In-Situ Nanoindentation Surface Topography of Lead-Free Piezoelectric Thin Films

Surface roughness significantly affects the performance of microelectromechanical systems (MEMS) and piezoelectric films. This study investigates the impact of surface roughness on the mechanical properties of thin piezoelectric films using nanoindentation and scanning probe microscopy (SPM). Four piezoelectric films with different thicknesses (220, 350, and 450 nm) and substrate configurations (LNO/SiO2/Si or LNO/Si) were analyzed. A discriminant analysis revealed that the fractal dimension is more effective than the arithmetic mean height (Sa) for distinguishing surfaces, with only 2% misclassification versus 25% for Sa. A multiscale analysis identified the Smr2 parameter with low-pass filtering at 140 nm as highly effective for surface discrimination, achieving only 0.1% misclassification. The analysis of the roughness parameter Sa at various scales showed that band-pass filtering at 500 nm yielded a 0.7% misclassification rate, indicating its relevance for fractal roughness characterization. Most relevant roughness parameters for mechanical property correlation were found: Smr2 with low-pass filtering at 500 nm correlated best with hardness (R2 = 0.82), and Vvc with low-pass filtering at 2 nm correlated best with reduced elastic modulus (R2 = 0.84). These results demonstrate that surface roughness features like valley volume and voids significantly impact the apparent mechanical properties of piezoelectric films.