In the present work, we address the problem of utilizing machine learning (ML) methods to predict the thermal properties of polymers by establishing "structure−property" relationships. Having focused on a particular class of heterocyclic polymers, namely polyimides (PIs), we developed a graph convolutional neural network (GCNN), being one of the most promising tools for working with big data, to predict the PI glass transition temperature T g as an example of the fundamental property of polymers. To train the GCNN, we propose an original methodology based on using a "transfer learning" approach with an enormous "synthetic" data set for pretraining and a small experimental data set for its fine-tuning. The "synthetic" data set contains more than 6 million combinatorically generated repeating units of PIs and theoretical values of their T g values calculated using the well-established Askadskii's quantitative structure−property relationship (QSPR) computational scheme. Additionally, an experimental data set for 214 PIs was also collected from the literature for training, fine-tuning, and validation of the GCNN. Both "synthetic" and experimental data sets are included into a PolyAskInG database (Polymer Askadskii's Intelligent Gateway). By using the PolyAskInG database, we developed GCNN which allows estimation of T g of PI with a mean absolute error (MAE) of about 20 K, which is 1.5 times lower than in the case of Askadskii QSPR analysis (33 K). To prove the efficiency and usability of the proposed GCNN architecture and training methodology for predicting polymer properties, we also employed "transfer learning" to develop alternative GCNN pretrained on proxy-characteristics taken from the popular quantumchemical QM9 database for small compounds and fine-tuned on an experimental T g values data set from PolyAskInG database. The obtained results indicate that pretraining of GCNN on the "synthetic" polymer data set provides MAE which is almost twice as low as that in the case of using the QM9 data set in the pretraining stage (∼41 K). Furthermore, we address the questions associated with the influence of the differences in the size of the experimental and "synthetic" data sets (so-called "reality gap" problem), as well as their chemical composition on the training quality. Our results state the overall priority of using polymer data sets for developing deep neural networks, and GCNN in particular, for efficient prediction of polymer properties. Moreover, our work opens up a challenge for the theoretically supported generation of large "synthetic" data sets of polymer properties for the training of the complex ML models. The proposed methodology is rather versatile and may be generalized for predicting other properties of different polymers and copolymers synthesized through the polycondensation reaction.
The possibility of predicting the coefficient of thermal expansion for the blends of polyvinyl chloride (PVC) with a number of organic polymers is shown. It was found that the higher the glass transition temperature of the polymer, the lower the coefficient of thermal expansion of the mixture of PVC with this polymer. The dependence of thermal expansion for composites based on wood of different species and bamboo is also analyzed. In all cases, the coefficient of thermal expansion is reduced, which allows the use of forecasting results for the development of new PVC-based building materials with improved thermal properties.
The abrasion of materials based on blends of ABS plastic with polyvinyl chloride (PVC) as well as terraced boards based on wood-polymer composites (DPC) has been studied. The measurements were carried out on a drum-type machine, and on a Taber's abrasimeter. For blends of ABS plastic with PVC at abrasion path length 600 m wear is 0.85%. For terracotta boards based on WPC, the wear during the test (loss of mass) was 0.0042 g. The abrasion of the sample was 9.29×10-5 g/cm2. Thus, the obtained blends should be recommended for application for floor coverings, since they possess negligible abrasion.
Diagrams of compatibility of water permeability with glass transition temperature, density, and cohesion energy are constructed. The computer program “Cascade”, INEOS RAS was used. Polyolefins, vinyl polymers, and polycarbonates were analyzed. Compatibility diagrams are constructed for the areas of lowest and highest water permeability for different classes of polymers. For polyolefins and vinyl polymers, the lowest water permeability is selected in the range from 0 to 100 barrer, and the highest water permeability is from 1000 to 2000 barrer. For polycarbonates, the intervals from 0 to 70 barrer and from 3000 to 7000 barrer are selected. These diagrams allow you to select polymers that meet the specified values of density, cohesion energy, glass transition temperature, and the onset of temperature of intense thermal degradation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.