SLOW SCAN CCD CAMERA ON OUR HVEM

点击翻译按钮获取中文摘要

A new Gatan lens-coupled, slow-scan CCD camera for our HVEM was installed in December 1997. In January, Paul Mooney from Gatan returned to help evaluate the performance of the camera with different scintillators, including a prototype YAG. Procedures and scripts for focussing the camera were worked out. We began to characterize camera performance by measuring its modulation transfer function (MTF), using an image of a high contrast edge. The best such object turned out to be the beam stop, a thin rod of dense metal that can be inserted just above the viewing screen. Our analysis has shown that the resolution of the camera is good but not excellent; it captures useful information about the specimen out to a spatial frequency of about 0.35-0.4 cycles/pixel (0.5 cycles/pixel is the Nyquist frequency, the highest that can be reported in a digital image). This means that the 1024 by 1024 pixel image from the camera has only as much information as a 640 by 640 pixel image recorded on film and digitized by our usual means. Some ability to montage images automatically will therefore be important for realizing the full potential of the camera for our studies of cellular structure. We have also assessed the camera's detective quantum efficiency (DQE) as a function of spatial frequency, using a method devised by Mooney for both CCD cameras and film. So far results have been obtained only at 500KV. At higher electron doses, such as we use for routine tomography, the camera performs as well as or better than film, except at the highest spatial frequencies. As the dose is reduced, the camera retains its good performance while film has a progressively worse DQE. This indicates that the camera may provide a significant advantage for cryo-microscopy, where dose must be minimized to avoid damage to the specimen. An experimental single-crystal YAG scintillator has also been evaluated at 500 KV. Its MTF is poor at this voltage, but its DQE at low frequencies was substantially higher than with the phosphor screen. Further experiments at 1 MeV will be carried out to evaluate whether a YAG scintillator with optimized thickness would provide a distinct advantage over phosphor.