Answers are grounded in this paper and cite their sources.
The basic premise underlying the use of the scanning electron microscope (SEM) for linewidth metrology in semiconductor research and production applications is that the video image acquired, displayed, analyzed, and ultimately measured accurately reflects the structure of interest. However, it has been clearly demonstrated that image distortions can be caused by the detected secondary electrons not originating at the point of impact of the primary electron beam and by the type and location of the secondary electron detector. These effects and their contributions to the actual image or linewidth measurement have not been fully evaluated. Effects due to uncertainties in the actual location of electron origination do not affect pitch (line center‐to‐center or similar‐edge‐location‐to‐similar‐edge‐location spacing) measurements as long as the lines have the same edge geometries and similar profiles of their images in the SEM. However, in linewidth measurement applications, the effects of edge location uncertainty are additive and thus give twice the edge detection error to the measured width. The basic intent of this work is to demonstrate the magnitude of the errors introduced by beam/specimen interactions and the mode of signal detection at a variety of beam acceleration voltages and to discuss their relationship to precise and accurate metrology.
No takes yet. Share an insight, caveat, or question.
Postek et al. (1988) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: