Specific Process Knowledge/Characterization/KLA-Tencor Surfscan 6420/Acceptance test results

From LabAdviser
Jump to navigation Jump to search

Calibration wafers

DTU Nanolab has two (earlier tree, but one has disappeared) calibration wafers for the KLA-Tencor Surfscan. The calibration wafers are 6" silicon wafers with a number of polystyrene latex spheres (PLS) uniformily spread over the surface.


Certified mean diameter and number PLS's on each wafer:

Calibration wafer 1:

  • PLS mean diameter: 0.204 µm +/- 0.008 µm
  • Number of PLS's: 7234 +/- 10%

Calibration wafer 2:

  • PLS mean diameter: 1.112 µm +/- 0.018 µm
  • Number of PLS's: 7700 +/- 10%

Calibration wafer 3:

  • PLS mean diameter: 4.000 µm +/- 0.043 µm
  • Number of PLS's: 7587 +/- 10%

Repeatability check

Acceptance criterion:

10 scans are done on each of the three PLS calibration wafers. The wafers are loaded/unloaded to the cassette every time. The repeatbility of +/- 1σ has to be less than 1%.

Results

Calibration wafer 1:

  • Measured average PLS diameter: 0.2063 µm. Repeatability: 0,24%
  • Average number of PLS's: 7065. Repeatability: 0.11%

Calibration wafer 2:

  • Measured average PLS diameter: 1.100 µm. Repeatability: 0%
  • Average number of PLS's: 7427. Repeatability: 0.20%

Calibration wafer 3:

  • Measured average PLS diameter: 3.970 µm. Repeatability: 0,24%
  • Average number of PLS's: 7101. Repeatability: 0.10%


Uniformity check

Acceptance criterion:

Measure the average PLS diameter by boxing an area in five different locations on a PLS wafer. The measurement has to be within 3%.

Results

Calibration wafer 2:

  • Non-uniformity: 1,36%

Calibration wafer 3:

  • Non-uniformity: 0.38%


Handling test

Acceptance criterion:

Run a cassette of 25 wafers. The machine should be able to error-free handle the cassette 40 times, i.e. it should be able to handle 1000 wafers without failing.

Results

6" cassette":

A cassette of 25 wafers was handled 40 times over night, i.e. the acceptance criterion was meet.

4" and 8" cassette:

A cassette of 25 wafers was handled two times. The scanning was then stopped to save time as handling of more cassettes is not expected to cause problems.