CONTRACT MANUFACTURING
2. Row- and column-based positioning
For users without a CAD drawing of the wafer layout, some image analysis programs offer a multiposition and repetitive measurement workfl ow. Figure 3 shows an example from OLYMPUS Stream software specifying multiple sample regions. First, the wafer layout (chip offset in x and y) as well as the alignment can be defi ned by three given chip positions. Users must pick up the repetitive structures (blue dots in Figure 3) in different chips, such as (3,3), (3,8), and (7,8).
The exact coordinates of each chip will be
defi ned based on this coordinate system. Then, users must defi ne a chip list and points of interest (POIs) per chip. This includes determining localisation and objective lenses for acquisition per POI (10X, 20X, 50X, etc.) and choosing whether to apply autofocus at each location. The settings will be accessed for all subsequent automatic movements to different chip positions.
Defect analysis on blank wafers Checking blank wafers for impurities is another important inspection. This testing can be challenging because the microscope must acquire - and the software must process - a large number of images. Fast multi position image acquisition software capabilities benefi t this inspection. To detect particles, some image analysis software enables users to set up a phase threshold to identify impurities. Other particle restrictions - such as shape and size limitations - can be set up as well. Phase analysis, particle counts, and size distributions can also be performed simultaneously with impurity detection.
Figure 5: Particle detection on a stitched image of an (exaggeratedly) contaminated wafer sample (left). A magnifi ed section of the image with a tabular display of the detected particles is shown on the right
Figure 4: A snapshot image of a blank wafer sample showing impurities in OLYMPUS Stream software. Imaged with an Olympus MX microscope and Olympus monochrome camera
Figure 4 shows an example of image analysis software detecting impurities on a contaminated sample and then displaying them in an Excel spreadsheet. Another reason wafer inspection can be slow is the sheer number of images that need to be captured. Capturing the images separately, processing them one after another, and saving them is slow and ineffi cient. Each step takes time, image storage takes up a lot of space, and human effort is involved at each step. In addition, under certain circumstances (e.g., repetitive measurements), the images all tend to look the same. Users are only interested in the images that contain relevant data, so there is rarely a need to save all of them. Modern image analysis software can combine multi-image-alignment acquisition with live image processing (phase analysis)
for impurity detection on the resulting stitched image. A typical workfl ow consists of: loading a wafer sample, performing the stitching acquisition process, setting up a threshold, and clicking the Particle Detection button. Additionally, users can defi ne some pass criteria, so the software can judge which particles to consider.
The table in Figure 5 shows the results of live particle detection on the stitched image. If your microscope is equipped with a motorised stage, clicking on a particle in the image or clicking on the results in the table will drive the stage to the particle location on the sample so that you can confi rm what the software detected. To facilitate the search for large particles, each table column can be sorted.
The current semiconductor chip shortage has caused manufacturers to explore ways to speed up all their processes, including quality control. This article demonstrates that choosing the right combination of microscopy hardware and image analysis software is important for making wafer inspection as effi cient as possible. Modern features, such as easy-to-use multiposition imaging and real-time analysis, greatly simplify and speed up the inspection process while providing repeatability and accuracy. Tailoring the analysis software to specifi c wafer navigation and measurement needs can further streamline inspections.
Olympus Cooperation
www.olympus-ims.com
34 MARCH 2022 | ELECTRONICS TODAY
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46
