Silicon Block Inspection with SWIR Imaging

The video below demonstrates light transmission from behind the silicon block imaged with an InGaAs camera.  The cracks and material inhomogeneities of this scrap block are clearly visible; a computer can easily map the location of the defects. Take note of the almost complete blocking of light transmission on the right end of the block. The lack of light identifies the top of the original casting, as carbon floats to the top of the molten silicon. 

{enclose Video-Flashlight-Silicon-block.flv}

The three photos in Figure 3 show how InGaAs detectors can image through the entire width of a silicon block before cutting it into wafers. The block in the image measures 156 x 156 x 254 mm (6.1 x 6.1 x 10"). It is completely solid and weighs 40 pounds. Despite that, the InGaAs camera is able to resolve detail in the small Air Force resolution chart taped to the back of the block.  The illumination comes from an incandescent flashlight positioned to project collimated light, creating a shadow pattern imaged by the camera through the block.  alt

In this image, one side of the block (the side facing the camera) is polished. Unpolished, the light rays would be scattered in many directions, obscuring the detail. With careful management of illumination and imaging, however, and by the use of InGaAs linescan cameras, coordinated with line illumination sources to map the defects within the block, the image processing software tells the diamond saws which parts to cut into wafers and which parts to recycle.


InGaAs detectors are also used to inspect blocks and wafers to gauge material quality. Using photoluminescence (PL), short-wavelength laser power irradiates the silicon and causes bright glow at the bandgap wavelength. Producers map good solar cell material with this technique. However, certain defects resulting from dislocation within the crystalline structure can actively quench the photovoltaic effect of the solar cell. The photoluminescence process will also reveal their locations by weakly emitting in the 1300 to 1600 nm range. As a result of Goodrich process improvements that reduce dark current and system-read noise in its imaging arrays, this very weak glow is now visible to their InGaAs cameras running at video frame rates.

Figure 4, for example, shows several areas of defect locations in a finished wafer. As PL is a contact-less form of inspection, it evaluates raw silicon blocks and freshly cut wafers for problems before proceeding with expensive processing. This is a clear advantage over electroluminescent testing of completed solar cells, though that remains a powerful inspection tool for monitoring product quality on the cell fabrication line, and for panel assemblers. alt

Other Applications

Beyond the solar industry, PL, at SWIR wavelengths, captures from outside of living small animals the weak glow from nano-tube structures tagged to tumor cells inside their internal organs. Other imaging applications that have been addressed with InGaAs technology include inspection of thin-film and triple-junction photovoltaic cells with both PL and EL techniques, hot hollow glassware inspection with direct imaging, and NIR spectroscopic-based sorting of agricultural, pharmaceutical, and recyclable products.  

Improved InGaAs-SWIR imaging has also benefited biomedical imaging with optical coherence tomography (OCT). This method detects disease in the eye, arteries, or organs, either with spectral-domain or full-field approaches. Currently developing is the use of SWIR transillumination to detect tooth decay and demineralization of tooth enamel. The sensor technology has also benefited military and homeland defense in areas such as covert surveillance, laser spotting, and hyperspectral imaging for camouflage and IED detection. Even the art world benefits from this technology; infrared reflectography techniques image paintings, looking under the paint layers of an artwork to see the artist’s original intent or detect counterfeits, or to see through browned varnish to see potential detail to be recovered in the restoration process.

This article was written by Doug Malchow, Manager, Business Development, Industrial Products for Sensors Unlimited-Goodrich ISR Systems. For more information, visit http://info.hotims.com/34461-142.