Improvement of Optical Efficiency and Crosstalk with A Trench Color Filter Scheme

碩士 === 國立交通大學 === 工學院半導體材料與製程設備學程 === 106 === CMOS image sensors (CIS) offer the advantages of lower cost, lowe power consumption, portability and system on chip (SoC) characteristics than Charge Coupled Device (CCD). In the past few years, CIS has been widely applied to consumer products, security...

Full description

Bibliographic Details
Main Authors: Lin, Che-Hung, 林哲宏
Other Authors: Pan, Fu-Ming
Format: Others
Language:zh-TW
Published: 2018
Online Access:http://ndltd.ncl.edu.tw/handle/snq97j
id ndltd-TW-106NCTU5686005
record_format oai_dc
spelling ndltd-TW-106NCTU56860052019-05-16T00:22:51Z http://ndltd.ncl.edu.tw/handle/snq97j Improvement of Optical Efficiency and Crosstalk with A Trench Color Filter Scheme CMOS影像感測器光學效率與干涉改善之研究 Lin, Che-Hung 林哲宏 碩士 國立交通大學 工學院半導體材料與製程設備學程 106 CMOS image sensors (CIS) offer the advantages of lower cost, lowe power consumption, portability and system on chip (SoC) characteristics than Charge Coupled Device (CCD). In the past few years, CIS has been widely applied to consumer products, security cameras, automotive and medical systems, Virtual Reality (VR) and mobile phones.Because of the strong demand for a smaller chip size and a higher resolution. To obtain a higher resolution at the same sensor size, the pixel size is needed to be shrunk. However, the scaling-down has two side effects: first, the amount of the incident light incident on each photodiode decreases, leading to a lower optical efficiency. Second, The optical crosstalk becomes more serious. In this thesis, we proposed an embeded trenched color filter structure to improve the optical efficiency and crosstalk. On the basis of the Snell’s law, we optimized the refractive index difference between the color filter and the surrounding dielectric films so that total internal reflection (TIR) of the incident light occurs, suppressing the optical crosstalk and increasing the optical efficiency. The Finite-Difference-Time-Domain (FDTD) software was used to simulate the optical efficiency and crosstalk in three filter structures: conventional, etched cavity and the embeded trenched structures to predict the performance of the image sensors. We then fabricated CIS sensors with the three structures at 0.11μm scale and measured the optical and crosstalk properties. The study shows that the embeded trenched structure outperforms the other two structures interms of the optical efficiency and the crosstalk. Compared with the conventional structure, the embedded trenched structure increases the optical efficiency by 35-39% and reduces the crosstalk by 0.6-0.7%. The embeded trenched structure has a higher optical efficiency that the etched cavity structure by 6-14% while is with a comparable crosstalk with the cavity one. Pan, Fu-Ming 潘扶民 2018 學位論文 ; thesis 96 zh-TW
collection NDLTD
language zh-TW
format Others
sources NDLTD
description 碩士 === 國立交通大學 === 工學院半導體材料與製程設備學程 === 106 === CMOS image sensors (CIS) offer the advantages of lower cost, lowe power consumption, portability and system on chip (SoC) characteristics than Charge Coupled Device (CCD). In the past few years, CIS has been widely applied to consumer products, security cameras, automotive and medical systems, Virtual Reality (VR) and mobile phones.Because of the strong demand for a smaller chip size and a higher resolution. To obtain a higher resolution at the same sensor size, the pixel size is needed to be shrunk. However, the scaling-down has two side effects: first, the amount of the incident light incident on each photodiode decreases, leading to a lower optical efficiency. Second, The optical crosstalk becomes more serious. In this thesis, we proposed an embeded trenched color filter structure to improve the optical efficiency and crosstalk. On the basis of the Snell’s law, we optimized the refractive index difference between the color filter and the surrounding dielectric films so that total internal reflection (TIR) of the incident light occurs, suppressing the optical crosstalk and increasing the optical efficiency. The Finite-Difference-Time-Domain (FDTD) software was used to simulate the optical efficiency and crosstalk in three filter structures: conventional, etched cavity and the embeded trenched structures to predict the performance of the image sensors. We then fabricated CIS sensors with the three structures at 0.11μm scale and measured the optical and crosstalk properties. The study shows that the embeded trenched structure outperforms the other two structures interms of the optical efficiency and the crosstalk. Compared with the conventional structure, the embedded trenched structure increases the optical efficiency by 35-39% and reduces the crosstalk by 0.6-0.7%. The embeded trenched structure has a higher optical efficiency that the etched cavity structure by 6-14% while is with a comparable crosstalk with the cavity one.
author2 Pan, Fu-Ming
author_facet Pan, Fu-Ming
Lin, Che-Hung
林哲宏
author Lin, Che-Hung
林哲宏
spellingShingle Lin, Che-Hung
林哲宏
Improvement of Optical Efficiency and Crosstalk with A Trench Color Filter Scheme
author_sort Lin, Che-Hung
title Improvement of Optical Efficiency and Crosstalk with A Trench Color Filter Scheme
title_short Improvement of Optical Efficiency and Crosstalk with A Trench Color Filter Scheme
title_full Improvement of Optical Efficiency and Crosstalk with A Trench Color Filter Scheme
title_fullStr Improvement of Optical Efficiency and Crosstalk with A Trench Color Filter Scheme
title_full_unstemmed Improvement of Optical Efficiency and Crosstalk with A Trench Color Filter Scheme
title_sort improvement of optical efficiency and crosstalk with a trench color filter scheme
publishDate 2018
url http://ndltd.ncl.edu.tw/handle/snq97j
work_keys_str_mv AT linchehung improvementofopticalefficiencyandcrosstalkwithatrenchcolorfilterscheme
AT línzhéhóng improvementofopticalefficiencyandcrosstalkwithatrenchcolorfilterscheme
AT linchehung cmosyǐngxiànggǎncèqìguāngxuéxiàolǜyǔgànshègǎishànzhīyánjiū
AT línzhéhóng cmosyǐngxiànggǎncèqìguāngxuéxiàolǜyǔgànshègǎishànzhīyánjiū
_version_ 1719164364365234176