Jonathan A. Malen

Understanding the formation of defects in Laser Powder Bed Fusion (L-PBF) is important for manufacturing parts that perform to desired specifications. The region of molten material underneath the melting laser beam and inside the vapor depression governs the formation of keyhole porosity in the final part as well as the alloy composition due to preferential evaporation of elements. Although this region is important to defect formation, it is not well quantified by thermal imaging methods because light emitted from a specified region of the surface undergoes reflections in the vapor depression that distort the perceived temperature at the camera. We present a method for simulating thermal imaging experiments on difficult concave melt pool surfaces that accounts for high-temperature reflections, reconciling discrepancies between experimental and simulated thermal fields, and addressing a key challenge in understanding defect formation in Laser powder bed fusion. We begin with bare plate computational fluid dynamics (CFD) temperature profiles and use radiative networks to predict local radiosities that incorporate reflections before passing the light through the simulated optical train to the camera sensor. This process enables an apples-to-apples comparison of CFD predictions of temperature and thermal images. For example, without considering reflections, the maximum deviation in the centerline temperature of 316L steel melt pools is up to 15%. However, with this correction, the maximum deviations are reduced to 4%–6%. Our findings indicate that longer imaging wavelengths reduce the impact of reflections on perceived experimental temperatures. By enabling a more rigorous validation of CFD temperature profiles, the method can help to control for keyhole porosity and powder feedstock alloy composition. Due to the method’s material and surface-agnostic nature, it will also find applications in other fields where understanding high surface temperatures is important such as in aerospace when imaging high-temperature turbine blades, or in renewable energy when working with concentrated solar power technologies.

The thermal transport community is increasingly interested in rigorous uncertainty quantification (UQ) of their measurements. In this work, we argue that Bayesian parameter estimation (BPE) represents a powerful framework for both analysis/fitting and UQ. We provide a detailed walkthrough of the technique (including code to duplicate our results) and example analysis based on measuring the thermal conductance of a gold/sapphire interface with FDTR. Comparisons are made against traditional analysis/UQ techniques adopted by the thermal transport community. Notable advantages of BPE include the interpretability of its results, including the capacity to indicate incorrect input assumptions, as well as a way to balance overall goodness of fit against prior knowledge of feasible parameter values. In some cases, incorporating this additional information can affect not only the magnitude of error bars but the inferred values themselves.

Laser beam directed energy deposition (DED-LB) is of interest as a manufacturing method for refractory complex concentrated alloys. However, in-situ mixing of metals with large differences in melting temperature and density can lead to inhomogeneities, which in turn can give rise to locally varying physical properties. To gain insight into these issues, we investigate the deposition of refractory metal Nb and Ta powders on like and unlike baseplates. The melting temperatures of Ta and Nb differ by 550 °C, while the density of Ta is twice that of Nb. When Ta powder is deposited on a Nb baseplate, un-melted Ta powder particles are observed. Bands of different compositions are also observed throughout the melt pools, both when Ta powder is deposited on Nb baseplate and when Nb powder is deposited on Ta baseplate. Deposition of the denser Ta atop a Nb baseplate results in a chaotic interface profile, while the interface is smooth in the opposite case. In autogenous melt pools, the grain sizes are about 10 times larger than those of the pure baseplates. However, the average melt pool hardnesses are larger by 6 % and 32 % than those of the Ta and Nb baseplates, respectively. The hardness increase is attributed to work hardening. This work provides a systematic study on process parameter dependence of inhomogeneities in DED-LB melt pools when physical properties of materials to be combined are significantly different.

AlxGa1−xN with a high Al composition (x) presents significant potential for advancing next-generation high-power electronic devices. To support the thermal design of AlxGa1−xN-based electronics, the thermal conductivity of AlxGa1−xN thin films was measured as a function of Al composition, temperature, and film thickness using time-domain thermoreflectance and frequency-domain thermoreflectance techniques. The measurement results were interpreted by modeling phonon transport in AlxGa1−xN films using the phonon Boltzmann transport equation. Phonon properties, including frequencies, group velocities, and lifetimes, were calculated using a virtual crystal approximation, with the effects of mass-disorder scattering incorporated via the Tamura model. The measured thermal conductivity of Al0.7Ga0.3N is an order of magnitude lower than those for GaN and AlN, exhibits an increase followed by saturation with temperature, and shows a modest decrease with a reduction in the film thickness. The modeling results agree with the measurement results and reveal that mass-disorder scattering and phonon-boundary scattering are the dominant mechanisms that reduce the thermal conductivity of AlxGa1−xN thin films.

Current research in additive manufacturing focuses on understanding the underlying physics of defect formation in parts. Multiphysics models are key to planning builds with minimal defects, but these models must be validated with experimental data to ensure meaningful predictions. We present experimental validation of a model for L-PBF melt pools using the OpenFOAM® open-source computational fluid dynamics (CFD) package that includes simulation of solid, liquid, and vapor phases. Solvers that simultaneously treat melt pool and vapor/gas flow physics are relatively new among existing L-PBF simulation tools. Solver parameters such as the Fresnel and mass accommodation coefficients are fit based on validation against ex-situ geometric data and in-situ thermal images. The results are compared to simulations using the semi-analytical Eagar-Tsai model, ANSYS Mechanical®, and FLOW-3D®. OpenFOAM simulations of 316L stainless steel melt pools show good agreement with both experimental results and similarly validated FLOW-3D simulations over a range of laser processing conditions. Vapor plume velocities are reported at over 250 m/s, and including a layer of pre-sintered powder shows increased melt pool variability and ejection of spatter particles. Results demonstrate the importance of validating models on multiple forms of experimental data and show the solver’s ability to simulate vapor and gas flow data. Future simulations of vapor and gas flow will lead to better understanding of L-PBF phenomena such as entrainment spatter, spatter trajectories within the gas flow, and vapor plume obstruction of the laser source.

Advanced process monitoring and model validation are essential for improving weld quality in both welding and welding-based additive manufacturing processes. Specifically, temperature is a key quantity of interest for understanding defect formation and microstructural evolution, which significantly impact mechanical properties. However, achieving accurate in-situ temperature imaging is challenging due to emissivity variations across the dynamic melt pool. To address this, we implemented a two-color imaging technique using a single commercial color camera to reduce temperature readings’ sensitivity to emissivity variations. High dynamic range images during melting were captured at various exposure times, and spatial and temporal filters were applied to minimize interference from the plasma arc emissions. The resulting temperature fields within the melt pool were then utilized to estimate cooling rates, which were further correlated to ex-situ hardness measurements. The strong correlation observed between cooling rates ranging from 20 to 600 K/s and hardness ranging between 250 to 400 HV demonstrated the potential of our easy-to-use two-color thermal imaging setup for preliminary evaluation of mechanical properties in a non-destructive manner. Beyond its significance for predicting mechanical properties, this technique provides a validated temperature measurement approach that can enhance the accuracy of physics-based models, such as those used to predict defect formation mechanisms, like porosity.

Abstract Laser powder bed fusion offers a high degree of geometric freedom for manufacturing with novel materials, yet failures during fabrication remain a critical barrier to achieving more complex components. Recoater blade collisions, cracking, and build plate delamination damage parts and performance, especially with hard and high-temperature materials. Cemented carbides are optimal for high-hardness machining and tooling parts, but high thermal gradients and complex composite behaviors exacerbate fabrication issues. Understanding the effects of build process parameters on macroscopic failure modes is critical to mitigate such issues. This study leverages thermomechanical modeling to investigate the effects of process parameter alterations on build stresses and deflection for WC-Ni part fabrication strategies. A comparative analysis revealed that reductions in laser energy density and part sizes reduced part deflection. The simulations reinforced that bed preheating reduced stresses and thermal gradients, but reductions in interlayer timing also benefited builds by adding additional interlayer heating. Sharp geometric features common in cemented carbide machining and tooling parts significantly increased deflection. Exploratory builds achieved high-density (> 97%) WC-17 wt% Ni parts, with delamination occurring only with the geometry with the highest simulated stress. Profilometry revealed over-melting around part edges and high surface roughness (arithmetic mean height up to 61.5 µm), indicating that localized features from scan strategies heavily contribute to recoater collisions. The successful fabrication of a range of parts, including drill bit geometries, demonstrated the effectiveness of parameter refinement as a tool to avoid catastrophic failure events with laser powder bed fusion.

Cemented carbides such as cemented tungsten carbide (WC) are known for their use in resilient wear-resistant applications where hardness and thermal stability are imperative. They are composed of carbide particles embedded in a metal binder. Laser Powder Bed Fusion (L-PBF) is a favorable method to form cemented carbides into complex geometries, but composites pose unique challenges relative to metals typically processed by L-PBF. Resolving the melt pool temperature distributions in L-PBF is key to understanding the underlying physics of the fusion process. Using a two-color thermal imaging method, melt pool thermal maps of WC 0.83 -Ni 0.17 were captured with linear energy densities ranging from 500–1750 J/m with and without powder. WC 0.83 -Ni 0.17 melt pools exhibit temperatures above 4000 K, which can lead to the generation of other WC phases. Compared to more common L-PBF materials such as 316L stainless steel (SS), WC 0.83 -Ni 0.17 melt pools reach higher temperatures. Our direct measurements find that the thermal conductivity of WC 0.83 -Ni 0.17 is 30 W/m-K at 300 K, which is higher than the thermal conductivity of 316L SS and suggests that other heat transfer limitations must cause the elevated melt pool temperatures. A FLOW-3D CFD model based on the composite properties was compared to both the melt pool centerline temperatures and width measurements of the samples fabricated by L-PBF. The simulations indicate that specifying the onset of fluidity is key to reproducing the high temperatures observed experimentally. Although Ni has a melting point of 1728 K, the simulations do not match experiments unless the onset of fluidity is set at the melting point of WC (3143 K). Within FLOW-3D, the onset of fluidity is controlled by the critical solid fraction, which is a uniquely important parameter for simulating composite materials.

While the effect of polymer chain tacticity on crystallinity, glass transition dynamics, and viscoelastic coefficients has been studied, its impact on thermal transport is unknown. Here, molecular dynamics simulations are used to determine the tacticity-dependent thermal conductivity of extended single chains of polypropylene and polystyrene. Chains with ordered tacticity show a threefold enhancement in thermal conductivity compared to chains with random tacticity. A kinetic theory-based model indicates that the enhancement results from a combination of large mean free paths and velocities of energy carriers. This work introduces tacticity control as a strategy to develop thermally conductive polymers.

The thermal conductivities of (100) γ-Ga2O3 films deposited on (100) MgAl2O4 substrates with various thicknesses were measured using frequency-domain thermoreflectance. The measured thermal conductivities of γ-Ga2O3 films are lower than the thermal conductivities of (2¯ 01) β-Ga2O3 films of comparable thickness, which suggests that γ-phase inclusions in the doped or alloyed β-phase may affect its thermal conductivity. The thermal conductivity of γ-Ga2O3 increases from 2.3−0.5+0.9 to 3.5±0.7 W/m K for films with thicknesses of 75–404 nm, which demonstrates a prominent size effect on thermal conductivity. The thermal conductivity of γ-Ga2O3 also shows a slight increase as temperature increases from 293 to 400 K. This increase in thermal conductivity occurs when defect and boundary scattering suppress signatures of temperature-dependent Umklapp scattering. γ-Ga2O3 has a cation-defective spinel structure with at least two gallium vacancies in every unit cell, which are the likely source of defect scattering.

High Resolution Image Download MS PowerPoint Slide Anisotropic thermal transport was measured in imine-linked two-dimensional polymer (2DP) films that were prepared by interfacial polymerization. Measurements of both in-plane ( k ∥ ) and cross-plane ( k ⊥ ) thermal conductivities relied on preparing free-standing 2DP films that were readily transferred for different measurement configurations. We polymerized two 2DP (Per-PDA and TAPPy-PDA) films at a liquid–liquid interface. These polycrystalline, imine-linked 2DP films are 100–200 nm in thickness and were measured by frequency domain thermoreflectance to extract k ⊥ and a suspended calorimetric platform technique to evaluate k ∥ . We find that k ∥ is larger than k ⊥ in both materials at room temperature, leading to anisotropy ratios ( k ∥ / k ⊥ ) as high as 2.3. We attribute this behavior to the fact that the stiff, in-plane covalent bonds of 2DPs transport heat more effectively than the flexible, supramolecular cross-plane interactions. Variable–temperature measurements revealed a positive correlation between temperature and thermal conductivity, which we attribute to phonon scattering from grain boundaries and defects in the polycrystalline 2DP films. Molecular dynamics simulations of pristine crystals predict larger thermal conductivities and anisotropy ratios exceeding 7. The simulations suggest that as higher quality 2DP films become available, higher thermal conductivities and anisotropy ratios will also manifest.

The thermal conductivities of WC-Ni 10 and WC-Ni 17 cermet powders used in powder bed additive manufacturing processes were measured using the transient hot wire method. The 10 % and 17 % Ni by weight powders have measured thermal conductivities of 0.13 ± 0.02 Wm −1 K −1 and 0.15 ± 0.02 Wm −1 K −1 respectively at a temperature of 295 K and a pressure of 101 kPa – more than 25 % lower than common metal powders used in additive manufacturing. The WC-Ni powder grains are irregularly shaped and have a low average circularity, which causes a lower packing fraction and different contact geometry compared to a spherical powder bed. This results in reduced heat transfer through the gas assisted solid pathway of the powder bed and a lower overall powder thermal conductivity. The pressure dependence on the powder bed thermal conductivity was measured from 1 kPa to 101 kPa with both N 2 and He used as infiltrating gases. The powder bed thermal conductivity increased as the ambient pressure increased, and the magnitude of this pressure dependence was found to be a function of the infiltrating gas, not the powder composition. At 101 kPa infiltrating N 2 pressure, the WC-Ni 10 (WC-Ni 17 ) powder bed thermal conductivities increased from 0.13 (0.15) Wm −1 K −1 at 295 K to 0.17 (0.19) Wm −1 K −1 at 435 K. With this knowledge, thermal management of WC-Ni powder bed additive manufacturing techniques can be more accurately addressed. • Thermal conductivities of WC-Ni powders are measured with the transient hot wire method. • Pressure and temperature dependence of powder thermal conductivity are studied. • Gas pressure and composition significantly impact WC-Ni powder thermal conductivity. • WC-Ni powders exhibit high void fractions, low circularity, and irregular morphologies. • WC-Ni powders have lower thermal conductivities than commonly used metal powders.

Simulated FDTR phase lag data for a gold on sapphire sample, for a range of possible values of 4 input parameters (laser spot size, substrate thermal conductivity, gold layer thickness, and interface thermal conductance). This dataset contains the primary simulation data of a research project, however it will not contain every post-processing file associated with the research. That will instead be stored on an associated GitHub repo, which will be linked here after it has been fully created. The data in this submission is analyzed and discussed in an associated research manuscript which is in preparation for submission.<br>

Simulated FDTR phase lag data for a gold on sapphire sample, for a range of possible values of 4 input parameters (laser spot size, substrate thermal conductivity, gold layer thickness, and interface thermal conductance). This dataset contains the primary simulation data of a research project, however it will not contain every post-processing file associated with the research. That will instead be stored on an associated GitHub repo, which will be linked here after it has been fully created. The data in this submission is analyzed and discussed in an associated research manuscript which is in preparation for submission.<br>

Anomalous melt pools during metal additive manufacturing (AM) can lead to deteriorated mechanical and fatigue performance. In-situ monitoring of the melt pool subsurface morphology requires specialized equipment that may not be readily accessible or scalable. Therefore, we introduce a machine learning framework to correlate in-situ two-color thermal images observed via high-speed color imaging to the two-dimensional profile of the melt pool cross-section. We employ a hybrid CNN-Transformer architecture to establish a correlation between single bead off-axis thermal image sequences and melt pool cross-section contours measured via optical microscopy. Specifically, a ResNet model embeds the spatial information contained within the thermal images to a latent vector, while a Transformer model correlates the sequence of embedded vectors to extract temporal information. The performance of this model is evaluated through dimensional and geometric comparisons to the corresponding experimental no-powder melt pool observations. Our framework is able to model the curvature of the subsurface melt pool structure, with improved performance in high energy density regimes compared to analytical models. Additionally, the use of ratiometric temperature estimates improves the accuracy of the model predictions compared to monochromatic imaging. This work establishes a framework extensible towards powder-based AM builds. • We predict 2D melt pool contours from high-speed thermal images using a hybrid CNN-Transformer model. • The model recreates subsurface melt pool structure, evaluated by geometric comparison. • Predictions align with multitrack melting experiments at varied hatch spacings. • Ratiometric temperature estimates yield more accurate melt pool contour predictions than monochromatic input. • Pre-training on simulation data reduces experimental data needed for accurate prediction.

Defects in Laser Powder Bed Fusion (L-PBF) parts often result from the meso-scale dynamics of the molten alloy near the laser, known as the melt pool. Experimental in-situ monitoring of the three-dimensional melt pool physical fields is challenging, due to the short length and time scales involved in the process. Multi-physics simulation methods can describe the three-dimensional dynamics of the melt pool, but are computationally expensive at the mesh refinement required for accurate predictions of complex effects. Therefore, we develop a generative deep learning model based on the probabilistic diffusion framework to map low-fidelity simulation information to the high-fidelity counterpart. By doing so, we bypass the computational expense of conducting multiple high-fidelity simulations for analysis by upscaling lightweight coarse mesh simulations. We demonstrate the preservation of key metrics of the melting process between the ground truth simulation data and the diffusion model output, such as the temperature field, the melt pool dimensions and the variability of the keyhole vapor cavity. We predict the melt pool depth within 3 μm based on low-fidelity input data 4× coarser than the high-fidelity simulations, reducing analysis time by two orders of magnitude.

Modern microprocessor performance is limited by local hot spots induced at high frequency by busy integrated circuit elements such as the clock generator. Locally embedded thermoelectric devices (TEDs) are proposed to perform active cooling whereby thermoelectric effects enhance passive cooling by the Fourier law in removing heat from the hot spot to colder regions. To mitigate transient heating events and improve temperature stability, we propose a novel analytical solution that describes the temperature response of a periodically heated hot spot that is actively cooled by a TED driven electrically at the same frequency. The analytical solution that we present is validated by experimental data from frequency domain thermal reflectance (FDTR) measurements made directly on an actively cooled Si thermoelectric device where the pump laser replicates the transient hot spot. We herein demonstrate a practical method to actively cancel the transient temperature variations on circuit elements with TEDs. This result opens a new path to optimize the design of cooling systems for transient localized hot spots in integrated circuits.

Insufficient overlap between the melt pools produced during Laser Powder Bed Fusion (L-PBF) can lead to lack-of-fusion defects and deteriorated mechanical and fatigue performance. In-situ monitoring of the melt pool subsurface morphology requires specialized equipment that may not be readily accessible or scalable. Therefore, we introduce a machine learning framework to correlate in-situ two-color thermal images observed via high-speed color imaging to the two-dimensional profile of the melt pool cross-section. Specifically, we employ a hybrid CNN-Transformer architecture to establish a correlation between single bead off-axis thermal image sequences and melt pool cross-section contours measured via optical microscopy. In this architecture, a ResNet model embeds the spatial information contained within the thermal images to a latent vector, while a Transformer model correlates the sequence of embedded vectors to extract temporal information. Our framework is able to model the curvature of the subsurface melt pool structure, with improved performance in high energy density regimes compared to analytical melt pool models. The performance of this model is evaluated through dimensional and geometric comparisons to the corresponding experimental melt pool observations.

Previous work on additively-manufactured oxide dispersion strengthened alloys focused on experimental approaches, resulting in larger dispersoid sizes and lower number densities than can be achieved with conventional powder metallurgy. To improve the as-fabricated microstructure, this work integrates experiments with a thermodynamic and kinetic modeling framework to probe the limits of the dispersoid sizes and number densities that can be achieved with powder bed fusion-laser beam. Bulk samples of a Ni–20Cr ＋ 1 wt% Y2O3 alloy are fabricated using a range of laser power and scanning velocity combinations. Scanning transmission electron microscopy characterization is performed to quantify the dispersoid size distributions across the processing space. The smallest mean dispersoid diameter (29 nm) is observed at 300 W and 1200 mm/s, with a number density of 1.0 × 1020 m−3. The largest mean diameter (72 nm) is observed at 200 W and 200 mm/s, with a number density of 1.5 × 1019 m−3. Scanning electron microscopy suggests that a considerable fraction of the oxide added to the feedstock is lost during processing, due to oxide agglomeration and the ejection of oxide-rich spatter from the melt pool. After accounting for these losses, the model predictions for the dispersoid diameter and number density align with the experimental trends. The results suggest that the mechanism that limits the final number density is collision coarsening of dispersoids in the melt pool. The modeling framework is leveraged to propose processing strategies to limit dispersoid size and increase number density.

polymers can be realized by enhancing the alignment, crystallinity, and intermolecular interactions. While a holistic mechanistic framework does not yet exist for thermal transport in polymeric materials, contemporary literature suggests that phonon-like heat carriers may be operative in macromolecules that meet the abovementioned criteria. In this review, we offer a perspective on how high thermal conductivity polymers can be systematically engineered from this understanding. Reports for several classes of macromolecules, including linear polymers, network polymers, liquid-crystalline polymers, and two-dimensional polymers substantiate the design principles we propose. Throughout this work, we offer opportunities for continued fundamental and technological development of polymers with high thermal conductivity.

Defects in laser powder bed fusion (L-PBF) parts often result from the meso-scale dynamics of the molten alloy near the laser, known as the melt pool. For instance, the melt pool can directly contribute to the formation of undesirable porosity, residual stress, and surface roughness in the final part. Experimental in-situ monitoring of the three-dimensional melt pool physical fields is challenging, due to the short length and time scales involved in the process. Multi-physics simulation methods can describe the three-dimensional dynamics of the melt pool, but are computationally expensive at the mesh refinement required for accurate predictions of complex effects, such as the formation of keyhole porosity. Therefore, in this work, we develop a generative deep learning model based on the probabilistic diffusion framework to map low-fidelity, coarse-grained simulation information to the high-fidelity counterpart. By doing so, we bypass the computational expense of conducting multiple high-fidelity simulations for analysis by instead upscaling lightweight coarse mesh simulations. Specifically, we implement a 2-D diffusion model to spatially upscale cross-sections of the coarsely simulated melt pool to their high-fidelity equivalent. We demonstrate the preservation of key metrics of the melting process between the ground truth simulation data and the diffusion model output, such as the temperature field, the melt pool dimensions and the variability of the keyhole vapor cavity. Specifically, we predict the melt pool depth within 3 $μm$ based on low-fidelity input data 4$\times$ coarser than the high-fidelity simulations, reducing analysis time by two orders of magnitude.

Abstract Powder bed fusion is a method of additive manufacturing (AM) where parts are constructed by iteratively melting metal cross-sections to build complex 3D structures. Defects often form during the printing process, where the dynamics of the melt pool can directly contribute to the formation of porous defects in the final part. For instance, insufficient overlap of the produced melt pools can result in unmelted regions of powder, while deep, unstable vapor depression cavities can lead to spherical voids becoming trapped in the substrate. Therefore, in situ of monitoring the melt pool during the melting process can telegraph the formation of defects and assist the creation of fully dense parts. Here, we augment data-driven-based monitoring techniques to enable the 3D visualization of the melt pool underneath the surface, based on the melt pool surface temperature and processing parameters. Specifically, a convolutional neural network (CNN) predicts the topography of the melt pool and keyhole cavity, based on the surface temperature data near the laser focal point and the nominal operating conditions. The data for the laser powder bed fusion process used to train the model is produced by full-field simulations of the meso-scale melting process, with the CFD software FLOW-3D. Data augmentation techniques are implemented to ensure generalizable performance in cases where the temperature data may be obscured and to ensure sharp, accurate predictions of the melt pool boundaries.

Previous work on additively-manufactured oxide dispersion strengthened alloys focused on experimental approaches, resulting in larger dispersoid sizes and lower number densities than can be achieved with conventional powder metallurgy. To improve the as-fabricated microstructure, this work integrates experiments with a thermodynamic and kinetic modeling framework to probe the limits of the dispersoid sizes and number densities that can be achieved with powder bed fusion-laser beam. Bulk samples of a Ni-20Cr $+$ 1 wt.% Y$_2$O$_3$ alloy are fabricated using a range of laser power and scanning velocity combinations. Scanning transmission electron microscopy characterization is performed to quantify the dispersoid size distributions across the processing space. The smallest mean dispersoid diameter (29 nm) is observed at 300 W and 1200 mm/s, with a number density of 1.0$\times$10$^{20}$ m$^{-3}$. The largest mean diameter (72 nm) is observed at 200 W and 200 mm/s, with a number density of 1.5$\times$10$^{19}$ m$^{-3}$. Scanning electron microscopy suggests that a considerable fraction of the oxide added to the feedstock is lost during processing, due to oxide agglomeration and the ejection of oxide-rich spatter from the melt pool. After accounting for these losses, the model predictions for the dispersoid diameter and number density align with the experimental trends. The results suggest that the mechanism that limits the final number density is collision coarsening of dispersoids in the melt pool. The modeling framework is leveraged to propose processing strategies to limit dispersoid size and increase number density.

The impact of rapid thermal annealing (1000 °C for 1 min) on the thermal transport properties of amorphous alumina (a-Al2O3) thin films grown by atomic layer deposition on β−Ga2O3 and amorphous silica (a-SiO2) substrates is determined using frequency-domain thermoreflectance measurements. The annealing more than doubles the a-Al2O3 thermal conductivity for both substrates (1.54 ± 0.13 to 3.14 ± 0.27 W m−1 K−1 for β−Ga2O3 and 1.60 ± 0.14 to 3.87 ± 0.33 W m−1 K−1 for a-SiO2) while keeping the film amorphous. The thermal conductivity increase is attributed to partial recrystallization and off-gassing of embedded impurities. Annealing halves the thermal boundary resistance of the a-Al2O3/a-SiO2 interface (10.5 ± 1.0 to 4.47 ± 0.42 m2 K GW−1), which is attributed to compositional mixing and structural reorganization that are enabled by the elastic matching of these two materials. The thermal boundary resistance of the a-Al2O3/β−Ga2O3 interface is not affected by annealing due to the elastic mismatch. Reducing the thermal resistance of a-Al2O3 dielectric films and adjacent interfaces by annealing will promote lateral heat spreading adjacent to hot spots and improve device longevity.

This work demonstrates an experimental technique to image melt pool temperature fields with a commercial color camera for laser-directed energy deposition (L-DED) processes. The technique relies on two-color thermography to construct spatially-resolved temperature fields and is demonstrated on 316 L stainless steel (SS) melt pools from solidus to near-vaporization temperatures. This two-color thermal imaging system negates the need for a priori knowledge of melt pool emissivity or the camera’s view factor and was validated with a calibrated tungsten filament lamp between temperatures of 1220 K and 2850 K. In-situ temperature measurements are combined with ex-situ cross-sectional geometry to determine the material’s effective absorptivity and coefficient of temperature-dependent surface tension, which is responsible for Marangoni convection, in a multi-physics computational fluid dynamics (CFD) model. Measured peak melt pool temperatures are important for understanding the molten convection within the melt pool, and measured cooling rates can be related to the resulting part microstructure. Peak temperatures from 1750 K to 3000 K show that below the boiling point, temperature increases with increasing laser power density and decreases less with increasing scanning velocity. Thermal images are used to estimate cooling rates in the melt pool tail, which seem to increase linearly with the ratio of scanning velocity to power. Our thermal imaging strategy will advance measurement science in L-DED processes by validating multi-physics CFD models, quantifying cooling rates, providing real-time feedback control, and allowing process mapping of critical melt pool behaviors.

We introduce an experimental method to image melt pool temperature with a single commercial color camera and compare the results with multi-physics computational fluid dynamic (CFD) models. This approach leverages the principle of two-color (i.e., ratiometric) thermal imaging, which is advantageous because it negates the need for a priori knowledge of melt pool emissivity, plume transmissivity, and the camera’s view factor. The color camera’s ability to accurately measure temperature was validated with a National Institute of Standards and Technology (NIST) blackbody source and tungsten filament lamp between temperatures of 1600 K and 2800 K. To demonstrate the technique, an off-axis high-speed color camera operating at 22,500 frames per second capturing a 2.8 mm × 2.8 mm area on the build plate was used to image both no-powder and powder single beads on a commercial laser powder bed fusion machine. Melt pool temperature fields for 316L stainless steel at varying processing conditions show peaks between 3300 K and 3700 K depending on the laser power and increased variability in the presence of powder. Measurements of nickel superalloy 718 and Ti-6Al-4V show comparable temperatures, with increased plume obstruction, especially in Ti-6Al-4V due to vaporization of aluminum. Multi-physics CFD models are used to simulate metal melt pools but some parameters such as the accommodation and Fresnel coefficients are not well characterized. Fitting a FLOW-3D® CFD model to ex-situ measurements of the melt pool cross-sectional geometry for 316L stainless steel identifies multiple combinations of Fresnel coefficient and accommodation coefficient that lead to geometric agreement. Only two of these combinations show agreement with the thermal images, motivating the need for thermal imaging as a means to advance validation of complex physics models. Our methodology can be applied to any color camera to better monitor and understand melt pools that yield high-quality parts.

The HTML version of this Article incorrectly omits Supplementary Movie 1 , Supplementary Movie 2 , Supplementary Movie 3 , and Supplementary Movie 4 . The Supplementary Movies can be found as Supplementary Information associated with the original article and this Correction, and their descriptions can be found in the file entitled ‘Description of Additional Supplementary Files’.

Sintering theory predicts no long-range mass transport or distortion for uniformly heated particles during particle coalescence. However, in sintering-based manufacturing processes, permanent part distortion is often observed. The driving forces and mechanisms leading to this phenomenon are not understood, and efforts to reduce distortion are largely limited to a trial-and-error approach. In this paper, we demonstrate that distortion during sintering results from mass-transport driven by nonhomogeneous temperature distribution. We then show that hitherto unknown mass transport mechanisms, working in the direction opposite to temperature gradient are the likely cause of distortion. The experimental setup, designed for this purpose, enables the quantification of distortion during sintering. Two possible mass transport mechanisms are defined, and the continuum model applicable to both is formulated. The model accurately predicts the transient and permanent distortion observed during experiments, including their size dependence. Methods to control distortion that can give rise to 4D printing are discussed.

High Resolution Image Download MS PowerPoint Slide Infrared-pump, electronic-probe (IPEP) spectroscopy is used to measure heat flow into and out of CdSe nanocrystals suspended in an organic solvent, where the surface ligands are initially excited with an infrared pump pulse. Subsequently, the heat is transferred from the excited ligands to the nanocrystals and in parallel to the solvent. Parallel heat transfer in opposite directions uniquely enables us to differentiate the thermal conductances at the nanocrystal/ligand and ligand/solvent interfaces. Using a novel solution to the heat diffusion equation, we fit the IPEP data to find that the nanocrystal/ligand conductances range from 88 to 135 MW m –2 K –1 and are approximately 1 order of magnitude higher than the ligand/solvent conductances, which range from 7 to 26 MW m –2 K –1 . Transient nonequilibrium molecular dynamics (MD) simulations of nanocrystal suspensions agree with IPEP data and show that ligands bound to the nanocrystal by bidentate bonds have more than twice the per-ligand conductance as those bound by monodentate bonds.

Powder-based additive manufacturing has transformed the manufacturing industry over the last decade. In the Laser Powder Bed Fusion (L-PBF) process, a specific part is built in an iterative manner in which two-dimensional cross-sections are formed on top of each other by selectively melting and fusing the desired areas of the powder bed. In this process, the temperature field and melt pool morphology play a major role in the quality of the manufactured part and its possible defects. Therefore, predicting these factors is of high importance. However, simulating such a complex phenomenon is usually very time-consuming and requires huge computational resources. In this work, we create three datasets consisting of single-trail L-PBF processes using the Flow-3D simulation software and use them to train a convolutional neural network capable of predicting the three-dimensional temperature field solely by taking the process parameters and the time step as input. The CNN achieves a relative Root Mean Squared Error less of than 5% for the temperature field in the solidifying region and an average Intersection over Union score of 80% to 90% in predicting the three-dimensional geometry of the melt pool. Moreover, since time is included as one of the inputs of the model, the temperature field can be obtained in a matter of a few seconds for any arbitrary time step without the need to iterate and compute all the steps.