New copolyhydrazides with anthrazoline fragments in the main chain: synthesis and optical properties.

Several new 2,8-diphenylpyrido[3,2-g]quinoline-4,6-dicarbohydrazides were synthesized and copolyhydrazides based on them were obtained. Molecular weight, thermal, stress-strain and optical properties were investigated. It was shown that all polymer films exhibit significant luminescence in the 450-650 nm region, the profile and intensity of which depended on the nature and position of substituents in the phenylene fragment.

Synthesis and photoluminescence properties of co-polyamides with anthrazoline-containing units in the main chain.

A series of anthrazoline-containing monomers are synthesized, and eight co-polyamides of different chemical structures, containing 1,9-anthrazoline fragments in the main chain, are obtained and investigated. Photoluminescent, stress-strain, and thermal properties of these polymers are studied. It is shown that polymers with fragments of 4,4'-(pyrido[3,2-g]-quinoline-2,8-diyl)dianiline and 4,4'-(10-methylpyrido[3,2-g]quinoline-2,8-diyl)dianiline possess an intense luminescence in the range 550-650 nm. The performed investigations made it possible to determine the effect of substituents of various natures in the anthrazoline cycle and the position of amide group (meta- and para-configurations) on optical, stress-strain, and thermal properties of copolymers, opening up a prospect for further developments of principles of design of polymers with optimal characteristics.

Machine Learning with Enormous "Synthetic" Data Sets: Predicting Glass Transition Temperature of Polyimides Using Graph Convolutional Neural Networks.

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 quantum-chemical 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.

Aromatic Copolyamides with Anthrazoline Units in the Backbone: Synthesis, Characterization, Pervaporation Application.

Copolyamides with anthrazoline units in the backbone (coPA) were synthesized and dense nonporous films were prepared by solvent evaporation. Glass transition temperature, density, and fractional free volume were determined for the dense nonporous films composed of polyamide and two of its copolymers containing 20 and 30 mol % anthrazoline units in the backbone. Transport properties of the polymer films were estimated by sorption and pervaporation tests toward methanol, toluene, and their mixtures. An increase in anthrazoline fragments content leads to an increasing degree of methanol sorption but to a decreasing degree of toluene sorption. Pervaporation of a methanol⁻toluene mixture was studied over a wide range of feed concentration (10⁻90 wt % methanol). Maximal separation factor was observed for coPA-20 containing 20 mol % fragments with anthrazoline units; maximal total flux was observed for coPA-30 with the highest fractional free volume.