Error-corrected AFM: a simple and broadly applicable approach for substantially improving AFM image accuracy

Abstract

Atomic force microscopy (AFM) has become an indispensable tool for imaging the topography and properties of surfaces at the nanoscale. A ubiquitous problem, however, is that optimal accuracy demands smooth surfaces, slow scanning, and expert users, contrary to many AFM applications and practical use patterns. Accordingly, a simple correction to AFM topographic images is implemented, incorporating error signals such as deflection and/or amplitude data that have long been available but quantitatively underexploited. This is demonstrated to substantially improve both height and lateral accuracy for expert users, with a corresponding 3–5 fold decrease in image error. Common image artifacts due to inexperienced AFM use, generally poorly scanned surfaces, or high speed images acquired in as fast as 7 s, are also shown to be effectively rectified, returning results equivalent to standard ‘expert-user’ images. This concept is proven for contact mode AFM, AC-mode, and high speed imaging, as well as property mapping such as phase contrast, with obvious extensions to many specialized AFM variations as well. Conveniently, as this correction procedure is based on either real time or post-processing, it is easily employed for future as well as legacy AFM systems and data. Such error-corrected AFM therefore offers a simple, broadly applicable approach for more accurate, more efficient, and more user-friendly implementation of AFM for nanoscale topography and property mapping.