cryogenic CCD operation

[David Lumb]

Tim Hardy wrote:

> Posted to CCD-world:
> -+-+-+-
> I have heard that CCDs cannot be operated at extremely cold
> temperatures (I'm looking at an application which would operate
> around 30K). Does anyone have any information on this?

Apart from the obvious problems with the package material, you run into a device
physics limit known as "carrier freeze-out" as temperatures approach liquid
nitrogen range (77 K). Basically your electrons won't want to move through the
silicon any more!

Actually that is an oversimplification: In Si, for a impurity concentration of
1e17 /cu cm the carrier concentration drops by an order of magnitude at
70K compared with room temperature. Complete freeze out occurs for T <40K. For
higher concentrations, the impurity energy level can become an energy "band",
eventually for high concentrations the band merges with the conduction band, and a
so-called degenerate semiconductor is formed. This degeneracy is likely to occur
in the source and drain regions of the on-chip MOSFETs. MOSFETs (unlike bipolar Si
devices) have in fact been considered for cryogenic applications.

Probably you may find that the FETs may work but start to get noisy, while the
transfer mechanism still works, but trapping becomes a problem (eg E K Banghart in
IEEE Trans. Electron Devices 38 (1991) )


>What is the failure mechanism? Has anyone tried it? What is the

> cutoff temperature?  Is there any published research on this topic?
>
> Thanks for your help,
>
> Tim
>
> ________________________________________________________________________

--
 David H Lumb
 XMM Instrument and Calibration Scientist
 Astrophysics Divn, SSD, ESTEC
 P.O. Box 299
 NL-2200 AG Noordwijk
 Netherlands
 phone: +31-71-5654446, fax: +31-71-5654690
 email: dlumb at astro.estec.esa.nl,   WWW: http://astro.estec.esa.nl


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.ctio.noao.edu/pipermail/ccd-world/attachments/19981117/6bc350f9/attachment.htm