No abstract

Polybenz‐3,1‐oxazinone‐4 (PBOZ) films are prepared via the thermal rearrangement of poly[(methylene‐bis‐anthranilamide) 4,4′‐diphenyloxidicarboxylic acid] (PAA) films by heating to 300 °C. PAA is synthesized by low‐temperature polycondensation. The PBOZ film exhibits excellent mechanical properties and a high glass‐transition temperature. Conversion of PAA to PBOZ leads to an increase in the fractional free volume as a result of thermal dehydration and cyclization during membrane formation and due to the removal of residual solvent. The excess free volume in the PBOZ film is approximately 3–4% of the total volume. The gas permeability and selectivity of the PBOZ film is higher than that of the PAA film. The thermal rearrangement of polymers is recognized as a suitable method to improve separation efficiency.

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.

A comparative study of metal − polymer complexes of Cu(I) with polybenzoxazinoneimide (PBOI) and its prepolymer imide-containing polyamic acid (PAA) as novel membrane materials for methyl tertiary butyl ether (MTBE) purification was undertaken. The structure, physical parameters and transport properties were characterized in detail to analyse the separation performance of the membranes and obtain new knowledge on the interdependence of the chemical structure and physical data with transport parameters. Thermally initiated conversion of PAA − Cu(I) to PBOI − Cu(I) was studied by TGA and DSC. The thermal conversion increases the polymer glass transition temperature and membrane density. Both polymers were applied to pervaporation separation of MTBE from methanol impurities. Membranes based on PAA are highly effective in MTBE purification and preferably permeate methanol. The transport properties of PAA − Cu(I) membrane are compared with those of known membranes.

No abstract