Nanotechnology World Conference 2026

Abstract

Titanium nitride (TiN) coatings are widely used in the field of cutting tools, biomedical implants, aerospace parts, and wear resistant applications because of their high hardness, corrosion resistance, and thermal stability. The coatings are fabricated through magnetron sputtering process and the performance of depends on their grain size, surface roughness and coating morphology which are determined by the deposition parameters during the sputtering process. The parameters are often optimized through a large number of trial and error experiments, which can be time consuming, material wasting, and expensive. Thus, the need for predictive computational models has grown in importance to rapidly design and optimize coatings.

In this work an Artificial Neural Network (ANN) model is designed to predict the characteristic grain size and surface morphology of TiN magnetron sputtered coatings based on the coating parameters including coating thickness, substrate bias voltage and deposition temperature. The considered coating thickness range is 250 and 1500 nm, substrate bias voltages 0 to 150 V and deposition temperatures 70 to 350 degrees C. These ranges were defined based on limitations of machine, Structure Zone Model (SZM) and relevant data reported in the literature.

Experimental and literature data is collected for training and validation purpose to learn the variables dependency and the microstructural evolution. Atomic Force Microscopy (AFM), Scanning Electron Microscope (SEM) and Grazing Incidence X ray Diffraction (GIXRD) are used to investigate the surface morphology and grain growth behavior. The effect of the foreseen microstructural characteristics on the mechanical performance was evaluated by nanoindentation to measure the hardness and elastic modulus.