Experimental and artificial neural network (ANN) method of microhardness and wear behaviour in Ni-P/AlN nanocomposite coatings


HÜKÜMDAR Ö., KESKİN A., KUMLU U., AKAR M. A., YAŞAR A., KILINÇÇEKER G.

Ceramics International, 2026 (SCI-Expanded, Scopus) identifier

  • Yayın Türü: Makale / Tam Makale
  • Basım Tarihi: 2026
  • Doi Numarası: 10.1016/j.ceramint.2026.01.016
  • Dergi Adı: Ceramics International
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Chemical Abstracts Core, Compendex, INSPEC
  • Anahtar Kelimeler: AlN, ANN, Electrodeposition, Microhardness, Ni-P, Wear
  • Çukurova Üniversitesi Adresli: Evet

Özet

In this study, Ni-P/AlN nanocomposite coatings were produced by electrodeposition on an St-37 steel substrate using different parameters and bath concentrations. The effects of aluminum nitride (AlN) ceramic particle and sodium hypophosphite concentrations, and current density on the structural, mechanical, and tribological properties of the deposited coatings were systematically investigated. Surface morphology and elemental composition were analyzed using SEM and EDS, while microstructural features were determined using XRD analysis. In addition, the complex relationship between the experimentally obtained microhardness and wear volume loss results and the production conditions was modeled using artificial neural networks (ANN). The results showed that the addition of AlN ceramic particles to the bath content significantly altered the surface morphology. XRD analysis revealed that increasing the AlN bath content decreased the average crystal grain size to 4.45 nm. Microhardness values increased from ∼260 HV for pure nickel to ∼480 HV for Ni-P/AlN coatings deposited with 2 g/L AlN and 20 g/L sodium hypophosphite at a current density of 50 mA/cm2, corresponding to an increase of approximately 92 %. Tribological analysis showed that the wear volume loss for pure nickel decreased from 0.561 mm3to 0.074 mm3for Ni-P/AlN composite coatings. The ANN model results showed good agreement between the experimental data and the predicted data (R = 0.9942 for microhardness and R = 0.99497 for wear volume loss).