Effect of chemical treatment on the mechanical and thermal performance of flax fiber reinforced aluminium 6082 laminate: a machine learning-enhanced investigation.

Authors

M Vinod, B S Nithyananda, Shrishail B Sollapur, Satish Gajbhiv, R N Chikkangoudar, Priya Dongare Jadhav, Shekhar Milind Mane, Rasika Gajendra Patil, Abhijit Bhowmik, Nagaraj Ashok

Published in

Scientific reports. Jun 18, 2026. Epub Jun 18, 2026.

Abstract

This work examined the role of alkaline and epoxy-based surface chemical treatments on the mechanical and thermal properties of flax fiber reinforced Aluminum Alloy 6082 hybrid Fiber Metal Laminates (FMLs), integrated with machine learning (ML) predictive frameworks. Flax fiber mats underwent surface modification via a 1% NaOH alkaline soak followed by an epoxy sizing treatment to promote stronger bonding with the aluminum matrix. Concurrently, aluminum sheets were treated with an NaOH-Na₂CO₃ alkaline bath to enhance metal-polymer interfacial compatibility. Composite laminates were manufactured through hand layup combined with compression molding at 30 bar and 70 °C curing for four hours. Tensile characterization followed ASTM D638 protocols, thermal conductivity measurements employed a guarded hot plate (GHP) system per ASTM C177, and corrosion evaluation relied on gravimetric immersion in 3.5 wt.% NaCl over 30 days. Predictive models using Artificial Neural Networks (ANN), Support Vector Regression (SVR), and Random Forest (RF) were developed on a structured 180-observation experimental dataset. Steady-state thermal conductivity values of 0.1039 W/m·K and 0.065 W/m·K were recorded for untreated and chemically treated FFAL laminates, respectively. Surface treatment yielded a 31% gain in tensile strength (168.5 MPa to 220.7 MPa) and a 34% rise in tensile modulus (12.3 GPa to 16.34 GPa), while thermal conductivity declined by 37.4%. Among all models, the Random Forest algorithm demonstrated superior predictive capability with R² = 0.972 for tensile strength and R² = 0.969 for thermal conductivity; MAPE values remained below 3.5%. All experimental trials were performed in triplicate with mean values reported alongside standard deviation. These findings substantiate that combining chemical surface modification with data-driven ML optimization constitutes a reliable design methodology for high-performance, sustainable natural fiber metal composites.

PMID:
42315743
Bibliographic data and abstract were imported from PubMed on 19 Jun 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.