QE effected by operating temperature?

[Sutherland, Scott]

Posted to CCD-world:
-+-+-+-
Hello all:

I was recently on a web page of a company called
Andor, a manufacturer of spectroscopic components,
among which are CCD cameras.  On their
web page they have a chart of wavelength dependent
quantum efficiencies as a function of the operating
temperature of the CCD.

See:  http://www.andor-tech.com/relqe.html

Now I have read that the transfer efficiency of charges
to the readout register are adversely affected by
very low temperatures (one reason for not cooling 
below LN2), and that readout noise can decrease
at low temperatures, and of course that dark
current is lower at lower temperatures,
but I was unaware of the dramatic effect temperature
has on QE.

Is the curve at the above web site typical of most CCD's?

What causes such a dramatic wavelength dependent 
QE versus temperature?

We work in the 850-1050 nm range, where the effect
is significant.  I had always considered the use of
lower temperatures (e.g. multi-stage TE coolers,
water circulation, etc.) as a trade off between cost
and complexity versus lower dark current.  However,
with this graph, I see that a significant trade off
has to be made between QE and dark current when
determining the operating temperature of the CCD.

Any additional info on this effect would be greatly appreciated.

Dr. Scott Sutherland
Senior Scientist
GAMMA-METRICS
-+-+-+-
For information about CCD-world, send email to owner-CCD-world at cfht.hawaii.edu.