Light Reflecting CMOS Image Sensor for High Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in manufacturing low-cost CMOS image sensors is to reduce pixel size and increase resolution while maintaining photodiode photon conversion efficiency, as the physical constants of semiconductor materials are mutually contradictory, requiring a compromise in semiconductor layer thickness.
Innovation Solution
Incorporating light-reflecting means in the dielectric layer opposite the photodiodes, which increases photon absorption and conversion efficiency, allowing for a reduction in semiconductor layer thickness without altering the existing structure, thereby minimizing electronic crosstalk and enabling smaller pixel widths and higher pixel density.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the semiconductor layer thickness is increased to maximize photon absorption, then the photodiode photon conversion efficiency is improved, but the pixel width cannot be reduced below approximately 2 μm due to lateral scattering of electrons
Solution Approach 1:
The patent introduces a light-reflecting means in the front-side dielectric layer to reflect oblique-incidence photons back into the semiconductor layer, effectively utilizing the vertical dimension to increase photon absorption path length without increasing lateral pixel dimensions. This allows maintaining thin semiconductor layers while improving photon conversion efficiency.
Solution Approach 2:
The patent converts the harmful effect of oblique-incidence photons that would otherwise be lost or cause lateral scattering into a beneficial effect by using light-reflecting means to redirect these photons back into the photodiode, thereby improving photon absorption efficiency without requiring larger pixel areas.
2Productivity
If the pixel width is reduced to increase the number of pixels on a semiconductor wafer, then the manufacturing cost is reduced and resolution is improved, but the effects of semiconductor thickness such as absorption of oblique-incidence photons and lateral scattering prevent this reduction
Solution Approach 1:
By adding light-reflecting means in the front-side dielectric layer, the patent extends the photon absorption path in the vertical dimension, allowing thinner semiconductor layers with smaller pixels to achieve the same or better photon absorption efficiency as thicker layers with larger pixels.
Solution Approach 2:
The patent changes the optical parameters of the front-side dielectric layer by introducing light-reflecting means, thereby altering the light propagation characteristics to enhance photon absorption in smaller, thinner pixels without sacrificing efficiency.
3Object-generated harmful factors
If the semiconductor layer thickness is reduced to minimize electronic crosstalk, then the collection of charges generated by photon absorption is improved, but the absorption of oblique-incidence photons is insufficient
Solution Approach 1:
The patent converts potentially lost oblique-incidence photons into useful absorbed energy by using light-reflecting means to redirect them back into the semiconductor layer, thereby maintaining efficient photon absorption even in thin semiconductor layers that minimize electronic crosstalk.
Solution Approach 2:
The light-reflecting means in the front-side dielectric layer acts as an intermediary that redirects oblique-incidence photons back into the semiconductor layer, enabling efficient energy absorption in thin layers without direct contact between the photons and the thick semiconductor material.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances photon absorption and charge collection, reduces electronic crosstalk, and improves modulation transfer function and color separation, enabling more pixels per surface area without disrupting sensor operation.
Implementation Method 1
light-reflecting means arranged in the dielectric layer, e.g., opposite the photodiodes, and capable of reflecting at least a portion of the light received by the sensor towards the photodiodes, or towards the semiconductor layer
Implementation Method 2
CMOS-type photodiodes and transistors produced in a layer containing at least one semiconductor... capable of reflecting at least a portion of the light received by the sensor towards the photodiodes
Implementation Method 3
photodiodes... conversion efficiency... photon absorption-generated charges
Data Source
AI summary
An image sensor comprising at least:CMOS-type photodiodes and transistors produced in a semiconductor layer having a thickness of between approximately 1 μm and 1.5 μm,a dielectric layer in which electrical interconnect layers are made, which are electrically connected to one another and/or to the CMOS photodiodes and/or transistors, said dielectric layer being arranged against a first face of the semiconductor layer opposite a second face of the semiconductor layer through which the light received by the sensor from the exterior is intended to enter,light-reflecting means arranged in the dielectric layer, opposite the photodiodes, and capable of reflecting at least a portion of the light received by the sensor towards the photodiodes.


