The increase in the number and complexity of process levels in semiconductor production has driven the need for the development of new measurement methods that can evaluate semiconductor devices at the critical dimensions of fine patterns and simultaneously inspect nanoscale contaminants or defects. However, conventional optical inspection methods often fail to resolve device patterns or defects at the level of tens of nanometers required for device development owing to their diffraction-limited resolutions. In this study, we used the stochastic optical reconstruction microscopy (STORM) technique to image semiconductor nanostructures with feature sizes as small as 30 nm and detect individual 20 nm-diameter contaminants. STORM imaging of semiconductor nanopatterns is based on the development of a selective labeling method of fluorophores for a negative silicon oxide surface using the charge interaction of positive polyethylenimine molecules. This study demonstrates the potential of STORM for nanoscale metrology and in-line defect inspection of semiconductor integrated circuits.
Keywords: semiconductor imaging; semiconductor metrology; stochastic optical reconstruction microscopy (STORM); wafer inspection.