Answers are grounded in this paper and cite their sources.
The charge-coupled device, as described originally by Boyle and Smith,1operates by moving minority carriers along the surface of a semiconductor with voltage pulses applied to metal electrodes which are separated from the semiconductor by an insulating layer. The transit from one electrode to the next is determined by the minority carrier transport under the influence of their own potential, fringing fields, and diffusion, and by the trapping properties of interface states. The transport limitations are largely determined by device geometry;2for 10-μm electrodes, thermal diffusion is predominantly responsible for transferring the last small amounts of charge forward and limits efficient operation to clock frequencies below 10 MHz. Surface state trapping is much less dispersive, and even at low frequencies 1011states/cm2eV can impose the requirement for regeneration after as few as 100 transfers.3,4In order to circumvent these problems, Boyle and Smith5have proposed a modified CCD structure in which the charges do not flow at the semiconductor surface; instead they are confined to a channel which lies beneath the surface. This buried channel device has the potential of eliminating surface state trapping. (Bulk trapping should be several orders of magnitude less important as a CCD loss mechanism.) Calculations show that this modification will give rise to increased fringing fields under the CCD electrodes. Diffusion is replaced as an important factor in the intrinsic transfer process by the more powerful field-aided transfer. This leads to fast, efficient transport even when very little charge remains to be transferred. In addition to these advantages, the higher mobility found in the bulk of the semiconductor should further enhance the speed of the device.2
No takes yet. Share an insight, caveat, or question.
Walden et al. (1972) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: