North Dakota State University

The glass transition temperature ($$\:{T}_{g}$$) is a pivotal parameter for amorphous polymers, influencing their behavior under different thermal properties. Assessing $$\:{T}_{g}$$ is crucial for evaluating material performance across varying temperatures. Our research integrates machine learning with cheminformatics framework to analyze $$\:{T}_{g}$$ values across a broad polymer dataset. This framework improves a comprehensive understanding of the quantitative correlations between the polymers’ structural attributes and their $$\:{T}_{g}$$. Utilizing a dataset of 250 polymers, we constructed a series of Machine Learning-based Quantitative Structure-Property Relationship (ML-QSPR) models. The initial ML-QSPR used Genetic Algorithm (GA) and Multiple Linear Regression (MLR) methods to identify an optimal set of molecular descriptors. Subsequently, various non-linear machine learning methods, including Support Vector Regression (SVR), Gaussian Process Regression (GPR), Random Forest (RF), and Multi-Layer Perceptron (MLP), were used for comparative purposes and predictive analysis. The results demonstrate that MLP, driven by seventeen (17) selected descriptors, yields the most accurate ML model, with training and external validation $$\:{R}^{2}\:$$values of 0.82 and 0.79, respectively. This model predicts with good performance the $$\:{T}_{g}$$ values of the polymers under study, showing the role of specific structural descriptors in refining polymer property predictions.

This study examined two assessment approaches—a case comparison prompt and a reasoning chain construction task to elicit students’ knowledge and application of green chemistry within the context of SN1 and SN2 reactions. The case comparison prompt presented students with a context-specific evaluation of two reactions, whereas the reasoning chain construction task required a general comparison of SN1 and SN2 reactions. Following each open-ended assessment, students were asked to select a conclusion regarding the overall greener reaction. The assessments were administered as an online survey, and student responses were analyzed using the twelve principles of green chemistry. Analysis of student responses showed that students consistently selected the same reaction type across assessments despite differences in task format. Across both assessments, students most frequently prioritized the use of safer solvents over considerations related to stereochemical outcomes, leading them to identify the SN1 reaction as the greener option. Overall, this study reveals how students apply green chemistry considerations to organic reactions across two open-ended assessment formats and demonstrates consistent patterns in their conclusions despite differences in the assessment formats. Importantly, the findings also revealed alternative lines of reasoning and misconceptions that are not readily captured by closed-ended assessments alone.