MIT/Lincoln Labs CCID20 W12C1 CCD Test Results

This is a standard resistivity phase 1 MIT/Lincoln Labs 2Kx4K CCD. The backside processing was completed before the new, improved processes were developed.

A summary report is available. Noteworthy points about this CCD include:

• This was a test CCD. It has a short that creates a very "hot" pixel.

Original postscript files are available from our anonymous ftp server and these provide better resolution and clarity than is usually possible on a web page. Here are a few figures to illustrate device highlights:

Dark

This image shows what happens with the shorted pixel in a long dark integration.

Brick wall

The CCD was given a UV flood treatment to see if the QE variations could be reduced. Partial success was achieved. However, the improvement we found turns out to have little to do with the UV flood. Instead, it was the backout of the CCD which we did before the UV flood that seems to make a difference. See the final note.

This first graph compares the overall QE with UV flood and without UV flood. Clearly some benefit is achieved. This improvement is achieved reducing the QE non-uniformity.

The following images compare the effects of UV flood on the QE variations. The images are 256x256 pixels. The image on the left was obtained without UV flood and the image on the right was obtained after a UV flood treatment. Two components of the QE variation are seen in these images. The darkest areas we identify as the picture frame pattern. The more rectangular, grey areas we identify as the brick wall pattern. The effect of the UV flood is to improve the QE in the picture frame.

The following plot compares the QE variation measure (MAX-MIN)/MEAN with UV flood and without UV flood. The picture frame and brick wall components of the QE variations are evaluated. The graph shows that the brick wall component undergoes very little change with UV flood. The picture frame component is improved by at least a factor of two. The image above illustrates that this improvement occurs by increasing the QE in the least sensitive areas of the CCD.

A subsequent series of tests showed that it was probably the heating of the CCD (to about 50^oC in dry oxygen over night) that produced the improvement in the brick wall pattern. The UV flood seems to play no role. The improvement persists after the CCD is warmed up. We suspect that removing moisture from the CCD surface is the important step. Probably any CCD with the Phase 1 brick wall pattern can be improved with a mild bake-out.

Back to Lincoln Labs Top Page