Machine intelligence accelerated design of conductive MXene aerogels with programmable properties

• Published in: Nature communications Vol. 15; no. 1; pp. 4685 - 14
• Main Authors: Shrestha, Snehi, Barvenik, Kieran James, Chen, Tianle, Yang, Haochen, Li, Yang, Kesavan, Meera Muthachi, Little, Joshua M., Whitley, Hayden C., Teng, Zi, Luo, Yaguang, Tubaldi, Eleonora, Chen, Po-Yen
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 01.06.2024; Nature Publishing Group; Nature Portfolio
• Subjects: 140/146; 147/135; 639/301/1023/1025; 639/925/357/1018; 639/925/357/537; Aerogels; Artificial neural networks; Automation; Cellulose; Compression; compression strength; Compressive strength; Data augmentation; Fabrication; finite element analysis; Freeze drying; Gelatin; Glutaraldehyde; Heating; Humanities and Social Sciences; Industrial robots; Inverse design; Learning algorithms; Machine learning; Manufacturing engineering; Mechanical properties; multidisciplinary; MXenes; Neural networks; Ohmic dissipation; Parameters; Physicochemical properties; Prediction models; Predictions; Resistance heating; Robot learning; Robotics; Science; Science (multidisciplinary); Structural integrity; Support vector machines; Thermal management; Workflow
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-024-49011-8

Designing ultralight conductive aerogels with tailored electrical and mechanical properties is critical for various applications. Conventional approaches rely on iterative, time-consuming experiments across a vast parameter space. Herein, an integrated workflow is developed to combine collaborative robotics with machine learning to accelerate the design of conductive aerogels with programmable properties. An automated pipetting robot is operated to prepare 264 mixtures of Ti 3 C 2 T x MXene, cellulose, gelatin, and glutaraldehyde at different ratios/loadings. After freeze-drying, the aerogels’ structural integrity is evaluated to train a support vector machine classifier. Through 8 active learning cycles with data augmentation, 162 unique conductive aerogels are fabricated/characterized via robotics-automated platforms, enabling the construction of an artificial neural network prediction model. The prediction model conducts two-way design tasks: (1) predicting the aerogels’ physicochemical properties from fabrication parameters and (2) automating the inverse design of aerogels for specific property requirements. The combined use of model interpretation and finite element simulations validates a pronounced correlation between aerogel density and compressive strength. The model-suggested aerogels with high conductivity, customized strength, and pressure insensitivity allow for compression-stable Joule heating for wearable thermal management. Machine learning-assisted robots produce MXene aerogels containing cellulose, gelatin, and glutaraldehyde, fabricating 162 compositions. Inverse design from resulting properties affords tailored compression-stable materials for Joule heating.