A new nano-accuracy AFM system for minimizing Abbe errors and the evaluation of its measuring uncertainty

• Published in: Ultramicroscopy Vol. 107; no. 4; pp. 322 - 328
• Main Authors: Kim, Dongmin, Lee, Dong Yeon, Gweon, Dae Gab
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.04.2007
• Subjects: Abbe error minimization; AFM; Flexure hinge stage; Mirror non-orthogonality; Pitch length; Uncertainty evaluation; Flexure hinge stage; AFM; Uncertainty evaluation; Mirror non-orthogonality; Pitch length; Abbe error minimization; 81.70.Fy; 83.85.Ei
• DOI: 10.1016/j.ultramic.2006.08.008
• ISSN: 0304-3991

A new AFM system was designed for the establishment of a standard technique of nano-length measurement in a 2D plane. In a long range (about several tens of micrometers), measurement uncertainty is dominantly affected by the Abbe error of the XY scanning stage. No linear stage is perfectly straight; in other words, every scanning stage is subject to tilting, pitch and yaw motions. In this paper, an AFM system with minimum offsets of XY sensing is designed. Moreover, the XY scanning stage is designed to minimize the rotation angle, as Abbe errors occur through multiple combination of the offset and the rotation angle. To minimize the rotation angle, an optimal design is performed by maximizing the ratio of the stiffness of the parasitic direction to the motion direction of each stage. This paper describes a design scheme of a full AFM system, in particular, the XY scanner. The full range of a fabricated XY scanner is 100 μm×100 μm. The tilting, pitch and yaw motions are measured by an autocollimator to evaluate the performance of the XY stage. The results show that the XY scanner have a 0.75 arcsec parasitic rotation about the maximum range, thus the uncertainty in terms of the Abbe errors are very small relative to other standard equipment. Using this AFM system, a 3 μm pitch specimen was measured. The measurement uncertainty of the total system was evaluated especially about pitch length. For a 1D evaluation, Abbe errors are the most dominant factor, and the expanded combined uncertainty ( k = 2 ) of system was ( 4.13 ) 2 + ( 5.07 × 1 0 - 5 × p ) 2 (nm). For a 2D evaluation, mirror non-orthogonality and Abbe errors are dominant factors, and expanded combined uncertainty ( k = 2 ) of the system was ( 4.13 ) 2 + ( 1.228 × 1 0 - 4 × p ) 2 in the X direction, and ( 6.28 ) 2 + ( 1.266 × 1 0 - 4 × p ) 2 in the Y direction (the unit is nanometers), where p is the measured length in nm.