Charge Coupled Devices were invented in 1969 by Willard S. Boyle and George E. Smith. In combination with the large apertures of modern telescopes they have gone on to revolutionise astronomy. They are characterised by their extreme sensitivity (greater than 90% peak), their very high cosmetic quality and low read noise. A variant, the electron multiplying CCD (EMCCD) uses a charge-domain multiplication mechanism that even allows single photons to be easily detected.
A popular analogy for the CCD structure is shown below. Here, each bucket represents a pixel, the conveyor belts represent the vertical and horizontal transfer structures and the measurement node represents the readout amplifier.
The conveyor belt structures are implemented in silicon as a repeating pattern of surface electrodes. These overlay the light sensitive part of the detector which is essentially a large photodiode. During integration the surface electrodes define a pattern of potential wells (one per pixel) in which detected photoelectrons are trapped. During readout the voltage of the surface electrodes is then modulated in a sequence that drives the pixel charge packets towards a readout amplifier positioned at the edge of the silicon , outside of the light sensitive area.
The conveyor belt analogy is continued below. The three graphics show the first row of the image being transferred vertically into the horizontal serial register. The serial register is then activated and transfers pixel charge, bucket at a time, into the measurement cylinder. The measurement is actually done using a very low value capacitor (a few tens of femto Farads) that converts the charge into a measurable voltage . A typical sensitivity is 7uV per electron.
Some introductory presentations on CCDs can be found here:
if you want to know more you should also consider my course. Details here.
Manufacturers’ websites:
An image gallery of a variety of CCD projects from the last 25 years is shown below.
