Scientific Charge-Coupled Device Imaging

Scientific imaging charge-coupled device (CCD) design and fabrication are driven primarily by performance, unlike commercial device fabrication for which cost is very important and constrains the effort that can be spent on performance. Because scientific and surveillance CCDs often are used to detect images from large optical systems, often a very large-area imager is required.


Figure 1. Historical growth in Lincoln Laboratory–fabricated CCDs.

Figure 1. Historical growth in Lincoln Laboratory–fabricated CCDs.

Figure 1 shows the historical growth in the size of CCDs designed and fabricated at Lincoln Laboratory over a 20-year span. A quarter is used to show the scale of the devices. The 10-megapixel device at the right is used in the Air Force Ground-based Electro-Optical Deep-Space Surveillance network and is a replacement for a Vidicon tube that had been used in this system since the 1970s. The optics size drove the size requirement for this device, which is approximately 3.8" in its diagonal measurement (the largest imager that could be built at Lincoln Laboratory in 1994 when 4" diameter wafers were used).

Another important attribute of scientific imagers is high sensitivity. This implies high (~100%) quantum efficiency and low read noise (which implies low output data rate). Search rate is another attribute that is sometimes important for Department of Defense (DoD) applications. This requirement drove us to use eight multiple outputs in the 10-megapixel device. This device is well suited for rapidly searching the sky for very faint objects, and has been very successfully used to search for asteroids in the Lincoln Near-Earth Asteroid Research (LINEAR) program.

Astronomy telescopes have large optics that create large prime image areas and that are used to look at dim objects; therefore, they have similar needs to defense surveillance CCDs.  

Figure 2. Eight-megapixel astronomy CCD.Figure 2. Eight-megapixel astronomy CCD.

Figure 2 shows a 2048 × 4096-pixel CCD imager mounted on a three-side-abuttable ceramic substrate with an attached flexprint for the input/output signals. This device may be abutted to similar devices on three sides to allow construction of very large image arrays that are suitable for astronomy.

The array shown in Figure 3 was constructed for the Canada-France-Hawaii Telescope (CFHT) on Mauna Kea, Hawaii, in 1998. It is composed of twelve 8.4-megapixel imagers, for a total focal-plane size of 100.7 megapixels. The physical size of this image array is about 12 × 18 cm. At the time, this was probably the largest CCD focal plane in existence.

Figure 3. One-hundred-one–megapixel CCD focal plane fabricated in 1998 for the CFHT on Mauna Kea, Hawaii.

Astronomy imaging from ground-based telescopes is degraded by atmospheric distortion. A large part of this distortion can be described as translational—the image moves back and forth across a small part of the imager. The orthogonal-transfer CCD (OTCCD) was conceived in the 1970s as a way to compensate for translational motion by moving the previously collected charge in unison with the dancing image to avoid motion blur during the image integration period. At Lincoln Laboratory, we have developed technology to build large OTCCD devices with high yield. Recently, we have developed large arrays of OTCCDs on the same chip, called orthogonal transfer arrays (OTAs), for use in ground-based astronomy.


top of page