confirms that the The structure obtained using PISM-GA achieves the highest drag-reduction rate. These results demonstrate that integrating physical information with data-driven methods improves surrogate-model accuracy, generalization, and optimization reliability, providing an effective approach for efficient drag-reducing microstructure design.

Partial-slip boundaries have been demonstrated to be effective at drag reduction of underwater vehicles. However, in practical applications, covering the entire body surface with slip materials is cost-inefficient. To address this issue, a deep learning-based approach is developed to optimize the spatial distribution of slip boundaries on the SUBOFF bare-hull model under prescribed engineering constraints. First, a convolutional neural network surrogate model is trained to predict drag efficiently, enabling rapid evaluation of different slip-boundary configurations. The optimal distribution is then determined through systematic exploration of the parameter space. Numerical simulations of the optimized configurations reveal two key findings: (1) slip boundaries significantly influence the skin-friction drag, whereas their effect on pressure drag remains negligible and (2) a well-designed distributed partial-slip boundary can effectively regulate the development of near-wall turbulence, leading to enhanced the overall drag reduction performance. This study provides a promising framework for the intelligent design of drag-reducing surfaces for underwater vehicles.

Directional liquid transport plays a critical role in many fields, especially in microfluidics and biochemical reagent analysis. How to achieve selective directional transport of different liquids through a simple microstructural surface is still a challenge. In the present paper, a novel open capillary tube with an egg-roll-like shape is designed and fabricated, which is composed of two semi-circular capillary tubes with different radii and an array of inclined microplates distributed on their inner surface. Different from the traditional capillary tube with constant radius, water transports along the inclined direction of the microplate, while the ethanol does in reverse in this curled open capillary tube. It is found that the selective transport direction is determined by the liquid-capillary contact configuration, which depends on both the inclination angle of the microplate and the contact angle of different liquids. The critical liquid contact angle that governs the selective direction and the transport displacement of liquids is predicted theoretically, which is well consistent with the experimental measurements. Based on the above mechanism, several potential application scenarios of the as-fabricated curled open capillary tube are further proposed, including the microfluidic channel diversion and reagent fusion reaction. The obtained results can provide an innovative prototype for the flexible manipulation of various liquids.

MoTe_{2}, a promising material for flexible electronics and straintronics, exhibits rapid strain-induced conductivity changes. However, the underlying mechanisms remain poorly understood. Here, through deep-learning molecular dynamics and density-functional theory calculations, we identify a reversible phase transition between the semiconductive T_{b}^{'} phase and the semimetallic T^{*} phase as the key mechanism driving this phenomenon. This transition, which is highly sensitive to the direction of applied strain, provides valuable insights for optimizing MoTe_{2}-based high-frequency nanoelectronic devices and reducing fabrication-related failures.

Droplet manipulations have important applications in many fields, especially droplet splitting and transport in aseptic operations or biochemical reagent analysis. However, droplet splitting or transport on existing functional surfaces is limited to predesigned microstructures or fixed patterns. It remains a challenge to realize reprogrammable surface microstructures for freely controllable droplet splitting and transport. In this study, a flexible technique for both the multimodal splitting and reciprocating transport of droplets on one surface is proposed. Such a surface is prepared with a facile fabrication method by premixing magnetic particles and softener into the polymer solvent matrix and immersing the solidified matrix in a lubricant. The movable wettability gradient is generated on the surface by an external magnetic field, which can act as an invisible "air knife" to split the droplet in multiple modes. The mechanism and critical conditions of droplet splitting are analyzed and revealed theoretically. Furthermore, the microstructural configurations and surface wettability can be reprogrammed by modulating the magnetic field strength and gradient. Accordingly, the splitting behavior of the droplet is transformed into the reciprocating transport behavior. The influencing factors of such behavior have also been analyzed. The reported reprogrammable manipulation of the droplet on one surface provides a versatile prototype for the actuation of droplets in microfluidic and biological analysis devices.