Detecting and addressing the surface following errors in the calibration of step heights by atomic force microscopy

Abstract

Step height standards are widely used for the calibration of the z-axis scale of scanning probe microscopes. NPL, in common with a number of other national measurement institutes, has constructed a metrological atomic force microscope (MAFM) that delivers traceable calibration of step height standards. The measurement uncertainty of the calibrations is a combination of those uncertainties associated with the metrology frame, whose values can be estimated with good accuracy, and those associated with the AFM head, whose values are more difficult to quantify. Estimates of these values are usually based on the repeatability of measurements, either in isolation for a particular sample, or in combination with historical data on instrument performance. That the estimates are reasonable may be verified by comparing the AFM measurements with measurements made on other independent, traceable systems. This paper describes how the verification process led to the detection of a significant following error in one of the heads employed on the NPL MAFM, which could then be evaluated and addressed. The value of measurement uncertainty for the step height measurements is also dependent on the physical characteristics of the standard. Results are presented for measurements of a fine scale step height standard that has been designed for optimum compatibility with the measurement procedure in order to reduce the measurement uncertainty.