CMOS Image Sensors

An image sensor is a device that converts a visual image to an electric signal, which is used mostly in digital cameras and other imaging devices. It is a set of charge-coupled device (CCD) or active-pixel sensors (APS) like CMOS sensors. The pixel responds to the light by accumulating charges - the more light, the more charge. Thereby the optical signal is transformed into an electric signal, which gets converted into digial information. Compared to CCD, CMOS sensors are faster, smaller, and cheaper as they are more integrated, which makes them also more power efficient.

For the manufacture of back-side illuminated image sensors, the CIS device wafer can be stacked onto a carrier wafer by the direct oxide bonding method, which depends on molecular forces that naturally attract both surfaces together when they are very smooth and flat. This process requires extremely smooth (RMS < 0.5 nm) and clean surfaces. When such SiO 2 surfaces are placed in contact, they initially form relatively weak van-der-Waals bonds. Subsequent heating to high temperatures (> 1000 ºC) is necessary to achieve high bond strength through the formation of covalent bonds. While direct oxide bonding is normally carried out at elevated temperatures to achieve a strong, robust bond interface, the higher temperatures required are not compatible with processed CMOS wafers.

Modifying the surface chemistry by plasma activation allows the formation of chemical bonds at significantly lower annealing temperatures (200 ºC - 400 ºC). The surface energies after being treated with various plasma activation conditions are 2 - 3 times higher than that without treatment. Optically aligned silicon direct bonding can be performed, with high throughput, within the SmartView® alignment system or GEMINI®FB.

Please see our related product GEMINI®FB for detailed information.


Direct oxide bonding. Source EVG.