Verification of Si wafer first-order phonon peaks for reliable calibration of Raman microscopes

Abstract

Raman microscopes are widely used in various fields and their spectral resolutions differ greatly depending on the system and optical components. Thus, the microscopes must be calibrated before measurement to obtain reliable results. Although the first-order phonon peak of Si wafers at ⁓520 cm⁻¹ is generally used as a calibrant of Raman microscopes, not only is it unclear how the positions of the first-order phonon peaks are comparable over Si wafers of different manufacturers, dopant types and crystal orientations, but they also shift with the temperature and residual stress. We examined the changes in the position of the first-order phonon peak at different temperatures using a HeNe laser at 633 nm and its plasma lines. Because a comparable linear relationship between the temperature and the wavenumber was obtained regardless of the Si wafer examined, most commercially available Si wafers can be used for the calibration of Raman microscopes. Although shifting of the peak was introduced by the laser power due to an increase in temperature at the laser spot, it was less sensitive than broadening of the peak width. A peak shift was observed with a 532-nm laser at 2.1 mW using a 100× air objective lens (numerical aperture: 0.9), but this did not occur with a 633- or 785-nm laser even at more than 10 mW. Thus, less laser power should be used to calibrate Raman microscopes using the first-order phonon peak of Si wafers under high-resolution conditions, especially for a 532-nm laser.