The shape and distribution of dislocation defects in semiconductor wafer preparation have a significant impact on the performance of electronic components. Due to differences in doping materials and preparation processes, the distribution of dislocations also varies. This device is used to inspect the morphology and distribution of wafer dislocations, providing data support for wafer material research and improving preparation processes. Suitable for 2-inch, 3-inch, 4-inch, and 6-inch gallium arsenide substrates.
Project |
Performance index |
Testing items |
Detect whether there are dislocation defects on the substrate through machine vision inspection technology and count the dislocation defects |
Dislocation characteristics: diamond shaped with long and short axis dislocations, black hexagonal dislocations, etc |
|
Card control requirements |
≤ 5000 dislocations/cm2, which means that the field of view per square centimeter requires less than 5000 dislocations, and more than 5000 dislocations are considered abnormal |
Counting method: (1) Whole piece imaging: comprehensively image the entire substrate, and then calculate the number of dislocations point by point; (2) 69 point imaging: Use a camera to image 69 1 millimeter long and wide points, calculate the number of dislocations, and then calculate the average and maximum single point values; |
|
Dislocation identification |
The dislocation features are diamond shaped with long and short axes or dark black similar to hexagons, and those without such features are not included in the count; In addition, dislocations of similar size are counted. When detecting large area dislocations, a key pop-up prompt will appear. |
Optical accuracy |
1um/Pixel |
Counting error |
≤± 3% (excluding sampling error) |
As the sample size increases, the counting error will gradually decrease |
|
Detection speed |
4-inch substrate 69 o'clock time: ≤ 50 seconds |
4-inch substrate whole piece time: ≤ 10 minutes |