Multitask Neural Network for Mapping the Glass Transition and Melting Temperature Space of Homo- and Co-Polyhydroxyalkanoates Using σ_Profiles Molecular Inputs

Abstract

Polyhydroxyalkanoates (PHAs) are an emerging type of bioplastic that have the potential to replace petroleum-based plastics. They are biosynthetizable, biodegradable, and economically viable and have a range of tunable properties. Despite their great potential, the structure and properties of PHA remain unexplored due to their theoretically infinite chemical space. Therefore, computational approaches for accurate predictions of their various properties need to be developed to effectively explore this large chemical space. For this purpose, this work presents a multitask artificial neural network (ANN) capable of predicting the glass transition temperature (T g ) and melting temperature (T m ) of PHA homopolymers and copolymers. The ANN inputs included the σ Profiles as molecular parameters describing the monomer chemistry and their composition. In contrast, the polymer molecular weight (M) and polydispersity index (PDI) were used to describe the polymer state. The results showed that after optimizing the hyperparameters, the selected ANN architecture was remarkable in predicting the T g and T m of PHA with R 2 values of 0.979 and 0.986 and average absolute relative deviation (AARD) of 0.476% and 0.520%, respectively. The proposed model represents an initiative to promote the development of robust, open source, and user-friendly models capable of predicting the properties of polymers based solely on molecular parameters (σ Profiles ), thereby saving time and resources for researchers worldwide. The framework described in this work is flexible so that it can be applied to a larger chemical space and incorporate other properties of polymers.

Enhancing precision in PANI/Gr nanocomposite design: robust machine learning models, outlier resilience, and molecular input insights for superior electrical conductivity and gas sensing performance

“…87 An additional advantage of σ-profiles lies in their capability to characterize molecules of varying sizes, facilitated by their unnormalized histograms comprising 61 points within the σ value range of [−0.030, +0.030] e Å −2 . 47 This inherent property ensures a consistent number of inputs for machine and deep learning applications, as it remains unaffected by variations in the molecular structure. Analyzing additives in the PANI/Gr system using σ-profiles provides a comprehensive view of their chemical composition and behavior.…”

“…35,36 Numerous successful endeavors have been undertaken by scientists to employ machine learning for investigating polymer synthesis and properties. 37 The application of ML in conjunction with mechanical properties of polymer composites, 38,39 liquid crystal behavior of copolyether, 40 thermal conductivity and dielectric properties, [41][42][43][44] glass transition, [45][46][47] melting and degradation temperature, as well as quantum physical and chemical properties, [48][49][50][51] has led to signicant achievement in prediction accuracy. PANI/graphene-based nanocomposites have risen as up-andcoming materials due to their distinctive electrical and gassensing properties, rendering them well-suited for diverse industrial applications.…”

Predicting the CO2 Capture Capability of Deep Eutectic Solvents and Screening over 1000 of their Combinations Using Machine Learning

“…The utilization of the σ-profile in this study was aimed at providing chemical information required for the training of MLP models. This tool serves as a quantitative method for describing the polarity differences of molecules and predicting their dispersal, electrostatic, and hydrogen-bonding interactions between the constituents of a mixture. , The σ-profile is considered as 3 primary areas: the HBA area, the nonpolar area, and the HBD area . The two dotted perpendicular lines located at −0.008 and +0.008 e/Å 2 indicate the boundaries between the nonpolar area and the two polar areas (HBD and HBA).…”

3,4-Dimethoxy phenyl thiosemicarbazone as an effective corrosion inhibitor of copper under acidic solution: comprehensive experimental, characterization and theoretical investigations

“…The visual depiction illustrates the following: regions designated as hydrogen bond acceptors (HBA) are colored red, while non-polar areas are represented by green; hydrogen bond donor sites (HBD) are represented by blue. [89][90][91][92] The s-prole, as used in COSMO-RS analysis, represents a probability distribution of surface area vs. charge density. [93][94][95] Graphically depicted, the s-prole is commonly divided into three discrete regions:…”