A stroboscopic scanning electron microscope (SEM) has been proposed as a means of generating still or slow-motion pictures of moving structures in microelectromechanical systems (MEMS). Such imaging is used in characterizing the dynamics of MEMS; characterization of the dynamics is a critical component of the MEMS development cycle.
Conventional strobed-illumination microscopy with visible or infrared light provides adequate temporal resolution but insufficient spatial resolution for measuring subwavelength motions in the main plane of a typical MEMS. A conventional SEM provides adequate spatial resolution, but is inadequate for resolving motions at frequencies greater than several tens of hertz because the illuminating electron beam is continuous. The proposed stroboscopic SEM would offer both the spatial resolution of a conventional SEM and the temporal resolution of conventional optical stroboscopy, making it possible to form crisp images of moving (e.g., vibrating) MEMS structures.
According to the proposal, a conventional SEM would be augmented with an electronic beam blanker that would be operated in coordination with a signal generator. The output of the signal generator would control the vibrational excitation of a MEMS device mounted in the SEM (see figure).
In one mode of operation, the blanking-pulse-repetition frequency would be set equal to the frequency of vibration, so that the resulting SEM image would "freeze" the motion at some phase in the vibration cycle. The phase could be varied by adjusting the phase offset between the vibration-waveform and blanking-pulse generators. In another mode of operation, the blanking-pulse-repetition frequency would be made to differ slightly (no more than a few hertz) from the vibration frequency, yielding a sequence of images at slightly different phases (in other words, a slow-motion picture). Freeze-motion images taken at different phases could be used to quantify the shape of a vibrational mode at the frequency of excitation, while slow-motion pictures could be used to obtain qualitative understanding of the motion.
This work was done by Kirill Shcheglov and Russell Lawton of Caltech for NASA's Jet Propulsion Laboratory.