Machine Learning-Assisted Design of Material Properties

The influence of different factors on the bond strength of lithium disilicate-reinforced glass-ceramics to Resin: a machine learning analysis

BACKGROUND

To assess the influence of various factors on the bond strength of glass-based ceramics and develop a model that can predict the bond strength values using machine learning (ML).

METHODS

The bond strength values of lithium disilicate-reinforced glass-ceramics were collected from existing literature. Nineteen features were listed, and 9 ML algorithms, including logistic regression, k-nearest neighbors, support vector machine, decision tree, ensemble methods (extra trees, random forest, gradient boosting, and extreme gradient boosting), and multilayer perceptron, were employed. Importance analysis was performed to determine the significance of the 19 features. A new data set comprising the top five contributing features was used for bond strength class prediction. Grid search cross-validation (CV) and stratified tenfold CV were employed for hyperparameter tuning and model performance assessments. The evaluation metrics used were the area under the receiver operating characteristic curve (AUC) and accuracy. Nested CV was also employed to assess the model performance and avoid untruly optimistic results.

RESULTS

A total of 193 bond strength values were collected. Hydrofluoric acid concentration and etching time, gamma-methacryloxypropyltrimethoxysilane or 10-methacryloxydecyldihydrogen phosphate in the primer, and Bis-GMA in the cement were the top five features contributing to the bond strength. Stratified CV produced AUC scores of 0.71-0.93 and accuracy scores of 0.64-0.83. Extreme gradient boosting achieved superior model performance and accuracy and demonstrated good performance in predicting the range of bond strength values.

CONCLUSIONS

ML shows promise as a data-driven tool for predicting the bond strength of glass-based ceramics to resin.

Machine Learning-Aided Inverse Design and Discovery of Novel Polymeric Materials for Membrane Separation

Polymeric membranes have been widely used for liquid and gas separation in various industrial applications over the past few decades because of their exceptional versatility and high tunability. Traditional trial-and-error methods for material synthesis are inadequate to meet the growing demands for high-performance membranes. Machine learning (ML) has demonstrated huge potential to accelerate design and discovery of membrane materials. In this review, we cover strengths and weaknesses of the traditional methods, followed by a discussion on the emergence of ML for developing advanced polymeric membranes. We describe methodologies for data collection, data preparation, the commonly used ML models, and the explainable artificial intelligence (XAI) tools implemented in membrane research. Furthermore, we explain the experimental and computational validation steps to verify the results provided by these ML models. Subsequently, we showcase successful case studies of polymeric membranes and emphasize inverse design methodology within a ML-driven structured framework. Finally, we conclude by highlighting the recent progress, challenges, and future research directions to advance ML research for next generation polymeric membranes. With this review, we aim to provide a comprehensive guideline to researchers, scientists, and engineers assisting in the implementation of ML to membrane research and to accelerate the membrane design and material discovery process.

AI-Driven Insight into Polycarbonate Synthesis from CO2: Database Construction and Beyond

Recent advancements in materials science have garnered significant attention within the research community. Over the past decade, substantial efforts have been directed towards the exploration of innovative methodologies for developing new materials. These efforts encompass enhancements to existing products or processes and the design of novel materials. Of particular significance is the synthesis of specific polymers through the copolymerization of epoxides with CO2. However, several uncertainties emerge in this chemical process, including challenges associated with successful polymerization and the properties of the resulting materials. These uncertainties render the design of new polymers a trial-and-error endeavor, often resulting in failed outcomes that entail significant financial, human resource, and time investments due to unsuccessful experimentation. Artificial Intelligence (AI) emerges as a promising technology to mitigate these drawbacks during the experimental phase. Nonetheless, the availability of high-quality data remains crucial, posing particular challenges in the context of polymeric materials, mainly because of the stochastic nature of polymers, which impedes their homogeneous representation, and the variation in their properties based on their processing. In this study, the first dataset linking the structure of the epoxy comonomer, the catalyst employed, and the experimental conditions of polymerization to the reaction's success is described. A novel analytical pipeline based on ML to effectively exploit the constructed database is introduced. The initial results underscore the importance of addressing the dimensionality problem. The outcomes derived from the proposed analytical pipeline, which infer the molecular weight, polydispersity index, and conversion rate, demonstrate promising adjustment values for all target parameters. The best results are measured in terms of the (Determination Coefficient) R2 between real and predicted values for all three target magnitudes. The best proposed solution provides a R2 equal to 0.79, 0.86, and 0.93 for the molecular weight, polydispersity index, and conversion rate, respectively. The proposed analytical pipeline is automatized (including AutoML techniques for ML models hyperparameter tuning), allowing easy scalability as the database grows, laying the foundation for future research.

Distribution of Single-Particle Resonances Determines the Plasmonic Response of Disordered Nanoparticle Ensembles

Understanding how colloidal soft materials interact with light is crucial to the rational design of optical metamaterials. Electromagnetic simulations are computationally expensive and have primarily been limited to model systems described by a small number of particles-dimers, small clusters, and small periodic unit cells of superlattices. In this work we study the optical properties of bulk, disordered materials comprising a large number of plasmonic colloidal nanoparticles using Brownian dynamics simulations and the mutual polarization method. We investigate the far-field and near-field optical properties of both colloidal fluids and gels, which require thousands of nanoparticles to describe statistically. We show that these disordered materials exhibit a distribution of particle-level plasmonic resonance frequencies that determines their ensemble optical response. Nanoparticles with similar resonant frequencies form anisotropic and oriented clusters embedded within the otherwise isotropic and disordered microstructures. These collectively resonating morphologies can be tuned with the frequency and polarization of incident light. Knowledge of particle resonant distributions may help to interpret and compare the optical responses of different colloidal structures, correlate and predict optical properties, and rationally design soft materials for applications harnessing light.

Sound Absorption Performance of Ultralight Honeycomb Sandwich Panels Filled with "Network" Fibers-Juncus effusus

This study investigates lightweight and efficient candidates for sound absorption to address the growing demand for sustainable and eco-friendly materials in noise attenuation. Juncus effusus (JE) is a natural fiber known for its unique three-dimensional network, providing a viable and sustainable filler for enhanced sound absorption in honeycomb panels. Microperforated-panel (MPP) honeycomb absorbers incorporating JE fillers were fabricated and designed, focusing on optimizing the absorber designs by varying JE filler densities, geometrical arrangements, and MPP parameters. At optimal filling densities, the MPP-type honeycomb structures filled with JE fibers achieved high noise reduction coefficients (NRC) of 0.5 and 0.7 at 20 mm and 50 mm thicknesses, respectively. Using an analytical model and an artificial neural network (ANN) model, the sound absorption characteristics of these absorbers were successfully predicted. This study demonstrates the potential of JE fibers in improving noise mitigation strategies across different industries, offering more sustainable and efficient solutions for construction and transportation.

Challenges and Strategies for Synthesizing High Performance Micro and Nanoscale High Entropy Oxide Materials

High-entropy oxide micro/nano materials (HEO MNMs) have shown broad application prospects and have become hot materials in recent years. This review comprehensively provides an overview of the latest developments and covers key aspects of HEO MNMs, by discussing design principles, computer-aided structural design, synthesis challenges and strategies, as well as application areas. The analysis of the synthesis process includes the role of high-throughput process in large-scale synthesis of HEOs MNMs, along with the effects of temperature elevation and undercooling on the formation of HEO MNMs. Additionally, the article summarizes the application of high-precision and in situ characterization devices in the field of HEO MNMs, offering robust support for related research. Finally, a brief introduction to the main applications of HEO MNMs is provided, emphasizing their key performances. This review offers valuable guidance for future research on HEO MNMs, outlining critical issues and challenges in the current field.

Illuminating Disorder: Optical Properties of Complex Plasmonic Assemblies

The optical properties of disordered plasmonic nanoparticle assemblies can be continuously tuned through the structural organization and composition of their colloidal building blocks. However, progress in the design and experimental realization of these materials has been limited by challenges associated with controlling and characterizing disordered assemblies and predicting their optical properties. This Perspective discusses integrated studies of experimental assembly of disordered optical materials, such as doped metal oxide nanocrystal gels and metasurfaces, with electromagnetic computations on large-scale simulated structures. The simulations prove vital for connecting experimental parameters to disordered structural motifs and optical properties, revealing structure-property relations that inform design choices. Opportunities are identified for optimizing optical property designs for disordered materials using computational inverse methods and tools from machine learning.

Machine Learning-Aided High-Throughput First-Principles Calculations to Predict the Formation Energy of μ Phase

The μ phase is a type of hard and brittle constituent that exists in high-temperature alloys. The formation energy is a crucial thermochemical datum, and the accurate calculation of the formation energy of the μ phase contributes to the material design of high-temperature alloys. Traditional first-principles calculations demand significant computational time and resources. In this study, an innovative machine learning (ML)-based approach to accurately predict the formation energy of the μ phase is proposed. This approach involves the utilization of six algorithms and two model evaluation methods to construct the ML models. Leveraging a comprehensive data set containing 1036 binary configurations of the μ phase, the model trained using a 10-fold cross-validation technique, and the multilayer perceptron (MLP) algorithm achieves a mean absolute error (MAE) of 23.906 meV/atom. To validate its generalization performance, the trained model is further validated on 900 ternary configurations, resulting in an MAE of 32.754 meV/atom. Compared with solely using traditional first-principles calculations, our approach significantly reduces the computational time by at least 52%. Moreover, the ML model exhibits exceptional accuracy in predicting the lattice parameters of the μ phase. The MAE values for the a and c parameters are 0.024 and 0.214 Å, respectively, corresponding to low error rates of only 0.479 and 0.578%. Additionally, the ML model was utilized to accurately predict the formation energy of all of the possible ternary configurations. To enhance accessibility to the formation energy data of the μ phase, a user-friendly graphical user interface (GUI) was developed, ensuring convenient usability for researchers and practitioners.

Pair-Variational Autoencoders for Linking and Cross-Reconstruction of Characterization Data from Complementary Structural Characterization Techniques

In materials research, structural characterization often requires multiple complementary techniques to obtain a holistic morphological view of a synthesized material. Depending on the availability and accessibility of the different characterization techniques (e.g., scattering, microscopy, spectroscopy), each research facility or academic research lab may have access to high-throughput capability in one technique but face limitations (sample preparation, resolution, access time) with other technique(s). Furthermore, one type of structural characterization data may be easier to interpret than another (e.g., microscopy images are easier to interpret than small-angle scattering profiles). Thus, it is useful to have machine learning models that can be trained on paired structural characterization data from multiple techniques (easy and difficult to interpret, fast and slow in data collection or sample preparation) so that the model can generate one set of characterization data from the other. In this paper we demonstrate one such machine learning workflow, Pair-Variational Autoencoders (PairVAE), that works with data from small-angle X-ray scattering (SAXS) that present information about bulk morphology and images from scanning electron microscopy (SEM) that present two-dimensional local structural information on the sample. Using paired SAXS and SEM data of newly observed block copolymer assembled morphologies [open access data from Doerk G. S.; et al. Sci. Adv.2023, 9 ( (2), ), eadd3687], we train our PairVAE. After successful training, we demonstrate that the PairVAE can generate SEM images of the block copolymer morphology when it takes as input that sample's corresponding SAXS 2D pattern and vice versa. This method can be extended to other soft material morphologies as well and serves as a valuable tool for easy interpretation of 2D SAXS patterns as well as an engine for generating ensembles of similar microscopy images to create a database for other downstream calculations of structure-property relationships.

In situ reductive dehalogenation of groundwater driven by innovative organic carbon source materials: Insights into the organohalide-respiratory electron transport chain

In situ bioremediation using organohalide-respiring bacteria (OHRB) is a prospective method for the removal of persistent halogenated organic pollutants from groundwater, as OHRB can utilize H₂ or organic compounds produced by carbon source materials as electron donors for cell growth through organohalide respiration. However, few previous studies have determined the suitability of different carbon source materials to the metabolic mechanism of reductive dehalogenation from the perspective of electron transfer. The focus of this critical review was to reveal the interactions and relationships between carbon source materials and functional microbes, in terms of the electron transfer mechanism. Furthermore, this review illustrates some innovative strategies that have used the physiological characteristics of OHRB to guide the optimization of carbon source materials, improving the abundance of indigenous dehalogenated bacteria and enhancing electron transfer efficiency. Finally, it is proposed that future research should combine multi-omics analysis with machine learning (ML) to guide the design of effective carbon source materials and optimize current dehalogenation bioremediation strategies to reduce the cost and footprint of practical groundwater bioremediation applications.

Emerging Trends in Machine Learning: A Polymer Perspective

In the last five years, there has been tremendous growth in machine learning and artificial intelligence as applied to polymer science. Here, we highlight the unique challenges presented by polymers and how the field is addressing them. We focus on emerging trends with an emphasis on topics that have received less attention in the review literature. Finally, we provide an outlook for the field, outline important growth areas in machine learning and artificial intelligence for polymer science and discuss important advances from the greater material science community.

Materials discovery of ion-selective membranes using artificial intelligence

Significant attempts have been made to improve the production of ion-selective membranes (ISMs) with higher efficiency and lower prices, while the traditional methods have drawbacks of limitations, high cost of experiments, and time-consuming computations. One of the best approaches to remove the experimental limitations is artificial intelligence (AI). This review discusses the role of AI in materials discovery and ISMs engineering. The AI can minimize the need for experimental tests by data analysis to accelerate computational methods based on models using the results of ISMs simulations. The coupling with computational chemistry makes it possible for the AI to consider atomic features in the output models since AI acts as a bridge between the experimental data and computational chemistry to develop models that can use experimental data and atomic properties. This hybrid method can be used in materials discovery of the membranes for ion extraction to investigate capabilities, challenges, and future perspectives of the AI-based materials discovery, which can pave the path for ISMs engineering.