Deep learning based approach for prediction of glass transition temperature in polymers

Goswami, Subhasish; Ghosh, Rajdeep; Neog, Arohan; Das, Biswajit

doi:10.1016/j.matpr.2021.02.730

Cited by 22 publications

(19 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As an exciting alternative, utilizing machine learning (ML) techniques and the increasing amount of polymer data sets offer a new opportunity to tackle the challenge in the polymer field. Successful polymer informatics attempts have touched upon the a number of property predictions like polymers’ electronic bandgap, , dielectric constant, refractive index, etc., but a lot more attention has been paid to the prediction of polymers’ glass transition temperatures. ,− This is primarily reflective of the facts that (1) the glass transition temperature is an important property controlling the phase transition and therefore the application of polymers and (2) the glass transition temperature T g is the most reported experimental measurement in publicly accessible databases like PoLyInfo, the Polymer Property Predictor and Database (NIST), and the CROW Polymer Properties Database …”

Section: Introductionmentioning

confidence: 99%

Benchmarking Machine Learning Models for Polymer Informatics: An Example of Glass Transition Temperature

Tao

Varshney

2021

J. Chem. Inf. Model.

128

View full text Add to dashboard Cite

In the field of polymer informatics, utilizing machine learning (ML) techniques to evaluate the glass transition temperature T g and other properties of polymers has attracted extensive attention. This data-centric approach is much more efficient and practical than the laborious experimental measurements when encountered a daunting number of polymer structures. Various ML models are demonstrated to perform well for T g prediction. Nevertheless, they are trained on different data sets, using different structure representations, and based on different feature engineering methods. Thus, the critical question arises on selecting a proper ML model to better handle the T g prediction with generalization ability. To provide a fair comparison of different ML techniques and examine the key factors that affect the model performance, we carry out a systematic benchmark study by compiling 79 different ML models and training them on a large and diverse data set. The three major components in setting up an ML model are structure representations, feature representations, and ML algorithms. In terms of polymer structure representation, we consider the polymer monomer, repeat unit, and oligomer with longer chain structure. Based on that feature, representation is calculated, including Morgan fingerprinting with or without substructure frequency, RDKit descriptors, molecular embedding, molecular graph, etc. Afterward, the obtained feature input is trained using different ML algorithms, such as deep neural networks, convolutional neural networks, random forest, support vector machine, LASSO regression, and Gaussian process regression. We evaluate the performance of these ML models using a holdout test set and an extra unlabeled data set from high-throughput molecular dynamics simulation. The ML model’s generalization ability on an unlabeled data set is especially focused, and the model’s sensitivity to topology and the molecular weight of polymers is also taken into consideration. This benchmark study provides not only a guideline for the T g prediction task but also a useful reference for other polymer informatics tasks.

show abstract

Section: Introductionmentioning

confidence: 99%

Benchmarking Machine Learning Models for Polymer Informatics: An Example of Glass Transition Temperature

Tao

Varshney

2021

J. Chem. Inf. Model.

128

View full text Add to dashboard Cite

show abstract

“…Looking at the T g values (Table 3), one can assume that the bigger -C 8 F 13 H 4 group in (CF)6 than -C 3 F 3 H 4 in (CF)1 might have reduced the chain mobility, whereas the longer -C 12 F 21 H 4 group in (CF)10 could induce a more flexible backbone chain, causing the lowering of the T g of this structure in comparison to (CF)6 but maintaining the same level as for (CF)0 [67]. On the other hand, the higher T g for the CF(6) sample may also be related to the higher thickness of the coating [67,68]. A similar phenomenon has been observed in other silica hybrids [69,70].…”

Section: Discussionmentioning

confidence: 94%

Structural and Functional Properties of Fluorinated Silica Hybrid Barrier Layers on Flexible Polymeric Foil

et al. 2021

View full text Add to dashboard Cite

The reported work was focused on sol–gel-derived organically modified and fluorinated silica coatings deposited on elastic polymeric foil. The structure and topography of the coatings were tested by infrared spectroscopy and microscopic studies. The functional properties were determined using thermal analysis, surface analysis, and oxygen permeability tests. The barrier feature of the investigated materials against oxygen was correlated with the properties of the coatings. The hybrid (organic–inorganic) structure of the coatings was proven, demonstrating the presence of a silica network modified with alkyl and fluoroalkyl groups since precursors with the isooctyl group or different lengths of the fluoroalkyl chains were used for the syntheses. The coatings were free of defects and had a smooth surface except for the sample containing the longest fluoroalkyl chain (perfluorododecyl group), which showed a wrinkle-like surface. The hydrophobic character of the coatings increased, whereas the oxygen permeation coefficient values decreased (reaching a fourfold lower coefficient in comparison to the bare substrate) with a higher content of the fluorinated carbon atoms in the structure. The results were enriched by the information from the thermomechanical analysis as well as nanoindentation and scratch tests giving values of the glass transition temperature, thermal expansion coefficient, coatings adhesion, and hardness of the investigated systems.

show abstract

“…According to the data in Table 5, PVC/nZB411 composites had lowest molecular or chain mobility. Agglomeration of nanoparticles caused an increase of free volume and Tg value of the composites decreased [36].…”

Section: Dynamic Mechanical Analysismentioning

confidence: 98%

Influence of Particle Size, Additive Ratio and Chemical Structure of Zinc Borate on Thermal and Mechanical Properties of Polyvinyl Chloride

İpek¹

2021

Journal of Boron

View full text Add to dashboard Cite

Zinc borate (ZB) is a flame retardant and smoke supressant compound of boron. It is widely used to achieve flame resistant polymer composites. Poly(Vinyl Chloride) (PVC) is one of the commercially important polymers that is mostly used for cable insulating, windows, doors, pipes, floorings and toys etc.. On the other hand, PVC releases toxic gases during fire as decomposition products. Zinc borate addition is thought to be a solution for this problem. However, this solution leads a new problem. Zinc borate effects the mechanical properties of the PVC. In the present study, to optimize the contribution of zinc borate, with different formulations (2ZnOwith different additive ratios (1% and 5% by weight) and different particle sizes (nano-sized and micro-sized) were added to the PVC structure and the mechanical and thermomechanical properties of each composite were investigated. Mechanical properties of composites were determined by tensile test, impact test, hardness test while thermomechanical behaviours were detected with dynamic mechanical analysis (DMA). In the perspective of mechanical strength of PVC-ZB composite, additive ratio of ZB was more effective than particle size and chemical formulation of ZB. The additive ratio more than 5% by weight can negatively affect the mechanical strength and quality of the PVC-ZB composite.

show abstract

Deep learning based approach for prediction of glass transition temperature in polymers

Cited by 22 publications

References 13 publications

Benchmarking Machine Learning Models for Polymer Informatics: An Example of Glass Transition Temperature

Benchmarking Machine Learning Models for Polymer Informatics: An Example of Glass Transition Temperature

Structural and Functional Properties of Fluorinated Silica Hybrid Barrier Layers on Flexible Polymeric Foil

Influence of Particle Size, Additive Ratio and Chemical Structure of Zinc Borate on Thermal and Mechanical Properties of Polyvinyl Chloride

Contact Info

Product

Resources

About