arXiv subjects

Kenneth Vecchio

Publications and source records attributed to Kenneth Vecchio.

9 recordsLinked to original sources

arXiv2026

Interpretable Material Spatial Intelligence for Discovery of Governing Microstructural Features

Many material systems exhibit complex spatial and temporal interactions across multiple length scales and modalities that govern macroscopic behavior. Although Machine Learning (ML) is widely used in materials science to predict this behavior, most approaches still rely on handcrafted descriptors or aggregated representations that overlook spatial organization, limiting insight into governing mechanisms. We introduce Materials Spatial Intelligence (MSI), a framework inspired by spatial intelligence that learns directly from multimodal spatial observations of material systems. MSI encodes high-resolution microstructural and deformation data into shared latent representations that preserve spatial relationships while supporting property prediction, interpretation, and optimization. By combining multimodal representation learning, MSI identifies the key features governing mechanical behavior and property trade-offs in structural alloys. Beyond prediction, MSI enables feature-driven microstructure optimization and mechanism discovery. More broadly, MSI establishes a foundation for applying spatial intelligence to materials science, leveraging interpretable ML systems to accelerate scientific discovery and materiel design

cond-mat.mtrl-sci↗

arXiv2025

Plasticity Encoding and Mapping during Elementary Loading for Accelerated Mechanical Properties Prediction

Encoding metal plasticity captured from high-resolution digital image correlation (DIC) can be leveraged to predict a wide range of monotonic and cyclic macroscopic properties of metallic materials. To capture the spatial heterogeneity of plasticity that develops in metals, latent space features describing plasticity of a small region are spatially mapped across a large field of view while maintaining the same spatial relationships as the experimental measurements. Latent space feature maps capture the complexity and heterogeneity of metal plasticity as a low-dimensional representation. These feature maps are then used to train a convolutional neural network-based model to predict monotonic and cyclic macroscopic properties. The approach is demonstrated on a large set of face-centered cubic metals, enabling rapid and accurate property prediction. The effects of hyperparameters and training strategies are analyzed, and the extension of the proposed approach to a broader range of metallic materials and loading conditions is discussed.

cond-mat.mtrl-sci↗

arXiv2025

Learning Metal Microstructural Heterogeneity through Spatial Mapping of Diffraction Latent Space Features

To leverage advancements in machine learning for metallic materials design and property prediction, it is crucial to develop a data-reduced representation of metal microstructures that surpasses the limitations of current physics-based discrete microstructure descriptors. This need is particularly relevant for metallic materials processed through additive manufacturing, which exhibit complex hierarchical microstructures that cannot be adequately described using the conventional metrics typically applied to wrought materials. Furthermore, capturing the spatial heterogeneity of microstructures at the different scales is necessary within such framework to accurately predict their properties. To address these challenges, we propose the physical spatial mapping of metal diffraction latent space features. This approach integrates (i) point diffraction data encoding via variational autoencoders or contrastive learning and (ii) the physical mapping of the encoded values. Together these steps offer a method offers a novel means to comprehensively describe metal microstructures. We demonstrate this approach on a wrought and additively manufactured alloy, showing that it effectively encodes microstructural information and enables direct identification of microstructural heterogeneity not directly possible by physics-based models. This data-reduced microstructure representation opens the application of machine learning models in accelerating metallic material design and accurately predicting their properties.

cond-mat.mtrl-sci↗

arXiv2024

Illustrating an Effective Workflow for Accelerated Materials Discovery

Algorithmic materials discovery is a multi-disciplinary domain that integrates insights from specialists in alloy design, synthesis, characterization, experimental methodologies, computational modeling, and optimization. Central to this effort is a robust data management system paired with an interactive work platform. This platform should empower users to not only access others data but also integrate their analyses, paving the way for sophisticated data pipelines. To realize this vision, there is a need for an integrative collaboration platform, streamlined data sharing and analysis tools, and efficient communication channels. Such a collaborative mechanism should transcend geographical barriers, facilitating remote interaction and fostering a challenge-response dynamic. In this paper, we present our ongoing efforts in addressing the critical challenges related to an accelerated Materials Discovery Framework as a part of the High-Throughput Materials Discovery for Extreme Conditions Initiative. Our BIRDSHOT Center has successfully harnessed various tools and strategies, including the utilization of cloud-based storage, a standardized sample naming convention, a structured file system, the implementation of sample travelers, a robust sample tracking method, and the incorporation of knowledge graphs for efficient data management. Additionally, we present the development of a data collection platform, reinforcing seamless collaboration among our team members. In summary, this paper provides an illustration and insight into the various elements of an efficient and effective workflow within an accelerated materials discovery framework while highlighting the dynamic and adaptable nature of the data management tools and sharing platforms.

cond-mat.mtrl-sci↗

arXiv2021

Enhancing plasticity in high-entropy refractory ceramics via tailoring valence electron concentration

Bottom-up design of high-entropy ceramics is a promising approach for realizing materials with unique combination of high hardness and fracture-resistance at elevated temperature. This work offers a simple yet fundamental design criterion - valence electron concentration (VEC) > ~9.5 e-/f.u. to populate bonding metallic states at the Fermi level - for selecting elemental compositions that may form rocksalt-structure (B1) high-entropy ceramics with enhanced plasticity (reduced brittleness). Single-phase B1 (HfTaTiWZr)C and (MoNbTaVW)C, chosen as representative systems due to their specific VEC values, are here synthesized and tested. Nanoindentation arrays at various loads and depths statistically show that (HfTaTiWZr)C (VEC=8.6 e-/f.u.) is hard but brittle, whilst (MoNbTaVW)C (VEC=9.4 e-/f.u.) is hard and considerably more resistant to fracture than (HfTaTiWZr)C. Ab initio molecular dynamics simulations and electronic-structure analysis reveal that the improved fracture-resistance of (MoNbTaVW)C subject to tensile and shear deformation may originate from the intrinsic material's ability to undergo local lattice transformations beyond tensile yield points, as well as from relatively facile activation of lattice slip. Additional simulations, carried out to follow the evolution in mechanical properties as a function of temperature, suggest that (MoNbTaVW)C may retain good resistance to fracture up to ~900-1200K, whereas (HfTaTiWZr)C is predicted to remain brittle at all investigated temperatures.

cond-mat.mtrl-sci↗

arXiv2019

Novel Remapping Approach for HR-EBSD based on Demons Registration

In this study, the possibility of utilizing a computer vision algorithm, i.e., demons registration, to accurately remap electron backscatter diffraction patterns for high resolution electron backscatter diffraction applications is presented. First, the angular resolution of demons registration is demonstrated to be lower than the conventional cross-correlation based method, particularly at misorientation angles > 9 degrees. In addition, GPU acceleration has been applied to significantly boost the speed of iterative registration between a pair of patterns with 10{\deg} misorientation to under 1s. Second, demons registration is implemented as a first-pass remapping, followed by a second pass cross-correlation method, which results in angular resolution of ~0.5*10-4 degrees, a phantom stress value of ~35MPa and phantom strain of ~2*10-4, on dynamically simulated patterns, without the need of implementing robust fitting or iterative remapping. Lastly, the new remapping method is applied to a large experimental dataset collected from an as-built additively-manufactured Inconel 625 cube, which shows significant residual stresses built-up near the large columnar grain region and regularly arranged GND structures.

physics.comp-ph↗

arXiv2019

Computer Vision Approach to Study Deformation of Materials

Characterization of the deformation of materials across different length scales has continuously attracted enormous attention from the mechanics and materials communities. In this study, the possibility of utilizing a computer vision algorithm to extract deformation information of materials has been explored, which greatly expands the use of computer vision approaches to studying mechanics of materials and potentially opens new dialogues between the two communities. The computer vision algorithm is first developed and tested on computationally deformed images, before evaluating experimentally collected images on speckle painted samples before and after deformation. Moreover, a virtual experiment has also shown the feasibility of mapping surface strain of a sample based on its natural pattern with significantly improved accuracy compared to DIC result, which provides new opportunities in experimentation and computer algorithms to study deformation mechanics of materials. Validation experiments include evaluating the performance of strain mapping using the computer vision approach in the uniaxial tensile test and three-point bending test, compared with extensometer reading and digital image correlation respectively.

physics.comp-ph↗

arXiv2019

Automated Reconstruction of Spherical Kikuchi Maps

An automated approach to reconstruct spherical Kikuchi maps from experimentally collected electron backscatter diffraction patterns and overlay each pattern onto its corresponding position on a simulated Kikuchi sphere is presented in this study. This work demonstrates the feasibility of warping any Kikuchi pattern onto its corresponding location of a simulated Kikuchi sphere and reconstructing a spherical Kikuchi map of a known phase based on any set of experimental patterns. The proposed method is the first automated approach to reconstructing spherical Kikuchi maps from experimental Kikuchi patterns. It potentially enables more accurate orientation calculation, new pattern center refinement methods, improved dictionary-based pattern matching, and phase identification in the future.

cond-mat.mtrl-sci↗

arXiv2019

A High-Entropy Silicide: (Mo0.2Nb0.2Ta0.2Ti0.2W0.2)Si2

A high-entropy metal disilicide, (Mo0.2Nb0.2Ta0.2Ti0.2W0.2)Si2, has been successfully synthesized. X-ray diffraction (XRD), energy dispersive X-ray spectroscopy (EDX), and electron backscatter diffraction (EBSD) collectively show the formation of a single high-entropy silicide phase. This high-entropy (Mo0.2Nb0.2Ta0.2Ti0.2W0.2)Si2 possesses a hexagonal C40 crystal structure with ABC stacking sequence and a point group of P6222. This discovery expands the known families of high-entropy materials from metals, oxides, borides, carbides, and nitrides to a silicide, for the first time to our knowledge, as well as demonstrating that a new, non-cubic, crystal structure (with lower symmetry) can be made into high-entropy. This (Mo0.2Nb0.2Ta0.2Ti0.2W0.2)Si2 exhibits high nanohardness of 16.7 +- 1.9 GPa and Vickers hardness of 11.6 +- 0.5 GPa. Moreover, it has a low thermal conductivity of 6.9 +- 1.1 W m-1 K-1, which is approximately one order of magnitude lower than that of the widely-used tetragonal MoSi2 and ~1/3 of those reported values for the hexagonal NbSi2 and TaSi2 with the same crystal structure.

cond-mat.mtrl-sci↗