Having a compact yet robust structurally based identifier or representation system is a key enabling factor for efficient sharing and dissemination of research results within the chemistry community, and such systems lay down the essential foundations for future informatics and data-driven research. While substantial advances have been made for small molecules, the polymer community has struggled in coming up with an efficient representation system. This is because, unlike other disciplines in chemistry, the basic premise that each distinct chemical species corresponds to a well-defined chemical structure does not hold for polymers. Polymers are intrinsically stochastic molecules that are often ensembles with a distribution of chemical structures. This difficulty limits the applicability of all deterministic representations developed for small molecules. In this work, a new representation system that is capable of handling the stochastic nature of polymers is proposed. The new system is based on the popular “simplified molecular-input line-entry system” (SMILES), and it aims to provide representations that can be used as indexing identifiers for entries in polymer databases. As a pilot test, the entries of the standard data set of the glass transition temperature of linear polymers (Bicerano, 2002) were converted into the new BigSMILES language. Furthermore, it is hoped that the proposed system will provide a more effective language for communication within the polymer community and increase cohesion between the researchers within the community.

The failure properties of a polymer network, including toughness, ultimate strain, and ultimate stress, are some of the most critical properties for network performance. The polymer networks often contain various topological defects, such as primary loops and dangling ends, which have a noticeable effect on these properties. This work focuses on understanding the effect of these defects on the fracture strength of a material by expanding the classical Lake−Thomas theory to account for such defects under the assumption that each defect is unaffected by the presence of other defects in its environment. A Flory−Stockmayer gel point criterion is combined with the improved theory to identify the incipience of fracture. The predictions demonstrate that although the presence of defects weakens the material by reducing the tearing energy, the overall network elongation depends strongly on the primary loop fraction. Specifically, a transition from a low ultimate-strain to a significantly high ultimate-strain behavior is predicted. The addition of a kinetic theory for bond scission predicts that the sharpness of this transition is a strong function of the strain rate. To experimentally test these predictions, a series of poly(ethylene glycol) (PEG) gels with previously characterized primary loop fractions were synthesized. Remarkably, the measured tearing energies agree quite well with the theoretical predictions and also suggest the onset of the low to high extensibility transition.

Physics-based models are the primary approach for modeling the phase behavior of block copolymers. However, the successful use of self-consistent field theory (SCFT) for designing new materials relies on the correct chemistry- and temperature-dependent Flory–Huggins interaction parameter χ AB that quantifies the incompatibility between the two blocks A and B as well as accurate estimation of the ratio of Kuhn lengths (b A/b B) and block densities. This work uses machine learning to model the phase behavior of AB diblock copolymers by using the chemical identities of blocks directly, obviating the need for measurement of χAB and b A/b B. The random forest approach employed predicts the phase behavior with almost 90% accuracy after training on a data set of 4768 data points, almost twice the accuracy obtained using SCFT employing χAB from group contribution theory. The machine-learning model is notably sensitive toward the uncertainty in measuring molecular parameters; however, its accuracy still remains at least 60% even for highly uncertain experimental measurements. Accuracy is substantially reduced when extrapolating to chemistries outside the training set. This work demonstrates that a random forest phase predictor performs remarkably well in many scenarios, providing an opportunity to predict self-assembly without measurement of molecular parameters.