AI-Guided Design of Multifunctional Epoxy Hybrid Nanocomposites with MWCNTs– Nanoclay Networks for Next-Generation Fire-Safe and Self-Sensing Coatings

This study reports the AI-guided design of epoxy/MWCNT/nanoclay hybrid nanocomposites that achieve a unique integration of structural reinforcement, flame retardancy, optical shielding, electrical conductivity, and multifunctional sensing. Response surface methodology (RSM) and Random Forest (RF) models were combined to explore a four-factor Box–Behnken design across 29 experimental runs. The optimized hybrids (≈ 0.8–1.0 wt% CNT and 3–4 wt% nanoclay) demonstrated superior fire resistance with limiting oxygen index (LOI) > 29% (vs. ~23% for neat epoxy) and peak heat release rate (pHRR) reduced to ~ 760 kW·m⁻² (down from ~ 1040 kW·m⁻²). Mechanical reinforcement was confirmed with tensile strength increasing to ~ 75 MPa and flexural modulus exceeding 4.5 GPa, representing ~ 25–30% gains over neat epoxy. Optical and barrier functions were enhanced with UV-blocking efficiencies above 90% and char yields up to 20%. Electrical conductivity improved by 4–5 orders of magnitude, stabilizing at ~ 10⁻³ S·m⁻¹, while multifunctional sensing was validated with thermal sensitivity above 0.04 ΔR/R·°C⁻¹ and NH₃ gas response approaching 0.11 ΔR/R. Long-term durability tests confirmed retention of > 80% of mechanical, fire, and electrical properties after 100 h UV irradiation, 200°C thermal aging, and 10⁵ cyclic loads. XPS and XRD analyses revealed strong interfacial bonding (C–O, Si–O) and exfoliated clay structures that underpin multifunctionality. AI-guided models achieved predictive accuracy above R² = 0.98, reducing trial-and-error efforts by nearly 50%. Collectively, these results establish the developed epoxy/MWCNT/nanoclay hybrids as next-generation multifunctional coatings with direct applicability to aerospace, automotive, and electronic protection.