LBNL CCD Packaging Program
Latest results: June 3, 2003
May 8, 2003 - A 300 micron thick silicon wafer was diced
to the size of a 2Kx4K CCD and glued to a piece of aluminum nitride. The assembled
package was measured for flatness at room temperature and while cooled in
our lab dewar.
There was little difference between the room temperature measurment and
the cold temperature measurement.
Here are the cold results:
Seen end-on it looks like this:
We used an LBNL CCD, working at 1.05 microns wavelength, to image the glueing
operation. At this wavelength we could "see" through the silicon wafer and
watch the glue wick into the gap between the silicon and the AlN. The mpeg
video of the glueing is
here.
May 9, 2003
We placed the same silicon/AlN package on a vacuum chuck with the AlN facing
the vacuum chuck, applied the vacuum, and remeasured the flatness. (Obviously
this is now out of the vacuum dewar!)
Here is the result:
The warp in the silicon is about 20 microns peak-to-peak when held in the
vacuum chuck. This suggests that the AlN is also warped and by about the same
amount as the silicon and that the AlN is pulled flat by the vacuum chuck.
This results in the doubling of the warp in the silicon.
When we release the vacuum on the vacuum chuck and measure flatness of the
silicon again we get this result:
The package relaxes to its original shape and the warp in the silicon returns
to about 10 microns peak-to-peak.
This is good news for the remaining packaging steps. The next step is go
glue a thick AlN mounting foot to the AlN side of our package. To do this
we will be holding the silicon/AlN package in a vacuum chuck, but with the
silicon facing the chuck. So we expect the silicon surface will be pulled
flat by the vacuum chuck. Our prediction is that it will remain flat, or nearly
so, when the vacuum is released because the foot is thick compared to the
Si/AlN package and the foot covers a large fraction of the surface of the
Si/AlN package.
June 3, 2003
Here is a photo of our glueing camera as assembled in our cleanroom:
The camera is looking at the silicon (30mm X 60mm piece about 250um thick)
and the image of the silicon can be seen on the monitor. We placed pieces
of paper beneith the silicon at top and bottom. They can be seen through
the silicon. The illumination comes from a small lamp which can be seen peaking
out from below the metal plate attached to the camera mounting. The metal
plate, which extends forward from the mounting, is actually flat black on
the bottom. The camera is tilted slightly so the image reflected from the
front surface of the silicon is the black face of that plate.