UV flood recipe

[Michael P. Lesser]

Hi everyone,

I have received a few comments about UV flooding recipes for Steward
CCD Laboratory devices, so I thought I would pass along some experiments
I have done in the last year or so.  This applies only to devices we have
backside treated here in the Lab.

You should follow the recipe in my home page first, 
ftp://frodo.as.arizona.edu/pub/ccdlab.html.  If that does not work,
please contact me as I would like to know what you are doing differently
or what subtle variations are important. 

I have tested the following sequences....

1) flooding in air and O2, with no difference.  I use laboratory air dried
by passing through a desiccant (not the desert!).  Untreated lab air (50%
humidty) also works, but is not recommended because it may not give
reproducible results.

2) flooding for 5 to 120 minutes in 5 minute increments.  No difference.  Full
QE (averaged over the entire CCD) is obtained in about 3 minutes, but the most
uniform flat fields take longer.  I use 15 minutes to be sure.

3) using Cd and Zn lamps.  No difference.

4) Flooding cold (from -20 to -100C).  Cold floods leave a high residual dark
current.  In general they are poor because of the dark current and because they
are done in vacuum.

5) Flooding in vacuum.  Good floods are NOT obtained.  Some oxygen is required.

6) Pumping with a dirty vacuum pump or in a dirty dewar.  Bad results,
the system must be clean to avoid contamination on the CCD surface.

7) Our latest results indicate Hg lamps do not flood very well, shorter
wavelengths are needed.

8) After many hours in vacuum (between flooding and cooling), the charge does
decay.  In air, the charge lasts for days.  If you have a slow vacuum system and
need to pump the dewar for hours, you can leave the flooding lamp on.  Flooding
in vacuum does not hurt, but alone is not enough to fully charge the surface.

9) Hydrogen will kill the charge on a CCD in seconds!

-Mike