A simulation method for charge transfer efficiency of multiple pulse readouts of pixel units in CMOS image sensors

By simulating the charge transfer process of photodiodes in the CMOS image sensor, the problem of incomplete charge transfer is solved, and the analysis of charge transfer efficiency and potential distribution is realized, which improves the performance of the image sensor.

CN114912278BActive Publication Date: 2024-12-06XINJIANG TECH INST OF PHYSICS & CHEM CHINESE ACAD OF SCI
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202210534852.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-05-17
Publication Date
2024-12-06
Estimated Expiration
2042-05-17

AI Technical Summary

Technical Problem

In CMOS image sensors, the potential distribution of the clamped photodiode is not ideal, resulting in incomplete charge transmission, affecting the linear responsiveness and quality of the image. It is difficult for the prior art to analyze the potential distribution in the transmission path through experimental measurements, and the restriction mechanism is not fully explained.

Method used

Reset the clamp photodiode before the exposure stage of the pixel unit of the CMOS image sensor, set the transmission gate voltage to 0V, simulate the collimated light source incident and light injection process, extract the number of photogenerated carriers, and at the floating diffusion node Set the high potential and pulse voltage until the number of remaining carriers does not change anymore, and calculate the charge transfer efficiency.

Benefits of technology

Through this method, the charge transfer efficiency can be accurately obtained, and the microscopic potential distribution during the charge transfer process can be visually displayed, the charge transfer limiting mechanism can be analyzed, and the performance design of the image sensor is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114912278B_ABST
    Figure CN114912278B_ABST
Patent Text Reader

Abstract

The present invention relates to a simulation method for multiple pulse reading charge transfer efficiency of a pixel unit of a CMOS image sensor. The method resets a clamped photodiode before the exposure phase of the pixel unit of the CMOS image sensor begins, sets the gate voltage of the transmission gate to 0V, and puts the photodiode in an accumulation state. Then, a collimated light source is used for incident in a simulation environment, and a light injection model is called to simulate the light generation process. The number of photogenerated carriers N1 in the clamped photodiode at this time is extracted, and a floating diffusion node is set to a high potential and maintained. At the same time, a pulse voltage is set on the transmission gate, and the number of remaining photogenerated carriers N2 in the clamped photodiode after each transmission gate is turned on is extracted. When the number of remaining photogenerated carriers N2 in the clamped photodiode no longer changes with time, the charge transfer efficiency of the pixel unit of the CMOS image sensor is calculated using the N2 value. The present invention can obtain the charge transfer efficiency, and can also intuitively see the microscopic potential distribution in the charge transfer process, thereby analyzing and obtaining the limiting mechanism of the charge transfer.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of image sensor pixel unit simulation, and in particular to a method for simulating charge transfer efficiency of multiple pulse readings of a CMOS image sensor pixel unit. Background Art

[0002] In CMOS image sensors, charge transfer efficiency is a key performance parameter that describes the charge transfer process. Ideally, all photogenerated electrons accumulated in the clamped photodiode should be transferred and read out. Incomplete charge transfer will cause image lag. Previous studies have focused on the degradation of charge transfer efficiency after irradiation of charge-coupled devices (CCDs). Since the readout of charge in charge-coupled devices (CCDs) requires transmission in multiple pixels, the charge transfer efficiency has a great impact on the readout information. Compared with charge-coupled devices (CCDs), 4TCMOS image sensors have added floating diffusion nodes, which separate the charge collection unit clamped photodiode and the readout unit floating diffusion nodes, greatly improving the charge transfer efficiency. However, with the demand for high sensitivity, the photosensitive area increases, making it difficult for the internal potential distribution of the clamped photodiode to achieve complete charge transfer, and it is impossible to transfer all electrons to the floating diffusion node, resulting in a decrease in the linear response of the CMOS image sensor and a deterioration in image quality.

[0003] Research on the charge transfer process of 4T CMOS image sensor pixel units is still in the development stage. Since the non-ideal potential distribution in the transmission path in actual devices is difficult to analyze through experimental measurement, the limiting mechanism of charge transfer has not been fully elucidated and requires in-depth analysis through simulation. Summary of the invention

[0004] The purpose of the present invention is to provide a simulation method for multiple pulse reading charge transfer efficiency of a pixel unit of a CMOS image sensor. The method resets the clamped photodiode before the exposure phase of the pixel unit of the CMOS image sensor begins, sets the gate voltage of the transmission gate to 0V, and puts the photodiode in an accumulation state. Then, a collimated light source is used for incident in a simulation environment, and a light injection model is called to simulate the light generation process. The number of photogenerated carriers N1 in the clamped photodiode at this time is extracted, and the floating diffusion node is set to a high potential and maintained. At the same time, a pulse voltage is set on the transmission gate, and the number of remaining photogenerated carriers N2 in the clamped photodiode after each transmission gate is turned on is extracted. When the number of remaining photogenerated carriers N2 in the clamped photodiode no longer changes, the charge transfer efficiency of the pixel unit of the CMOS image sensor is calculated using the N2 value. The method described in the present invention can obtain the correct charge transfer efficiency, and can also intuitively see the microscopic potential distribution in the charge transfer process, thereby analyzing the limiting mechanism of charge transfer.

[0005] The simulation method of the charge transfer efficiency of a CMOS image sensor pixel unit for multiple pulse readings according to the present invention is performed in the following steps:

[0006] a. Before the exposure phase of the pixel unit of the CMOS image sensor begins, the clamped photodiode is reset and the transmission gate voltage is set to 0V, so that the photodiode is in the accumulation state;

[0007] b. Using a collimated light source in the TCAD software simulation environment, call the Ray Tracing light injection model in the TCAD software to simulate the light generation process and extract the number N1 of photogenerated carriers in the clamped photodiode at this time;

[0008] c. Set the floating diffusion node to a high potential and maintain it, and set a pulse voltage on the transmission gate at the same time to extract the number N2 of remaining photogenerated carriers in the clamped photodiode after each transmission gate is turned on;

[0009] d. When the number N2 of remaining photogenerated carriers in the clamped photodiode no longer changes with time, the charge transfer efficiency CTE of the pixel unit of the CMOS image sensor is calculated using the value of N2, that is, CTE=1-N2 / N1×100%.

[0010] The present invention provides a method for simulating the charge transfer efficiency of a CMOS image sensor pixel unit by multiple pulse readings, wherein:

[0011] a. Before the exposure phase of the pixel unit of the CMOS image sensor begins, the clamped photodiode is reset and the transmission gate voltage is set to 0V, so that the photodiode is in the accumulation state;

[0012] b. In the simulation environment, a collimated light source is used for incident light, and the light injection model is called to simulate the light generation process, and the number of photogenerated carriers N1 in the clamped photodiode at this time is extracted;

[0013] c. Set the floating diffusion node to a high potential and maintain it, and set a pulse voltage on the transmission gate at the same time to extract the number N2 of remaining photogenerated carriers in the clamped photodiode after each transmission gate is turned on.

[0014] d. When the number N2 of the remaining photogenerated carriers in the clamped photodiode no longer changes, the charge transfer efficiency CTE of the pixel unit of the CMOS image sensor is calculated using the value of N2, that is, CTE=1-N2 / N1×100%.

[0015] The simulation environment in step b refers to the simulation environment in the TCAD software. The light injection model calls the Ray Tracing model in the TCAD software. The light generation rate G in the model, that is, the generation rate of electron-hole pairs in silicon, is expressed as:

[0016]

[0017]

[0018] P is the ray intensity factor, which includes reflection, transmission, and damage caused by material absorption on the light path. η0 is the internal quantum efficiency, which indicates the number of carrier pairs generated by each photon, y is the relative distance of the ray, h is Planck's constant, λ is the wavelength, and c is the speed of light. α is the absorption coefficient, and k is the imaginary part of the refractive index of light;

[0019] During the simulation, G is coupled to the Poisson equation and the current continuity equation, and the light intensity and carrier concentration at the corresponding grid points are transiently solved:

[0020]

[0021]

[0022]

[0023] Where ψ is the electrostatic potential, ε is the Si dielectric constant, and ρ is the local space charge density. The local space charge density is the sum of all mobile charges and fixed charges, including electrons, holes, and ionized impurities. n and p are the electron and hole concentrations, q is the unit charge, and J n , J p are the electron and hole current densities, μ n and μ p are the electron and hole mobilities, D n and D p are the electron and hole diffusion coefficients, respectively. ie is the effective intrinsic carrier concentration, T L is the lattice temperature.

[0024] Extract the number N1 of photogenerated carriers in the clamped photodiode at this time;

[0025] c. Set the floating diffusion node to a high potential and maintain it, and set a pulse voltage on the transmission gate at the same time to extract the number N2 of remaining photogenerated carriers in the clamped photodiode after each transmission gate is turned on;

[0026] d. When the number N2 of the remaining photogenerated carriers in the clamped photodiode no longer changes, the charge transfer efficiency CTE of the pixel unit of the CMOS image sensor is calculated using the value of N2, that is, CTE=1-N2 / N1×100%.

[0027] The method for simulating the charge transfer efficiency of a CMOS image sensor pixel unit by multiple pulse readings described in the present invention is applicable to the simulation of the charge transfer efficiency of any type of CMOS image sensor. The present invention can obtain the charge transfer efficiency and can also intuitively see the microscopic potential distribution during the charge transfer process, thereby analyzing the restriction mechanism of the charge transfer.

[0028] Therefore, the present invention is suitable for use by device development units, scientific research institutes and aerospace payload units that need to master the performance of CMOS image sensors. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a graph showing the change in the number of photogenerated carriers in the clamped photodiode over time. DETAILED DESCRIPTION

[0030] Example

[0031] a. Before the exposure phase of the pixel unit of the CMOS image sensor begins, the clamped photodiode is reset and the transmission gate voltage is set to 0V, so that the photodiode is in the accumulation state;

[0032] b. In the TCAD software simulation environment, a collimated light source with a wavelength of 530 nm is used for incident light. The parameters X.ORIGIN and Y.ORIGIN are set to determine the position of the light source. The parameter ANGLE specifies the propagation direction of the light beam relative to the X-axis. The parameters MIN.WINDOW and MAX.WINDOW jointly specify the incident range of the light. In the Atlas module, the light beam will be automatically divided into a series of light rays within the specified incident range according to the geometric shape of the device. After the light is incident, the light injection model (Ray Tracing model) is called to simulate the light generation process through formulas (1)-(5). The light generation rate G in the model, that is, the generation rate of electron-hole pairs in silicon, is expressed as:

[0033]

[0034]

[0035] P is the ray intensity factor, which includes reflection, transmission, and damage caused by material absorption on the light path. η0 is the internal quantum efficiency, which indicates the number of carrier pairs generated by each photon, y is the relative distance of the ray, h is Planck's constant, λ is the wavelength, and c is the speed of light. α is the absorption coefficient, and k is the imaginary part of the refractive index of light;

[0036] During the simulation, G is coupled to the Poisson equation and the current continuity equation, and the light intensity and carrier concentration at the corresponding grid points are transiently solved:

[0037]

[0038]

[0039]

[0040] Where ψ is the electrostatic potential, ε is the Si dielectric constant, and ρ is the local space charge density. The local space charge density is the sum of all mobile charges and fixed charges, including electrons, holes, and ionized impurities, n and p are the electron and hole concentrations, q is the unit charge, and J n , J p are the electron and hole current densities, μ n and μ p are the electron and hole mobilities, D n and D p are the electron and hole diffusion coefficients, n ie is the effective intrinsic carrier concentration, T L is the lattice temperature;

[0041] The number of photogenerated carriers N1 in the clamped photodiode at this time is extracted as 10805;

[0042] c. Set the floating diffusion node to a high potential and maintain it, and set a pulse voltage on the transmission gate at the same time to extract the number N2 of remaining photogenerated carriers in the clamped photodiode after each transmission gate is turned on;

[0043] d. Until the number of remaining photogenerated carriers N2 in the clamped photodiode no longer changes, at this time N2 is 1351, and the charge transfer efficiency CTE of the pixel unit of the CMOS image sensor is calculated using the N2 value (CTE=1-N2 / N1×100%), CTE=87.5%, and the charge transfer time is 495ns.

[0044] Since the non-ideal potential distribution in the transmission path of the pixel unit of the CMOS image sensor is difficult to analyze through experimental measurement, the limiting mechanism of charge transfer has not yet been fully elucidated and needs to be analyzed in depth through simulation. The simulation method proposed in the present invention can be used to obtain the charge transfer efficiency of the pixel unit of the CMOS image sensor, and the microscopic potential distribution during the charge transfer process can be intuitively seen, and the limiting mechanism of charge transfer can be analyzed and obtained, providing a strong theoretical basis for the design of high-performance CMOS image sensors.

[0045] The above is only a specific implementation of the simulation of charge transfer efficiency of a CMOS image sensor described in the present invention, but the protection scope of the present invention is not limited thereto. Any replacement or addition or reduction that can be understood by anyone familiar with the technology within the technical scope disclosed by the present invention should be included in the scope of the present invention.

Claims

1. A simulation method for multiple pulse reading charge transfer efficiency of a CMOS image sensor pixel unit, characterized in that Follow these steps: a. Before the exposure phase of the pixel unit of the CMOS image sensor begins, the clamped photodiode is reset and the transmission gate voltage is set to 0V, so that the photodiode is in the accumulation state; b. Use collimated light source incident in the TCAD software simulation environment, call the Ray Tracing light injection model in the TCAD software to simulate the light generation process, and extract the number N1 of photogenerated carriers in the clamped photodiode at this time; c. Set the floating diffusion node to a high potential and maintain it, and set a pulse voltage on the transmission gate at the same time to extract the number N2 of remaining photogenerated carriers in the clamped photodiode after each transmission gate is turned on; d. Until the number N2 of remaining photogenerated carriers in the clamped photodiode no longer changes with time, the charge transfer efficiency CTE of the pixel unit of the CMOS image sensor is calculated using the N2 value, that is, CTE=1-N2 / N1×100%.

Citation Information

Patent Citations

  • High-speed CMOS image sensor

    CN104835825A

  • Test method for transferring efficiency of photo-generated charges in photodiode

    CN109212400A