ABSTRACT Continuous carbon fiber‐reinforced thermoplastic composites (CCFRTCs) have garnered significant attention in the aerospace sector due to their exceptional mechanical properties. However, manufacturing components with complex shapes remains challenging because of varying geometric and functional requirements. Additive manufacturing (AM) reduces material waste and improves efficiency. Thus, AM has become a transformative solution for producing complex CCFRTC components. Traditional AM techniques rely on conventional heat sources, which face limitations in precision, energy efficiency, and process controllability. In contrast, laser‐assisted automated fiber placement (L‐AFP) enables localized and rapid heating. This ensures precise thermal management during thermoplastic matrix consolidation, greatly improving production efficiency and quality. Motivated by the need to summarize technical gaps and guide fabrication of high‐performance CCFRTCs for demanding applications, this study seeks to explore L‐AFP technology for CCFRTCs. It elaborates on four aspects: (1) L‐AFP technology principles and equipment; (2) CCFRTC prepreg preparation and modification techniques; (3) L‐AFP process parameter responses; (4) Common defects in formed components. Furthermore, the advantages and limitations of L‐AFP are critically analyzed, and future research directions are proposed to address existing technology shortcomings.

Hybrid perovskites are among the most important photon-to-electricity materials, whose performance is strongly limited by poor heat dissipation. This is due to their significantly low thermal conductivity, and the origin of such low thermal conductivity is still in debate and much less well-documented. In this paper, we experimentally observe that the thermal conductivity of hybrid perovskites is low and possesses a weak temperature dependence. We further show that this ultralow thermal conductivity of hybrid perovskites is caused by the disorder of organic molecules using atomistic simulations. This disorder strongly scatters the lattice vibrations and hinders the thermal transport in hybrid perovskites. Our calculations further show that the temperature dependence of the vibrational thermal conductivity results from the lattice framework transits from ∼T0 to ∼T-1, and the value doubles at room-temperature and increases four-fold at 150 K when the disorder introduced by organic molecules is excluded. Meanwhile, based on atomistic simulations, the thermal energy in hybrid perovskites is found to be transferred by vibrations of the lattice framework (∼48% to ∼65%) and the organic molecules (∼35% to ∼52%). We map heat transport through inorganic octahedral cages and organic molecules, clarify their mutual coupling and interference, and guide thermal management in perovskites.

The precise determination of micromechanical parameters in fiber-reinforced composites remains challenging and subject to significant uncertainty. We introduce a unified Bayesian-renormalization group method (RGM) multiscale inversion framework to address these challenges by integrating physics-based RGM modeling, Bayesian inference, and an artificial neural network (ANN) surrogate. By replacing computationally expensive RGM evaluations, the ANN surrogate enables an acceleration of over 30,000 times, reducing the total Bayesian inference from approximately 4800 h to about 8 min without compromising predictive accuracy. Virtual experiments are first conducted to verify the internal consistency and reliability of the inversion process, demonstrating that the posterior predictions accurately reproduce the target strength distribution. Validation against experimental data further confirms that the reference values fall within the 95% credible intervals of the inferred distributions for in-situ fiber and interfacial strength parameters, highlighting the importance of probabilistic identification under realistic scenarios. The results also reveal the physical differences between insitu and ex-situ mechanical behaviors. Moreover, comparisons with existing inversion methods highlight the superior accuracy and reduced uncertainty achieved by the proposed Bayesian-RGM framework. This work establishes an efficient probabilistic framework for micromechanical characterization of composite materials, with strong potential for broader application in related fields.

Inferring material properties and mechanical parameters from full-field measurements has been a central challenge in mechanics, with wide applications ranging from elastography to aerospace damage detection. The inevitable measurement noise exacerbates the ill-posedness of these inverse problems, necessitating robust uncertainty quantification method. Bayesian inference offers a principled methodology to recast these inverse problems within a statistical framework. However, it faces significant difficulties when the inferred spatial distributions of mechanical properties are represented in high-dimensional discrete forms (“curse of dimensionality”) and/or when highly nonlinear systems with limited sensitivity information for posterior approximations are involved. To address these two challenges, we develop a novel method, named frequency-domain Bayesian inference. This method projects a high-dimensional Bayesian problem into a low-dimensional representation in the frequency domain, and leverages differentiable programming techniques to enable automatic sensitivity analysis during posterior sampling of mechanical parameters. The proposed method operates without any prior datasets and achieves accurate inference with reliable uncertainty estimates, even when only a single experimental measurement is available. Its generality and accuracy are validated across four categories of inverse mechanical problems, including elastic modulus identification, plastic yield stress field identification, external load reconstruction, and thermal conductivity estimation. It is worth noting that the identification of plastic yield stress fields demonstrates the method’s capability to handle history-dependent circumstance with strong nonlinearities. Additional experiments with digital image correlation measurements are further conducted to highlight the practical applications of the proposed method.

Integrating tunable building-blocks, control, and actuation systems in mechanical metamaterials enables automatic, precise, and fast performance reprogramming for physical intelligence. However, limited by the mixing rule of composite materials, large-range and precise reprogramming typically requires manipulating each building-block, dramatically increasing the complexity of control and actuation systems. Here, an ultrahigh local stiffening-sensitivity mechanism is revealed in the rotating-squares-based structure with quasi-single degree-of-freedom, and is harnessed to drastically reprogram the mechanical performance by manipulating only few local points. The metamaterial's macroscopic modulus increases up to 23.5× by stiffening only 2.5% of local points, corresponding to a tuning sensitivity that is two orders of magnitude higher than that of conventional materials. Guided by the structure-performance relation established with a neural network, the metamaterial's performance can be automatically adjusted to match different targets, using ≤4 built-in stiffening actuation switches within 0.1-0.5 s. This work presents local stiffening-sensitive mechanical metamaterials and establishes a new design strategy for efficient mechanical performance reprogramming.

It is a long-standing challenge to harness ideal stress plateau with unobvious peak, large capacity and high steadiness when designing impact-resistant structures. To achieve this goal, bionic hybrid hierarchical lattices (BHHL) inspired by muscles and shells of the lobsters are proposed, which integrates high-capacity stretchdominated and high-stability bending-dominated units. Prescribable deformation pattern and ideal stress plateau are observed from experiments and simulations. Energy absorption efficiency, force efficiency and force steadiness of BHHL respectively reach 54.8 %, 92.9 % and 84.6 % under quasi-static loads, which averagely outperforms five classical structures of same mass by 12.7 %, 32.8 % and 66.7 %. Notably, these superiorities almost remain unaffected in strong impact experiments with energies of 13.1-25.6 kJ. A plastic hinge model with relative error less than 3.8 % is developed to estimate the plateau stress, and can be used to tailor ideal target stress plateau under both quasi-static and dynamic loads. Due to integration of stretch- and bending-dominated mechanisms, specific energy absorption of BHHL is inferior to stretch-dominated but obviously superior to bending-dominated structures of same mass, whilst its other indicators are notably higher than all eight comparative models. This works provides a new pathway to quickly tailor ideal stress plateau for impact-resistant lattices based on metallic constituent material and novel structural design.

By optimizing the BN/SiC dual-layer interphase in SiCf/SiC composites, this work significantly enhances high-temperature mechanical properties. The thickness of the SiC layer was designed to be 0 nm (000T), 500 nm (500T) and 900 nm (900T). Introducing a SiC interphase improved crack deflection, with thickness having little effect on this function. The 900T specimen exhibited excellent ultimate tensile strength (267 +/- 7 MPa) and failure strain (0.89 +/- 0.03 %) at 1350 degrees C in air. Its lower density and Young's modulus increased the proportional limit stress and reduced the crack opening displacement (COD), minimizing the oxidation-induced fiber damage. In-situ tensile tests confirmed smaller COD in 900T than in 500T. Acoustic emission data indicated that an appropriate SiC layer thickness delays fiber fracture, maintaining mechanical properties in oxidative environments. This work provides new and deep insights for the low-cost and efficient preparation of high-performance SiCf/SiC composites.

Designing lightweight protective materials that simultaneously achieve high stiffness, efficient energy dissipation, and damage tolerance under high-velocity impact remains a grand challenge. This study designs a functionally graded, sandwich-like graphene/CNT networks/graphene (Gr/CNTNs/Gr) continuous laminate and utilizes coarse-grained molecular dynamics simulations to resolve its impact response. The laminate dissipates energy through a coordinated sequence involving rapid fragmentation of the top graphene, bending and inter-tube sliding within the CNT networks, and shear-controlled deformation of the bottom graphene that suppresses fragment ejection. Impact velocity governs a transition from global dissipative deformation to localized covalent bond rupture. With the interlayer CNT networks serving as a tunable damping medium, a strict geometric and mass normalization analysis isolates the coupled thickness effects, revealing that both the energy absorption per unit thickness and the specific energy absorption (SEA, 10.4 MJ/kg) simultaneously converge at a unified optimal density of . At , the laminate doubles the SEA of graphene and reduces hazardous fragment kinetic energy by 67% compared to CNT films. Crucially, the laminate retains a load-bearing framework post-impact, enabling effective repair via simple graphene patching to withstand secondary strikes. This balanced performance surpasses steel, Kevlar, and graphene/polymer composites, offering architecture-level guidelines for hierarchical carbon laminates.

In glaciated icing conditions, when the aircraft impacts ice particles in the cloud, the ice particles may rebound, fragment or stick to the aircraft surface, further affecting the ice accretion process. However, the fragmentation mechanism particularly near the minor fragmentation zone is complex and not clear. Here, two particle diameter levels (1.24 mm and 1.56 mm) and three impact velocity levels (7 m/s, 10 m/s, 16 m/s) were set to perform ice particle impact experiment on the self-developed high-speed ice particle impact experimental setup. A Kalman filter tracking algorithm and a projective stereological method were improved and used to track and calculate the volume of the fragments, respectively. According to the statistical analysis of the experimental results, it was found that the impact character number xi is a valid dimensionless number for measuring the degree of fragmentation. The results of the fragment volume calculation were then verified and it was found that the addition of the semi-ellipsoid shape improves the accuracy of the calculation, reducing the mean relative error by more than 9 %. Finally, a correlation between the impact character number xi and the distribution exponent Psi was obtained. Combined with the fragmentation mode probability model and the estimate correlation for the upper cut-off position, a possible model for the fragment volume distribution near the minor fragmentation zone was constructed. Due to the existence of scale invariance and the dimensionless number in this model, it may be applicable to a higher range of xi.

The inverse identification methods of heterogeneous mechanical properties from measured displacements/ strains play a critical role in various engineering fields, ranging from aerospace to medical diagnostics. However, the commonly presence of large-scale measurement missingness significantly exacerbates the ill-posedness of the inverse problem. This dual challenge not only imposes critical limitations on the solution accuracy of conventional methods, but also creates growing requirements of effective uncertainty quantifications for the identification results. In this paper, a novel deep learning in frequency domain (DLfd)-based inverse identification framework is proposed to resolve large-scale and arbitrary distributed measurement missing scenarios. The framework transforms the inferred variables from high-dimensional discretized elastic properties to reduced missing displacement/strain components. This dimensionality reduction process effectively mitigates the challenges in solving the inverse problem and, most importantly, enables uncertainty quantification through the Bayesian inference method. Results demonstrate that even with more than 15% missing data, the L1-error remains as low as 3.846%, and two standard deviation confidence intervals effectively encompass the ground truth, ensuring a reliable evaluation. Furthermore, the identification method is validated on phantom experiment data, successfully reconstructing both the position and shape of the inclusion, confirming the applicability of our framework in practical circumstances.

Non-pneumatic tires (NPTs) represent a revolutionary advancement in the tire industry, owing to their puncture-proof nature and high design flexibility. To achieve both high load-bearing capability and low, evenly distributed ground pressure, the shear band structure of NPTs must balance high compressive stiffness with shear flexibility. To address this challenge, this work proposes a novel shear band structure based on tensegrity metamaterials. By adjusting the stiffness of two types of springs, the proposed shear band enables independent control of its compressive and shear stiffness. The shear band is subsequently integrated with various types of spokes, and the effects of multiple design parameters on the tire’s load-bearing capacity, peak material stress, and contact pressure are systematically analyzed. To achieve performance customization, a Bayesian optimization framework is developed to autonomously explore parameter sets that satisfy performance requirements while minimizing stress and contact pressure. Finally, the feasibility of the tensegrity metamaterial-based shear band design is validated through its successful integration into a four-wheel-drive vehicle model. The proposed shear band offers a new approach for the customized performance and practical application of NPTs.

This paper introduces an innovative multilayer metasheets deformation strategy that enables the deployment of two-dimensional (2D) flat materials into complex three-dimensional (3D) curved surfaces, leveraging the differential Poisson’s ratio mechanism. Based on the design principles, we initially designed and established an analytical model for multilayer metasheets concept. Finite element simulations are then utilized to investigate the impact of varying Poisson’s ratio characteristics among different layers at the unit cell level on the curvature and mechanical characteristics of the metasheets. Building on this unit cell characteristic research, we further explored the potential of metasheets in deploying general curved surfaces through in-plane combinatorial design methods, and developed a more precise shape inverse design approach for deformable metasurfaces constructed from multilayer metasheets. We verify the 3D curved surface deployment mechanism of metasheets concept through a series of quasi-static tensile experiments, which present a good agreement with our simulation results. Finally, this paper further discusses the potential applications of the multilayer metasheets concept. The multilayer metasheets concept offers a fresh perspective for introducing mechanical metamaterial into the 3D shape-shifting techniques, broadening the path for the application of curved surface deployment in more general engineering scenarios.

Solid-state lithium metal batteries have garnered considerable interest as next-generation energy storage devices owing to higher energy density and safety. However, uneven deposition at the Li anode/solid electrolyte (SE) interface during charging induces the growth of Li dendrites, posing significant safety risks due to potential short circuits. The interface evolution is intrinsically coupled with mechanical contact between the Li anode and SE, where external pressure plays a critical role. In this paper, we develop a mechano-electrochemical bi-coupled phase-field model to simulate Li dendrite growth under various loading conditions, thereby elucidating and quantifying the impact of external pressure - including both stack and lateral pressures on Li dendrite growth. Our key findings include: 1) The lateral widening and vertical penetration of Li dendrites can be inhibited under lateral pressure and stack pressure, respectively. Notably, the length and width of the Li dendrites are considerably reduced when stack and lateral pressures are simultaneously applied. The direction of inhibition is closely associated with the regions/branches which maintain higher stress and smooth surface. 2) Larger external pressure decreases the Li dendrite area and enhances space utilization, due to the reduced overall reaction rate at the Li dendrite/SE interface. 3) The uniformity of electrochemical reaction at Li dendrite/SE interface is improved under equal large stack and lateral pressures. These insights provide essential guidance for pressure management strategies in battery design.

To accurately predict the tensile stress-strain behavior of unidirectional fiber-reinforced ceramic matrix composites (FRCMCs) considering interphase and Coulomb friction, this paper develops a comprehensive micro-mechanics model through in-depth analyses of micro-damage evolutions, including matrix cracking, interfacial debonding, and fiber fragmenting. The critical role of interfacial friction in the nonlinear tensile response of FRCMCs is highly emphasized in this model. Thereby, Coulomb friction, instead of the typically assumed constant friction, is adopted, and meanwhile, the effects of interphase thickness, Poisson effect, interfacial roughness, and residual stress are carefully incorporated. Comparison with previous experimental results indicates that the model successfully predicts the tensile response for various interphase thicknesses and theoretically elucidates the mechanisms behind the non-monotonic influence of interphase thickness on ultimate strength. Based on this model, the impacts of interfacial characteristics, interphase properties, and temperature on tensile behavior are systematically analyzed. The findings indicate that elevating interfacial friction significantly enhances the mechanical performance of FRCMCs, and a relatively thin (similar to 100 nm) and low-textured interphase is preferred when no brittle fracture occurs. Moreover, the study analyzes the length-dependent strength in scenarios of interfacial separation, exhibiting a distinct decrease-and-increase trend with composite length due to fiber pull-out effects. The study provides valuable guidance for the further interphase design of FRCMCs.

This study examines the isothermal precipitation behavior of V(C,N) in V-containing high manganese steel through stress relaxation experiments and TEM analysis. By comparing experimental and theoretical PTT curves, it is shown that deformation shifts the curves left due to stored energy. For 0.1V steel, the PTT curve displays a "C" shape with a nose temperature of 830 degrees C and a minimum incubation time of 4.2 s. In 0.2V steel, a "double-C" shape emerges, reflecting N-rich and C-rich V(C,N) precipitation. Grain boundaries dominate nucleation, but decreasing temperature increases the driving force, enabling dislocation and rare homogeneous nucleation. Higher V content accelerates precipitation kinetics, yielding more and larger precipitates, enhancing the precipitation potential under similar deformation conditions.

Metamaterials programmed with target rate-dependent mechanical properties are efficient platforms for realizing advanced functionalities. Yet, the loading rate-dependent mechanical property programming has received limited attention. Here, the “stair-building” strategy is employed in the rate domain by combining the bistability with viscoelasticity. An arbitrary target curve in the programmable space can be approximated by a “stair” built by two kinds of “bricks”. The “bricks” can be realized by a dual-bistable unit, constructed by two bistable structures in series. The dual-bistable unit can switch between two efficient stable phases without inducing changes in the global morphology. Such a unit exhibits N-shaped stress-strain curves at both efficient stable phases with different peak values, resulting in different heights of “bricks”. Moreover, the N-shaped curves have rate-dependent peak values, indicating that the heights of “bricks” change with loading rate. The “stair-building” strategy is realized by array-structured mechanical metamaterials based on dual-bistable units. Different stress-strain curves under various loading rates can be reprogrammed in the same piece of metamaterial by intentionally selecting the efficient stable phases of units. Besides, the rate effect of the metamaterial can also be tuned by reprogramming stress-strain curves under both low and high loading rates, respectively. This reprogrammable metamaterial is promising in smart vibration isolators and adaptive energy absorbers.

To better predict the stress-strain relationship during unloading/reloading cycles in unidirectional fiber-reinforced ceramic matrix composites (FRCMCs), a comprehensive micromechanical hysteresis loop model is proposed, considering Coulomb friction and incorporating effects of interphase thickness, Poisson effect, residual thermal stress (RTS), and interfacial roughness. Based on the model, the hysteresis behavior is categorized into three distinct domains, i.e., small debonding energy (SDE), large debonding energy (LDE), and overlarge debonding energy (OLDE). For FRCMCs with low debonding toughness and thin interphase, the SDE scenario is more prevalent. By applying the model to interface characterization, we propose a more scientific set of parameters for interfacial performance description, including frictional coefficient, initial interfacial radial pressure, debonding toughness, and axial RTS. Furthermore, a method for obtaining the interfacial parameters based on tensile hysteresis tests is established, taking into account all the three domains. Especially, an innovative two-stage fitting approach is designed to derive interfacial properties for the SDE case, addressing the gaps in prior research. The comparison of the current model with experimental data for Cf/PyC/SiC and Nicalon/CAS composites demonstrates the reliability in hysteresis testing analysis.

Understanding the thermal transport of colloidal quantum dots (QDs) in aqueous environments is critical for applications such as bioimaging, and nanomedicine. However, the intricate interplay between surface wettability and the structure of the interfacial water layer complicates the thermal transport mechanisms. This study employs molecular dynamics simulations to investigate the thermal transport behavior of QD superlattices in water, focusing on the effect of surface wettability and water concentration. Our results reveal that variations in surface wettability significantly influence interfacial thermal transport properties by altering hydrogen bond formation and the spacial dirstribution of interfacial water molecules. Hydrophilic QDs exhibit the highest thermal conductivity and interfacial thermal conductance, attributed to the penetration of water molecules into the void regions between QDs. The thermal conductivity decomposition further indicates a transition from conduction-dominated to a convection-dominated mechanism as the water concentration increases. These findings provide a theoretical framework for understanding the role of wettability in QD thermal transport and offer guidance for optimizing thermal properties through tailored surface modification of QDs.