2026 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2026): Posters
Program Organizers: David Leigh, University of Texas at Austin
Tuesday 4:00 PM
August 4, 2026
Room: Zlotnick Ballroom Foyer
Location: AT&T Center
A Framework for Additive Manufacturing Defense Against Cybersecurity Attack Vectors: Adam Imran1; 1NA
Additive manufacturing systems, particularly network-connected fused deposition modeling (FDM) printers, are increasingly exposed to cybersecurity threats that can compromise data confidentiality, file and object integrity, and device availability. Existing research has identified attack vectors, without a structured way to relate them to defenses. This paper introduces a framework that organizes attacks and defenses across the FDM printing lifecycle. The model uses a two-dimensional matrix in which lifecycle stages (from design to post-processing) are mapped against impacts on confidentiality, integrity, and availability (CIA). Defense categories are aligned with documented attack vectors and established cybersecurity frameworks, including MITRE and NIST. Although not experimentally validated, the introduced framework provides a structured reference for analyzing vulnerabilities and identifying appropriate defense domains in networked 3D printing environments, supporting risk assessment, defensive guardrails, and future research.
A Neural ODE Approach for Thermomechanical Field Prediction in Directed Energy Deposition: Dhruba Aryal1; Praveen Vulimiri1; Todd Sparks1; Albert To1; 1University of Pittsburgh
In metal directed energy deposition, accurate prediction of thermal and mechanical fields such as temperature, residual stress and displacement enables improved design and optimization of parts and process parameters without relying on expensive experimental trials. High fidelity thermo-mechanical simulations using the Finite Element Method can provide such predictions. However, these simulations remain computationally expensive. This work proposes a deep learning method by training a neural ordinary differential equation (ODE) model that leverages features derived from the governing coupled thermo-mechanical equations to predict both thermal and mechanical responses. The model demonstrates excellent agreement with unseen datasets for both fields, accurately capturing residual stresses and deformations while achieving significantly faster inference compared to full finite element simulations. Additionally, the model enables localized prediction of thermo-mechanical fields at regions of interest without requiring full-domain simulation, offering further computational savings for targeted analysis.
A Rheological Approach to Measuring Cure Depth of Soft Photopolymers: Daniel Rau1; CHARLES YEBOAH1; 1University of Wyoming
Vat photopolymerization (VP) builds 3D structures through the layer-by-layer photocuring of liquid photopolymer resins, and the accurate measurement of UV exposure versus cure depth is critical for high-fidelity printing. However, measuring the cure depth of soft resins is extremely difficult as conventional measurements deform the soft and thin films, corrupting the measurement. We advance a photorheology-based approach to measure the cure depth of soft elastomers under process-relevant UV irradiance and wavelengths. To ensure accuracy, we replace the conventional normal-force based measurements with a torque-based detection method to eliminate compression of the soft materials. Because photocuring behavior is highly dependent on the specific UV source used, we also explore using resin absorption and UV source spectra to translate curing behavior between different UV sources. Together, these advances help translate rheometer measurements of cure depth to printer-specific conditions, supporting reliable VP of soft photopolymers across a range of UV sources.
A Sparse 3D Transformer-Based Inherent Strain Surrogate for Mechanical Field Prediction in Laser Powder Bed Fusion: Shane Garner1; Praveen Vulimiri1; Albert To1; 1University of Pittsburgh
Laser powder bed fusion (LPBF) is a widely used additive manufacturing process for producing complex metal components. However, localized thermal gradients during fabrication generate residual stresses that often lead to distortion, cracking, and build failure. While inherent strain simulations can estimate these effects, they are computationally expensive when applied to large-scale or complex design spaces. In this work, a data-driven surrogate model based on a sparse 3D transformer architecture is developed to provide a rapid approximation of such simulations. The model learns the mapping between voxelized geometries and the resulting mechanical fields, including stress, strain, and displacement. By leveraging a local attention mechanism over structured voxel neighborhoods, the model captures both short-range load transfer and long-range mechanical interactions driven by geometry and boundary conditions. The proposed framework significantly reduces computational cost compared to finite element simulations, providing a scalable tool for process-aware design and evaluation of complex LPBF geometries.
Adaptive Height Control for Improved Geometry in Wire Arc Additive Manufacturing
: Eli Landeche1; 1Georgia Southern University
Wire Arc Additive Manufacturing (WAAM) suffers from dimensional inaccuracies due to variations in bead geometry and layer height accumulation during deposition. In this study, the dimensional accuracy of WAAM-fabricated walls was evaluated by comparing two strategies: a conventional constant layer height and a sensor-assisted deposition. The constant height maintains a fixed increment throughout the build, whereas the sensor-assisted approach adjusts the deposition height based on real-time feedback from monitoring sensors.Thin wall structures were fabricated using identical welding parameters to isolate the influence of the height control strategy. Measurements along the build direction were obtained using post-process metrology and compared with the programmed geometry. The results indicate that the constant height method leads to cumulative deviation as the build progresses due to variations in bead height and thermal distortion. Comparatively, the sensor-assisted approach reduces layer variations by compensating for deposition irregularities, resulting in improved geometric conformity and reduced height variation.
Assembly-Aware Print Pose Optimization in 3D Printing for Multi-Robot Assembly
: Arya Haria1; Ronnie Frank Pires Stone1; Zhenghui Sha1; 1University of Texas at Austin
Robotic-enabled assembly processes generally require components to be held in precise, predetermined grasps. Traditionally, this is achieved using custom-made assembly jigs and multiple robotic arms to reorient parts. In the context of additive-assembly cells, however, where the assembled parts are 3D printed in real-time, a unique advantage exists: control over the initial pose of the manufactured parts. In this paper, we leverage this idea and present a novel end-to-end computational pipeline for an additive-assembly cell, starting from a given assembly task, selecting the base part (i.e., passive jig) printing orientation, and providing the final assembly sequence. We find the best print orientation for the base part by considering possible supportless print orientations and the grasp required for downstream assembly. As such, this work demonstrates how a simple additive-assembly cell can use a single robotic manipulator to execute complex assembly sequences, thus reducing the number of steps and resources needed.
Assessment of Mesh Densities and Increment Sizes Between Goldak and Uniform Heat Source Models Using Finite Element Analysis: Elise Roberson1; Clark Hensley1; Logan Betts1; Matthew Priddy1; 1Mississippi State University
Wire-arc directed energy deposition (arc-DED) involves large temperature gradients and complex melt pool geometries, which are challenging to properly and efficiently simulate. The primary moving heat source models used for arc-DED modelling are the Goldak model and the uniform model. The Goldak model provides highly accurate results, but its complex melt pool geometry requires a fine mesh, entailing long runtimes. Contrarily, the uniform model provides a simpler melt pool representation, allowing for coarser meshes and shorter runtimes, but reducing accuracy. This study assessed the trade-off of accuracy and computational expense between these models. Several mesh densities and increment sizes were simulated for each model, and the resultant thermal data was compared alongside experimentally acquired thermal images and temperature probe values. The resultant accuracy levels and runtimes for each heat source and parameter combination were compared. This knowledge will assist in the determination of future simulation parameters for desired runtimes.
Characterization of Polypropylene/Polylactic Acid Polymer Blends Fabricated by Fused Granulate Fabrication: Alvarado Jacqueline1; David Roberson1; 1University of Texas El Paso
This research investigates the ability to use additive manufacturing to print polymer blends composed of extremely dissimilar constituents. Different polymer blends of polypropylene (PP) and polylactic acid (PLA) with varying PP concentrations of 5%, 10%, and 20%. The blends were created by way of melt-compounding and were characterized using tensile testing, dynamic mechanical analysis (DMA), and creep testing. The tensile and DMA specimens were fabricated using fused granulate fabrication (FGF), while the creep specimens were made using a hot press. Scanning electron microscopy (SEM) was performed on the fracture surfaces of the tensile samples to analyze microstructural features after mechanical testing and to observe how each polymer blend behaved after testing. Features such as elasticity, fracture behavior, and surface morphology were examined and compared between the different blends. The results demonstrated the ability to fabricate a blend system composed of dissimilar materials (a polyester and a polyolefin) using FGF.
Comparison of Microsample and Bulk Tensile Properties in the Directed Energy Deposited 316L Stainless Steel.: Prudhvi Raj Pola1; Reyna Simpson2; Olukayode Fatoki3; Santhosh Parupelli3; Salil Desai3; Ranji Vaidyanathan1; 1Oklahoma State University; 2University of Florida; 3North Carolina A&T State University
Metal additive manufacturing (AM) offers significant design flexibility, enabling the fabrication of complex geometries without the need for costly tooling. However, the qualification of metal AM parts remains time-consuming and requires extensive, expensive testing. This difficulty is primarily due to property variations across the part, which depend on geometry and process conditions, particularly the thermal history. In this study, micro-specimens (3 mm x 1 mm x 0.25 mm) are extracted from the gage section of standard samples, and their results are compared to standard samples reported in the literature. The study also examines how properties vary within the gage section as a consequence of thermal history. Micro-sample testing can identify critical regions within the component that may be overlooked by bulk testing. This approach provides a more comprehensive understanding of spatial property differences, thereby offering a stronger basis for assessing part performance.
Correlation Between Rheology, Print Quality, and Porosity in Robocasted Silicon Nitride Ceramics: Christian Mendez Moreno1; Carmen Rocha1; Abigail Ortega1; Francisco Medina1; 1University of Texas at El Paso
This study investigates the development of silicon nitride (Si₃N₄) ceramic slurries for robocasting through the evaluation of various binders, dispersants, and balancing agents. Multiple slurry formulations were prepared using different additive concentrations to determine their influence on rheological behavior, printability, and final part quality. Rheological testing was conducted to evaluate viscosity and flow characteristics required for successful extrusion during the robocasting process. The optimized formulations were subsequently printed into test geometries and analyzed for density and porosity after processing. Results demonstrated that additive selection and concentration significantly affected slurry stability, extrusion consistency, and structural integrity of the printed components. Porosity measurements ranged from approximately 20% to 10%, indicating notable improvements in densification and print quality among the optimized formulations. The findings highlight the importance of slurry formulation design in achieving enhanced rheological performance and improved physical properties for silicon nitride ceramic components fabricated through material extrusiontechniques
Design of Interlocking Joints for Multi-Material Interface Using Topology Optimization: Myung Kyun Sung1; Albert To1; 1University of Pittsburgh
Traditional joining methods like adhesives, welding, and bolting often fail under environmental stress and lack versatility across diverse materials or geometries. Modern applications require resilient, rapidly fabricated joints for dissimilar materials that can survive extreme conditions. However, current techniques cannot precisely tune strength or failure modes through localized geometry. This study addresses this by using topology optimization to design mechanically interlocked interfaces for multi-material additive manufacturing. By accounting for interfacial weak points during the design process, this approach enables robust, assembly-free joining of dissimilar materials.
Development and Validation of a Low-Cost Sensing System for Powder Dispersion Monitoring in Binder Jetting: Sam Erickson1; Max Gunn1; Nathan Crane1; 1Brigham Young University
Powder handling in binder jetting creates significant health and equipment hazards due to airborne dust, yet commercial sensors often lack the sampling frequency and data-logging flexibility required to monitor transient dispersion events. This paper presents the design and verification of a custom, low-cost (~$33), compact air particulate sensing system. Built using a PMS7003 laser diffraction sensor and an ESP32 microcontroller, the device monitors six particle size bins (0.3 to 10 µm) at a high sampling frequency of 0.8 Hz. Data is then logged directly to an onboard, removable microSD card in accessible formats for statistical analysis. Verification tests demonstrated high consistency between sensors, with a standard deviation of less than 6%. This scalable, open-source hardware approach provides a robust solution for quantifying powder dispersion and improving environmental safety monitoring in additive manufacturing facilities.
Development of Digital Twins in Additive Manufacturing: Sung-Heng Wu1; Usman Tariq1; Ranjit Joy1; Braden Mclain1; Frank Liou1; 1Missouri University of Science & Technology
Digital Twin, originally introduced by NASA in 2003, has gained increasing attention in additive manufacturing due to recent advances in sensing, computation, and data-driven modeling. This research presents a framework for developing a digital twin to improve and optimize additive manufacturing processes. By integrating virtual representation, real-time monitoring, predictive modeling, control strategies, and continuous optimization, the proposed framework aims to enhance process understanding while reducing experimental iterations, material waste, and development costs. This poster demonstrates how these key components are integrated into the additive manufacturing workflow and presents preliminary results from in-house system development.
Diffusion Study on Highly Porous 3D Printed Structures Made Using Vat Photopolymerization and In-Situ Phase Separation: Rachel Lamb1; Nikitha Garlapati1; Xiangyu Gao1; Makayla Makuvise1; Michael Cullinan1; 1University of Texas at Austin
Absorption and release of concentrated solutions are important concepts across many industries, from small implants for pharmaceutical drug delivery to larger systems for recycling rare materials. Current commercial systems built with traditional manufacturing approaches are limited in their ability to selectively optimize individualized solutions. Using a combination of vat photopolymerization and photopolymer-induced phase separation (PIPS), porous polymer membranes are printed with controllable porosity ranges, enabling improved diffusion-based mass transport. This study highlights the diffusion properties of 25 to 50µm thick membranes printed with varying pore sizes measured using a Franz side-by-side diffusion cell and UV-Vis spectroscopy. These diffusion results are correlated with scanning electron microscope (SEM) images of surface pores for each resin composition. Recommendations are provided for future structures and printing parameters to achieve ideal diffusion properties.
Effect of Microstructural and Geometrical Features on Low Cycle Fatigue Properties of Additively Manufactured Hastelloy X: Ajay Kushwaha1; Ritam Pal1; Brandon Kemerling2; Daniel Ryan2; Sudhakar Bollapragada2; Amrita Basak1; 1Pennsylvania State University; 2Solar Turbines Incorporated
Nickel-based superalloys such as Hastelloy X are widely used for high-temperature turbine components. While laser powder bed fusion (LPBF) enables fabrication of complex geometries, it introduces structural vulnerabilities like anisotropic microstructures and stochastic defects. This research investigates how macroscopic geometric features and microscopic heterogeneities influence the low cycle fatigue (LCF) behavior of LPBF Hastelloy X. Using an integrated approach, notched LCF tests quantified life reduction from stress concentrators, while Bayesian-calibrated crystal plasticity (CP) simulations isolated microstructural influences in un-notched machined specimens. Results show that notched failure is dominated by multi-site initiation at surface-connected lack of fusion defects. Additionally, surface roughness varies within the geometry depending on local build angle. CP simulations revealed plastic strain localization within coarse grains and near twin boundaries. These results provide a baseline for identifying geometric and microstructural fatigue drivers and support future investigations of surface-microstructure interactions in LPBF Hastelloy X.
Effect of Spatial Beam Profile Shaping on Denudation Zone in PBF-LB of Stainless Steel: Tadashi Yamaguchi1; Keuisuke Takenaka2; Yorihiro Yamashita3; Takahiro Kunimine4; Yuji Sato2; Masahiro Tsukamoto2; 1Graduate School of Engineering, The University of Osaka; 2Joining and Welding Institute, The University of Osaka; 3Faculty of Engineering, University of Fukui; 4Faculty of Mechanical Engineering, Institute of Science and Engineering, Kanazawa University
Powder Bed Fusion with Laser Beam (PBF-LB) is a metal additive manufacturing using a laser. However, PBF-LB still faces several issues including rough surfaces and void formation. Denudation zones (DZs), formed along the melt track during laser irradiation, are considered one of the major factors of these defects. Spatial beam profile shaping has attracted attention as a promising approach for suppression of DZ. Gaussian beam exhibits high peak power density at the beam center, which promotes excessive metal vapor generation. In this study, laser beam profile was shaped into a flat-top distribution to reduce the peak power density while maintaining the same laser power and spot diameter. The results demonstrated that a flat-top beam significantly reduced both DZ formation and metal vapor generation compared with Gaussian beam.
Fused Granulate Fabrication of a PLA/Graphene Oxide-Reinforced PVA Polymer Blend: David Roberson1; Jazlyn Alcala1; 1University of Texas El Paso
In this work a polymer blend system composed of polylactic acid (PLA) reinforced with a polyvinyl acetate (PVA)/graphene oxide (GO) composite was melt-compounded. PVA was used to facilitate incorporation of the GO into the composite because the GO was in an aqueous solution. It was found during the melt compounding process that a filament with a homogenous diameter could not be made due to phase separation of the two constituents. However, it was noted that pellets of the material could be made. Sample pools of different PVA/GO composite loading were made and the materials were characterized by tensile testing, dynamic mechanical analysis (DMA), and scanning electron microscopy (SEM). All test specimens were fabricated using fused granulate fabrication (FGF). The work presented here demonstrates an advantage of FGF that is not often highlighted, the ability to fabricate components from a material system that cannot be extruded into a filament.
Generalizing the Computational Fluid Dynamics Imposed Finite Element Method (CIFEM) to Varied-Geometry, Multi-Layer Simulations of Laser Powder Bed Fusion: Sierra Stevenson1; Seth Strayer1; Albert To1; 1University of Pittsburgh
Functional heat source models in the finite element method fail to capture thermal fields from the laser powder bed fusion process as accurately as high-fidelity computational fluid dynamics (CFD) simulations, but CFD simulations are prohibitively expensive at the part scale. To address this, the CFD-imposed Finite Element Method (CIFEM) was developed, which employed a surrogate model trained on single-layer, constant-length multi-track CFD data to rapidly predict thermal fields and impose them in a finite element method simulation. This work extends CIFEM’s predictive capabilities to a wider range of build geometries by training on varied-length scan track simulations as well as multi-layer simulations, yielding a significant increase in accuracy to the previous CIFEM model while preserving the same speedup from CFD. A revised local thermal environment calculation is also presented, improving robustness to varied scan geometries.
Investigating the Effect of Disturbances on the Dimensional Quality of Parts from Directed Energy Deposition(DED): Emmanuel Bamido1; Michael Cullinan1; 1University of Texas at Austin
Metal based Additive manufacturing processes like Directed Energy Deposition experience limited industrial adaptation due to dimensional inaccuracies of the final part. One source of quality defects is the dynamic changes occurring within the process. In this research, a multiphysics model was developed to analyze the effects of wave-like disturbances on the final dimension of the part. The disturbances at varying frequencies were integrated into the substrate geometry, and two nozzles sprayed metallic powder(Inconel) on the substrate at constant laser power. The final height of the part was measured at different frequencies, and it was observed that the amplitude of the response reduced with increasing frequency.
Kinematic Control of Deposition Volume via Spatially Adaptive Micro-Meander Toolpaths in 5-Axis Wire-DED: M. Ali Yikilmaz1; Albert To1; 1University of Pittsburgh
In Wire Arc Additive Manufacturing (WAAM), changing the height of a layer usually requires adjusting the machine's speed or the rate at which wire is fed. However, fluid dynamics create a problem: making a weld bead taller naturally makes it wider. This often leads to poor overlapping between tracks, uneven shapes, and unpredictable heat buildup. This research offers a mechanical solution instead of changing welding settings. We use "micro-meander" toolpaths that change shape to control how much metal is deposited. A specialized algorithm automatically adjusts the width and frequency of these small zig-zag patterns based on their location. This ensures the edges of each track line up perfectly. By separating the machine’s movement speed from the amount of metal being added, we keep the welding arc stable and the temperature consistent. This method allows for smooth surfaces and variable heights without the physical issues found in standard WAAM.
Effect of Oxygen Contamination in Shielding Environment on Laser Metal Deposited Ti-5Al-5Mo-5V-3Cr alloy: Ranjit Joy1; Sung-Heng Wu1; Frank Liou1; 1Missouri University of Science and Technology
Laser Metal Deposition (LMD) can promote oxygen pickup from the shielding environment, causing variation in tensile behavior of β Ti-5553 deposits. Ti-5553 powder was deposited in an enclosed volume having different oxygen concentrations and by conventional Localized Shielding Gas (LSG) strategy. The critical oxygen concentration that causes ductile-to-brittle transition was systematically investigated using tensile testing. Oxygen pickup in LMD processed thin walls increased with increasing oxygen concentration. Tensile testing revealed only incremental increase in yield, and ultimate tensile strengths with corresponding increase in oxygen pickup, primarily due to solid solution strengthening. However, the post-yield ductility was significantly influenced by oxygen pickup. Gradual reduction in total elongation was registered from 1 PPM to 1000 PPM while exhibiting significant ductility, which was followed by a transition to mixed mode at an oxygen concentration of 5000 PPM. Interestingly, LSG processing induced a fully brittle failure, attributed to increased α precipitation at grain boundaries.
LPBF CP1 Velo 3D – Ultrasonic Fatigue Testing (USF): Regina Almazan Pichardo1; 1W.M Keck Center for 3d Innovation/UTEP
CP1 is a high-performance Al–Fe–Zr alloy developed by Constellium Aheadd, specifically for laser powder bed fusion (LPBF) systems. The alloy is designed for demanding aerospace applications due to its high thermal conductivity and elevated strength. In this study, CP1 specimens were fabricated using an LPBF Velo3D system and stress relieved following the manufacturer’s recommendation (4h at 400 °C). The builds were machined into tensile and ultrasonic fatigue (USF), where fatigue specimen geometry was defined using Shimadzu USF software based on material properties. Fractography was conducted using a JEOL JSM-IT500 SEM, and Vickers microhardness (HV) was measured using a Qness 30 CHD Master+ system. This work investigates the validity of USF as an accelerated alternative for fatigue characterization.
Cancelled
Mechanism-Based Defect Analysis and Process Window Development for Directed Energy Deposition of Al6061: MohammadHassan Kalantari1; Chang Hwan Choi1; 1Stevens Institute of Technology
Directed energy deposition (DED) of Al6061 is limited by a narrow processing window, in which both insufficient and excessive heat inputs can lead to defect formation. In this work, multilayer Al6061 deposits were fabricated by varying energy density to investigate the development of crack networks, porosity, and microstructural features. Cumulative crack density and pore morphology were quantitatively evaluated from metallographic cross sections using image-based analysis. Cracks were primarily observed in reheated partially melted zones, at melt pool boundaries, and within solute-enriched interdendritic regions. These results demonstrate that both liquation during layer reheating and solidification cracking due to limited interdendritic feeding contribute to crack formation. In the conduction-mode regime, the optimized window appeared near the threshold beyond which higher heat input promoted keyhole-mode instability and increased porosity. In Al6061, these thermal conditions promote cracking and porosity by limiting interdendritic feeding and gas escape, thereby defining the reduced-defect processing window during DED.
Multi-Nozzle Molten Metal Jetting for High Throughput Metal Additive Manufacturing: Kareem Tawil1; Irtaza Razvi1; Chris Chungbin1; Andrew Greeley1; Gabriel Stash1; Daniel Cormier1; David Trauernicht1; Denis Cormier1; 1Rochester Institute of Technology
Single nozzle Molten Metal Jetting (MMJ) has demonstrated ability to produce metal parts with high precision and strength, but at lower deposition rates than competing technologies. In this poster, we highlight a novel multi-nozzle MMJ system which can match or exceed the volumetric deposition rates of other metal AM technologies, without compromising on feature resolution. A prototype system is built and tested with eight individually addressable nozzles, at a nozzle pitch of 3mm. The droplet diameter and velocity at each nozzle can be tuned within 1.6% and 1.4%, respectively. Maximum stable volumetric deposition rates exceeding 225 cm3/hr have been achieved. We discuss jetting waveform control, printing strategies, and design methodology for multi-nozzle MMJ systems. The approach discussed highlights an exciting path forward, allowing MMJ to compete with other metal AM technologies on deposition rates, feature resolution, and part strength.
Numerical Simulation of DLP 3D-Printed Photopolymers: Siyuan He1; Kubra Sekmen2; Daniel Weisz-Patrault3; Andrei Constantinescu3; 1Laboratoire Navier, ENPC, Institut Polytechnique de Paris, Université Gustave Eiffel, CNRS; 2Virginia Tech; 3Laboratoire de Mécanique des Solides, CNRS, École Polytechnique, Institut Polytechnique de Paris
Digital Light Processing (DLP) 3D printing enables rapid fabrication of complex geometries by selectively curing liquid photopolymer resin layer by layer. However, UV light penetration into previously cured layers induces depth-dependent shrinkage and viscoelastic deformation, which can generate residual stresses. To predict these effects, we propose a multiphysics finite element approach for DLP 3D-printed photopolymers. The model incorporates a macroscopic viscoelastic constitutive law in which the degree of cure serves as the internal variable and evolves as a function of the irradiation conditions, specifically UV light intensity and exposure time. After experimental calibration, the model is implemented through recurrence relations in an incremental, memory-efficient formulation. This approach enables effective prediction of the spatiotemporal residual stress distribution across the part cross-section by solving a sequence of two-dimensional plane deformation problems.
Optimization of SiC Powder Packing Through Bimodal Distribution and Mechanical Milling for Binder Jet Additive Manufacturing
: Owen Hernandez1; Abigail Ortega1; Carmen Rocha1; Hector Alaniz1; Francisco Medina1; 1W.M. Keck Center for 3D Innovation
Silicon carbide (SiC) fabrication via Binder Jetting (BJT) is limited by low green and sintered densities, typically around 40%. This study proposes a bimodal particle size strategy to improve densification using coarse (60 µm) and fine (16 µm) SiC powders. The coarse powder is subjected to high-energy ball milling (24 h, 150 rpm, 10:1 ball-to-powder ratio) with 5 mm SiC media to refine particle morphology while minimizing contamination. Experimental trials investigate A:B weight ratios of 75:25, 50:50, and 25:75 to evaluate the effects of bimodal mixing and milling on packing efficiency. Samples are fabricated using a 50 µm layer thickness and 60% binder saturation ratio with a solvent-based binder to reduce green-part brittleness. Pressureless sintering is performed at 1600°C for 2 h, targeting densities between 50% and 60%. Characterization includes three-point flexural strength testing and density analysis using geometric measurements and Archimedes’ principle.
Powder Quality and SLS Processability of ZrB₂/PA12 Composites for Neutron Shielding: Melannie Wagoner1; Arturo Hernandez-Barreto2; Desiderio Kovar2; Sheldon Landsberger2; 1Student; 2University of Texas at Austin
Traditional neutron shielding materials such as concrete are increasingly impractical for mass- and space-constrained applications, driving demand for lightweight alternatives. Polymer–ceramic composites offer a compelling solution by integrating hydrogen-rich matrices for fast neutron moderation with neutron-absorbing ceramic fillers. In this work, zirconium diboride (ZrB₂) was incorporated into nylon 12 (PA12) at loadings of 10–60 vol% for fabrication via selective laser sintering (SLS). PA12 moderates fast neutrons while boron within ZrB₂ provides thermal neutron absorption, and ZrB₂ may further enhance thermal conductivity during processing. Thermogravimetric Analysis (TGA) verified filler content in printed parts, with larger-scale furnace burn-off analyses currently underway to improve measurement representativeness. Differential Scanning Calorimetry (DSC) revealed that the SLS processing window (ΔT) remained stable at approximately 36 °C across all compositions, ruling out window narrowing as the cause of reduced printability at high loadings and implicating other limiting factors.
Process Modeling and Design of Magnetoactive Elastomers in Reactive Extrusion Additive Manufacturing: Kaleb Washington1; Brandon Yu1; Mary Frecker2; Carolyn Seepersad1; 1Georgia Institute of Technology; 2Pennsylvania State University
Magneto-active elastomers (MAEs) are smart materials capable of shape programming, large deformations and large magnetic actuation forces when exposed to an external magnetic field. Applications of these capabilities include actuators, soft grippers and a variety of other shape-changing devices. Additive manufacturing processes, such as the novel Reactive-Extrusion Additive Manufacturing (REAM) process utilized in this work, are capable of spatially customizing the composition and magneto-active response of these materials by blending multiple feedstocks on demand. Prior work has demonstrated this capability via physics-based modeling and experimental fabrication, but the printing capabilities of the underlying process have not been considered. The methodology of this work consists of creating a user-defined part and shape matching with an accompanying target profile. This part is defined by multiple raster layers with a magnetic fraction function that models the transient response of the REAM process alongside Euler-Bernoulli theory to produce manufacturable magneto-active complex parts with precision.
Process Sensitivity Assessment of a Build-Integrated Witness Artifact for Laser Powder Bed Fusion: Sina Nejati Eghteda1; Albert To1; 1University of Pittsburgh
Process variability in laser powder bed fusion (LPBF) can introduce defect populations that degrade mechanical performance, yet current qualification methods rely on costly post-build testing. This work investigates the sensitivity of a compact, build-integrated witness artifact to process-induced material variations in LPBF. The artifact is fabricated concurrently with the build and yields a mechanical response that reflects the material condition produced during manufacturing. Builds are conducted under systematically varied LPBF parameters spanning conditions expected to produce distinct defect characteristics. Response metrics extracted from the artifacts are analyzed across the investigated process window and compared with conventional fracture specimens to evaluate consistency with established fracture resistance measures. Results indicate that the artifact can distinguish between process conditions associated with different levels of material integrity, supporting its potential as a rapid screening tool for LPBF build quality.
Runtime-Reconfigurable RAPID Framework for ABB Arc-DED Systems: David Johnson1; Jeffery Betts1; Charlotte Thompson1; Matthew Priddy1; 1Mississippi State University
In Wire Arc-Directed Energy Deposition (arc-DED) systems, deposition strategies impact thermal history, performance and geometry. While select geometries (e.g., Thinwall, Block, Cylinder) are commonly used to fabricate test specimens on similar arc-DED systems, disparate methods for tool path generation result in variability and prevent direct comparison across machines and facilities. This framework provides a deployable solution to fabricate standard geometries for arc-DED systems utilizing ABB Robots and Fronius Welders. By standardizing toolpath implementation through dynamically generated ABB RAPID routines, end users can focus on process parameter development, novel materials, and deposition techniques while improving build-to-build repeatability. This work provides a set of standard reference geometries that can be dynamically reconfigured within ABB RAPID at runtime, minimizing the influence of motion and toolpath differences and enabling direct comparisons of process parameters across materials, systems, and facilities.
The 10th Annual 3D Printed Aircraft Competition: Expanding the Scope of Student Learning Through Competition: John Sistrunk1; Robert Taylor1; Thomas Allsup1; Jeffrey Liu1; 1University of Texas Arlington
The University of Texas at Arlington (UTA) hosted the 10th Annual 3D Printed Aircraft Competition on July 11, 2026. For a decade, this annual event has provided student teams from many universities a competitive forum to learn and apply principles of aircraft design, lightweight structural design, and design for additive manufacture. The competition challenges students to design, build, and fly a fully 3D printed airframe. A limitation on power duration during flight further challenges students to develop lightweight designs that effectively integrate knowledge of structural mechanics with 3D printing process mechanics. Adding to fixed and rotary wing flight categories, the event this year developed a new competition category for energy absorption. A student team from UTA developed metrics and rules for this new category and designed and flew a vehicle with energy absorbing meta-materials. This poster discusses competition rules, categories, student designs, process considerations, lessons learned, and event results.
Thermal Annealing-Induced Property Enhancement in Additively Manufactured PC-CF Composites: Kazi Md Masum Billah1; Caleb Cannon1; Josiah White1; Youssef K Hamidi1; 1University of Houston-Clear Lake
Polycarbonate–carbon fiber (PC-CF) composites fabricated through material extrusion additive manufacturing offer high strength, thermal resistance, and lightweight characteristics for engineering applications. However, residual stresses and weak interlayer adhesion developed during the printing process can negatively affect mechanical performance and dimensional stability. This study investigates the effect of post-processing thermal annealing on the mechanical and structural behavior of 3D printed PC-CF composites. Printed specimens were annealed at different temperatures under controlled conditions and evaluated using tensile, hardness, and impact testing. Dimensional changes and fracture behavior were also analyzed to assess the influence of annealing on structural integrity. The results are expected to demonstrate that moderate thermal annealing improves interlayer bonding and mechanical performance through stress relaxation and enhanced polymer chain mobility, while excessive annealing may introduce warpage and dimensional deformation. The findings provide practical guidelines for optimizing thermal post-processing conditions in high-performance thermoplastic composite additive manufacturing.
Toward a Physics-Resolving Digital Twin for Metal AM: Graph Neural Network Surrogates for Thermomechanical Field Prediction and Sensor-Driven Defect Detection: Usman Tariq1; Sung-Heng Wu1; Frank Liou1; 1Missouri University of Science and Technology
Reliable qualification of metal additive manufacturing parts requires accurate prediction of thermal history, residual stress, and distortion. Physics-based finite element simulation captures this accurately but takes hours to days per build, making it incompatible with real-time control. This work develops a physics-aware two-stage graph neural network surrogate that converts finite element meshes into graphs with physics-informed features including laser proximity, boundary distance, and element birth timing. A DeeperGCN thermal model predicts full-field temperature across time steps and generalizes to unseen geometries without retraining. A recurrent graph neural network maps the predicted thermal history to full-field residual stress and displacement, preserving causal thermomechanical structure. These models enable rapid pre-build risk assessment orders of magnitude faster than finite element analysis. Looking ahead, the surrogate is designed to assimilate in-situ sensor data layer by layer, enabling subsurface inference, anomaly detection, and closed-loop process control as a physics-resolving digital twin for metal additive manufacturing.
Towards a Universal Scaling Law and Thermal Modeling Framework for Wetting Behavior in Laser Metal Additive Manufacturing: Peter Morcos1; Sameh Tawfick1; 1University of Illinois Urbana Champaign
Wetting behavior plays a critical role in melt pool stability, interlayer bonding, and defect formation in laser metal additive manufacturing (AM). In this work, a combined physics-informed scaling-law and computational modeling framework is developed to predict contact angle and characterize melt pool behavior across multiple metallic material systems. Dimensionless geometric descriptors derived from measurable melt pool features, including melt pool depth, width, height, and cross-sectional area, are used to formulate generalized power-law scaling relationships for contact angle prediction. In parallel, computational fluid dynamics (CFD)-based thermal simulations are performed to model melt pool evolution and thermal behavior during processing. Experimental single-track data are used to validate the numerical predictions and evaluate the transferability of the proposed framework across different materials and processing conditions. The results demonstrate the potential of combining data-driven scaling laws with thermal modeling to support process understanding and parameter optimization in laser metal AM.