Evaluation of Errors in the Measurement of Surface Roughness at High Spatial Frequency by Atomic Force Microscopy on a Thin Film

Abstract

Factors such as sample deformation, which comes from the applied force, and the probe shape, which results in image dilation, lead to the errors in the measurement of roughness by atomic force microscopy (AFM). We explored the roughness errors that result from the applied imaging force, different probe materials, and the probe radius in the roughness measurements of a polysilicon film surface. Structures with high spatial frequencies, which are strongly affected by the probe shape, were separated by comparing radial power spectrum density curves. A geometrical model was established to describe the roughness–probe radius relationship, which was compared with experimental results under optimized imaging conditions. For a surface with a small correlation length WCL = 14.1 nm comparable to the radius of a commercial probe (Rt<10 nm), a probe with a 7 nm radius contributes an error of around 43.7%. To achieve a precise measurement of the roughness of such a surface, it is important to develop a suitable model that describes the roughness distribution well.