WO2022000142A1

WO2022000142A1 - Image sensor and manufacturing method therefor, imaging apparatus, and terminal device

Info

Publication number: WO2022000142A1
Application number: PCT/CN2020/098604
Authority: WO
Inventors: 姚国峰; 沈健
Original assignee: 深圳市汇顶科技股份有限公司
Priority date: 2020-06-28
Filing date: 2020-06-28
Publication date: 2022-01-06
Also published as: CN114207826A

Abstract

An image sensor (20), comprising a pixel array having a plurality of basic units. The basic units comprise first pixels (111), second pixels (112), and third pixels (113), wherein the first pixels (111) are shortwave band light pixels, and the second pixels (112) are medium wave band light pixels. The pixel array comprises, in a vertical direction, a color filter layer (120), a first substrate layer (130), a dielectric layer (140), and a second substrate layer (150). First recesses (137) corresponding to the first pixels (111) and extending toward the first substrate layer (130) and second recesses (138) corresponding to the second pixels (112) and extending toward the first substrate layer (130) are formed on the front surface of the first substrate layer (130), and the depth of the first recesses (137) is greater than that of the second recesses (138). The image sensor can improve the accuracy of color image formation.

Description

Image sensor and its manufacturing method, imaging device and terminal equipment

technical field

Embodiments of the present application relate to the technical field of image sensors, and more particularly, to an image sensor and a manufacturing method thereof, as well as an imaging device and terminal device using the image sensor.

Background technique

Complementary Metal Oxide Semiconductor (CMOS) image sensor is a typical solid-state imaging sensor, which is widely used in digital cameras, mobile terminals, security monitoring terminals and other equipment. The basic working principle of the CMOS image sensor is that the incident light from the outside irradiates the pixel array of the image sensor, and a photoelectric effect occurs, and corresponding charges are generated in the pixels. The electric charge is transmitted to the analog signal processing unit and the digital-to-analog converter through the readout circuit, and is converted into a digital image signal for output. For black and white digital images, the number corresponding to each pixel reflects the gray value of that pixel.

If a color digital image is to be collected, a color filter layer arranged in a specific pattern needs to be arranged on the pixel array of the CMOS image sensor to form a corresponding color pixel array. The Bayer array is a common method, see Figure 1, the basic unit of the Bayer array is a 2 × 2 four-pixel array, a four-pixel array contains 1 red (Red, R) pixel, 1 blue (Blue (Blue) pixel array. , B) pixel and 2 green (Green, G) pixels. The color filter layer similar to the Bayer array also includes a RYYB array, and the basic unit of the RYYB array includes 1 red pixel, 1 blue pixel and 2 yellow (Yellow, Y) pixels, in addition, there are The basic unit of the RGBW array includes 1 red pixel, 1 blue pixel, 1 green pixel and 1 white (White, W) pixel.

In the color pixel array, pixels for detecting the intensity of visible light in the short-band (called short-band light pixels, such as blue pixels) and pixels for detecting the intensity of visible light in the middle-band (called mid-band light pixels, Such as green pixels) have a certain response to the red light in the 650nm ~ 700nm band, but this response is not desired, so it is called a false response. The false response of short-wavelength light pixels and medium-wavelength light pixels to red light in the wavelength range of 650 nm to 700 nm is caused by the intrinsic characteristics of the color filter layer material. The erroneous response of the short-wavelength light pixel and the mid-wavelength light pixel to red light in the 650nm-700nm wavelength band will cause color distortion of the color image reproduced by the image sensor.

In view of this, it is necessary to provide an imaging device and equipment that can reduce the false response of short-wavelength light pixels or mid-wavelength light pixels to red light in the 650nm-700nm wavelength band, thereby improving the accuracy of color image imaging.

SUMMARY OF THE INVENTION

Embodiments of the present application provide an image sensor and a manufacturing method thereof, as well as an imaging device and a terminal device using the CMOS image sensor. The image sensor and its manufacturing method, the imaging device and the terminal device can solve the above problems.

A first aspect of the embodiments of the present application provides an image sensor, the image sensor includes a pixel array having a plurality of basic units, the basic unit includes a first pixel, a second pixel and a third pixel; the pixel array In the vertical direction, it includes: a color filter layer, including a first filter sub-area, a second filter sub-area and a third filter sub-area, the first filter sub-area corresponds to the first pixel, and is used to transmit light in the first wavelength band. The second filter sub-region corresponds to the second pixel and is used to transmit the light of the second wavelength band and filter out the light of other wavelength bands except the second wavelength band, The third filter sub-region corresponds to the third pixel, and is used to transmit light in a third wavelength band and filter out light in other wavelength bands except the third wavelength band, wherein the wavelength of the first wavelength band is smaller than the second wavelength band the wavelength of the second wavelength band is smaller than the wavelength of the third wavelength band; the first substrate layer, on the front side of the first substrate layer away from the color filter layer, is formed with a direction toward the first substrate layer an extended first groove corresponding to the first filter sub-region and a second groove corresponding to the second filter sub-region, formed between the first groove and the first filter sub-region There is a first photoelectric conversion region, a second photoelectric conversion region is formed between the second groove and the second filter subregion, and a region corresponding to the first substrate layer and the third filter subregion is formed There is a third photoelectric conversion area, wherein the depth of the first groove is greater than the depth of the second groove; the dielectric layer includes a body, a first embedded part and a second embedded part connected to the body, the The body is disposed on the side of the first substrate layer away from the color filter layer, the first embedded portion is located in the first groove, and the second embedded portion is located in the second groove; a floating diffusion region, a second floating diffusion region and a third floating diffusion region, the first floating diffusion region, the second floating diffusion region and the third floating diffusion region are formed in the first floating diffusion region In the substrate layer; a first gate, a second gate and a third gate, the first gate, the second gate and the third gate are arranged in the dielectric layer and are connected with the The first substrate layer is in contact with the first substrate, the first gate is located between the first photoelectric conversion region and the first floating diffusion region, and the second gate is located between the second photoelectric conversion region and the first floating diffusion region. between the second floating diffusion regions, the third gate is located between the third photoelectric conversion region and the third floating diffusion region, and the first gate is used to control the first photoelectric The charge in the conversion area is transferred to the first floating diffusion area, the second gate is used to control the charge in the second photoelectric conversion area to transfer to the second floating diffusion area, the third The gate is used to control the transfer of charges in the third photoelectric conversion region to the third floating diffusion region; the second substrate layer is arranged on one side of the dielectric layer.

In the CMOS image sensor provided by the embodiments of the present application, first grooves and second grooves with reasonable depths are respectively arranged in the regions of the first substrate layer corresponding to the first pixels and the second pixels. The effective thickness of the region of the first substrate layer corresponding to the pixel and the second pixel, so as to ensure that the light of the target wavelength band of the first pixel and the second pixel has higher absorption within the effective thickness of the corresponding region of the first substrate layer ratio and photoelectric conversion efficiency, and reduce the absorption ratio and photoelectric conversion efficiency of the non-target wavelength band light (red light of 650nm-700nm) of the first pixel and the second pixel within the effective thickness of the corresponding region of the first substrate layer , reducing the interference of the light in the non-target wavelength band (red light of 650nm-700nm) on the first pixel and the second pixel, thereby improving the color image imaging accuracy of the CMOS image sensor.

According to the above first aspect, in an optional embodiment of the present application, the light in the first wavelength band is visible light in the short wavelength band.

According to the above-mentioned first aspect, in an optional embodiment of the present application, the light in the second wavelength band is visible light in the middle wavelength band.

Corresponding pixel arrays, such as Bayer arrays, RYYB arrays, etc., can be selected according to requirements.

According to the above first aspect, in an optional embodiment of the present application, the first gate includes a first part, a second part and a third part, the first part is located outside the first groove, The upper end of the second part is in contact with one end of the first part, the second part is in contact with the inner wall of the first groove, and the lower end of the second part is in contact with one end of the third part In contact, the third part is located at the bottom of the first groove.

According to the above-mentioned first aspect, in an optional embodiment of the present application, the second gate includes a fourth part, a fifth part and a sixth part, and the fourth part is located at the bottom of the second groove Externally, the upper end of the fifth part is connected to one end of the fourth part, the fifth part is in contact with the inner wall of the second groove, and the lower end of the fifth part is connected to the sixth part One end of the parts is connected, and the sixth part is located at the bottom of the second groove.

Extending the first gate and the second gate to the bottoms of the first groove and the second groove, respectively, can improve the transfer efficiency of the first gate to control the transfer of charges from the first photoelectric conversion region to the first floating diffusion region , and the second gate controls the transfer efficiency of the charge from the second photoelectric conversion region to the second floating diffusion region, further improving the detection accuracy of the first pixel and the second pixel, and improving the color image imaging of the CMOS image sensor. accuracy.

According to the above first aspect, in an optional embodiment of the present application, the CMOS image sensor is a backside illuminated image sensor or a stacked image sensor.

A second aspect of the embodiments of the present application provides an imaging device, where the imaging device includes the CMOS image sensor described in any one of the above, a circuit board, a cutoff filter, a lens assembly, and a bracket body;

The lens assembly is arranged above the cut-off filter through the bracket body for condensing incident light; the cut-off filter is arranged above the CMOS image sensor through the bracket body for use in Filtering the incident light condensed by the lens assembly; the CMOS image sensor is arranged on the circuit board, and is used for receiving the light filtered by the cut-off filter and performing imaging.

According to the above second aspect, in an optional embodiment of the present application, the cut-off filter is a 700 nm cut-off filter, and the cut-off filter is used to filter out light with a wavelength of 700 nm or more.

In the CMOS image sensor of the imaging device, the distances from the photoelectric conversion areas of different pixels to the color filter layer are reasonably set, and the first grooves and the second photoelectric conversion areas are respectively provided below the first photoelectric conversion area and the second photoelectric conversion area. grooves, thereby reducing the influence of red light in the band of 650nm to 700nm on the first pixel and the second pixel. At the same time, a 700nm infrared cut-off filter is set above the CMOS image sensor, which can filter out light with a wavelength above 700nm and avoid The amount of light entering the third pixel is too large, which leads to overexposure of the red pixel, thereby improving the accuracy of color image imaging of the CMOS image sensor.

A third aspect of the embodiments of the present application provides a terminal device, where the terminal device includes the CMOS image sensor described in any of the above or the imaging device described in any of the above.

A fourth aspect of the embodiments of the present application provides a method for fabricating a CMOS image sensor, where the CMOS image sensor is the CMOS image sensor described in any one of the above, and the method for fabricating the CMOS image sensor includes: in a first substrate layer A first groove corresponding to the first pixel and a second groove corresponding to the second pixel are formed on the front side of the first substrate layer, and the depth of the first groove is greater than that of the second groove; A first gate corresponding to the first pixel, a second gate corresponding to the second pixel and a third gate corresponding to the third pixel are formed on the front surface of the substrate layer; The substrate layer is implanted with ions to form a first photoelectric conversion region and a first floating diffusion region, and ions are implanted into the first substrate layer on both sides of the second gate to form a second photoelectric conversion region and a second floating diffusion region. Ions are implanted into the first substrate layer on both sides of the gate to form a third photoelectric conversion region and a third floating diffusion region, wherein the first photoelectric conversion region corresponds to the first groove, and the second photoelectric conversion region corresponds to the second groove corresponding to the grooves; depositing a dielectric material to cover the front surface of the first substrate layer to form a first sublayer of the dielectric layer, and smoothing the surface of the first sublayer; forming a metal interconnection layer; connecting the second substrate layer and the metal interconnection layer The backside of the first substrate layer is thinned; an anti-reflection layer, a color filter layer and a microlens are formed on the backside of the first substrate layer 130 .

According to the above fourth aspect, in an optional embodiment of the present application, the first gate corresponding to the first pixel and the second gate corresponding to the second pixel are formed on the front surface of the first substrate layer and the third gate corresponding to the third pixel includes:

The first gate includes a first part, a second part and a third part, the first part is located outside the first groove, and the upper end of the second part is connected to one end of the first part, so The second part is fitted with the inner wall of the first groove, the lower end of the second part is connected to one end of the third part, and the third part is located at the bottom of the first groove; The second gate includes a fourth part, a fifth part and a sixth part, the fourth part is located outside the second groove, and an upper end of the fifth part is opposite to one end of the fourth part Then, the fifth part is attached to the inner wall of the second groove, the lower end of the fifth part is connected to one end of the sixth part, and the sixth part is located in the second groove bottom of.

In the manufacturing method, by respectively setting a first groove and a second groove with reasonable depths in the region of the first substrate layer corresponding to the first pixel and the second pixel, the first and second pixels are reasonably arranged. The effective thickness of the corresponding region of the first substrate layer, thereby ensuring that the light of the target wavelength band of the first pixel and the second pixel has a higher absorption ratio and photoelectric conversion efficiency within the effective thickness of the corresponding region of the first substrate layer, In addition, the absorption ratio and photoelectric conversion efficiency of the light of the non-target wavelength band (red light of 650 nm to 700 nm) of the first pixel and the second pixel within the effective thickness of the corresponding first substrate layer region are reduced, and the non-target wavelength band is reduced. The light (red light of 650nm-700nm) interferes with the first pixel and the second pixel, thereby improving the precision of color image imaging of the CMOS image sensor.

Description of drawings

1 is a schematic diagram of a basic unit RGGB array of a Bayer array;

Fig. 2 is a sectional view of the RGGB array shown in Fig. 1 along dotted line A-A';

FIG. 3 is a schematic structural diagram of a CMOS image sensor provided by an embodiment of the present application;

FIG. 4 is a graph of the absorption depth of light of different wavelengths in a single crystal silicon material;

FIG. 5 is a schematic structural diagram of a CMOS image sensor provided by another embodiment of the present application;

FIG. 6 is an imaging device provided by an embodiment of the present application;

7 to 13 are schematic structural diagrams formed by each step of the manufacturing method of a CMOS image sensor provided by an embodiment of the present application;

FIG. 14 is a schematic structural diagram formed by a possible implementation of step 2 of the method for fabricating a CMOS image sensor provided by an embodiment of the present application.

One or more embodiments are exemplified by the pictures in the corresponding drawings, and these exemplifications do not constitute limitations of the embodiments, and elements with the same reference numerals in the drawings are denoted as similar elements, Unless otherwise stated, the figures in the accompanying drawings do not constitute a scale limitation.

detailed description

The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

A CMOS image sensor includes a pixel array having a plurality of basic units including a first pixel, a second pixel and a third pixel, the basic unit being the smallest unit for imaging. Exemplarily, a common pixel array is a Bayer array, and the basic unit of the Bayer array is a 2*2 four-pixel array RGGB. As shown in Figure 1, an RGGB array includes 1 red pixel, 1 blue pixel and 2 green pixels. Please refer to FIG. 2. The cross-sectional view of the RGGB array shown in FIG. 1 along the dotted line AA' is shown in FIG. 2. The pixel array of the CMOS image sensor includes a color filter layer 20 and a first substrate layer in the vertical direction. 30 , a dielectric layer 40 and a second substrate layer 50 . Specifically, the CMOS image sensor 10 is a backside illuminated image sensor or a stacked image sensor.

The color filter layer 20 includes a first filter sub-area 21, a second filter sub-area 22 and a third filter sub-area 23. The first filter sub-area 21 corresponds to the first pixel 11 and is used to transmit the light of the first wavelength band and filter the light. Except for the light of other wavelength bands except the first wavelength band, the second filter sub-region 22 corresponds to the second pixel 12 and is used to transmit the light of the second wavelength band and filter out the light of other wavelength bands except the second wavelength band. The three filter sub-regions 23 correspond to the third pixels 13, and are used for transmitting light in the third wavelength band and filtering out light in other wavelength bands except the third wavelength band. Specifically, the first pixel 11 is a blue pixel, and the corresponding first filter sub-region 21 is a blue filter sub-region for transmitting blue light; the second pixel 12 is a green pixel, and the corresponding second filter sub-region 21 is a blue pixel. The photon area 22 is a green filter sub-area for transmitting green light; the third pixel 13 is a red pixel, and the corresponding third filter sub-area 23 is a red filter sub-area for transmitting red light.

The back side of the first substrate layer 30 is the light incident surface, the back side of the first substrate layer 30 is in contact with the color filter layer 20 , and a first photoelectric conversion area 31 and a second photoelectric conversion area are formed in the first substrate layer 30 33 . The third photoelectric conversion region 35 , the first floating diffusion region 32 , the second floating diffusion region 34 , and the third floating diffusion region 36 . The first photoelectric conversion region 31 and the first floating diffusion region 32 correspond to the first filter sub-region 21; the second photoelectric conversion region 33 and the second floating diffusion region 34 correspond to the second filter sub-region 22; The third photoelectric conversion region 35 and the third floating diffusion region 36 correspond to the third filter sub-region 23 . Specifically, the first substrate layer 30 may be a single crystal silicon substrate with a first doping type, usually a silicon epitaxial layer. Through the ion implantation process, the photoelectric conversion region and the floating diffusion region can be respectively formed in the regions of the first substrate layer 30 corresponding to different pixels. Exemplarily, when the first substrate layer 30 is doped with P-type boron, N-type doping elements with different concentrations may be implanted into two sub-regions of the region of the first pixel 11 corresponding to the first substrate layer 30, such as Phosphorus or arsenic can respectively form the first photoelectric conversion region 31 and the first floating diffusion region 32; the second photoelectric conversion region 33, the second floating diffusion region 34, the third photoelectric conversion region 35 and the third floating diffusion region 36 can also be formed in the same way.

The CMOS image sensor 10 further includes a first gate 41 , a second gate 42 and a third gate 43 , and the first gate 41 , the second gate 42 and the third gate 43 are arranged in the dielectric layer 40 , and is in contact with the first substrate layer 30 . The first gate 41 is located between the first photoelectric conversion region 31 and the first floating diffusion region 32, and is used to control the transfer of charges in the first photoelectric conversion region 31 to the first floating diffusion region 32; the second gate 42 is located between the second photoelectric conversion region 33 and the second floating diffusion region 34, and is used to control the transfer of charges in the second photoelectric conversion region 33 to the second floating diffusion region 34; the third gate 43 is located in the third photoelectric conversion region Between the region 35 and the third floating diffusion region 36 , the charge in the third photoelectric conversion region 35 is controlled to migrate to the third floating diffusion region 36 . Specifically, the material of the first gate 41 , the second gate 42 and the third gate 43 may be polysilicon, and the first gate 41 , the second gate 42 and the third gate 43 are the same as the first substrate layer 30 The connected structure can be fabricated by the deposition of the polysilicon layer, the photolithography of the polysilicon layer, and the etching of the polysilicon layer.

The dielectric layer 40 includes a body 47 and a first recessed portion 44 , a second recessed portion 45 and a third recessed portion 46 disposed on the body 47 . The body 47 is in contact with the first substrate layer 30 , the first gate 41 is located in the first recess 44 , the second gate 42 is located in the second recess 45 , and the third gate 43 is located in the third recess 46 . Generally speaking, the dielectric layer 40 includes a metal interconnection layer inside, and the metal interconnection layer can constitute a functional circuit, for example, to apply control signals to the first gate 41 , the second gate 42 and the third gate 43 and to read Functions such as the charge quantity of the first floating diffusion region 32 , the second floating diffusion region 34 and the third floating diffusion region 36 are displayed.

The second substrate layer 50 is disposed on one side of the dielectric layer 40 . The second substrate layer 50 may function as a support.

In a practical application scenario, the incident light from the outside illuminates the pixel array of the CMOS image sensor 10 , the first filter sub-region 21 of the first pixel 11 transmits blue light to filter out light other than blue light, and the second pixel The second filter sub-region 22 of 12 transmits green light and filters out light except green light, and the third filter sub-region 23 of the third pixel 13 transmits red light and filters out light except red light. The blue light, green light and red light respectively transmitted by the color filter layer 20 enter the first substrate layer 30 and are absorbed and converted into corresponding charges, wherein the first charges located in the first photoelectric conversion region 31 are The second charge in the second photoelectric conversion region 33 migrates to the second floating diffusion region 34 under the control of the second gate 42 and is located in the second floating diffusion region 34 under the control of the first gate 41 . The third charges in the third photoelectric conversion region 35 are transferred to the third floating diffusion region 36 under the control of the third gate electrode 43 . A readout circuit (not shown in FIG. 2 ) reads the charge amounts in the first floating diffusion region 32 , the second floating diffusion region 34 and the third floating diffusion region 36 respectively, and transmits the information to the calculation processing circuit ( 2), the digital signals corresponding to the first pixel 11, the second pixel 12 and the third pixel 13 can be obtained after processing by the calculation and processing circuit, and the magnitude of the digital signal represents the amount of light irradiated to the outside of the pixel. The intensity of the corresponding target band light in the incident light. Specifically, the magnitude of the digital signal corresponding to the red pixel represents the intensity of the red light in the incident light irradiated to the outside of the red pixel, and the magnitude of the digital signal corresponding to the green pixel represents the intensity of the green light in the incident light irradiated to the outside of the green pixel. , the magnitude of the digital signal corresponding to the blue pixel represents the intensity of blue light in the incident light irradiated to the outside of the blue pixel. After the CMOS image sensor obtains the digital signal of each pixel, it also needs to perform post-processing, such as demosaicing, white balance correction, color space conversion, etc., and finally obtain a color digital image.

Ideally, the size of the digital signal corresponding to each pixel only represents the intensity of the corresponding target band light in the incident light irradiated to the outside of the pixel. At this time, the image sensor can better reproduce the color image. However, both short-wavelength light pixels (such as blue pixels) and medium-wavelength light pixels (such as green pixels) have a certain false response to red light in the 650nm-700nm band, which is caused by the intrinsic characteristics of the color filter layer material. impact. The false response will reduce the accuracy of color image imaging of the CMOS image sensor.

Referring to FIG. 3 , an embodiment of the present application provides a CMOS image sensor, and the CMOS image sensor can improve the accuracy of color image imaging. The CMOS image sensor 20 includes a pixel array having a plurality of basic units including a first pixel 111 , a second pixel 112 and a third pixel 113 . Specifically, in the embodiment of the present application, the pixel array is a Bayer array, the first pixel 111 is a blue pixel, the second pixel 112 is a green pixel, and the third pixel 113 is a red pixel. The light is the light of the non-target wavelength band of the first pixel 111 and the second pixel 112. The first pixel 111 and the second pixel 112 have a certain degree of false response to the light of the 650nm-700nm band, which should be suppressed as much as possible, while the third pixel 113 The response to light in the 650nm-700nm band is an effective response. The pixel array of the CMOS image sensor 20 includes a color filter layer 120 , a first substrate layer 130 , a dielectric layer 140 and a second substrate layer 150 in a vertical direction.

The color filter layer 120 includes a first filter sub-region 121 , a second filter sub-region 122 and a third filter sub-region 123 . The first filter sub-region 121 corresponds to the first pixel 111 and is used to transmit light in the first wavelength band and filter out light in other wavelength bands except the first wavelength band; the second filter sub-region 122 corresponds to the second pixel 112 , which is used to transmit the light of the second wavelength band and filter out the light of other wavelength bands except the second wavelength band; the third filter sub-region 123 corresponds to the third pixel 113 and is used to transmit the light of the third wavelength band and filter out the light of other wavelength bands. Light in other wavelength bands except the third band. Wherein, the wavelength of the first waveband is smaller than the wavelength of the second waveband, and the wavelength of the second waveband is smaller than the wavelength of the third waveband; The second filter sub-region 122 transmits green light and filters out light in other wavelength bands except green light, and the third filter sub-region 123 transmits red light and filters out light in other wavelength bands except red light, wherein red light The wavelength of light is the longest, green light is second, and blue light has the shortest wavelength.

The back side of the first substrate layer 130 is connected to the color filter layer 120 , and a first groove 137 extending toward the first substrate layer 130 and corresponding to the first filter sub-region 121 is formed on the front side of the first substrate layer 130 . and a second groove 138 corresponding to the second filter sub-region 122, a first photoelectric conversion region 131 is formed between the first groove 137 and the first filter sub-region 121, the second groove 138 and the second filter sub-region 121 A second photoelectric conversion region 133 is formed between the regions 122, and a third photoelectric conversion region 135 is formed in the region corresponding to the third pixel 113 in the first substrate layer 130, wherein the depth of the first groove 137 is greater than that of the second photoelectric conversion region 137. Depth of groove 138.

The dielectric layer 140 includes a body 146, a first embedded portion 144 and a second embedded portion 145 connected to the body 146, the first embedded portion 144 is located in the first groove 137, and the second embedded portion 145 is located in the second groove 138;

The CMOS image sensor also includes a first floating diffusion region 132, a second floating diffusion region 134, and a third floating diffusion region 136. The first floating diffusion region 132, the second floating diffusion region 134, and the third floating diffusion region region 136 is formed within first substrate layer 130;

The CMOS image sensor further includes a first gate 141 , a second gate 142 and a third gate 143 , the first gate 141 , the second gate 142 and the third gate 143 are all disposed in the dielectric layer 140 , and the first gate 141 , the second gate 142 and the third gate 143 A gate 141 , a second gate 142 and a third gate 143 are all connected to the first substrate layer 130 . The first gate 141 is located between the first photoelectric conversion region 131 and the first floating diffusion region 132 in the horizontal direction, and is used to control the charge in the first photoelectric conversion region 131 to migrate to the first floating diffusion region 132; The second gate 142 is located between the second photoelectric conversion region 133 and the second floating diffusion region 134 in the horizontal direction, and is used to control the charge in the second photoelectric conversion region 133 to migrate to the second floating diffusion region 134; the third The gate 143 is located between the third photoelectric conversion region 135 and the third floating diffusion region 136 in the horizontal direction, and is used for controlling the transfer of charges in the third photoelectric conversion region 135 to the third floating diffusion region 136 .

The second substrate layer 150 is adhered to the dielectric layer 140 and can function as a support.

Exemplarily, the first substrate layer 130 is a single crystal silicon substrate with a first doping type, which is usually a silicon epitaxial layer, and may specifically be a P-type boron doped silicon substrate. A first groove 137 is provided in a region corresponding to the first pixel 111 on the front surface of the first substrate layer 130, and a second groove 138 is provided in a region corresponding to the second pixel 112 on the front surface of the first substrate layer 130; The N-type first photoelectric conversion region 131 and the N-type first floating region are formed by injecting N-type doping elements with different concentrations into two sub-regions of the first substrate layer 130 corresponding to the first pixel 111 respectively. Diffusion region 132; by injecting N-type doping elements with different concentrations into two sub-regions in the region of the first substrate layer 130 corresponding to the second pixel 112, respectively, to form an N-type second photoelectric conversion region 133 and an N-type The first floating diffusion region 134; the N-type third photoelectric conversion region 135 is formed by injecting N-type doping elements with different concentrations into two sub-regions of the first substrate layer 130 corresponding to the third pixel 113 respectively. and an N-type third floating diffusion region 136;

The material of the first gate 141 , the second gate 142 and the third gate 143 may be polysilicon, and the structure in which the first gate 141 , the second gate 142 and the third gate 143 are in contact with the first substrate layer 130 It can be made by deposition of polysilicon layer, photolithography of polysilicon layer and etching of polysilicon layer.

The material of the dielectric layer 140 can be silicon oxide, silicon nitride, silicon oxynitride, fluorosilicate glass, phosphosilicate glass, borophosphosilicate glass and their combination materials, which are usually formed by a chemical vapor deposition (Chemical Vapor Deposition, CVD) process. To realize the deposition of the material, the surface needs to be smoothed after the deposition, and the smoothing usually adopts a chemical mechanical polishing (Chemical Mechanical Polishing, CMP) process; the dielectric layer 140 also includes a metal interconnection layer, and the metal interconnection layer specifically includes contact holes , metal layers, conductive vias, etc., can be used to apply control signals to the first gate 141 , the second gate 142 and the third gate 143 and to read out the first floating diffusion region 132 and the second floating diffusion The function of the charge quantity of the region 134 and the third floating diffusion region 136, etc.

The second substrate layer 150 is bonded to the dielectric layer 140, and the bonding process may be a bonding process; the second substrate layer 150 may be a bare silicon wafer (Bare Si wafer) or an image signal processing wafer (ISP wafer).

In a practical application scenario, the incident light from the outside illuminates the CMOS image sensor 20 , and the first filter sub-region 121 corresponding to the first pixel 111 transmits the light of the first wavelength band and filters out other wavelengths except the first wavelength band. The signal of the wavelength band, the second filter sub-region 122 corresponding to the second pixel 112 transmits the light of the second wavelength band and filters out the light of other wavelength bands except the second wavelength band, and the third filter sub-region 122 corresponding to the third pixel 113 123 transmits the light of the third wavelength band and filters out the light of other wavelength bands except the third wavelength band. Exemplarily, the first filter sub-region 121 of the first pixel 111 transmits blue light and filters out light other than blue light, and the second filter sub-region 122 of the second pixel 112 transmits green light and filters out the green light. The third filter sub-region 123 of the third pixel 113 transmits the red light and filters out the light other than the red light. After the light in the first wavelength band, the light in the second wavelength band, and the light in the third wavelength band enter the first substrate layer 130, they are absorbed by the first substrate layer 130 and converted into corresponding charges, wherein the light in the first photoelectric conversion The first charges in the region 131 are transferred to the first floating diffusion region 132 under the control of the first gate 141 , and the second charges in the second photoelectric conversion region 133 are transferred to the first floating diffusion region 132 under the control of the second gate 142 . Two floating diffusion regions 134 , and the third charges located in the third photoelectric conversion region 135 are transferred to the third floating diffusion region 136 under the control of the third gate 143 . The reading circuit (not shown in FIG. 3 ) reads the charge signals in the first floating diffusion region 132 , the second floating diffusion region 134 and the third floating diffusion region 136 respectively, and transmits the information to the calculation processing circuit ( 3), the digital signals of the first pixel 111, the second pixel 112 and the third pixel 113 can be obtained after processing by the calculation processing circuit. Based on the digital signal, color image imaging can be performed.

In the embodiment of the present application, grooves with different depths are reasonably arranged in the regions of the first substrate layer 130 corresponding to the first pixels 111 and the second pixels 112, thereby reducing the pair of the first pixels 111 and the second pixels 112 by 650 m The absorption and photoelectric conversion of light in the ∼700 nm band further reduces the interference of the light in the 650 m∼700 nm band to the first pixel 111 and the second pixel 112 , and improves the accuracy of color image imaging of the CMOS image sensor 20 .

The incident light from the outside passes through the color filter layer 120 and then enters the first substrate layer 130. As the depth of the first substrate layer 130 gradually increases, the absorption ratio of the incident light in the first substrate layer 130 gradually decreases, that is, The attenuation ratio of the intensity of the incident light decreases as the depth of the first substrate layer 130 increases, and at the surface of the first substrate layer 130 , the attenuation ratio of the incident light is the largest. When the attenuation ratio of the incident light in the first substrate layer 130 is 1/e (about 37%), the depth at which the incident light is located in the first substrate layer 130 is called the absorption depth. Lights of different wavelengths have different decay rates in the first substrate layer 130 , so the absorption depths of lights of different wavelengths in the first substrate layer 130 are different. Taking the common single crystal silicon as the first substrate layer 130 as an example for illustration, the absorption depths of light of different wavelengths in the single crystal silicon material are shown in FIG. 4 .

Exemplarily, the thickness of the first substrate layer 130 is 5 μm, and a first groove 137 with a depth of 4.2 μm is provided in the region of the first substrate layer 130 corresponding to the first pixel 111 . The distance between the backside of a substrate layer 130 is referred to as the effective thickness of the region of the first substrate layer 130 corresponding to the first pixel 111 , and the effective thickness of the first pixel 111 is 0.8 μm. It can be seen from FIG. 4 that the absorption depth of blue light (the main wavelength is 400 nm to 450 nm) in the single crystal silicon material is about 0.4 μm to 0.8 μm, so the target wavelength blue light of the first pixel 111 is in the first substrate layer 130 . The region has higher absorption ratio and photoelectric conversion efficiency. Since the absorption depth of red light in the 650nm-700nm band is about 3.5um-5um, in the region where the depth of the first substrate layer is 0.8um-5um, the red light in the 650nm-700nm band can be absorbed and converted into photoelectricity in large quantities, but Since the effective thickness of the region of the first substrate layer 130 corresponding to the first pixel 111 is 0.8 um, the light in the non-target wavelength band (red light in the 650 nm-700 nm band) of the first pixel 111 is in the area of the first substrate layer 130 . The absorption ratio and the photoelectric conversion efficiency in the region are low, thereby reducing the influence of the light in the non-target band (red light in the 650nm-700nm band) on the first pixel 111 and improving the detection accuracy of the first pixel 111;

Similarly, the thickness of the first substrate layer 130 is 5um, and the region of the first substrate layer 130 corresponding to the second pixel 112 is provided with a second groove 138 with a depth of 3.3um. The distance between the backside of the substrate layer 130 is referred to as the effective thickness of the region of the first substrate layer 130 corresponding to the second pixel 112 , and the effective thickness of the second pixel 112 is 1.7 μm. It can be seen from FIG. 4 that the absorption depth of green light (main wavelength is 500nm-560nm) in the single crystal silicon material is about 0.9um to 1.7um, so the green light of the target wavelength band of the second pixel 112 is in the first substrate layer. In the region of 130, there is a higher absorption ratio and photoelectric conversion efficiency. Since the absorption depth of red light in the band of 650nm-700nm is about 3.5um-5um, in the region where the depth of the first substrate layer is 1.7um-5um, the red light in the band of 650nm-700nm can be absorbed and photoelectrically converted in large quantities, but Since the effective thickness of the region of the first substrate layer 130 corresponding to the second pixel 112 is 1.7 μm, the light in the non-target wavelength band (red light in the 650 nm˜700 nm band) of the second pixel 112 is in the area of the first substrate layer 130 . The absorption ratio and photoelectric conversion efficiency in the region are low, thereby reducing the influence of the light in the non-target band (red light in the 650nm-700nm band) on the second pixel 112 and improving the detection accuracy of the second pixel 112;

For the third pixel 113 , the red light in the band of 650 nm˜700 nm is the light in the target band, so the region of the first substrate layer 130 corresponding to the third pixel 113 does not need to be provided with a groove, and the front surface of the first substrate 130 to the The distance of the back surface is the effective thickness of the region of the first substrate layer 130 corresponding to the third pixel 113 .

Therefore, in the embodiment of the present application, by reasonably setting the effective thicknesses of the regions of the first substrate layer 130 corresponding to different pixels, the effects of the first pixels 111 and the second pixels 112 on the red light in the non-target wavelength band of 650-700 nm are reduced. Therefore, the accuracy of color image imaging of the CMOS image sensor 20 is improved.

In addition, when the basic unit of the pixel array of the CMOS image sensor includes pixels that can detect long-wavelength light and short-wavelength light at the same time, or includes pixels that can detect long-wavelength light and middle-wavelength light at the same time, the long-wavelength light can be simultaneously detected at this time. The region of the first substrate layer corresponding to the pixels of short-wavelength light or the pixels capable of simultaneously detecting long-wavelength light and medium-wavelength light does not need to be provided with grooves. Exemplarily, when the pixel array of the CMOS image sensor is an RYYB array, the basic unit of the RYYB array includes 1 red pixel, 2 yellow pixels and 1 blue pixel, and the yellow pixel can transmit red light and green light at the same time. The red light is light with a long wavelength band. At this time, the response of the yellow pixel to the red light is an effective response, so there is no need to provide a groove in the region of the first substrate layer corresponding to the yellow pixel. Therefore, the RYYB array only needs to provide grooves with reasonable depths in the first substrate layer corresponding to the blue pixels.

A CMOS image sensor provided by another embodiment of the present application is shown in FIG. 5 . On the basis of the CMOS image sensor shown in FIG. 3 , the pixels corresponding to the grooves provided on the front surface of the first substrate layer are The gate of the groove extends to the bottom of the groove along one side of the groove, thereby improving the transfer efficiency of the charge transfer from the photoelectric conversion area corresponding to the pixel to the corresponding floating diffusion area, thereby improving the detection accuracy of the pixel. The color accuracy of the color image imaging of the CMOS image sensor is improved.

As shown in FIG. 5 , the CMOS image sensor 30 includes a pixel array having a plurality of basic units, and the basic units include a first pixel 211 , a second pixel 212 and a third pixel 213 . Specifically, in this embodiment of the present application, The pixel array is a Bayer array, the first pixel 211 is a blue pixel, the second pixel 212 is a green pixel, and the third pixel 213 is a red pixel. The pixel array of the CMOS image sensor 30 includes a color filter layer 220 , a first substrate layer 230 , a dielectric layer 240 and a second substrate layer 250 in a vertical direction.

The color filter layer 220 includes a first filter sub-region 221, a second filter sub-region 222 and a third filter sub-region 223. The first filter sub-region 221 corresponds to the first pixel 211 and is used to transmit light of the first wavelength band and filter out light in other wavelength bands except the first wavelength band, the second filter sub-area 222 corresponds to the second pixel 212 for transmitting the light in the second wavelength band and filtering out the light in other wavelength bands except the second wavelength band , the third filter sub-region 223 corresponds to the third pixel 213 and is used to transmit light in the third band and filter out light in other bands except the third band, wherein the wavelength of the first band is smaller than that of the second band wavelength, the wavelength of the second band is smaller than the wavelength of the third band. Specifically, the first filter sub-region 221 transmits blue light and filters out light in other wavelength bands except blue light, the second filter sub-region 222 transmits green light and filters out light in other wavelength bands except green light, and the third filter sub-region 222 filters out light in other wavelength bands except green light. The photon region 223 transmits red light and filters out light in other wavelength bands except red light. Among them, red light has the longest wavelength, green light is the second, and blue light has the shortest wavelength.

The back side of the first substrate layer 230 is connected to the color filter layer 220 , and a first groove 237 extending toward the first substrate layer 230 and corresponding to the first filter sub-region 221 is formed on the front side of the first substrate layer 230 and a second groove 238 corresponding to the second filter sub-region 222, a first photoelectric conversion region 231 is formed between the first groove 237 and the first filter sub-region 221, the second groove 238 and the second filter sub-region 221 A second photoelectric conversion region 233 is formed between the regions 222, and a third photoelectric conversion region 235 is formed in the region of the first substrate layer 230 corresponding to the third pixel 213, wherein the depth of the first groove 237 is greater than that of the second groove Depth of slot 238.

The dielectric layer 240 includes a body 246 and a first embedded portion 244 and a second embedded portion 245 connected to the body 246 . The first embedded portion 244 is located in the first groove 237 , and the second embedded portion 245 is located in the second groove 238 ;

The CMOS image sensor 30 also includes a first floating diffusion region 232, a second floating diffusion region 234, and a third floating diffusion region 236. The first floating diffusion region 232, the second floating diffusion region 234, and the third floating diffusion region a diffusion region 236 is formed within the first substrate layer 230;

The CMOS image sensor 30 further includes a first gate 241, a second gate 242 and a third gate 243, the first gate 241 includes a first part, a second part and a third part, the first part is located on the first substrate layer On the front side of the first substrate layer 240 , one side of the second part is in contact with the first part, and the second part is in contact with the sidewall of the first groove 237 the third part is connected with the other side of the second part, and the third part is in contact with the bottom of the first groove 237, the first part, the second part and the The third part is a whole, and the first gate 241 is located between the first photoelectric conversion region 231 and the first floating diffusion region 232 in the horizontal direction, and is used to control the charge generated by the first photoelectric conversion region 231 to migrate to The first floating diffusion region 232; similarly, the second gate 242 includes a fourth part, a fifth part and a sixth part, the fourth part is located on the front surface of the first substrate layer 240, and is connected with the first substrate layer 240 The front side of the fifth part is in contact with the fourth part, the fifth part is in contact with the side wall of the second groove 238, and the sixth part is in contact with the fourth part. The other side of the fifth part is connected, and the sixth part is in contact with the bottom of the second groove 238, the fourth part, the fifth part and the sixth part are a whole, and The second gate 242 is located between the second photoelectric conversion region 233 and the second floating diffusion region 234 in the horizontal direction, and is used to control the charge generated by the second photoelectric conversion region 233 to migrate to the second floating diffusion region 234; The three gates 243 are attached to the front surface of the first substrate layer 240, and the third gate 243 is located between the third photoelectric conversion region 235 and the third floating diffusion region 235 in the horizontal direction, for controlling the third photoelectric The charges generated by the conversion region 235 are transferred to the third floating diffusion region 236 .

The second substrate layer 250 is attached to one side of the dielectric layer 240 .

Exemplarily, the first substrate layer 230 is a single crystal silicon substrate with a first doping type, which is usually a silicon epitaxial layer, and may specifically be a P-type boron doped silicon substrate. A first groove 237 is provided in a region corresponding to the first pixel 211 on the front surface of the first substrate layer 230, and a second groove 238 is provided in a region corresponding to the second pixel 212 on the front surface of the first substrate layer 230; The N-type first photoelectric conversion region 231 and the N-type first floating region are formed by implanting N-type doping elements with different concentrations into two sub-regions of the first substrate layer 230 corresponding to the first pixel 211 respectively. Diffusion region 232; by injecting N-type doping elements of different concentrations into two sub-regions in the region of the first substrate layer 230 corresponding to the second pixel 212, respectively, to form an N-type second photoelectric conversion region 233 and an N-type The first floating diffusion region 234; the N-type third photoelectric conversion region 235 is formed by injecting N-type doping elements with different concentrations into two sub-regions of the first substrate layer 230 corresponding to the third pixel 213 respectively. and an N-type third floating diffusion region 236;

The material of the first gate 241 , the second gate 242 and the third gate 243 may be polysilicon, and the first gate 241 , the second gate 242 and the third gate 243 are in contact with the front surface of the first substrate layer 230 The structure can be fabricated by the deposition of polysilicon layer, photolithography of polysilicon layer, and etching of polysilicon layer.

The material of the dielectric layer 240 can be silicon oxide, silicon nitride, silicon oxynitride, fluorosilicate glass, phosphosilicate glass, borophosphosilicate glass and their combination materials, and chemical vapor deposition process is usually used to realize the deposition of materials. The surface also needs to be polished, and the polishing usually adopts a chemical mechanical polishing process; the dielectric layer 240 also includes a metal interconnection layer, and the metal interconnection layer specifically includes structures such as contact holes, metal layers, and conductive vias, which can provide The first gate 241 , the second gate 242 and the third gate 243 apply control signals and read out the charge amount of the first floating diffusion region 232 , the second floating diffusion region 234 , the third floating diffusion region 236 , etc. Features.

The second substrate layer 250 is bonded to the dielectric layer 240, and the bonding process may be a bonding process; the second substrate layer 250 may be a bare silicon wafer (Bare Si wafer) or an image signal processing wafer (ISP wafer).

In a practical application scenario, the incident light from the outside irradiates the CMOS image sensor 30, and the first filter sub-region 221 corresponding to the first pixel 211 transmits the light of the first wavelength band and filters out the light of the first wavelength band. For signals of other wavelength bands, the second filter sub-region 222 corresponding to the second pixel 212 transmits the light of the second wavelength band and filters out signals of other wavelength bands except the light of the second wavelength band, and the first filter sub-region 222 corresponding to the third pixel 213 The filter sub-region 223 transmits light in the third wavelength band and filters out signals in other wavelength bands except for the light in the third wavelength band. Exemplarily, the first filter sub-region 221 of the first pixel 211 transmits blue light and filters out light other than blue light, and the second filter sub-region 222 of the second pixel 212 transmits green light and filters out light other than green light. light, the third filter sub-region 223 of the third pixel 213 transmits red light and filters out light other than red light. The light of the first wavelength band, the light of the second wavelength band, and the light of the third wavelength band are respectively absorbed and converted into the first charge, the second charge and the third charge after entering the first substrate layer 230. The first charges in a photoelectric conversion region 231 migrate to the first floating diffusion region 232 under the control of the first gate 241 , and the second charges in the second photoelectric conversion region 233 are under the control of the second gate 242 Transferred to the second floating diffusion region 234 , the third charges located in the third photoelectric conversion region 235 are transferred to the third floating diffusion region 236 under the control of the third gate 243 . The reading circuit reads the charge signals in the first floating diffusion area 232, the second floating diffusion area 234 and the third floating diffusion area 236 respectively, and transmits the information to the calculation processing circuit, and the calculation processing circuit is processed by the calculation processing circuit. Then the digital signals of the first pixel 211 , the second pixel 212 and the third pixel 213 can be obtained. Based on the digital signal, color image imaging can be performed.

In the embodiment of the present application, the CMOS image sensor 30 is configured by disposing first grooves 237 and second grooves 238 with reasonable depths in the regions of the first substrate layer 230 corresponding to the first pixels 211 and the second pixels 212, respectively. The effective thickness of the region of the first substrate layer 230 corresponding to the first pixel 211 and the second pixel 212 is reasonably set, so as to ensure that the light of the target wavelength band of the first pixel 211 and the second pixel 212 is in the corresponding first substrate layer 230 There is a higher absorption ratio and photoelectric conversion efficiency within the effective thickness of the region, and the non-target wavelength light (red light of 650nm-700nm) of the first pixel 211 and the second pixel 212 is reduced in the corresponding first substrate layer. The absorption ratio and the photoelectric conversion efficiency within the effective thickness of the region of 230 reduce the interference of the light in the non-target band (red light of 650 nm to 700 nm) on the first pixel 211 and the second pixel 212, thereby improving the CMOS image sensor 30 The accuracy of color image imaging.

In addition, in the embodiment of the present application, one side of the first gate 241 extends along the sidewall of the first groove 237 to the bottom of the first groove 237 , thereby reducing the number of the first gate 241 and the first gate 241 . The distance between a photoelectric conversion region 221 can improve the transfer efficiency of the charge in the first photoelectric conversion region 221 to the first floating diffusion region 222, thereby improving the detection accuracy of the first pixel 211; One side of the gate electrode 242 extends along the sidewall of the second groove 238 to the bottom of the second groove 238, thereby reducing the distance between the second gate electrode 242 and the second photoelectric conversion region 223, thereby improving the second The transfer efficiency of the charge in the photoelectric conversion region 223 is transferred to the second floating diffusion region 224 , thereby improving the detection accuracy of the second pixel 212 .

In addition, by extending one side of the first gate 241 and the second gate 242 to the bottom of the first groove 237 and the bottom of the second groove 238, respectively, one side of the first photoelectric conversion region 221 and the first The structure in which the gate electrode 241 extends to the side of the first groove 237 is easier to realize in the process. Similarly, the side of the second photoelectric conversion region 223 and the second gate electrode 242 extend to the side of the second groove 238. Aligned structures are also easier to implement in the process.

An embodiment of the present application provides an imaging device, as shown in FIG. 6 . The imaging device 40 includes an image sensor 304, a circuit board 300, a cut filter 301, a lens assembly 302, and a bracket body 303. The image sensor 304 is the CMOS image sensor in any of the above-mentioned embodiments of the present application. The image sensor 304 is disposed on the circuit board 300 , the cut-off filter 301 and the lens assembly 302 are placed above the CMOS image sensor 304 through the bracket body 303 , and The lens assembly 302 is located above the cut-off filter 301 . Exemplarily, the lens assembly 302 includes a convex lens and/or a concave lens, which can condense the incident light, and can focus the shooting scene on the back focal plane of the lens assembly 302, that is, the plane where the image sensor 304 is located; The filter 301 is used to filter the light beams collected by the lens assembly 302, and can be used to filter out invalid light information; the image sensor is used to receive and image the light filtered by the cut-off filter 301. Exemplarily, the imaging device 40 may be a camera module of a camera.

As a preferred embodiment, the cut-off filter 301 may be an infrared cut-off filter. Specifically, the infrared cut-off filter may be a 700 nm cut-off filter, which is used to filter out light above 700 nm.

Since the filters of the short-wavelength light pixels and the medium-wavelength light pixels have a certain false response to the light in the 650nm-700nm band, in order to eliminate this false response, the prior art usually sets a 650nm cut-off filter above the image sensor. , the 650nm cut-off filter can filter out light with wavelengths above 650nm, while maintaining high transmittance for visible light below 650nm. Although the 650nm cut-off filter can solve the color distortion problem of color images caused by the false response of the short-band light pixels and the mid-band light pixels to the red light in the 650nm-700nm band, it will also cause the red pixels to enter the The decrease in the amount of light deteriorates the imaging performance in dark scenes. In the image sensor of the embodiments of the present application, grooves with reasonable depths are arranged on the first substrates of the short-wavelength light pixels and the medium-wavelength light pixels, so as to ensure that the light in the target wavelength bands of the short-wavelength light pixels and the medium-wavelength light pixels can be fully absorbed and absorbed. Under the premise of photoelectric conversion, the absorption ratio of the first substrate of the short-wavelength light pixel and the medium-wavelength light pixel to the light in the 650nm-700nm wavelength band is reduced, and the short-wavelength light pixel and the medium-wavelength light caused by the intrinsic characteristics of the filter are reduced. The false response of the wavelength band light pixel to the light in the 650nm-700nm band. Therefore, in this embodiment of the present application, it is not necessary to set a 650 nm cut-off filter above the pixel array in order to reduce the false response of the short-wavelength light pixel and the mid-wavelength light pixel to the red light in the 650 nm-700 nm wavelength band. Since light with wavelengths above 700nm has limited effective information for red pixels, and light above 700nm can pass through the filter of red pixels, this will make the amount of light entering red pixels too much, so that the photoelectric conversion area is saturated. state, but reduces the accuracy of the CMOS image sensor to restore the color of the image. Therefore, a 700nm infrared cut-off filter is arranged above the CMOS image sensor, which can filter out light with a wavelength above 700nm, avoid excessive light input of red pixels and lead to overexposure of red pixels, thereby improving the CMOS image sensor to restore image color. Accuracy.

An embodiment of the present application provides a terminal device, where the terminal device includes the image sensor or imaging device in the above-mentioned embodiments of the present application. Exemplarily, the terminal device may be a video camera, a video camera, or a surveillance camera.

An embodiment of the present application provides a method for fabricating a CMOS image sensor. The fabrication method can be used to fabricate the image sensor shown in FIG. 4 . Specifically, please refer to FIGS. 7 to 13 . The fabrication method includes:

Step 1: A first groove corresponding to the first pixel and a second groove corresponding to the second pixel are formed on the front surface of the first substrate layer, and the depth of the first groove is greater than that of the second the depth of the groove;

Referring to FIG. 7 , for example, the first substrate layer 230 may be a single crystal silicon substrate, and the thickness of the first substrate layer 230 may be 4um. The first pixel 211 may be a blue pixel for detecting blue light. The second pixel 212 may be a green pixel for detecting green light. Since the absorption depth of blue light in the single crystal silicon substrate is 0.4um～0.8um, and the absorption depth of green light in the single crystal silicon substrate is 0.9um～1.7um, the depth of the first groove 237 can be set to 3.2 um um, the depth of the second groove 238 can be set to 2.3um, the first groove 237 and the second groove 238 can be formed on the first substrate layer 230 by dry etching or wet etching process, and the first groove The depth of the groove 237 is greater than the depth of the second groove 238 .

Step 2: forming a first gate corresponding to the first pixel, a second gate corresponding to the second pixel and a third gate corresponding to the third pixel on the front surface of the first substrate layer;

8, please refer to FIG. 8, the first gate 241 includes a first part, a second part and a third part, the first part of the first gate 241 is located outside the first groove 237, the first gate 241 The upper end of the second part is connected to one end of the first part of the first gate 241 , the second part of the first gate 241 is in contact with the inner wall of the first groove 237 , and the second part of the first gate 241 is in contact with the inner wall of the first groove 237 . The lower end is in contact with one end of the third part of the first gate 241, and the third part of the first gate 241 is located at the bottom of the first groove 237; the second gate 241 includes a fourth part, a fifth part and a sixth part part, the fourth part of the second gate 241 is located outside the second groove 238, the upper end of the fifth part of the second gate 241 is in contact with one end of the fourth part of the second gate 241, the second gate The fifth part of the second gate 241 is in contact with the inner wall of the second groove 238; the lower end of the fifth part of the second gate 241 is in contact with one end of the sixth part of the second gate 241. Six parts are located at the bottom of the second recess 238 ; the third gate 243 is located on the front side of the first substrate layer 230 .

Step 3: Implant ions into the first substrate layer on both sides of the first gate to form a first photoelectric conversion region and a first floating diffusion region, and implant ions into the first substrate layer on both sides of the second gate to form a second The photoelectric conversion region and the second floating diffusion region are implanted into the first substrate layer on both sides of the third gate to form the third photoelectric conversion region and the third floating diffusion region, wherein the first photoelectric conversion region and the first The grooves correspond to the second photoelectric conversion regions, and the second photoelectric conversion regions correspond to the second grooves.

Referring to FIG. 9, exemplarily, when the first substrate layer 230 is a P-type boron-doped silicon substrate, the implanted ions may be N-type doping elements such as phosphorus or arsenic. Specifically, the first photoelectric conversion region 231 is provided between the first groove 237 and the back surface of the first substrate layer 230 ; the second photoelectric conversion region 233 is provided between the second groove 238 and the back surface of the first substrate layer 230 ; The third photoelectric conversion region 235 is arranged in the region of the first substrate layer 230 corresponding to the third pixel 213 .

Step 4: depositing a dielectric material to cover the front surface of the first substrate layer to form a first sub-layer of the dielectric layer, and smoothing the surface of the first sub-layer;

Referring to FIG. 10, exemplarily, the formation of the first sub-layer 2401 of the dielectric layer usually adopts a chemical vapor deposition process and requires multiple depositions. Silicon nitride with a thickness of about 50 nanometers, and finally silicon oxide with a thickness of about 5 to 6 microns is deposited to fill all the recesses, the recesses include the first groove 237 and the second groove 238, and then the first sub-layer 2401 The surface is smoothed, usually by chemical mechanical polishing.

Step 5: forming a metal interconnection layer;

Referring to FIG. 11 , for example, the dielectric layer 240 includes a metal interconnection layer 2402, the metal interconnection layer 2402 includes contact holes, a metal layer and conductive vias, and the metal interconnection layer 2402 may include a plurality of sub-levels. They are isolated from each other by insulating dielectric materials. The metal interconnection layer 2402 can function to apply control signals to the first gate 241, the second gate 242 and the third gate 243 and to read out the first floating diffusion region 232, the second floating diffusion region 234, Functions such as the amount of charge of the third floating diffusion region 236;

Step 6: bonding the second substrate layer with the metal interconnect layer, and thinning the back of the first substrate layer;

Referring to FIG. 12 , exemplarily, after the front surface of the second substrate layer 250 is planarized, the metal interconnection layer 2402 is bonded, and the back surface of the first substrate layer 230 is thinned.

Step 7: forming an anti-reflection layer, a color filter layer and a microlens on the back of the first substrate layer.

Referring to FIG. 13, exemplarily, the anti-reflection layer 2301 and the microlens 2303 can be used to increase the amount of incident light, and the pattern of the color filter layer 2302 can be set according to requirements, for example, a Bayer array can be used.

In the manufacturing method provided by the embodiment of the present application, by respectively setting the first grooves 237 and the second grooves 238 with reasonable depths in the regions of the first substrate layer 230 corresponding to the first pixels 211 and the second pixels 212, it is reasonable to The effective thickness of the region of the first substrate layer 230 corresponding to the first pixel 211 and the second pixel 212 is set, so as to ensure that the light of the target wavelength band of the first pixel 211 and the second pixel 212 is in the corresponding first substrate layer 230. The effective thickness of the region has a higher absorption ratio and photoelectric conversion efficiency, and reduces the non-target wavelength light (red light of 650nm-700nm) of the first pixel 211 and the second pixel 212 in the corresponding first substrate layer 230. The absorption ratio and the photoelectric conversion efficiency within the effective thickness of the region can reduce the interference of the light in the non-target band (red light of 650 nm to 700 nm) on the first pixel 211 and the second pixel 212, thereby improving the performance of the CMOS image sensor 30. Accuracy of Color Image Imaging.

In addition, by extending the first gate 241 and the second gate 242 to the bottoms of the first groove 237 and the second groove 238, respectively, the first gate 141 can improve the charge transfer of the first photoelectric conversion region 231 controlled by the first gate 141 to The transfer efficiency of the first floating diffusion region 232 and the transfer efficiency of the second gate 242 to control the transfer of charges from the second photoelectric conversion region 233 to the second floating diffusion region 234 further improve the first pixel 211 and the second pixel 212 The detection accuracy is improved, and the accuracy of the color image restoration of the CMOS image sensor is improved.

As a possible embodiment, the embodiment of the present application provides another possible implementation manner of step 2. Please refer to FIG. 14 , exemplarily, the first gate 141 is disposed outside the first groove 237 , and one side of the first gate 141 is aligned with one side of the inner wall of the first groove 237 ; the second gate 142 is disposed Outside the second groove 238 , and one side of the second gate 142 is aligned with one side of the inner wall of the second groove 238 ; the third gate 143 is disposed on the front surface of the first substrate layer 230 . The material of the first gate 141, the second gate 142 and the third gate 143 may be polysilicon, and the first gate 141, the second gate 142 and the third gate 143 may be formed by deposition of a polysilicon layer, photolithography and polysilicon layer etching.

It should be noted that, on the premise of no conflict, each embodiment described in this application and/or the technical features in each embodiment can be arbitrarily combined with each other, and the technical solution obtained after the combination should also fall within the protection scope of this application .

It should be understood that the specific examples in the embodiments of the present application are only to help those skilled in the art to better understand the embodiments of the present application, rather than limiting the scope of the embodiments of the present application, and those skilled in the art can Various improvements and modifications can be made, and these improvements or modifications all fall within the protection scope of the present application.

The above are only specific embodiments of the present application, but the protection scope of the present application is not limited to this. should be covered within the scope of protection of this application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

An image sensor, characterized in that the image sensor includes a pixel array with a plurality of basic units, the basic unit includes a first pixel, a second pixel and a third pixel; the pixel array includes in the vertical direction:

The color filter layer includes a first filter sub-area, a second filter sub-area and a third filter sub-area, wherein the first filter sub-area corresponds to the first pixel and is used to transmit the light of the first wavelength band and filter out the first filter sub-area. For light in other wavelength bands than one wavelength band, the second filter sub-region corresponds to the second pixel, and is used to transmit light in the second wavelength band and filter out light in other wavelength bands except the second wavelength band, and the third filter sub-region corresponds to the second pixel. The area corresponds to the third pixel, which is used to transmit light in the third wavelength band and filter out light in other wavelength bands except the third wavelength band, wherein the wavelength of the first wavelength band is smaller than the wavelength of the second wavelength band, and the wavelength of the first wavelength band is smaller than that of the second wavelength band. The wavelength of the second band is smaller than the wavelength of the third band;

a first substrate layer, a first groove extending toward the first substrate layer and corresponding to the first filter sub-region and a first groove corresponding to the first filter sub-region and A second groove corresponding to the second filter sub-region, a first photoelectric conversion region is formed between the first groove and the first filter sub-region, and the second groove is connected to the second filter sub-region. A second photoelectric conversion region is formed between the photonic regions, and a third photoelectric conversion region is formed between the regions corresponding to the first substrate layer and the third filter subregion, wherein the depth of the first groove is greater than that of all the first grooves. the depth of the second groove;

The dielectric layer includes a body, a first embedded part and a second embedded part connected to the body, the body is arranged on the side of the first substrate layer away from the color filter layer, and the first embedded part is located on the side of the first substrate layer away from the color filter layer. in the first groove, the second embedded portion is located in the second groove;

A first floating diffusion region, a second floating diffusion region and a third floating diffusion region, the first floating diffusion region, the second floating diffusion region and the third floating diffusion region are formed in the first floating diffusion region. in a substrate layer;

a first gate, a second gate and a third gate, the first gate, the second gate and the third gate are arranged in the dielectric layer and are connected with the first lining the bottom layer is connected, the first gate is located between the first photoelectric conversion region and the first floating diffusion region, and the second gate is located between the second photoelectric conversion region and the second floating diffusion region between the floating diffusion regions, the third gate is located between the third photoelectric conversion region and the third floating diffusion region, and the first gate is used to control the voltage in the first photoelectric conversion region Charges are transferred to the first floating diffusion region, the second gate is used for controlling the transfer of charges in the second photoelectric conversion region to the second floating diffusion region, and the third gate is used for controlling the transfer of charges in the third photoelectric conversion region to the third floating diffusion region;

The second substrate layer is disposed on one side of the dielectric layer.
The image sensor according to any one of claims 1, wherein the light in the first wavelength band is visible light in a short wavelength band.
The image sensor according to claim 2, wherein the light in the second wavelength band is visible light in the middle wavelength band.
The image sensor according to claim 3, wherein the pixel array is a Bayer array.
The image sensor according to any one of claims 1-4, wherein the first gate comprises a first part, a second part and a third part, and the first part is located in the first groove Externally, the upper end of the second part is connected to one end of the first part, the second part is in contact with the inner wall of the first groove, and the lower end of the second part is connected to the third part one end of the first groove is connected, and the third part is located at the bottom of the first groove.
6. The image sensor of claim 5, wherein the second gate comprises a fourth part, a fifth part and a sixth part, the fourth part is located outside the second groove, the The upper end of the fifth part is in contact with one end of the fourth part, the fifth part is in contact with the inner wall of the second groove, and the lower end of the fifth part is in contact with one end of the sixth part Then, the sixth part is located at the bottom of the second groove.
The image sensor according to any one of claims 1-4 or 6, wherein the image sensor is a backside illuminated image sensor or a stacked image sensor.
An imaging device, characterized in that the imaging device comprises the image sensor, circuit board, cut-off filter, lens assembly and bracket body according to any one of claims 1-4 or 6;

The lens assembly is disposed above the cut-off filter through the bracket body, and is used for condensing incident light;

The cut-off filter is arranged above the image sensor through the support body, and is used for filtering the incident light that is converged by the lens assembly;

The image sensor is arranged on the circuit board and is used for receiving the light filtered by the cut-off filter and performing imaging.
The imaging device according to claim 8, wherein the cut-off filter is a 700 nm cut-off filter, and the cut-off filter is used to filter out light with a wavelength of 700 nm or more.
A terminal device, characterized in that the terminal device comprises the image sensor according to any one of claims 1-4 or 6 or the imaging device according to any one of claims 8-9.
A manufacturing method of a CMOS image sensor, wherein the image sensor is the image sensor according to any one of claims 1-4 or 6, and the manufacturing method of the image sensor comprises:

A first groove corresponding to the first pixel and a second groove corresponding to the second pixel extending toward the first substrate layer are formed on the front surface of the first substrate layer, and the depth of the first groove is greater than that of the second groove. depth;

forming a first gate corresponding to the first pixel, a second gate corresponding to the second pixel, and a third gate corresponding to the third pixel on the front surface of the first substrate layer;

Ions are implanted into the first substrate layer on both sides of the first gate to form a first photoelectric conversion region and a first floating diffusion region, and ions are implanted into the first substrate layer on both sides of the second gate to form a second photoelectric conversion region and the second floating diffusion region, and implanting ions into the first substrate layer on both sides of the third gate to form a third photoelectric conversion region and a third floating diffusion region, wherein the first photoelectric conversion region is in the same relationship with the first groove. Correspondingly, the second photoelectric conversion region corresponds to the second groove;

depositing a dielectric material to cover the front surface of the first substrate layer to form a first sub-layer of the dielectric layer, and smoothing the surface of the first sub-layer;

forming a metal interconnect layer;

bonding the second substrate layer to the metal interconnect layer and thinning the backside of the first substrate layer;

An anti-reflection layer, a color filter layer and a microlens are formed on the back surface of the first substrate layer 130 .
The method for fabricating an image sensor according to claim 11, wherein the first gate corresponding to the first pixel and the second gate corresponding to the second pixel are formed on the front surface of the first substrate layer The pole and the third gate corresponding to the third pixel include:

The first gate includes a first part, a second part and a third part, the first part is located outside the first groove, and the upper end of the second part is connected to one end of the first part, so The second part is in contact with the inner wall of the first groove, the lower end of the second part is connected to one end of the third part, and the third part is located at the bottom of the first groove;

The second gate includes a fourth part, a fifth part and a sixth part, the fourth part is located outside the second groove, and the upper end of the fifth part is opposite to one end of the fourth part Then, the fifth part is attached to the inner wall of the second groove, the lower end of the fifth part is connected to one end of the sixth part, and the sixth part is located in the second groove bottom of.