Accurately predicting the onset of thermal degradation (T d ) in polyimides is critical to materials development in high-temperature aerospace and electronics applications. Direct measurement of T d via thermogravimetric analysis requires costly synthesis, necessitating alternate routes to rapidly screen monomer chemistries for promising new materials. In this work, we develop an integrated framework combining reactive molecular dynamics with machine learning models to rapidly and accurately estimate the 5% mass loss T d of polyimide chemistries. Beginning from a curated data set of polyimide repeat units with experimentally validated T d values, we perform reactive molecular dynamics with ReaxFF to simulate thermal degradation as the polymers are heated from 300 to 4000 K. We use the simulations to create features related to the onset of thermal degradation, which we combine with chemical fingerprint information obtained from RDKit to serve as the input to XGBoost models which are trained to predict T d . We find the performance of such models is highly dependent on the choice of features, and the means by which such features are combined. While purely chemistry-based models perform well, we find we can significantly enhance the accuracy of T d prediction by training meta-models, in which the predictions of separate XGBoost models trained on chemistry information and simulation information, respectively, are used to featurize new machine learning models to make final predictions. Our work provides a foundation upon which new chemistries can be evaluated for thermal degradation performance, leading to the generation and design of new polyimide chemistries with very high T d .

In this study, we investigate the thermo-mechanical properties and strain accommodation mechanisms in novel thermoplastic helicene-siloxane hybrid copolymers via molecular dynamics simulations to explore their viability as embedded sensors in helicene based polymers. We systematically vary helicene size and siloxane linker flexibility to understand their impact on thermo-mechanical behavior. The simulations reveal that helicene molecules exhibit local ordering at lower temperatures and transition to random distribution at elevated temperatures. Thereby, a helicene rearrangement temperature (Th) is identified, marking the onset of significant rotational fluctuations of the helicene units. The glass transition temperature (Tg) is also determined for all systems, showing an increase with increasing helicene content. We observe Th to be consistently higher than Tg, a behavior analogous to order-disorder transitions in liquid crystalline polymers, highlighting that helicene rearrangement requires greater thermal activation than bulk polymer segmental motion. Under uniaxial tensile deformation, the helicenes show minimal individual elongation (approximately 0.5% to 2% local strain at 300% applied strain). Instead, the macroscopic strain is observed to be primarily accommodated by the stretching of the flexible siloxane component and the reorientation of the helicene units along the loading axis in all modeled systems.

Carbon fiber-reinforced composites (CFRC) are pivotal in advanced engineering applications due to their exceptional mechanical properties. A deep understanding of CFRC behavior under mechanical loading is essential for optimizing performance in demanding applications such as aerospace structures. While traditional Finite Element Method (FEM) simulations, including advanced techniques like Interface-enriched Generalized FEM (IGFEM), offer valuable insights, they can struggle with computational efficiency. Existing data-driven surrogate models partially address these challenges by predicting propagated damage or stress-strain behavior but fail to comprehensively capture the evolution of stress and damage throughout the entire deformation history, including crack initiation and propagation. This study proposes a novel auto-regressive composite U-Net deep learning model to simultaneously predict stress and damage fields during CFRC deformation. By leveraging the U-Net architecture’s ability to capture spatial features and integrate macro- and micro-scale phenomena, the proposed model overcomes key limitations of prior approaches. The model achieves high accuracy in predicting the evolution of stress and damage distribution within the microstructure of a CFRC under unidirectional strain, offering a speed-up of over 150 times compared to IGFEM.

MXenes combine rich surface chemistry, mechanical strength, and high conductivity for a multitude of emerging applications. Predictive modeling supports accelerated materials designs and has been limited by the absence of validated and transferable force fields. Here, we introduce an interpretable, reactive INTERFACE force field (IFF and IFF-R) for Ti3C2Tx MXenes that is trained based on chemical knowledge and achieves quantitative agreement with experiments across lattice parameters (<0.5%), density (<0.2%), liquid contact angles, Raman spectra, and the in-plane elastic modulus (∼320 GPa). The models cover surface terminations from hydroxyl (-OH) to fluorine (-F) groups and are extensible to other chemistries. We introduce pH-resolved surface chemistry and identify dopamine adsorption mechanisms at MXene-aqueous interfaces supported by QCM-D and UV-Vis experiments. The data reveal coplanar and perpendicular binding modes and concentration-dependent multilayer assembly. We predict previously inaccessible properties, including termination-dependent cleavage energies, interlayer shear moduli and dynamic shear failure, nanoindentation and brittle fracture, anisotropic in-plane and out-of-plane thermal conductivities, including the role of defects. Agreement with available experimental data is consistently close and exceeds DFT accuracy across the benchmark properties examined. The IFF/IFF-R model is compatible with CHARMM, AMBER, OPLS, and CVFF force fields for simulations of MXenes with diverse surface terminations, electrolyte interfaces, biointerfaces, and polymer composites without additional parameters. Parameter sets, 3D models, and analysis scripts are provided for community use. The validated, reactive, and transferable IFF framework facilitates predictive design of MXene-based films, membranes, sensing interfaces, and composites.

Carbyne is a one-dimensional chain of carbon atoms that has high elastic modulus and thermal conductivity. However, its mechanical properties vary with temperature. We use molecular dynamics to investigate the bond structure of polymers formed from cumulenic and polyynic carbyne pyrolyzed at high temperatures after quenching and predict the polymers’ mechanical properties. We observe that nanostructures begin to form during pyrolysis at 1,000K, and there is a major transformation from sp-hybridized carbyne to sp2-, and sp3-hybridized polymers after heating the carbyne up to 3,000K. Pyrolyzed cumulene forms an amorphous carbon polymer with graphitic structures that become more crystalline and porous with an increase in temperature. However, pyrolyzed polyyne forms amorphous carbon polymer with no indication of graphitization and lower density than pyrolyzed cumulene when heated above 1,000K. We perform compression testing of the pyrolyzed carbyne polymers after quenching them to 300K and observe that the graphitization of the pyrolyzed cumulene leads to a significant increase in compression modulus at 2,000K. However, the less stable nanostructures and lower density of pyrolyzed polyyne at 2,000K result in a decrease of compressive modulus along all axes. We find that the pyrolysis-driven reorientation of bonds in the carbyne polymers contributes to the directional dependence of the compression modulus with respect to the initial axis of the carbyne at 300K. This is most notable after pyrolysis of the cumulene and polyyne at 3,000K; the modulus of the polymers decreases along the fiber axis and increases along axes perpendicular to the fiber axis as bonds are reoriented at high temperatures.

Helicenes are a class of helically chiral, aromatic molecules that are often functionalized and are of interest for a variety of applications due to their axial chirality. However, their syntheses are typically conducted under high dilution conditions to prevent undesirable side reactions and require large volumes of solvent, which makes scaling up a challenge. This study discusses the challenges of scaling helicene syntheses and offers facile strategies to address some of these challenges. The increased interest for using helicenes to address materials, sensing, and electronic applications necessitates that strategies for scaling them effectively with high purity need to be developed. It is well-known that flow chemistry facilitates more reproducible, scalable, safe, and efficient options for chemical synthesis, making it a valuable tool in both academic and industrial settings, as it allows for precise control over reaction conditions such as stoichiometry, mixing, temperature, and reaction time, leading to greater yields and better selectivity for a variety of reaction classes. Using quantitative 1H-NMR and isolated yields of the desired product and notable side products, we evaluated three reactor systems: 1L-batch, 5L-batch, and flow reactors towards the synthesis of a [5]-helicene tetraester (5HLTE). After initial optimization, the optimal conditions were used to demonstrate the scalability and provided throughput of similar to 5 g/day in a 5 mL reactor flow system, scaling linearly with reactor volume. Discrete control of purity is vital for these applications in that impurities may provide incorrect structure-property conclusions when applied to organic electronics and polymer mechanical properties.

Carbon-carbon (C/C) composites are attractive materials for high-speed flights and terrestrial atmospheric reentry applications due to their insulating thermal properties, thermal resistance, and high strength-to-weight ratio. It is important to understand the evolving structure-property correlations in these materials during pyrolysis, but the extreme laboratory conditions required to produce C/C composites make it difficult to quantify the properties in situ. This work presents an atomistic modeling methodology to pyrolyze a crosslinked phenolic resin network and track the evolving thermomechanical properties of the skeletal matrix during simulated pyrolysis. First, the crosslinked resin is pyrolyzed and the resulting char yield and mass density are verified to match experimental values, establishing the model's powerful predictive capabilities. Young's modulus, yield stress, Poisson's ratio, and thermal conductivity are calculated for the polymerized structure, intermediate pyrolyzed structures, and fully pyrolyzed structure to reveal structure-property correlations, and the evolution of properties are linked to observed structural features. It is determined that reduction in fractional free volume and densification of the resin during pyrolysis contribute significantly to the increase in thermomechanical properties of the skeletal phenolic matrix. A complex interplay of the formation of six-membered carbon rings at the expense of five and seven-membered carbon rings is revealed to affect thermal conductivity. Increased anisotropy was observed in the latter stages of pyrolysis due to the development of aligned aromatic structures. Experimentally validated predictive atomistic models are a key first step to multiscale process modeling of C/C composites to optimize next-generation materials.

Molecular generative models based on deep learning have increasingly gained attention for their ability in de novo polymer design. However, there remains a knowledge gap in the thorough evaluation of these models. This benchmark study explores de novo polymer design using six popular deep generative models: Variational Autoencoder (VAE), Adversarial Autoencoder (AAE), Objective-Reinforced Generative Adversarial Networks (ORGAN), Character-level Recurrent Neural Network (CharRNN), REINVENT, and GraphINVENT. Various metrics highlighted the excellent performance of CharRNN, REINVENT, and GraphINVENT, particularly when applied to the real polymer dataset, while VAE and AAE show more advantages in generating hypothetical polymers. The CharRNN, REINVENT, and GraphINVENT models were further trained on real polymers utilizing reinforcement learning methods, targeting the generation of hypothetical polymers with high glass transition temperatures. The findings of this study provide critical insights into the capabilities and limitations of each generative model, offering valuable guidance for future endeavors in polymer design and discovery.

Vitrimer is an emerging class of sustainable polymers with self-healing capabilities enabled by dynamic covalent adaptive networks. However, their limited molecular diversity constrains their property space and potential applications. Recent development in machine learning (ML) techniques accelerates polymer design by predicting properties and virtually screening candidates, yet the scarcity of available experimental vitrimer data poses challenges in training ML models. To address this, we leverage molecular dynamics (MD) data generated by our previous work to train and benchmark seven ML models covering six feature representations for glass transition temperature (Tg) prediction. By averaging predicted Tg from different models, the model ensemble approach outperforms individual models, allowing for accurate and efficient property prediction on unlabeled datasets. Two novel vitrimers are identified and synthesized, exhibiting experimentally validated higher Tg than existing bifunctional transesterification vitrimers, along with demonstrated healability. This work explores the possibility of using MD data to train ML models in the absence of sufficient experimental data, enabling the discovery of novel, synthesizable polymer chemistries with superior properties. The integrated MD-ML approach offers polymer chemists an efficient tool for designing polymers tailored to diverse applications.

Significant advancements in machine learning have accelerated and improved structure–property predictions for materials discovery. However, data are often scarce due to large parameter spaces consisting of chemistry, structure, synthesis, and processing variables. At small data limits, theory can be leveraged to inform machine learning (ML) models with domain knowledge and improve generalization. Here, we determine how the accuracy of first-principles calculations affects theory-informed ML predictions of the experimental solvation free energy ΔGsolvexp in both “small” (101–102) and “large” (103–104) data size limits. We compare several existing theory-informed techniques to a baseline (no theory) model: feature-informed, difference, and ratio. At small data limits, all theory-informed models exhibit lower RMSE, reducing training data size by more than 65% compared to the baseline model. With larger training set sizes and as theory prediction accuracy declines, the difference and ratio models exhibit larger errors than the baseline model, indicating negative transfer occurs. No negative transfer is observed in the feature-informed model; however, the model is unable to extrapolate outside of the known ΔGsolvexp distribution, whereas the difference and ratio models exhibit lower extrapolation error. Finally, we employ each model in an active learning algorithm and compare two exploration acquisition functions: maximum difference and maximum variance. Sampling with the maximum difference policy reduces RMSE and variance of the feature-informed model faster than maximum variance early in the exploration campaign, as it identifies label bounds in fewer iterations. Our findings highlight the balance between leveraging theory and relying on data-driven models in high-throughput materials discovery.

Carbon–carbon composites are a material commonly used in high heat flux heat environments, such as space missions for terrestrial re-entry. Phenolic resins have been used as carbon matrix precursors due to high char yields of 50 – 55

Graphene-based nanostructures hold immense potential as strong and lightweight materials, however, their mechanical properties such as modulus and strength are difficult to fully exploit due to challenges in atomic-scale engineering. This study presents a database of over 2,000 pristine and defective nanoscale CNT bundles and other graphitic assemblies, inspired by microscopy, with associated stress-strain curves from reactive molecular dynamics (MD) simulations using the reactive INTERFACE force field (IFF-R). These 3D structures, containing up to 80,000 atoms, enable detailed analyses of structure-stiffness-failure relationships. By leveraging the database and physics- and chemistry-informed machine learning (ML), accurate predictions of elastic moduli and tensile strength are demonstrated at speeds 1,000 to 10,000 times faster than efficient MD simulations. Hierarchical Graph Neural Networks with Spatial Information (HS-GNNs) are introduced, which integrate chemistry knowledge. HS-GNNs as well as extreme gradient boosted trees (XGBoost) achieve forecasts of mechanical properties of arbitrary carbon nanostructures with only 3 to 6% mean relative error. The reliability equals experimental accuracy and is up to 20 times higher than other ML methods. Predictions maintain 8 to 18% accuracy for large CNT bundles, CNT junctions, and carbon fiber cross-sections outside the training distribution. The physics- and chemistry-informed HS-GNN works remarkably well for data outside the training range while XGBoost works well with limited training data inside the training range. The carbon nanostructure database is designed for integration with multimodal experimental and simulation data, scalable beyond 100 nm size, and extendable to chemically similar compounds and broader property ranges. The ML approaches have potential for applications in structural materials, nanoelectronics, and carbon-based catalysts.

Two-dimensional MXenes combine mechanical strength, high conductivity, and rich surface chemistry; however, predictive modeling has been limited by the absence of validated and transferable force fields. We introduce an interpretable, reactive INTERFACE/IFF-R model for Ti3C2Tx MXenes that achieves quantitative agreement with experiments across lattice parameters (<0.5%), density (<0.2%), water and diiodomethane contact angles, Raman spectra, and the in-plane elastic modulus (~320 GPa). The model covers the full spectrum of surface terminations from hydroxyl (–OH) to fluorine (–F) groups and is readily extensible to additional chemistries. Using this validated foundation, we predict previously inaccessible properties, including termination-dependent cleavage energies, interlayer shear moduli and dynamic shear failure, reactive nanoindentation leading to brittle fracture, and anisotropic thermal conductivities spanning both in-plane and out-of-plane directions. We further establish pH-resolved surface chemistry and identify dopamine adsorption mechanisms at MXene–aqueous interfaces, supported by QCM-D and UV–Vis experiments, revealing preferred coplanar and perpendicular binding modes and concentration-dependent multilayer assembly. The model is fully compatible with CHARMM, AMBER, OPLS, CVFF, and IFF, enabling realistic simulations of MXene–polymer composites, electrolytes, biointerfaces, and diverse surface terminations without additional parameters. Complete parameter sets, 3D models, and analysis scripts are provided for community use. This framework delivers a validated, reactive, and transferable platform for predictive design of MXene-based films, composites, membranes, and sensing interfaces.

Viscosity is a crucial material property that influences a wide range of applications, including three-dimensional (3D) printing, lubricants, and solvents. However, experimental approaches to measuring viscosity face challenges such as handling multiple samples, high costs, and limited compound availability. To address these limitations, we have developed computational models for viscosity prediction of small organic molecules, utilizing machine learning (ML) and nonequilibrium molecular dynamics (NEMD) simulations. Our ML framework, which includes feed-forward neural networks (FNN) and physics-informed neural networks (PINN), is based on the largest data set of small molecule viscosities compiled from the literature. The PINN model, in particular, incorporates temperature dependence through a four-parameter model, allowing for the direct prediction of continuous temperature-dependent viscosity curves. The ML models demonstrate exceptional prediction accuracy for the viscosity of various organic compounds across a wide range of temperatures. External validation of our models further confirms that the ML prediction models outperform the NEMD approach in predicting viscosity across a diverse range of organic molecules and temperatures. This highlights the potential of ML models to overcome limitations in traditional MD simulations, which often struggle with accuracy for specific molecules or temperature ranges. Our further feature importance analysis revealed a strong correlation between molecular structure and viscosity. We emphasize the key role of substructures in determining viscosity, offering deeper molecular insights for material design with tailored viscosity.

Vitrimers are a novel class of sustainable polymers with dynamics covalent adaptive networks driven by bond exchange reactions between different constituents, making vitrimers reprocessable and recyclable. Current modeling approaches of bond exchange reactions fall short in realistically capturing the complete reaction pathways, which limit our understanding the viscoelastic properties of vitrimers. This research addresses these limitations by extending and employing Accelerated Reactive Molecular Dynamics (ReaxFF) technique, thus enabling a more accurate representation of vitrimer viscoelas- tic behavior at the molecular level. Bayesian optimization is employed to select force field parameters within the Accelerated ReaxFF framework, and an empirical function is proposed to model temperature dependency, thereby controlling reaction probabil- ities under varying temperatures. The extended framework is employed to simulate non-isothermal creep behavior of vitrimers under different applied stress levels, heating rates and numbers of reactions. The simulation results agree with experimental findings in literature, validating the robustness of the framework.

Vitrimers are a distinct category of polymers that could conduct dynamic cross-linking in response to temperature stimuli. In this work, using a molecular dynamics simulation model, we explore self-healing and the thermal transport behavior of vitrimer-graphene composites to overcome its inherent slow self-healing process. The temperature-dependent reversible cross-link mechanisms, which possess the capacity to dynamically modify the mechanical properties of these substances, are studied by explicitly integrating the temperature-dependent reaction probabilities. This modeling approach efficiently predict the changes in the mechanical properties of vitrimers when undergoing temperature cycling both above and below the topological freezing point, as well as the damage healing and subsequent recuperation of mechanical behaviors. The heat transport behavior of the graphene-vitrimer composite is investigated using non-equilibrium molecular dynamics in conjunction with self-healing models. The thermal conductivity of vitrimer-graphene composites, calculated by including a bilayer graphene sheet with varied flake size and interlayer spacings, exhibits a considerable enhancement as compared to standalone vitrimers. Notably, the thermal conductivity is subject to change when the separation distance changes. These results shed light on the self-healing and heat transmission in vitrimer and open the door to possible applications in several fields, including electronics, energy efficiency, aerospace, and materials research.

The structural integrity of MXene and MXene-based materials is important across applications from sensors to energy storage. While MXene processing has received significant attention, its structural integrity for real-world applications remains challenging due to its flake-like structure. Here the mechanical response of layered MXene-polymer nanocomposites (MPC) with high MXene concentration ( > 70 %) and bioinspired nacre-like brick-and-mortar architecture is investigated to offer insights for MPC design and processing. An automated finite element analysis (FEA) framework is developed to analyze MPC models with randomized geometries and multiple combinations of the parameter space. Specifically, the influence of concentration, aspect ratio (AR), flake thickness, flake distribution, and interfacial strength is investigated. The results reveal property trends such as increasing elastic modulus, strength, and toughness with increasing cohesive strength and concentration for lower AR ( =40, 60) but a decreasing trend at higher AR of 75. Local structural features like flake distribution, overlapping MXene lengths, and interconnected polymers in adjacent layers was found a critical determinant of performance. For example, stronger cohesive interaction showed 6X high toughness (291 +/- 226 KJ/m(3)) compared to weaker case (50 +/- 24 KJ/m(3)), but the large scatter highlighted the impact of microstructural features. The results are compared and validated with theoretical, computational, and experimental work. The findings provide valuable guidance for optimizing MPC design and their processing. Finally, the automation of the framework allows the design to be extended beyond the current system and chosen material combinations.

Vitrimers, an emerging class of polymer materials, are thermosets with dynamic covalent cross-linkers, allowing for topology rearrangement at elevated temperatures. However, vitrimers have several drawbacks, such as slow response times and often lack photothermal catalytic activity. Herein, we demonstrate that embedding functional nanofillers, i.e., hierarchically assembled plasmonic gold nanoparticles (AuNPs) on graphene nanoplatelets (GNPIs) into a vitrimer matrix, induces an ultrafast photothermal healing response. Unlike previous research that mainly focused on bulk materials, our exploration of vitrimer nanocomposite films uncovers unique advantages, such as optical transparency in the visible wavelength, flexibility, and ultrafast localized healing upon exposure to a 532 nm wavelength laser. These remarkable properties of vitrimer nanocomposite films were demonstrated with three various filler compositions and concentrations, where AuNPs/GNPls serve as a powerful filler. Photothermally activated self-healing of these hybrid materials is demonstrated by taking advantage of the localized surface plasmon resonance (LSPR) of AuNPs and the broad absorbance wavelength and high thermal conductivity of GNPls. Furthermore, profilometry is utilized to quantify the volume percent recovery of healing, providing quantitative evidence of increased healing with a higher filler concentration and laser dosage. This localized, ultrafast healing is pivotal for future coating applications, where bulk heating could lead to undesirable deformations. Our comprehensive understanding of the role of filler composition, filler concentration, and laser dosage in the self-healing properties of films opens up a wide array of potential applications for these light-responsive functional materials. The potential applications of these materials span from self-healing coatings to flexible electronics, inspiring a new era of innovative solutions.