Related Experiment Video
Updated: Jan 16, 2026
11:20
Fabrication of Superhydrophobic Metal Surfaces for Anti-Icing Applications
Published on: August 15, 2018
9.0K
Machine learning-driven stability analysis of eco-friendly superhydrophobic graphene-based coatings on copper
Himanshu Prasad Mamgain1, Maria Vittoria Diamanti2, Pravat Ranjan Pati3
1Department of Physics, Applied Science, School of Advanced Engineering, UPES, Dehradun, 248007, Uttarakhand, India. himanshuhm1111@gmail.com.
Scientific Reports
|October 3, 2025
Summary
Machine learning models predict the durability of superhydrophobic graphene coatings on copper. XGBoost and Random Forest excel at predicting coating stability under various stresses, enhancing material design.
Area of Science:
- Materials Science
- Computational Materials Science
- Surface Engineering
Background:
- Developing durable, eco-friendly superhydrophobic (SHP) coatings is crucial for protecting materials like copper.
- Graphene incorporation enhances coating resilience and service life.
- Predicting coating stability under various stress conditions is challenging using traditional methods.
Purpose of the Study:
- To integrate machine learning (ML) with materials science for predicting the stability of SHP graphene-based coatings on copper.
- To evaluate the performance of various ML and regression models in predicting contact angle (CA) retention under stress.
- To accelerate the design and analysis of durable SHP coatings by reducing experimental testing.
Main Methods:
- Applied ML models including XGBoost, polynomial regression, Random Forest (RF), K-Nearest Neighbours (KNN), and Support Vector Regression (SVR).
- Predicted contact angle (CA) stability under stress conditions: NaCl immersion, abrasion, tape peeling, sand impact, and open-air exposure.
- Compared the predictive accuracy and generalization capabilities of different ML algorithms.
Main Results:
- Ensemble learning models (XGBoost, RF) and higher-order polynomial regression showed superior predictive accuracy compared to SVR and KNN.
- ML models accurately predicted long-term CA values, highlighting graphene's benefit in preserving superhydrophobicity under mechanical stress.
- XGBoost and RF effectively captured nonlinear relationships between stress parameters and CA retention.
Conclusions:
- Advanced predictive models like XGBoost and higher-degree polynomial regression are essential for understanding SHP coating stability.
- ML integration significantly accelerates the development cycle for durable, eco-friendly SHP coatings.
- Graphene-based coatings demonstrate enhanced resilience and extended service life, validated by ML predictions.
