Aug 2026· Journal of Physical Chemistry C· 0 citations· 70 references
Abstract
The glass transition temperature (Tg) is a critical descriptor governing the morphological stability, emitter orientation, and interfacial integrity of amorphous thin films in organic electronics. However, experimental Tg measurements suffer from high resource costs and interlaboratory variability, while machine learning models are bottlenecked by scarce, noisy data sets. Here, we establish a physics-based atomistic molecular dynamics (MD) protocol to predict the Tg of 160 diverse organic electronic materials. To study computational throughput and predictive accuracy, we systematically benchmarked nine configurations spanning system sizes (5,000, 10,000, and 15,000 atoms) and cooling step relaxation times (5, 10, and 15 ns). Extracted via an automated, bias-free hyperbolic fitting scheme, our preferred standalone workflow (15,000 atoms, 15 ns) yields a correlation of R2 = 0.89 and a mean absolute error (MAE) of 10.9 K relative to experiment. Structural descriptor analysis confirms that accuracy remains uniform regardless of molecular weight or heteroatom density, establishing this transferable workflow as a digital sieve to accelerate the discovery of next-generation organic electronics.
Determining the glass transition temperature Tg in materials science in general and for amorphous polymer systems in particular is a delicate matter due to the uncertainty in the definition of Tg and the complexity of the glass transition phenomenon itself. Machine learning (ML) provides powerful approaches for analyzing complex, high-dimensional data to reveal hidden patterns. Recently, we have applied an unsupervised ML technique to identify Tg of a polymer melt of weakly semiflexible bead-spring chains using the time evolvement of pairwise internal distances between monomers along chains, as input features. Here, we investigate the change of individual internal chain relaxation as the polymer melt transforms from the liquid to glassy state. The average overall relaxation remains unchanged and displays the usual temperature dependence. However, for some individual pair distances, scattered throughout the sample, relaxation is significantly delayed, which serves as a robust indicator of approaching the glass transition. Moreover, these changes and the first principle components are highly correlated. This is an evidence that the ML technique indeed captures a significant indicator of the approach of the glass transition. Typical experiments, which average over the whole sample, cannot identify such features.
Atreyee Banerjee, Hsiao-Ping Hsu, Scott T. Milner et al.· Journal of Chemical Physics· 0 citations
The glass transition temperature (Tg) of polyimides is a critical parameter determining their processability and application performance. Traditional experimental methods for measuring Tg are time‐consuming and costly, while existing machine learning prediction models predominantly rely on manually defined molecular descriptors, which often fail to fully capture detailed molecular structural information, limiting their prediction accuracy and generalization capability. To address this, this study proposes a hybrid feature engineering strategy combining Morgan fingerprints and molecular descriptors to comprehensively represent the chemical structure of polyimides. Based on a dataset of 1257 polyimide samples from a public database, we systematically compared six feature selection methods and employed multiple mainstream machine learning algorithms for modeling. The results show that the CATB model performed best, achieving a coefficient of determination (R2) of 0.882 and a mean absolute error (MAE) of 17.34 °C on an independent test set, with fivefold cross‐validation further confirming the model's robustness. SHAP interpretability analysis revealed the significant influence of key features such as the number of rotatable bonds, ether bonds, and ether‐linked oxyethylene units on Tg, providing clear guidance for molecular design. External validation demonstrated the model's strong generalization ability. This study not only achieves high‐precision and robust Tg prediction but also highlights the importance of hybrid feature strategies in polymer property modeling, offering a data‐driven foundation for the rational design of polyimides.
Peishuai Xing, Xiaodong Guo, Yang Wang et al.· Molecular Informatics· 0 citations
Accurate description of structural relaxation processes is essential for predicting long-term stability of chalcogenide glasses utilized in memory, photonic, and infrared-optical applications. Although the Tool–Narayanaswamy–Moynihan (TNM) model remains one of the most effective frameworks for describing the glass transition kinetics, determination of its nonlinearity (x) and nonexponentiality (β) related features is often hindered by instrumental distortions and instability of the conventional curve-fitting procedures. Here, a robust and user-friendly implementation of the simulation-comparative method for extraction of the x and β TNM parameters is presented. An extensive library of precomputed core data sets is introduced, covering physically relevant ranges of enthalpy relaxation activation energy (200–500 kJ·mol–1) and glass transition temperatures (−50–450 °C) for chalcogenide glasses. Utilization of these data sets in terms of the presented simulation-comparative method enables rapid, fitting-free estimation of x and β by direct comparison with experimental calorimetric data, using only basic data-processing tools. The predictive performance of the method was systematically validatedusing defined and randomly generated theoretically simulated data sets, as well as experimental calorimetric data extracted from the literaturethe method predicts the parameters x and β with an uncertainty of ±0.05, further decreasing this uncertainty to ±0.02 for the most relevant cases of the relaxation behavior. By eliminating the need for specialized software and unstable optimization routines, the presented approach substantially lowers the barrier for routine TNM analysis and enables reliable monitoring of subtle structural relaxation trends in chalcogenide glasses. The provided core data sets, together with accompanying MATLAB code, establish a practical and broadly applicable platform for structural relaxation studies in amorphous chalcogenide materials.
The prediction of stable alloys forming solid-state solutions across large portions of the composition space is a serious theoretical challenge, since one has to evaluate the Gibbs free energy, including both configurational and vibrational contributions. This requires an energy theory capable of extremely high throughput. By taking the Ni-Pd system as prototype, we construct an efficient Jacobi-Legendre machine-learning potential based on density-functional-theory data, which provides accurate energies and forces across the entire composition space. Based on a cluster expansion up to three-body terms and only 873 trainable parameters, this allows us to compute the partition function by directly integrating all accessible microstates, differing for composition, atomic configuration and thermal agitation. We confirm that Ni and Pd are fully miscible, forming an $fcc$ solid-state solution. This is only metastable at room temperature, while becomes thermodynamically stable at around 600~K, with the stability achieved first at the Pd-rich end of the composition range. Interestingly, entropy and heat capacity analysis reveal a competition between the solid-state solution and two intermetallic phases with long-period L1$_0$ structure for NiPd and NiPd$_3$. All in all, our approach offers a powerful and high-throughput workflow for the study of disordered alloys, an approach that can be extended to multi-component systems such as high-entropy alloys.
Rutchapon Hunkao, U. Patil, S. Sanvito· 0 citations
Chemical short-range order (CSRO), the non-random local arrangement of atoms in solid solutions, strongly affects the phase stability and performance of medium- and high-entropy alloys (M/HEAs). Despite its importance, the fundamental nature of CSRO formation remains contested: is it a formal thermodynamic transition? Here, investigating CoCrNi as a model system, we indicate that the main CSRO transformations observed in alloys might not be classical thermodynamic transitions, but instead a kinetic arrest phenomenon analogous to the glass transition. Combining atomistic simulations and in situ synchrotron dilatometry experiments enabled the study of CSRO evolution and its structural impact across multiple length scales. For example, CSRO-driven changes in bond lengths and bond distribution significantly impact the observed lattice parameter and volume, allowing Warren-Cowley parameter values to be determined over a full experimental temperature range. We demonstrate that the degree of CSRO and the apparent transition temperatures, defined here as the komplex reaction temperatures (Tkr), are not intrinsic material constants. Rather, they are path-dependent quantities governed by thermal history and diffusional constraints, directly reflecting the frozen CSRO state. Our findings clarify the thermodynamic and kinetic mechanisms underlying CSRO evolution and establish a framework to distinguish thermodynamic transitions under kinetic constraints from genuine kinetically arrested phenomena. Understanding this distinction is crucial for controlling CSRO during alloy design and processing and provides a foundation for future investigations exploring the implications of CSRO in advanced materials. In situ synchrotron dilatometry and atomistic simulations indicate that changes between chemical short-range order states (evolving local chemical distributions) are governed by a kinetic arrest phenomenon rather than a thermodynamic transition.
G. C. Stumpf, Yifan Cao, V. Bacurau et al.· Nature Communications· 0 citations
Melting point (MP) is an important thermophysical property for the chemical process industry, yet accurate prediction of MP for organic compounds in the absence of experimental data remains challenging due to the complex interplay between molecular packing, intermolecular interactions, and electronic structure. Traditional group contribution and quantitative structure-property relationship models, which rely primarily on static molecular descriptors, often fail to capture these critical condensed-phase effects. In this study, we present a hybrid machine learning framework that integrates cheminformatics descriptors with quantum chemical features and dynamic condensed-phase descriptors derived from molecular dynamics (MD) simulations. Using a curated subset of the DIPPR 801 database, multiple machine learning architectures, including light gradient boosting machine (LightGBM) and graph convolutional networks, were evaluated with feature sets of increasing physical fidelity. The best-performing model, based on LightGBM trained on Dragon descriptors augmented with MD and quantum chemical features, achieves a mean absolute error of 22.5 K, outperforming descriptor-only models and structure-based deep learning baselines. Shapley additive explanations interpretability analysis reveals that melting behavior is governed primarily by molecular topology, surface-area-weighted electronic descriptors, and condensed-phase interaction properties. In contrast, many isolated functional group and single molecule electronic descriptors contribute negligibly once these effects are accounted for. These results demonstrate that incorporating physics-informed, multi-scale descriptors enables more accurate and physically interpretable MP predictions.
Frank T. Mtetwa, N. Giles, W. Wilding et al.· Journal of Chemical Physics· 0 citations