Evolution of Pixel Technology in CMOS Image Sensor

By January 29, 2020 February 13th, 2020 No Comments

Picture This

Mobile devices – and their rapidly improving camera capabilities – are a huge part of our everyday lives. With the incredible technology contained in handheld devices today, it’s easier than ever before to capture, record and store imagery of ourselves and the world around us. As common as it might seem to us today, this capability didn’t arrive overnight. It is the result of many years, and decades, of research and development.
And as demand for higher quality images and smaller size devices continues, so does innovation, breakthrough and advancement. Below, we learn more about CMOS image sensor (CIS) technology development and SK hynix’s work with CIS in a contributed piece by Hoon Sang Oh, Fellow of SK hynix’s CIS Business.

Image Quality of CIS and Pixel

Most of the mobile devices we use today, such as cell phones, tablet PCs and laptop computers, are equipped with at least one or more camera sensors. The quality of the images we take on these devices is determined by an electric mechanism in the sensors called “pixel,” a key component which converts light into electric signal.

Among many indicators of the image quality, the most representative one is a definition and measurement procedure called “Image Signal-to-Noise Ratio (SNR)” (Fig. 1). In order to get high image SNR, we need to increase signal term and reduce noise term, which are mostly determined by major pixel performance indices such as full well capacity, sensitivity, pixel dark noise, readout-circuit noise and pixel crosstalk.

Under bright conditions, full well capacity and pixel crosstalk are the major factors in the image SNR, while under dark conditions, many factors, including sensitivity, pixel dark noise, readout-circuit noise and pixel crosstalk, contribute to image SNR. Therefore, technically, it is more difficult to get high image SNR, i.e. better image quality, under dark environment, since we need to control and improve many performance indices of pixel.

Fig. 1. Definition and Measurement procedure of Image SNR

Image Download

History of CIS Pixel Evolution

There have been constant needs for higher resolution sensor over the past decade, and continuous efforts to develop smaller pixel. The evolution history of CIS pixel technology is depicted in Fig. 2, where we see how the technical hurdles for pixel size scaling were overcome. Three phases have led the technical innovations for smaller pixel: (1) SENSITIVITY (2) CROSSTALK and (3) QUAD (or TETRA) PIXEL technology.

In the first phase, pixel engineers concentrated on making up for the sensitivity loss inevitably caused by the reduction of pixel size, developing many innovative technologies including on-chip-lens (or micro-lens), deep photodiode with thicker silicon, and backside illumination technology. As the pixel size reached about one micrometer, the second phase, which focused more on crosstalk reduction, began. During this time, novel technologies such as metal grid structure in color filter layer and deep trench isolation process for Si photodiode were developed in order to suppress the optical and electrical crosstalk, respectively.

As pixel size continued to shrink to sub-micrometer domain, a new concept of pixel operation mode based on QUAD or TETRA pixel structure was proposed as a solution to the sensitivity issue of small pixel under low illumination environment. The basic operation principle of the QUAD pixel is illustrated in Fig. 3. It is expected that all the sub-micrometer sized pixels to be developed in near future will also adopt this QUAD pixel scheme.

Fig 2. Evolution of CIS Pixel Technology

Image Download

Fig 3. QUAD pixel operation principle

Image Download

Image Sensors & SK hynix

SK hynix started its CIS business in 2007 and has produced 8-inch wafer-based image sensor products for more than a decade. With the advent of IoT and 5G-based ICT world, the image sensor market is growing rapidly today, and SK hynix is fostering the CIS business as the next growth engine beyond its memory business in order to meet this soaring demand. After a couple of years of preparation for the 12-inch fab CIS process, SK hynix launched a 12-inch wafer-based 1.0um product in 2019, which adopted its proprietary Black Pearl pixel technology and received good reviews from customers due to the highly competitive pixel performance. Black Pearl boasts improved noise characteristics, implementing clear images with little noise under low illumination conditions.

Looking Ahead

Continuously working to provide top-tier level sensor performance and productivity, SK hynix is committed to taking its efforts to the next level in the coming decade. SK hynix is now expanding its CIS product portfolio spectrum from mid-end to high-end products based on both performance and price competitiveness. Through its investment in the CIS space in 2020 and beyond, the company aims to quickly become one of the world’s leading CIS companies.



ByHoon Sang Oh

Fellow of CIS Business at SK hynix