Abstract

This paper presents a complementary metal-oxide-semiconductor (CMOS) image sensor (CIS) capable of capturing UV-selective and visible light images simultaneously by a single exposure and without employing optical filters, suitable for applications that require simultaneous UV and visible light imaging, or UV imaging in variable light environment. The developed CIS is composed by high and low UV sensitivity pixel types, arranged alternately in a checker pattern. Both pixel types were designed to have matching sensitivities for non-UV light. The UV-selective image is captured by extracting the differential spectral response between adjacent pixels, while the visible light image is captured simultaneously by the low UV sensitivity pixels. Also, to achieve high conversion gain and wide dynamic range simultaneously, the lateral overflow integration capacitor (LOFIC) technology was introduced in both pixel types. The developed CIS has a pixel pitch of 5.6 µm and exhibits 172 µV/e− conversion gain, 131 ke− full well capacity (FWC), and 92.3 dB dynamic range. The spectral sensitivity ranges of the high and low UV sensitivity pixels are of 200–750 nm and 390–750 nm, respectively. The resulting sensitivity range after the differential spectral response extraction is of 200–480 nm. This paper presents details regarding the CIS pixels structures, doping profiles, device simulations, and the measurement results for photoelectric response and spectral sensitivity for both pixel types. Also, sample images of UV-selective and visible spectral imaging using the developed CIS are presented.

Highlights

  • The rise of applications for new technologies such as artificial intelligence (AI), machine vision, and Internet-of-Things (IoT) has led to an increase on image sensor performance demands, for better data acquisition and analysis [1,2,3,4,5]

  • The technology was further developed for pinned photodiode with full charge transfer capability, suitable for CIS implementation, and we report an optical filter-less CIS outputting UV-selective and visible light images simultaneously in a single exposure by employing the differential spectral response method

  • We developed a 648H × 488V pixels (640H × 480V effective) CIS with high and low UV light sensitivity pixels arranged in a checker pattern

Read more

Summary

Introduction

The rise of applications for new technologies such as artificial intelligence (AI), machine vision, and Internet-of-Things (IoT) has led to an increase on image sensor performance demands, for better data acquisition and analysis [1,2,3,4,5]. Previous works achieved color RGB information for each pixel without demosaicing by stacking vertically B-, G-, and R-sensitive organic photoconductive films [24], or by using silicon (Si) with a triple-well structure to stack vertically three layers of pixels, leveraging the relationship between light penetration depth with the wavelength in Si [25] Another approach combines the depth sampling with filter arrays by stacking two CIS with color filters and reflectors, achieving high color fidelity in low light [26]. The technology was further developed for pinned photodiode with full charge transfer capability, suitable for CIS implementation, and we report an optical filter-less CIS outputting UV-selective and visible light images simultaneously in a single exposure by employing the differential spectral response method.

Circuit Architecture
Pixel Structure and Implant Profiles
Device Simulation
Chip Manufacturing and Measurement Results
Findings
Conclusions

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.