Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xmm.vilspa.esa.es/user/socpub/ccd_review/node8.html
Дата изменения: Fri Aug 30 15:55:14 1996
Дата индексирования: Tue Oct 2 11:53:10 2012
Кодировка:

Поисковые слова: arp 220
Front Illumination next up previous
Next: Back Illumination Up: Detection Efficiency Previous: Deep Depletion

Front Illumination

The standard CCDs have been made with polycrystalline silicon electrodes, typically 0.5 tex2html_wrap_inline261 m thick, and usually with a passivating oxide layer of similar thickness on top. A total of tex2html_wrap_inline245 m represents a significant attenuation layer for sub-keV photons (the silicon attenuation length at the softest energies of the traditional X- ray imaging band is tex2html_wrap_inline265 's tex2html_wrap_inline267 .). These electrode thicknesses have been defined by the need to ensure low resistance to allow rapid propagation of clock signals along the electrodes, particularly important for high rate, TV applications. However, astronomical applications require only relatively slow clocking rates and the electrodes may be made as thin as fabrication yields allow, in order to reduce the attenuation of the dead layers. EEV, Loral and Lincoln Labs are developing electrode thicknesses of tex2html_wrap_inline269 m, for example.

The first attempt to produce a front-illuminated CCD with a thin electrode structure was the Virtual phase technology [Hynecek]. In this case, two of the electrode phases were removed, and the potential well in that area defined by a surface implant. Charge transfer was effected by pulsing the voltages on the remaining electrode alternately above and below the potential necessary to form the same surface potential as in the implanted phase. The implant was made asymmetric to ensure uni-directional charge transfer. While good blue and soft X-ray performance was demonstrated, the manufacturer experienced great difficulty in forming the implants with reproducible properties, with recurring problems such as noise generated by avalanching in high electric field regions, and poor charge transfer efficiency at very low signal levels. Finally it was found that radiation damage created large dark current generation rates and the build up of interface charge which prevented correct operation of the implants. The methods which have to be employed to circumvent these problems for the Solar-A soft X-ray instrument are described by [Acton et al].

The Open Pinned Phase [Janesick et al 1989b] technology uses the implanted structure concept of the virtual phase device, but requires a less complex implant because only one out of three electrodes is replaced with an implant. It should be simpler to manufacture with reproducible performance, but it still remains to be seen if the requirement to pass signal charge through the implanted section will limit the low signal level performance after radiation damage.

For enhanced hard X-ray response, the front-illuminated technology has the advantage of allowing as thick a detection layer as possible. Photons are absorbed preferentially near the surface collection potential wells with the minimum radial spread of charge between pixels. There is no requirement fully to deplete the silicon substrate, and photons or particle tracks generating charge outside the depleted volume may produce signals which are detected but split between many pixels. Hard X-ray detection sensitivities are maximised, but with loss of spectroscopic information.


next up previous
Next: Back Illumination Up: Detection Efficiency Previous: Deep Depletion

Dave Lumb
Fri Aug 30 07:43:04 MET DST 1996