Before any transistor is laid down, the incoming silicon wafer must be analyzed for flatness and defects. From this inspection, wafers can be classified to allow the best wafers to be used for the smallest technology node. Typically measurement techniques such as interferometry are used for this. Extremely stable cameras with low noise are needed for accurate measurements. This is too general to be really helpful with camera selection. With consideration of the application and the measurement method, we are able to prioritize the most important camera parameters to consider.
The interferometry measurements would be very difficult and expensive if the entire full size wafer (300 mm) were imaged in one view. The optics costs alone would be exorbitant. Therefore typical interferometry methods (i.e. Mirau or Fizeau) use small size optics to perform many accurate measurements that are stitched together to one flatness map.
Figure 1. Fizeau Interferometer (Source: Wikipedia)
Figure 2. Combined Wafer Flatness Map (Source: TU Eindhoven)
This technique works because when two waves with the same frequency combine, the resulting pattern is determined by the phase difference between the two waves. Waves that are in phase will undergo constructive interference and waves that are out of phase will undergo destructive interference. In practice the light beams will have different intensities, so the result will not be true 200% - 0%. It will therefore exhibit a much lower contrast.
In order to then prioritize the camera requirements for this application, it is helpful to first think of the image or images required for the measurement. For interferometry, several images are required, information in the entire image is used, and there is limited contrast in the images. Therefore the most important camera specifications to consider are:
- Dynamic range to provide detailed information in low contrast images
- Image uniformity for accurate data over the entire image
- Full well capacity since shot noise is the dominant noise source
- Frame rate to not limit throughput since many images are necessary for one measurement
- Image-to-image stability because multiple images are combined for the measurement
- Mechanical / Thermal stability to support image-to-image stability