arXiv subjects

Edgar D. Zanotto

Publications and source records attributed to Edgar D. Zanotto.

6 recordsLinked to original sources

arXiv2021

Machine learning unveils composition-property relationships in chalcogenide glasses

Due to their unique optical and electronic functionalities, chalcogenide glasses are materials of choice for numerous microelectronic and photonic devices. However, to extend the range of compositions and applications, profound knowledge about composition-property relationships is necessary. To this end, we collected a large quantity of composition-property data on chalcogenide glasses from SciGlass database regarding glass transition temperature ($T_g$), Young's modulus ($E$), coefficient of thermal expansion (CTE), and refractive index ($n_D$). With these data, we induced predictive models using three machine learning algorithms: Random Forest, K-nearest Neighbors, and Classification and Regression Trees. Finally, the induced models were interpreted by computing the SHAP (SHapley Additive exPlanations) values of the chemical features, which revealed the key elements that significantly impacted the tested properties and quantified their impact. For instance, Ge and Ga increase $T_g$ and $E$ and decrease CTE (three properties that depend on bond strength), whereas Se has the opposite effect. Te, As, Tl, and Sb increase $n_D$ (which strongly depends on polarizability), whereas S, Ge, and P diminish it. Knowledge about the effect of each element on the glass properties is precious for semi-empirical compositional development trials or simulation-driven formulations. The induced models can be used to design novel chalcogenide glasses with required combinations of properties.

cond-mat.mtrl-sci↗

arXiv2021

Predicting and interpreting oxide glass properties by machine learning using large datasets

With the advent of powerful computer simulation techniques, it is time to move from the widely used knowledge-guided empirical methods to approaches driven by data science, mainly machine learning algorithms. We investigated the predictive performance of three machine learning algorithms for six different glass properties. For such, we used an extensive dataset of about 150,000 oxide glasses, which was segmented into smaller datasets for each property investigated. Using the decision tree induction, k-nearest neighbors, and random forest algorithms, selected from a previous study of six algorithms, we induced predictive models for glass transition temperature, liquidus temperature, elastic modulus, thermal expansion coefficient, refractive index, and Abbe number. Moreover, each model was induced with default and tuned hyperparameter values. We demonstrate that, apart from the elastic modulus (which had the smallest training dataset), the induced predictive models for the other five properties yield a comparable uncertainty to the usual data spread. However, for glasses with extremely low or high values of these properties, the prediction uncertainty is significantly higher. Finally, as expected, glasses containing chemical elements that are poorly represented in the training set yielded higher prediction errors. The method developed here calls attention to the success and possible pitfalls of machine learning algorithms. The analysis of the SHAP values indicated the key elements that increase or decrease the value of the modeled properties. It also estimated the maximum possible increase or decrease. Insights gained by this analysis can help empirical compositional tuning and computer-aided inverse design of glass formulations.

cond-mat.mtrl-sci↗

arXiv2021

Designing optical glasses by machine learning coupled with a genetic algorithm

Engineering new glass compositions have experienced a sturdy tendency to move forward from (educated) trial-and-error to data- and simulation-driven strategies. In this work, we developed a computer program that combines data-driven predictive models (in this case, neural networks) with a genetic algorithm to design glass compositions with desired combinations of properties. First, we induced predictive models for the glass transition temperature ($T_g$) using a dataset of 45,302 compositions with 39 different chemical elements, and for the refractive index ($n_d$) using a dataset of 41,225 compositions with 38 different chemical elements. Then, we searched for relevant glass compositions using a genetic algorithm informed by a design trend of glasses having high $n_d$ (1.7 or more) and low $T_g$ (500 °C or less). Two candidate compositions suggested by the combined algorithms were selected and produced in the laboratory. These compositions are significantly different from those in the datasets used to induce the predictive models, showing that the used method is indeed capable of exploration. Both glasses met the constraints of the work, which supports the proposed framework. Therefore, this new tool can be immediately used for accelerating the design of new glasses. These results are a stepping stone in the pathway of machine learning-guided design of novel glasses.

cond-mat.mtrl-sci↗

arXiv2021

Is the Structural Relaxation of Glasses Controlled by Equilibrium Shear Viscosity?

Knowledge of relaxation processes is fundamental in glass science and technology because relaxation is intrinsically related to vitrification, tempering as well as to annealing and sev-eral applications of glasses. However, there are conflicting reports -- summarized here for different glasses -- on whether the structural relaxation time of glass can be calculated using the Maxwell equation, which relates relaxation time with shear viscosity and shear modulus. Hence, this study aimed to verify whether these two relaxation times are comparable. The structural relaxation kinetics of a lead metasilicate glass were studied by measuring the re-fractive index variation over time at temperatures between 5 and 25 K below the fictive temperature, which was initially set 5 K below the glass transition temperature. Equilibrium shear viscosity was measured above and below the glass transition range, expanding the current knowledge by one order of magnitude. The Kohlrausch equation described very well the experimental structural relaxation kinetics throughout the investigated temperature range and the Kohlrausch exponent increased with temperature, in agreement with studies on other glasses. The experimental average structural relaxation times were much longer than the values computed from isostructural viscosity, as expected. Still, they were less than one order of magnitude higher than the average relaxation time computed through the Maxwell equation, which relies on equilibrium shear viscosity. Thus, these results demon-strate that the structural relaxation process is not controlled by isostructural viscosity, and that equilibrium shear viscosity only provides a lower boundary for structural relaxation kinetics.

cond-mat.soft↗

arXiv2020

Which glass stability parameters can assess the glass-forming ability of oxide systems?

Glass forming ability (GFA) is a property of utmost importance in glass science and technology. In this paper, we used a statistical methodology---involving bootstrap sampling and the Wilcoxon test---to find out which glass stability parameters can better predict the glass forming ability. We collected or measured the necessary data for twelve stoichiometric oxide glasses that underwent predominant heterogeneous nucleation (the most common case). We found that some GS parameters could predict the GFA of these oxide glasses quite well, whereas others perform poorly. Parameter $K_w$ was the top ranked, closely followed by the $K_H$, $γ$, $H^{'}$, $ΔT_{rg}$, and $K_{cr}$. Our results corroborate previous reports carried out using a smaller number of glasses, much less GS parameters, and less rigorous statistics. We also found that using $T_c$ instead of $T_x$ improved the predictive power of these parameters. Finally, the Jezica, the only parameter considered here that predicts the GFA without requiring the production of a glass piece (i.e., without relying on any crystallization information), ranked reasonably well in our analysis.

cond-mat.soft↗

arXiv2018

Rebuttal to Schmelzer and Tropin: Glass Transition, Crystallization of Glass-Forming Melts, and Entropy

In a recent article, Schmelzer and Tropin [Entropy 20 [2] 103 (2018)] presented an unfounded, confusing critique of several aspects of modern glass science. Relying on pre-Socratic Greek philosophy and state-of-the-art scientific understanding from the 1920s-1930s, Schmelzer and Tropin propagate an antiquated view of glass physics that is at odds with well-accepted knowledge in the field from both theory and experiments conducted in the post-World War II era. The objective of this short letter is to elucidate and extinguish their critique. This rebuttal is directed to our colleagues and especially to students who might otherwise become confused.

cond-mat.mtrl-sci↗