The Frame CCDs are devices in which the total area of the CCD is available for sensing incoming photons during the exposure period. During readout of the CCD, charge is shifted sequentially across the array necessitating the use of a shutter to prevent smearing for almost all exposure lengths. (Technically,if the exposure time is much longer than the actual readout rate, then the level of smearing can be quite small.) This format has 100% fill factor, which means that 100% of each pixel area is being utilized to detect photons during the exposure. The pixels are typically square in format so there is no image distortion inherent to the detector. These devices are available with pixel sizes that range from 6.8 microns square (Kodak 1400 CCD) up to 24 microns square (SITe 512 CCD) and with pixel formats of 512 × 512 up to 2k × 3k. The full frame CCDs can come designed for front side illumination or for back side illumination. With the front illuminated CCD, the light must traverse the SiO 2 gate structures overlying the photosensitive silicon called the depletion layer. These gate structures are essential for pixel formation in the CCD, but the differences in the indices of refraction between the gates and the silicon causes shorter wavelength light to reflect off the surface of the CCD, reducing the QE for those wavelengths.