GLDP000009 In-House Research

Rapid, Autonomous High-throughput Characterization of Hydrogel Rheological Properties via Automated Sensing and Physics-guided Machine Learning

Authors

Junru Zhang a, Yang Liu a b, Durga Chandra Sekhar.P c, Manjot Singh a, Yuxin Tong a, Ezgi Kucukdeger a, Hu Young Yoon b, Alexander P. Haring a b, Maren Roman b d, Zhenyu (James) Kong a, Blake N. Johnson a b e f

Affiliation

a Grado Department of Industrial and Systems Engineering, Virginia Tech, Blacksburg, VA 24061 USA

b Macromolecules Innovation Institute, Virginia Tech, Blacksburg, VA 24061 USA

c Department of Computer Science, Virginia Tech, Blacksburg, VA 24061 USA

d Department of Sustainable Biomaterials, Virginia Tech, Blacksburg, VA, 24061, USA

e Department of Materials Science and Engineering, Virginia Tech, Blacksburg, VA 24061 USA

f Department of Chemical Engineering, Virginia Tech, Blacksburg, VA 24061 USA

<p>Rapid, Autonomous High-throughput Characterization of Hydrogel Rheological Properties via Automated Sensing and Physics-guided Machine Learning</p>
Click to Enlarge
Chronicle Description

  • Rapid, autonomous high-throughput characterization of hydrogels in 96-well plate formats via automated sensing and machine learning.

  • High-throughput characterization of composition-property relations and percolation processes in hydrogel libraries.

  • Physics-guided supervised machine learning-driven classification of material phase using sensor physics-based feature augmentation.

Associated Publications

Rapid, autonomous high-throughput characterization of hydrogel rheological properties via automated sensing and physics-guided machine learning

2022
Zhang, Junru; Liu, Yang; Sekhar.P, Durga Chandra; Singh, Manjot; Tong, Yuxin; Kucukdeger, Ezgi; Yoon, Hu Young; Haring, Alexander P.; Roman, Maren; Kong, Zhenyu; Johnson, Blake N.
Applied Materials Today
Data Files & Attachments
1 File(s) Available
File Name Description File Type Size Action
Pluronic F127 Rheometer Data Rheometer data serves as the gold standard for characterizing materials' rheoloigical property. Here we provide the Pluronic F127 rheometer data as benchmarking data that shows the data generated from our method exhibit strong correlation with rheometer data. It suggests that our method can serves as an alternative for rheometer to characterize rheological property and percolation process in an autonomous and high throughput fashion. XLSX 9.29 KB Login to Download
Research Keywords & Tags
Chemical structure Machine learning Mathematical methods
Related Chronicles

Protein−Ligand CH−π Interactions: Structural Informatics, Energy Function Development, and Docking Implementation

View Chronicle

Knowledgebra: An Algebraic Learning Framework for Knowledge Graph

View Chronicle