Most vat photopolymerisation processes remain limited to a single material, so lattice structures combining a brittle and ductile polymer within one part have not been demonstrated. This work introduces a multi-material stereolithography platform with a novel dual-blade coating system that processes two photopolymers within a single vat and can be retrofitted to existing systems. A custom voxel-based pipeline enables material assignment down to [Formula: see text] voxel resolution. Using two chemically compatible photopolymers with a strong mechanical contrast ([Formula: see text] compared to [Formula: see text], [Formula: see text] compared to[Formula: see text]), the aim was to quantify how unit cell topologies govern the tensile performance of multi-material cellular solids.From a design space of [Formula: see text] configurations (five cell topologies with two sizes, three volume fractions and two lattice material combinations), a D-optimal subset of [Formula: see text] configurations was used to train different models, which were assessed against an unseen validation set. The models predicted maximum stress, corresponding strain, and stress at break with an [Formula: see text] up to [Formula: see text] and MAPE of [Formula: see text]. A rule-of-mixtures response dominates, with material switching at low lattice fractions shifting [Formula: see text] up to [Formula: see text]. The demonstrated framework provides a validated workflow for explorative statistics-driven design of multi-material lattices.
Laser powder bed fusion (L-PBF) and laser directed energy deposition (L-DED) are widely applied across industries owing to their high geometric freedom and manufacturing precision. However, porosity and residual stress can arise from complex interactions among process parameters, degrading the mechanical performance and reliability of manufactured parts. This makes accurate prediction, monitoring, and quality assessment essential. Conventional analytical methods rely on simplified assumptions that limit their ability to capture complex, coupled physical phenomena, while data-driven approaches remain data-dependent and lack interpretability. Physics-informed machine learning (PIML), which integrates physical laws with data-driven learning, has recently attracted considerable interest as a means of overcoming these challenges. This review investigates PIML utilisation strategies across the process lifecycle of L-PBF and L-DED. First, the relationships between process parameters and the formation mechanisms of porosity and residual stress are analysed. PIML approaches applied in the pre-, in-, and post-process stages are then comparatively reviewed. Finally, current challenges and future research directions are discussed in terms of prediction accuracy, computational efficiency, generalisation capability, and industrial applicability. This provides an integrated perspective on PIML's application in next-generation intelligent laser metal additive manufacturing (LAM).
Laser beam powder bed fusion (PBF-LB) is reshaping high-end manufacturing. However, the non-equilibrium metallurgical process involves transient phenomena governed by multiphysics coupling, and their correlation with build quality remains unclear. Consequently, defect control still relies on empirical trial-and-error. This paper analyzes the formation mechanisms of critical physical phenomena during PBF-LB. It assesses the principles, capabilities, and limitations of in situ monitoring techniques, including synchrotron X-ray, high-speed optical, schlieren, and acoustic emission. Mechanistic modelling approaches, spanning macroscale thermomechanical modelling, mesoscale melt pool dynamics, and microstructural evolution, together with their applications in defect prediction, are critically analyzed. Furthermore, data-mechanism dual-driven methods, integrating multimodal data and physics-informed machine learning, link melt flow, keyhole instability, vapour recoil, and spatter to porosity, lack of fusion, cracking, and segregation. Finally, recent advances in defect-mitigation strategies are analyzed, including process-window optimisation, spatiotemporal beam shaping, and multiphysical-field assistance, to support defect-free PBF-LB manufacturing.
Enhancing both strength and ductility in commercially pure titanium (CP-Ti) still poses significant technical challenges. Here, a micro-laser powder bed fusion (micro-LPBF) followed by annealing strategy was developed to achieve an exceptional strength-ductility combination in CP-Ti. The resulting microstructure consists of fine rod-like α/α′ crystallographic units, a dense network of Burgers orientation relationship (BOR)-related α/α′ inter-variant boundaries, and abundant < c + a > dislocations. These features arise from preferential formation and retention of α/α′ variant during the non-equilibrium β → α/α′ transformation under the highly localised and rapid heating–cooling cycling inherent to micro-LPBF. The CP-Ti achieves an excellent yield strength of 793 MPa together with a uniform elongation of 8.7%. The high strength originates primarily from the reduced effective dislocation mean free path provided by the fine α/α′ units, with grain-boundary (GB) strengthening contributing approximately 507 MPa. The exceptional ductility is enabled by heterogeneous deformation accommodation among α/α′ crystallographic units with different characteristic sizes, geometrically necessary dislocation (GND) accumulation, and progressive activation of < c + a > dislocations, which provide additional c-axis strain accommodation and sustain strain hardening during plastic deformation. This work offers novel perspectives on the regulation of transformation crystallography and deformation mechanisms in additively manufactured CP-Ti for developing high-performance Ti materials.
Four-dimensional (4D) printing enables additively manufactured structures to morph under external stimuli, but its practical use remains limited by the lack of efficient inverse design methods that directly generate manufacturable material layouts. Existing approaches often depend on iterative optimisation or continuous material fields that require post-processing before fabrication. Here, we present a non-iterative inverse design framework that reformulates 4D printing material programming as a segmentation-based discrete assignment problem. After offline training, the model directly converts a prescribed target deformation into a fabrication-ready bilayer orientation map in a single design-inference pass, without target-specific iterative simulation, smoothing, or topology processing. Using temperature-responsive liquid crystal elastomers (LCEs) as a representative system, we construct an experimentally validated finite element dataset and fine-tune a pretrained SegFormer model through a sparse-to-dense training route using an approximately 3,000-sample simulation-data scale. The predicted discrete orientation layouts are directly compatible with direct ink writing (DIW). Across different spatial resolutions and label granularities, the inverse-designed structures achieve mean relative displacement errors below 7% in experimental validation. Demonstrations from simple bending to complex morphing surfaces with void features confirm the scalability of the method under realistic manufacturing constraints. This work provides a route toward automated, fabrication-ready inverse design for 4D printing.
The combination of additive manufacturing (AM) and topology optimisation enables lightweight designs. However, structures optimised purely for mechanical stiffness often fail during printing due to severe heat accumulation, while those designed solely for laser-induced heat accumulation lack load-bearing capacity. Approximating and controlling the local constraints associated with these conflicting demands presents a numerical challenge. Conventional static aggregation methods frequently suffer from numerical oscillations, failing to reconcile global topological evolution and local peak suppression. To address these limitations, this paper proposes a dynamic adaptive weighted [Formula: see text]-norm (DAW-PN) method. The core innovation lies in a response-dependent Sigmoid weighting mechanism embedded in the [Formula: see text]-norm aggregate, which redistributes aggregation weights from broad design exploration toward high-response local regions during optimisation. DAW-PN is applied to suppress AM-induced thermal hotspots and mitigate thermo-mechanical stress concentrations. Numerical benchmarks show that DAW-PN improves convergence stability and reduces oscillations in the tested cases. These results show that DAW-PN provides a flexible aggregation strategy for improving the approximation and control of high-response local fields in the tested cases. Its scalability is illustrated through a normalised 3D rocket-engine-mount-inspired design example, where the algorithm synthesises a spatial truss network that maintains stiffness while reducing process-induced thermal hotspots in the adopted surrogate model.
Nanoclays have emerged as potent biomaterials for regenerative medicine, characterised by a unique disk-like morphology and a high surface-to-volume ratio, endowing them with inherent bioactivity. These mineral-based nanoparticles can modulate cellular signalling, making them particularly effective for directing osteogenic differentiation. While three-dimensional (3D) bioprinting has revolutionised the fabrication of complex tissue scaffolds, the primary bottleneck remains the development of bioinks that simultaneously provide high structural fidelity and a supportive microenvironment for cells. Consequently, nanoclay-based formulations have moved to the forefront of research, bridging this gap to provide ideal bioink candidates for bone tissue engineering (BTE). This review explores the application of nanoclay to prepare multibiofunctional nanocomposite bioinks, specifically BTE. It begins by providing a comprehensive overview of nanoclay characteristics and their biomedical applications, followed by a detailed analysis of nanoclay-based bioink formulations tailored for bone tissue constructs. It examines various bioprinting techniques and the use of different nanoclay types, emphasising their impact on bioink properties, including improved shear-thinning and enhanced mechanical strength, for bone tissue repair and regeneration. This review further presents the potential of optimised nanoclay-reinforced bioinks to advance bone tissue regeneration and facilitate successful clinical translation by critically evaluating current limitations and outlining future research directions.
Spatter is inherent in laser beam powder bed fusion (PBF-LB), especially large-sized spatter, which is too massive to be effectively removed by the shielding gas flow and thus poses a persistent threat, degrades surface quality. The study proposes an in-situ monitoring informed solid–liquid–gas coupling simulation to reveal two new spatter mechanisms for NiTi alloy, termed Backward-Large (B-L) and Frontward-Large (F-L) spatter. B-L spatter arises not from the previously reported vapor recoil pressure but from sudden vapor impact force, generating a liquid column at the rear keyhole wall that undergoes Rayleigh breakup. F-L spatter originates from forward-jetting vapor or remelted powder, driven by self-generated recoil pressure (distinct from that previously reported of the molten pool) and downward gas flow. Defect formation mechanisms are further elucidated in relation to large spatters. With low initial vertical velocity, B-L spatter falls back onto the melt track, creating protrusion defects exceeding layer thickness, disrupting powder spreading, and potentially terminating prints. A real-time control method using high-speed imaging dynamically adjusts parameters upon detecting spatter fallback, effectively suppressing B-L spatter and preventing ultra-high protrusions. Printed sample characterisation validates these findings, clarifying distinct generation pathways and quality impacts of B-L and F-L spatter.
To enhance the customisation capability of 3D food printing, this study leveraged continuous switching 3D printing and the surimi and beef gels to investigate the mechanism by which soft–hard gel heterostructures influence texture attributes and establish a programmable textural strategy. First, the relationship between the printing process and the consistency of two-phase filaments was examined. Hydrodynamic analysis revealed that the viscoelasticity and the interaction during switching contributed to a reduction in extrusion velocity. Consequently, a variable speed printing strategy was implemented to ensure uniform line width. A patterned printing mode was then employed to precisely control the spatial arrangement of the two slurries, enabling accurate printing of soft-hard gel heterostructures. Furthermore, using voxel structures with varying unit sizes of soft and hard gels as a model system, the influence of gel spatial distribution on texture was studied. Simulation analysis identified that the preferential deformation of soft gel, the skeletal effect of hard gel, and the role of spatial connectivity determine mechanical and textural properties. Based on these insights, interlayer and helix structures were designed to program texture. These architectures endowed the printed products with textural enhancements such as a soft-followed-by-hard texture and improved elasticity.
Physics-informed grain structure engineering in fusion-based laser additive manufacturing offers great promise for producing high-performance high-entropy alloys, especially for low-cost and compositionally-customized laser directed energy deposition (DED). However, detrimental columnar grains prevalently form in DED due to relatively low solidification rates and extremely high temperature gradients, making them difficult to disrupt and still poorly understood. Here, combining in-situ high-speed synchrotron X-ray imaging and diffraction, thermal imaging, and ex-situ electron microscopy approaches, we reveal a dynamic recrystallization (DRX) mechanism associated with the disruption of epitaxial columnar grains during DED of an exemplary high-entropy Cantor alloy. Multimodal melt pool monitoring enables quantitative analysis of solidification parameters and melt flow characteristics with high temporospatial resolution. Experimental observations combined with thermo-kinetic calculations indicate that both intragranular and intergranular DRX can be activated by localised micro-strains. DRX promotes grain refinement and homogeneity by tailoring energy input with optimally increasing laser power and reducing scan speed, and contributes to a tensile-to-compressive strain reversal during real-time deposition. The results of solidification-dependent grain structure evolution down to melt pool scale provide deeper understanding of in-process grain growth kinetics and enable effective methods for disrupting columnar grain growth during DED processing of novel alloys.
Programming TiB reinforcement patterns in laser directed energy deposited titanium matrix composites (TMCs) holds considerable promise for enhancing mechanical properties, but it remains challenging due to unsatisfactory cooling. This study realised in-situ modulation of TiB distributions, including prior β-Ti grain interiors and boundaries in 0.1 wt.% B-modified TiB/Ti6Al4V composites manufactured under low laser energy density (LED) parameters (25.0 J/mm²–40.4 J/mm²) through laser power modulation. Unique intragranular TiB (IG-TiB) was assembled by multiple fine whiskers into cage-like structures. The composites modified by IG-TiB and grain boundary TiB (GB-TiB) displayed excellent strength-ductility synergy at room temperature, particularly at 750 W, which achieved an ultimate tensile strength of 1126.1 MPa and elongation to fracture of 10.6%. In-situ analysis revealed that GB-TiB caused significant local strain concentration at the prior β-Ti grain boundaries and ultimately led to intergranular failure. However, the dispersed IG-TiB retarded strain localisation in single areas by inducing cooperative deformation across multiple soft-oriented α colonies. Improved intergranular bonding and reasonable multiregional-distributed strain partitioning synergistically facilitated TMCs ductility. Strength enhancement was attributed to grain refinement, cooperative load-transfer effects by IG-TiB/GB-TiB, and Orowan strengthening mechanisms. This work proposed new insights into customising intragranular and grain boundary TiB patterns in additively manufactured TMCs.
Numerical simulation is a powerful tool for analysing thermal and mechanical behaviour in laser powder bed fusion (LPBF). Goldak’s heat source model is widely used in additive manufacturing simulations; however, its four heat source parameters are process-dependent and require case-specific calibration, which is time-consuming and resource-intensive. To address this limitation, this study developed a machine learning-based Gaussian process regression (GPR) model to predict Goldak heat source parameters directly from process parameters. The model was trained using data generated from single-track LPBF experiments conducted under 30 laser power and scan speed combinations, each repeated experimentally twice. For each condition, heat source parameters were obtained through a previously developed melt pool profile-based optimisation framework, and six acceptable calibrated parameter sets were retained, resulting in 180 input-output samples. The GPR model was then evaluated on unseen process conditions to assess its predictive capability. Furthermore, thermal simulations using the predicted parameters reproduced experimental melt pool profiles with less than 10% deviation. A comparative study against neural networks and twin Gaussian processes (TGP) was also performed to benchmark predictive performance. The proposed approach provides a data driven framework for estimating heat source parameters from process conditions and supports more efficient calibration workflows in LPBF simulations.
Extrusion accuracy is essential in extrusion-based additive manufacturing because it governs filament-width gradients and multi-material patterning, which can tune permeability and mechanical performance in hydrogel constructs. In progressive cavity pump (PCP) systems, volumetric efficiency is reduced during steady extrusion by leakage through the rotor -stator clearance, while flow is delayed during start-stop events by pressure build-up and compliance in the pump-nozzle system. These effects decouple commanded and delivered flow rates and increase printing errors. A moving-mesh computational fluid dynamics model of a PCP was developed, and reduced-order models were derived to predict steady-state output flow and transient response lag. The models were embedded in a dynamic feedforward compensation strategy (DFCS), through which G-code was augmented with inlet-pressure and rotor-speed commands. Continuous filament-width gradients from 100 to 600 μm were enabled by steady-state compensation, while start -stop deposition in dot arrays was improved by transient compensation, yielding area-based deposition errors of 14–24%. Overall, PCP flow dynamics are linked to executable toolpath commands through DFCS, thereby improving gradient printing and start-stop-intensive hydrogel printing.
The performance optimisation of Triply Periodic Minimal Surfaces (TPMS) structures currently faces a critical bottleneck involving a transition from a density-driven to a geometry-driven paradigm. Traditional design strategies for TPMS lattices are commonly interpreted using the Gibson-Ashby framework, which effectively describes density-dependent mechanical scaling. However, this framework does not explicitly distinguish the effects of local geometric parameters on mechanical properties. Concurrently, constrained by their inherent mathematical frameworks, existing TPMS-based structures struggle to achieve precise customisation of local curvature, which significantly limits the potential for optimisation of TPMS-based structures in high-performance, lightweight design applications. Here, we propose a TPMS-based design method that achieves controlled local curvature modulation through discrete geometric reconstruction in Gyroid lattices, and establish a Curvature-Volume Fraction (CV) model that quantitatively correlates curvature with mechanical properties. To validate this, we fabricated curvature-modulated Gyroid scaffolds using WE43 magnesium alloy via laser powder bed fusion (LPBF). The results demonstrate that curvature modulation reconfigures failure modes from shear band failure to layer-by-layer failure, achieving a ∼33% increase in yield strength and ∼29% increase in elastic modulus compared to standard Gyroids. This work establishes curvature as a design dimension for TPMS-based lattice architectures, opening new avenues for the customisation of additively manufactured lattices.
Spheroid-based bioprinting is essential for reproducing high-cell-density tissues where cell–cell interactions are critical, such as pancreatic tissue. Although spheroids more effectively mimic the in vivo microenvironment compared to conventional culture systems, nozzle clogging remains a major technical challenge that compromises printing resolution and consistency. In this study, we integrated two-way coupled fluid-particle simulations with experimental validation to systematically analyze key factors influencing clogging and resolution stability in high-concentration spheroid printing. Using pancreatic-derived extracellular matrix (pdECM)-based bioink, we analyzed wall shear stress (WSS) and relative velocity distributions under various nozzle geometries and pneumatic conditions. PDMS molding techniques were applied to visualise internal nozzle structures, enabling precise comparisons between experimental and simulation results. Tapered nozzles, particularly 25G and 22G configurations, exhibited more uniform flow characteristics than cylindrical nozzles of equivalent gauge, resulting in reduced shear stress and minimised spheroid aggregation near the nozzle outlet. Optimised nozzle geometry and pneumatic control promoted uniform spheroid distribution while maintaining structural integrity. Notably, the 25G tapered nozzle maintained comparable printing resolution to the 24G cylindrical nozzle while demonstrating significantly more stable extrusion with reduced clogging risk, confirming that nozzle geometry optimisation can simultaneously address both resolution and clogging challenges.
Heat treatment is essential for meeting mechanical-performance requirements in additively manufactured components, but it consumes significant furnace energy and involves coupled time-temperature interactions that make trial-and-error development inefficient. We demonstrate a multi-objective Bayesian optimization (MOBO) protocol for Laser Powder Bed Fused (LPBF) Inconel 718 (IN718) that simultaneously maximises ultimate tensile strength (UTS) and minimises estimated cycle energy consumption. Profilometry-Based Indentation Plastometry (PIP) provides rapid UTS estimates during the campaign, while a calibrated furnace model estimates per-cycle energy consumption (EC). Relative to the average seed baseline, the campaign achieved a 21% increase in UTS and a 118% improvement in the energy efficiency objective (1/EC), with a Pareto hit ratio of 0.33. The identified schedules favour short solution annealing (15 min, 950-1034 degrees C) paired with moderate aging, and microstructural observations are consistent with the inferred trends, indicating that heat treatment primarily modifies segregation products and precipitate populations rather than grain morphology. Within practical processing constraints, the proposed workflow makes resource intensity an explicit optimisation target while retaining metallurgical interpretability and practical experimental cadence.
Multi Jet Fusion (MJF) is a powder bed fusion additive manufacturing technique increasingly adopted in commercial production due to its efficient material utilisation and high throughput. However, MJF-printed objects still suffer from anisotropic reinforcement or low stiffness, and hybrid powders designed to enhance the mechanical performance of parts remain commercially unavailable. In this work, a novel glass bead/glass fibre/polyamide 12 (GB/GF/PA12) hybrid composite was designed to improve powder printability and provide significant isotropic reinforcement. The incorporation of particulate GB improved the powder flowability and decreased part porosity, thus contributing to increased ultimate tensile strength of 55.04, 69.43, and 42.67 MPa in the X, Y, and Z orientations. The designed GB/GF/PA12 composite demonstrated an over 400% increase in the tensile modulus in the Y orientation, reaching up to 7.20 GPa and significantly outperforming PA12-based specimens fabricated via selective laser sintering and MJF. This composite also demonstrated good compatibility with the commercial MJF printer, yielding specimens with good dimensional stability and comparable tensile properties to the specimens printed on MJF testbed. This work not only provides a commercially viable hybrid-reinforced powder but also serves as a guideline for developing future high-performance polymer composites tailored for advanced additive manufacturing applications.
The dynamic of functional microsystems, such as microrobotics, depends on the precise monolithic fabrication and functional integration of microscale kinematic pairs. Although point-scanning technology endows two-photon lithography (TPL) with unparalleled geometric freedom and manufacturing precision, structures fabricated in liquid photoresists inherently suffer from mechanical instability while lacking mechanical support during the forming process. This fundamental constraint hinders the realisation of truly freeform three-dimensional structures capable of a high degree of freedom (DOF) relative motion. To overcome this bottleneck, we have developed a high-viscosity epoxy photoresist with substantial yield stress for the temporary immobilisation of unsupported components. This epoxy photoresist exhibits solid-like behaviour to provide robust support for overhanging structures and skips the fully continuous stacking, while a surface fluorosilane grafting ensures controlled relative motion. A testing array of the hollow cone and suspended sphere structure was constructed to determine the minimum accuracy between dynamic components and static constraint. The monolithically integrated 3D kinematic pair was demonstrated, including multilayer suspended ball shells, nested hollow balls, gear bearings, and a bio-inspired swinging hair follicle, which are challenging to achieve with liquid two-photon resists. This work provides a robust materials strategy for two-photon volumetric 3D printing, enabling the creation of ready-to-use integrated micro-mechanical devices.
