Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Olefin Metathesis Polymerization: Overview01:13

Olefin Metathesis Polymerization: Overview

2.1K
Recently, the development of olefin metathesis polymerization advanced the field of polymer synthesis. Simply put, the reorganization of substituents on their double bonds between two olefins in the presence of a catalyst is known as the olefin metathesis reaction. The use of metathesis reaction for polymer synthesis is called olefin metathesis polymerization.
Ruthenium-based Grubbs catalyst is the most commonly used catalyst for olefin metathesis polymerization. Grubbs catalyst consists...
2.1K
Polymers: Molecular Weight Distribution01:10

Polymers: Molecular Weight Distribution

3.4K
For any given polymer, the weight average molecular weight (Mw) is higher than, if not equal to, the number average molecular weight (Mn). The only situation in which the weight average molecular weight and the number average molecular weight are equal is when a polymer consists only of chains with equal molecular weight. However, this never happens in a synthetic polymer, since it is difficult to control the polymerization process up to a molecular level with accuracy to a hundred percent.
3.4K
Types of Step-Growth Polymers: Polyesters01:20

Types of Step-Growth Polymers: Polyesters

2.2K
The introduction of polyesters has brought major development to the textile industry. The wrinkle-free behavior of polyester blends has eliminated the need for starching and ironing clothes.
Polyesters are commonly prepared from terephthalic acid and ethylene glycol; the crude product is known as poly(ethylene terephthalate) or PET. However, polyesters are synthesized industrially by transesterification of dimethyl terephthalate with ethylene glycol at 150 °C. The two reactants and the...
2.2K
Olefin Metathesis Polymerization: Ring-Opening Metathesis Polymerization (ROMP)01:16

Olefin Metathesis Polymerization: Ring-Opening Metathesis Polymerization (ROMP)

2.6K
Ring-opening metathesis polymerization or ROMP involves strained cycloalkenes as starting materials. The mechanism of ROMP proceeds by reacting cycloalkene with Grubbs catalyst to give metallacyclobutane intermediate which undergoes a ring-opening reaction to form new carbene. The new carbene reacts with another molecule of cycloalkene. Repetition of these steps leads to the formation of an unsaturated open-chain polymer product. All these steps are reversible, however, relieving the ring...
2.6K
Polymer Classification: Stereospecificity01:26

Polymer Classification: Stereospecificity

2.4K
Polymerization generates chiral centers along the entire backbone of a polymer chain. Accordingly, the stereochemistry of the substituent group has a significant effect on polymer properties. Polymers formed from monosubstituted alkene monomers feature chiral carbons at every alternate position in the polymer backbone. Relative to the predominant orientation of substituents at the adjacent chiral carbons, the polymer can exist in three different configurations: isotactic, syndiotactic, and...
2.4K
Polymer Classification: Architecture01:14

Polymer Classification: Architecture

2.7K
Polymers are classified as linear or branched on the basis of their chain architecture. The polymer chains in linear polymers have a long chain-like structure with minimal to no branching at all. Even if a polymer features large substituent groups on the monomer, which appear as branches to the skeleton, it is not considered a branched polymer. A branched polymer contains secondary polymer chains that arise from the main polymer chain. The branching occurs when the polymer growth shifts from...
2.7K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Framework-templated gas lattices in metal-organic frameworks.

Nature communications·2026
Same author

Methanol-Ethanol Discrimination and Selective Sensing Enabled by Molecular Sieving in Conductive MOFs.

Advanced materials (Deerfield Beach, Fla.)·2026
Same author

LLMB: AI Agent for Lithium Metal Battery Research Using Large Language Model.

ACS central science·2026
Same author

Bioelectronic synthesis of hydrogen sulfide enables spatiotemporal regulation of protein modification and cellular redox.

Science advances·2026
Same author

Universal cryogenic transfer of liquid metal particles in polymers for wafer-scale stretchable integrated electronics.

Nature communications·2026
Same author

Photoactivated conductive MOF thin film arrays on micro-LEDs for chemiresistive gas sensing.

Nature communications·2025

Related Experiment Video

Updated: Jun 30, 2025

Synthesis of Cyclic Polymers and Characterization of Their Diffusive Motion in the Melt State at the Single Molecule Level
06:55

Synthesis of Cyclic Polymers and Characterization of Their Diffusive Motion in the Melt State at the Single Molecule Level

Published on: September 26, 2016

7.9K

Multimodal Transformer for Property Prediction in Polymers.

Seunghee Han1, Yeonghun Kang1, Hyunsoo Park2

  • 1Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of Science and Technology, Daejeon 34141, Republic of Korea.

ACS Applied Materials & Interfaces
|March 19, 2024
PubMed
Summary

This study introduces a multimodal transformer model that improves polymer property prediction by combining molecular representations. The model using SMILES and dimer inputs significantly outperformed others across five key polymer properties.

Keywords:
SMILESgraph neural network (GNN)machine learningmultimodalpolymerstransformer

More Related Videos

Self-assembling Morphologies Obtained from Helical Polycarbodiimide Copolymers and Their Triazole Derivatives
09:22

Self-assembling Morphologies Obtained from Helical Polycarbodiimide Copolymers and Their Triazole Derivatives

Published on: February 7, 2017

7.8K
Cross-Modal Multivariate Pattern Analysis
13:51

Cross-Modal Multivariate Pattern Analysis

Published on: November 9, 2011

20.0K

Related Experiment Videos

Last Updated: Jun 30, 2025

Synthesis of Cyclic Polymers and Characterization of Their Diffusive Motion in the Melt State at the Single Molecule Level
06:55

Synthesis of Cyclic Polymers and Characterization of Their Diffusive Motion in the Melt State at the Single Molecule Level

Published on: September 26, 2016

7.9K
Self-assembling Morphologies Obtained from Helical Polycarbodiimide Copolymers and Their Triazole Derivatives
09:22

Self-assembling Morphologies Obtained from Helical Polycarbodiimide Copolymers and Their Triazole Derivatives

Published on: February 7, 2017

7.8K
Cross-Modal Multivariate Pattern Analysis
13:51

Cross-Modal Multivariate Pattern Analysis

Published on: November 9, 2011

20.0K

Area of Science:

  • Materials Science
  • Computational Chemistry
  • Polymer Science

Background:

  • Predicting polymer properties is crucial for material design and application.
  • Traditional methods often struggle with the complexity of polymer structures and their diverse properties.
  • Deep learning offers a promising avenue for accurate property prediction.

Purpose of the Study:

  • To develop and evaluate a multimodal transformer model for enhanced polymer property prediction.
  • To investigate the impact of combining different molecular representations (SMILES, monomer, dimer) on prediction accuracy.
  • To provide deeper insights into the relationship between deep learning models and polymer attributes through attention analysis.

Main Methods:

  • Designed a multimodal transformer architecture integrating Simplified Molecular Input Line Entry System (SMILES) and molecular graph representations.
  • Trained and fine-tuned three models with distinct input embeddings: SMILES, SMILES + monomer, and SMILES + dimer.
  • Evaluated model performance across five key polymer properties: density, glass-transition temperature (Tg), melting temperature (Tm), volume resistivity, and conductivity.
  • Analyzed model attention scores to understand feature importance and model interpretability.

Main Results:

  • The multimodal transformer incorporating both SMILES and dimer representations significantly outperformed the SMILES-only model across all five tested polymer properties.
  • The inclusion of multimodal features demonstrably enhanced the predictive accuracy of the transformer architecture.
  • Attention score analysis revealed specific relationships between molecular features and predicted polymer properties.

Conclusions:

  • Multimodal transformer models integrating diverse molecular representations show significant potential for advancing polymer property prediction.
  • Combining SMILES and dimer inputs offers a superior approach compared to using SMILES alone for predicting a range of polymer properties.
  • This work opens new avenues for understanding and optimizing polymer behavior using advanced deep learning techniques.