This study presents algorithms for additively manufacturing semi-woven carbon fiber composite skins over singly and doubly curved non-planar surfaces through an innovative remapping/wrapping algorithm. The algorithm efficiently generates 5-axis continuous fiber deposition paths from 2D fiber layout patterns in a way that reduces geometric distortion in fiber placement that normally occurs when planar 2D toolpaths are linearly projected onto non-planar surfaces. A customized 5-axis 3D printer equipped with a modified continuous carbon fiber deposition head is used to validate feasibility of the algorithm by printing a semi-woven carbon fiber skin onto a hemi-spherical surface. Toolpaths are smoothened prior to printing to reduce jerking motions in the 5-axis machine's A/B axes.
Composite winding processes offer the possibility to produce lightweight pressure vessels, a key element in hydrogen mobility, with the potential to reduce global carbon dioxide emissions. Digitizing winding processes has significantly gained attention because of the potential to improve the ecological and economic sustainability of both the process and the product. Numerous process parameters influence manufacturing performance and overall sustainability. This necessitates precise parameter adjustment. However, computing process parameters is not trivial and often imprecise, especially with regard to process-related fluctuations. Consequently, inline measurement systems are needed to acquire actual parameter values and enable process simulations, the generation of digital twins, and closed-loop controls. This paper provides an overview of measurement principles and sensors for monitoring composite winding processes, addressing their implementation, as well as studies on their general feasibility and accuracy. Presented are principles for monitoring the input material, fiber tension, temperature, length and speed, as well as methods for acquiring impregnation properties (resin or volatile content), tack, and fiber deposition geometry (winding angle, gaps, overlaps). The review critically compares the methods’ advantages and restrictions concerning inline use in composite winding and assesses their technology readiness levels as well as future potential for industrial inline process monitoring in composite winding. A decision aid is provided for selecting suitable sensor principles according to the targeted process or material parameters. The early publications presented in this overview address simple measurements and date back to the mid-1990s, while data-driven technologies such as artificial intelligence are the subject of current research.
Carbon fiber-reinforced polymers are increasingly employed in load-bearing structures such as thick reservoirs, pipes or flywheels where cure-induced defects can impair performance and dimensional accuracy. This study investigates whether interlaminar performance and microstructure quality are preserved in tow-wound composites fabricated via partial pre-curing and in situ compaction, which proved to mitigate the risk of thermal overshoot and stress build-up. A proof-of-concept manufacturing line is developed to produce wound rings under varying pre-cure and compaction conditions. Interlaminar performance is evaluated through short-beam and double cantilever beam tests, supported by microscopy. Results show void contents below 3%, although the fiber volume fraction (33%-41%) remains lower than for prepreg references (53%-58%). Interlaminar properties are comparable between tow-wound and prepreg laminates, with no significant degradation due to pre-curing or compaction. The results demonstrate that partial pre-curing combined with local compaction does not degrade interlaminar performance, supporting the feasibility of such strategies for efficient and automated manufacturing.
Precise control of gripping forces is essential for automated preforming of technical textiles to prevent material damage while ensuring reliable handling. This study investigates the holding force thresholds causing shear deformation in four glass fiber-based fabrics: one woven fabric and three non-crimp fabrics (NCFs). A novel experimental setup combining a universal testing machine, a coanda gripper, and a laser profile scanner was used to detect the onset of relative motion and determine the critical holding forces. Preforming tests with two coanda grippers evaluated the maximum forming height using two different pressure configurations. The results demonstrate a linear relationship between applied pressure and holding force, with critical forces inversely correlated to air permeability. The analysis shows that friction properties and flexural stiffness are critical factors for bond reliability, with flexural stiffness exhibiting a strong negative correlation (HR = 0.013) with the risk of failure. The regression analysis also shows that optimal alignment is crucial. These results underscore that strategic material selection and optimal fiber alignment can positively influence the forming height.
This work evaluates tooling-reduced ribbing by printing thermoplastic stiffeners onto thermoset composite skins via co-cured, directly printable thermoplastic interphases. A helicopter door demonstrator provides the application context for localized reinforcement features in regions where SMC processing is constrained by limited flow and tooling-driven draft angles. Two material routes were investigated on M18/1 CFRP coupons as repeatable screening platforms. PEI served as a high-temperature reference interphase for printing PEI and PPSU, while PC was evaluated as a lower-temperature interphase for printing PETG. Differential scanning calorimetry was used to define substrate-temperature levels, and single-lap-shear tests provided first-order indicators of interface quality and process sensitivity. The results reveal a coupled thermal process window: insufficient activation leads to weak bonding, whereas excessive thermal exposure reduces lapshear strength despite successful deposition. An analytical model further quantifies the lightweighting potential of eliminating demolding draft angles, showing double-digit crosssectional mass savings for typical rib proportions.
This study evaluates the low-velocity impact response of aerospace-grade epoxy laminates (M18/1) interleaved with four distinct thermoplastic films (PEI, PES, PSU, PC). Impact performance was correlated with quasi-static fracture toughness to assess the transferability of intrinsic polymer properties to dynamic loading. The results reveal a fundamental trade-off governed by chemical compatibility. Chemically compatible interphases (PEI, PES) successfully bridged the stiffness mismatch, increasing the damage initiation threshold by up to 23%, although total energy absorption plateaued due to matrix saturation. Conversely, the incompatible Polycarbonate (PC) system exhibited a mechanism shift: despite high intrinsic ductility, low Mode I adhesion (GIc) caused immediate decoupling, triggering a pseudoplastic sliding mode that increased energy absorption (+10%) at the cost of structural integrity. Polysulfone (PSU) revealed an inefficient compromise, suffering from premature adhesive failure without providing significant energy dissipation. These findings challenge established propagation-based models, identifying Mode I initiation toughness as the critical ‘gatekeeper’ for structural toughening in hybrid laminates.
This study compares amorphous PEI, PES, PSU, and PC films co-cured as functional interlayers in aerospace-grade CFRP laminates based on HexPly (R) M18/1. A unified quasi-static test matrix comprising single lap shear, interlaminar shear strength, and Mode I and Mode II fracture toughness was applied to establish a directly comparable mechanical ranking. All interlayer systems preserved structural integrity, increasing single-lap shear strength by 6.2% to 13.3% to about 16 to 18 MPa. Interlaminar shear strength remained in a range of 48 to 59 MPa, but depended on interlayer placement. Fracture testing revealed two distinct response types. PEI and PES showed the most effective toughening, with PEI reaching the highest Mode I toughness at 590 J/m & sup2; (+59.0%) and PES the highest Mode II toughness at 6180 J/m & sup2; (+518.6%). In contrast, PSU and PC showed poor opening-mode resistance. Overall, the results demonstrate that interlayer selection governs both the fracture mechanism and achievable damage tolerance.
The efficacy of silane coupling agents in polymer nanocomposites is often restricted by limited reactive sites and insufficient physical engagement with the matrix. In this work, a multi-amino covalent organic framework-derived silane coupling agent (COF-SCA) was synthesized via a one-pot solvothermal method to engineer a stable filler-matrix interface. Unlike conventional small-molecule silanes, this macromolecular agent exhibits a porous, amino-rich framework. COF-SCA was applied to functionalize carbon nanotubes (CNTs) and hexagonal boron nitride (h-BN). Structural analysis indicates the formation of a continuous, rough coating on the nanofillers. Mechanical testing shows enhanced load-bearing capability, with tensile strength increases of 23% (CNTs) and 22% (h-BN) compared to APTES-treated systems. This enhancement is attributed to synergistic effects from abundant amino groups and the rigid porous framework.
Carbon-fibre-reinforced polymer (CFRP) robotic grinding is strongly anisotropic and stage-dependent, hindering reliable surface-roughness prediction. This study proposes a physically constrained multimodal framework for predicting Sa by fusing Z-axis vibration and post-grinding surface texture. Experiments at nine grinding angles were segmented into Entry, Stable, Exit and Whole stages. Grey Relational Analysis was used to select informative vibration and texture features, and a fifth-order polynomial augmentation strategy expanded the dataset from 36 to 148 samples. An early-fusion artificial neural network was trained and compared with vibration-only and texture-only models. On an independent test set, the fusion model achieved an MAE of 0.111 mu m, an RMSE of 0.131 mu m and an R & sup2; of 0.898, reducing MAE by 56.6% and 45.6%, respectively. Within +/- 0.30 mu m tolerance, it achieved a 100% pass rate, demonstrating robust and engineering-reliable Sa prediction for CFRP robotic grinding.
Studies on radiation-resistant electrical insulation for superconducting magnets of nuclear fusion power plants are discussed. Metal oxide nanoparticle-incorporated organic polymer composite (MONAP) insulation films were manufactured and studied for their dielectric breakdown strength after exposure to neutrons and proton radiation at varying levels. Samples with up to 5 wt% of nanoparticles (NPs) of SiO2, MgO, and ZrO2 were studied. The samples were exposed to radiation of up to 2 & times; 1016 neutrons/cm2 and 200 Mrad of protons. The addition of the nanoparticles improved the radiation resistance of the MONAP, and the performance depended on the type of NPs and the radiation exposure. Dielectric breakdown measurements at 77 K showed little degradation. Enhancement of dielectric strength was observed in some instances.
A multiple thermally assisted piercing process has been developed as a method of making equally spaced holes in thermoplastic composites. The consequences for the mechanical properties of the composite of introducing a limited set of inline holes into cross-ply laminates have been investigated. Open-hole tension and Iosipescu shear testing has been carried out on specimens containing drilled or pierced holes aligned with the direction of loading; microscopy and digital image correlation techniques have also been used to investigate local changes in fiber orientation and strain distributions under load. The strain fields for inline holes in drilled and pierced specimens under tensile loading can be understood in terms of local changes to the modulus as a consequence of the piercing or drilling process; in addition, some features of the strain fields can be predicted with the aid of a shear-lag model developed for modeling matrix cracking in cross-ply laminates. Although significant differences were found between the strain fields of the drilled and pierced specimens, no consistent improvement in strength was observed for the pierced composites compared to drilled composites for different holes spacings. Under shear loading, the pierced composites were found to have a significantly poorer response compared to drilled composites, which is related to the premature collapse of the holes in shear due to (a) localized fractures in regions of low fiber volume fraction and (b) intact fibers being pulled across the holes causing hole collapse.
To address the issues of low photocatalytic efficiency and poor hydrophobic durability of traditional coatings. A photocatalytic self-cleaning superhydrophobic coating was prepared by compounding anatase TiO2 with silane reagents and SiO2, and using the sol-gel method combined with a hydrothermal synthesis process. Performance tests show that the photocatalytic degradation efficiency of this coating reaches 96.7%, the contact Angle is maintained above 140°, the mass loss is only 0.5% after 28 days of acid and alkali erosion, the composition contents of O, Si, Ti and C elements are 51.2%, 21.6%, 16.4% and 10.2% respectively, and the pore size is 10.1 nm. The specific surface area is 16.4 m2 /g. This coating, through the coupling of photocatalysis and superhydrophobic properties, achieves the optimization of long-term self-cleaning, corrosion resistance, anti-aging and durability of building exterior wall materials.
Aiming at the complex mechanical behavior of polypropylene self-reinforcing composites at different temperatures and the insufficient prediction accuracy of existing models, this study proposed an improved phenomenological constitutive model considering temperature dependence. The experimental results show that the strength of materials drops sharply at high temperatures ranging from 40 degrees C to 120 degrees C, while the stiffness and strength increase significantly at low temperatures ranging from -30 degrees C to 0 degrees C. Model verification shows that this model can accurately capture tensile, bending and shearing behaviors at different temperatures: the overall matching rate between the tensile curve and the experimental value reaches 98.5%, and the prediction deviation of shear performance at 20 degrees C is only 0.2 MPa, demonstrating excellent multi-stage temperature transition prediction ability. This research provides a reliable mechanical property prediction tool for the engineering application of this material under complex temperature conditions.
Several studies have shown that using Automated Fiber Placement (AFP) technology can improve productivity and resource efficiency in manufacturing sandwich structures. In this process the bridging effect occurs, which negatively impacts the processability and component quality. The aim of our work is to investigate the reduction of the bridging effect by analyzing the influence of prepreg and adhesive film tackiness, path planning parameters, and AFP process parameters on the bridging effect. To achieve this, the tackiness was first characterized in the probe tack test and then AFP lay-up tests were carried out. Suitable tack levels were identified to reduce the bridging effect. The results indicate that the unbonded area can be reduced by a factor of up to 6.9 by adjusting path planning parameters and process parameters statistically significantly influence the bridging effect. In conclusion, guidelines were deduced for sandwich structure's productive, sustainable, and high-quality AFP manufacturing.
Machine learning approaches that integrate physical laws with data-driven models are transforming process optimization and quality assurance in polymer matrix composite manufacturing. This review synthesizes recent developments in neural metamodels for injection molding, spatio-temporal digital twins for resin infusion, and symbolic-regression surrogates for vacuum networks. Article identifies remaining challenges—such as extension to semicrystalline systems, uncertainty quantification under real-world noise, and deployment on industrial platforms—and outline strategies for addressing them. Building on these insights, a unified physics-informed surrogate concept is proposed that leverages temporal encoders, recurrent propagation, and multi-output decoders with embedded conservation constraints. This model is designed for rapid prediction of part quality metrics, cure state, flow front progression, and temperature fields, and supports gradient-based inversion for closed-loop control in advanced composite processing.
Transportation emissions are a major driver of global warming, making vehicle greenhouse gas reduction essential. Lightweight design, such as hollow shafts and tubes, lowers energy use by optimizing stiffness-to-mass ratios. Fiber-reinforced polymers, especially thermoplastic variants, excel in these applications due to their high specific stiffness, customizable mechanical properties, and scalable manufacturing. This study introduces two novel methods for producing braided hollow carbon fiber-reinforced polyamide 6 profiles: rotational molding for straight preforms and bladder-assisted molding for curved preforms. Numerical simulations of braiding were compared to actual braid architectures, revealing both the capabilities and current limitations of the simulation software for tape-based braiding. Analyses included fiber angle, cover factor, and CT-based wall thickness measurements. The potential for increasing consolidation pressure in rotational molding is shown by means of a theoretical analysis. Bladder-assisted molding produced fully consolidated, minimally wrinkled curved profiles, proving the feasibility of manufacturing high-quality curved braided profiles without post-consolidation forming.
Thick-section composites (10-100 mm) are increasingly used in structurally demanding applications. Growing interest in sustainable alternatives has driven the development of recyclable, room-temperature-processable liquid thermoplastic resins to replace thermosets in vacuum-infused composites. However, managing the thermal effects of polymerisation to avoid boiling and defects remains a challenge in thick-laminate manufacturing . While low-exotherm grades are available, their behaviour in thick laminates remains poorly understood. This study examines the exothermic polymerisation of Elium (R) 188 XO, a low-exotherm thermoplastic resin, in laminates with thicknesses of 9.5 mm, 17.9 mm and 26.4 mm. Process times are presented to support implementation. Using five embedded thermocouples, maximum temperatures of 86.5 degrees C, 92.6 degrees C and 93.9 degrees C were recorded, all remaining below the resin's boiling point. The results indicate a progressive increase in interlaminar temperature with increasing laminate thickness. Ambient-temperature-adjusted data showed peak increases of 2.7 degrees C and 3.1 degrees C between successive laminate thicknesses. These findings provide critical insights into polymerisation behaviour, informing process optimisation and industrial adoption.
Representative volume element (RVE) models have been widely used to study the influence of additive manufacturing parameters on the mechanical properties of 3D-printed components. However, prior work primarily focused on simple infill patterns, often neglecting the complexities of interwoven geometries. This study introduces a methodology that integrates finite element analysis (FEA) with a statistical approach to predict the mechanical properties of novel interwoven structures produced by the z-stitching technique. Enhanced performance characteristics are explored by strategically aligning and stitching filaments in multiple planes. The FEA approach is grounded in meso-mechanical analyses using RVEs to predict effective orthotropic properties, specifically evaluating stress-strain behavior, modulus of elasticity, and strength. Mechanical properties derived from FEA-based homogenization were validated against experimental tensile tests. The combined use of numerical modeling and statistical analysis enables an efficient, iterative design process for complex 3D-printed structures, reducing computational demands and experimental efforts.
Distortion in carbon fiber woven fabrics significantly impacts composite mechanical performance through defective fiber tow distribution. This work proposes a machine vision method to locate defective areas, identify defects, and describe fiber tow distribution patterns. A back-lighting imaging system was designed to minimize surface reflection interference, enabling high-quality fabric image acquisition. We developed the Light Transmission Algorithm (LT) analyzing voids at fiber tow intersections to calculate void-to-fabric ratios, providing qualitative and quantitative distribution indicators. A defect recognition method combining isometric and random feature sampling enables segmentation of abnormal fiber distribution regions through standard sample comparisons. The system achieves 95%-100% identification accuracy. The proposed methods demonstrate strong interpretability and robustness in assessing the quality of carbon fiber woven fabrics, addressing critical challenges in local defect detection while enabling comprehensive distribution analysis.
