Cmos image sensor and method of manufacturing the same
By employing a vertically stacked three-layer N/P+/N doped photodiode in a CMOS image sensor, the capacity and collection efficiency problems of traditional photodiode structures in CMOS image sensors are solved, achieving a wider dynamic range and higher quantum efficiency.
Patent Information
- Application Number
- CN202611185048.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-05
- Publication Date
- 2026-11-17
AI Technical Summary
In the photodiode structure of traditional CMOS image sensors, the potential well capacity is limited, photogenerated electrons overflow, collection efficiency is low, and dark current is large, resulting in a narrow dynamic range and limited imaging performance.
A photodiode employing a vertically stacked three-layer N/P+/N doped structure forms a multi-layer reverse-biased PN junction, realizing a vertical charge storage channel, increasing the volume of the vertical depletion region, separating photogenerated electron-hole pairs, and reducing dark current.
It significantly improves the full-well capacity and charge collection efficiency of pixels, broadens the dynamic range of the sensor, improves quantum efficiency in the visible and near-infrared bands, and reduces dark current.
Smart Images

Figure CN123124480A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of semiconductor manufacturing technology, and in particular to a CMOS image sensor and its manufacturing method. Background Technology
[0002] A typical CMOS image sensor has a 4T pixel unit, such as Figure 1 As shown, the sensor includes a photodiode (PD), a transmission transistor (TG), a reset transistor (RST), a source follower transistor (SF), a row select transistor (SEL), and a floating drain (FD). The photodiode (PD) is connected to the floating drain (FD) via the transmission transistor (TG). One node of the floating drain (FD) is connected to one end of the source / drain of the reset transistor (RST), and the other end is connected to the gate of the source follower transistor (SF). The source / drain of the source follower transistor (SF) is connected to the operating power supply (VDD) and the row select transistor (SEL), respectively. The row select transistor (SEL) outputs the pixel unit readout signal. Among them, the photodiode (PD) is the core device for the pixel to realize photoelectric conversion and store photogenerated charge, directly determining the sensor's full-well capacity, quantum efficiency, charge collection efficiency, dark current, and other key photoelectric properties.
[0003] Traditional photodiode (PD) structures are formed by a single N-type ion implantation to create a monolayer N-type charge collection trap, as shown in the diagram. Figure 2 As shown. This single-layer N-type depletion region potential well has a limited volume and the following drawbacks: (i) The potential well capacity (full well charge capacity FWC) is limited. Under strong light conditions, photogenerated electrons are very easy to overflow, resulting in high light saturation, image trailing and high light blooming defects, and the dynamic range of the sensor is low. (ii) The collection path of photogenerated electrons in silicon is short and the depletion region is not deep enough. The electron-hole pairs excited by deep photons are not completely separated, resulting in poor charge collection efficiency and low weak photon efficiency.
[0004] (iii) Silicon surface has many interface defects, resulting in significant dark current and image lag, which restricts the improvement of imaging performance of CMOS image sensors. Summary of the Invention
[0005] The technical problem to be solved by the present invention is to provide a CMOS image sensor and a method for manufacturing the same, which can increase the full-well capacity of pixels, improve charge collection efficiency, and broaden the dynamic range of the sensor.
[0006] To solve the above-mentioned technical problems, the CMOS image sensor provided by the present invention includes a photodiode PD, a transmission transistor TG, a reset transistor RST, a source follower transistor SF, a row select transistor SEL, and a floating drain FD in its pixel unit. The photodiode PD adopts a vertical three-layer N / P+ / N stacked doped structure; The stacked doped structure includes a bottom N-doped layer 103, a middle P+ doped layer 104 and an upper N-doped layer 105 formed in the P-doped silicon epitaxial layer 101 of the pixel unit region, which are stacked sequentially from bottom to top. The stacked doped structure has a right N-doped region 107 adjacent to its right side. The left part of the TG polysilicon gate 109 is located directly above the right part of the right N-doped region 107, and the right part of the TG polysilicon gate 109 is located to the right of the right N-doped region 107. A gate dielectric layer 108 is formed between the transport transistor TG polysilicon gate 109, the right N-doped region 107, and the P-doped silicon epitaxial layer 101. An N+-doped floating drain FD is formed on the silicon epitaxial layer 101 of the pixel cell region to the right of the polysilicon gate 109 of the transmission tube TG.
[0007] Preferably, a P+ doped pinning layer 110 is formed on the upper surface of the upper N-doped layer 105 and the upper surface of the left part of the right N-doped region 107.
[0008] Preferably, the stacked doped structure has a left N-doped region 106 adjacent to its left side.
[0009] A P+ doped pinning layer 110 is formed on the upper surface of the upper N-doped layer 105, the upper surface of the left N-doped region 106, and the upper surface of the left part of the right N-doped region 107.
[0010] Preferably, the bottom of the left N-doped region 106 and the right N-doped region 107 are at the same position as the bottom of the bottom N-doped layer 103.
[0011] Preferably, the lateral width of the left N-doped region 106 is smaller than the lateral width of the right N-doped region 107.
[0012] Preferably, the intermediate P+ doped layer 104 and the P+ doped pinning layer 110 are doped with boron, with a concentration greater than 1×10¹. 8 atom / cm³; The floating drain 112 is doped with arsenic at a concentration greater than 1 × 10¹ 8 atom / cm³.
[0013] Preferably, the gate dielectric layer 108 is silicon oxide.
[0014] To solve the above-mentioned technical problems, the present invention provides a method for manufacturing the CMOS image sensor, which includes the following steps: S1. Ion implantation is performed in the preset region of the photodiode PD in the pixel unit region of the P-doped silicon epitaxial layer 101 to form a bottom N-doped layer 103, a middle P+ doped layer 104 and an upper N-doped layer 105 stacked from bottom to top, forming a vertical three-layer N / P+ / N stacked doped structure. S2. N-type ion implantation is performed on the silicon epitaxial layer 101 of the pixel unit region on both sides of the stacked doped structure to form a right N-doped region 107 adjacent to the right side of the stacked doped structure. S3. Sequentially deposit the gate dielectric layer 108 and the polysilicon layer, and then photolithographically etch the polysilicon layer to form the transport transistor TG polysilicon gate 109. The left part of the transport transistor TG polysilicon gate 109 is located directly above the right part of the right N-doped region 107, and the right part of the transport transistor TG polysilicon gate 109 is located to the right of the right N-doped region 107. S4. N+ doped ions are implanted into the silicon epitaxial layer 101 of the pixel unit region to the right of the polysilicon gate 109 of the transmission tube TG to form an N+ doped floating drain FD. S5. Proceed with subsequent processes to complete the manufacturing of the CMOS image sensor.
[0015] Preferably, in step S4, P+ doping is implanted into the upper part of the N-doped layer 105 and the upper left part of the right N-doped region 107 to form a P+ doped pinning layer 110.
[0016] Preferably, in step S2, N-type ion implantation is performed on the silicon epitaxial layer 101 of the pixel unit region on the left and right sides of the stacked doped structure to form a left N-doped region 106 and a right N-doped region 107 adjacent to the left and right sides of the stacked doped structure. In step S4, P+ doping is implanted in the upper part of the left N-doped region 106, the upper part of the upper N-doped layer 105, and the upper left part of the right N-doped region 107 to form a P+ doped pinning layer 110.
[0017] Preferably, the bottom of the left N-doped region 106 and the right N-doped region 107 are at the same position as the bottom of the bottom N-doped layer 103; The lateral width of the left N-doped region 106 is smaller than that of the right N-doped region 107; The intermediate P+ doped layer 104 and the P+ doped pinning layer 110 are doped with boron at a concentration greater than 1 × 10¹. 8 atom / cm³; The floating drain 112 is doped with arsenic at a concentration greater than 1 × 10¹ 8 atom / cm³.
[0018] Preferably, in step S1, the thickness of the upper N-doped layer 105 is greater than the thickness of the bottom N-doped layer 103 and the thickness of the intermediate P-doped layer 104.
[0019] The CMOS image sensor of this invention has a vertically stacked three-layer N / P+ / N doped structure for its pixel unit photodiode PD. Multiple reverse-biased PN junctions are formed within the P-doped silicon epitaxial layer 101, realizing a vertical charge storage channel and forming a basic vertical depletion region. This significantly expands the volume of the vertical depletion region, allowing photogenerated electrons to be captured and stored by the multiple N-wells, significantly improving the potential well storage capacity and increasing the full-well capacity of the pixel. This effectively suppresses high light saturation and charge overflow tailing, broadening the sensor's dynamic range. The vertically multi-layered built-in electric field separates photogenerated electron-hole pairs, allowing carriers excited by photons deep within the silicon to be rapidly collected by the N-type wells, improving charge collection efficiency and significantly enhancing quantum efficiency in the visible and near-infrared bands. Attached Figure Description
[0020] Figure 1 This is a schematic diagram of a typical 4T pixel unit circuit in a CMOS image sensor. Figure 2 A schematic diagram of the cross-sectional structure of a traditional CMOS image sensor photodiode (PD). Figure 3 This is a schematic cross-sectional view of the N / P+ / N stacked doped structure formed by ion implantation in the manufacturing method of the CMOS image sensor of the present invention. Figure 4 This is a schematic diagram of the cross-sectional structure of the right N-doped region formed by ion implantation in the manufacturing method of the CMOS image sensor of the present invention. Figure 5 This is a schematic diagram of the cross-sectional structure of the left N-doped region formed by ion implantation in the manufacturing method of the CMOS image sensor of the present invention. Figure 6 This is a schematic diagram of the cross-sectional structure of the polysilicon gate forming transmission tube in the manufacturing method of the CMOS image sensor of the present invention. Figure 7 This is a schematic diagram of the cross-sectional structure of the N+-doped floating drain electrode in the manufacturing method of the CMOS image sensor of the present invention.
[0021] Explanation of reference numerals in the attached figures: 101 P-doped silicon epitaxy; 103 Bottom N-doped layer; 104 Middle P+ doped layer; 105 Top N-doped layer; 106 Left N-doped region; 107 Right N-doped region; 108 Gate dielectric layer; 109 Polysilicon gate for transport transistor; 110 P+ doped pinning layer. Detailed Implementation
[0022] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention.
[0023] It should be noted that the illustrations provided in this embodiment are only schematic representations of the basic concept of the present invention. Although the illustrations only show components related to the present invention and are not drawn according to the actual number, shape and size of the components, the shape, quantity and proportion of each component can be arbitrarily changed in actual implementation, and the layout of the components may also be more complex. Example 1
[0024] A CMOS image sensor such as Figure 7 As shown, its pixel unit includes a photodiode PD, a transmission transistor TG, a reset transistor RST, a source follower transistor SF, a row select transistor SEL, and a floating drain FD. The photodiode PD adopts a vertical three-layer N / P+ / N stacked doped structure; The stacked doped structure includes a bottom N-doped layer 103, a middle P+ doped layer 104 and an upper N-doped layer 105 formed in the P-doped silicon epitaxial layer 101 of the pixel unit region, which are stacked sequentially from bottom to top. The stacked doped structure has a right N-doped region 107 adjacent to its right side. The left part of the TG polysilicon gate 109 is located directly above the right part of the right N-doped region 107, and the right part of the TG polysilicon gate 109 is located to the right of the right N-doped region 107. A gate dielectric layer 108 is formed between the transport transistor TG polysilicon gate 109, the right N-doped region 107, and the P-doped silicon epitaxial layer 101. An N+-doped floating drain FD is formed on the silicon epitaxial layer 101 of the pixel cell region to the right of the polysilicon gate 109 of the transmission tube TG.
[0025] The CMOS image sensor in Example 1 features a vertically stacked three-layer N / P+ / N doped structure for its pixel photodiodes (PDs). Multiple reverse-biased PN junctions are formed within the P-doped silicon epitaxial layer 101, creating a vertical charge storage channel and forming a basic vertical depletion region. This significantly expands the volume of the vertical depletion region, allowing photogenerated electrons to be captured and stored in the multiple N-wells, resulting in a significant increase in potential well storage capacity and pixel full-well capacity. This effectively suppresses high light saturation and charge overflow tailing, broadening the sensor's dynamic range. The built-in electric fields in the vertical layers separate photogenerated electron-hole pairs, allowing carriers excited by photons deep within the silicon to be rapidly collected by the N-type wells, improving charge collection efficiency and significantly enhancing quantum efficiency in the visible and near-infrared bands. The right N-doped region 107 on the side of the photodiode PD forms a lateral N-doped channel, ensuring complete charge transfer when the transmission transistor TG is turned on. Example 2
[0026] Based on the CMOS image sensor of Embodiment 1, a P+ doped pinning layer 110 is formed on the upper surface of the upper N-doped layer 105 and the upper surface of the left part of the right N-doped region 107.
[0027] Preferably, the stacked doped structure has a left N-doped region 106 adjacent to its left side.
[0028] A P+ doped pinning layer 110 is formed on the upper surface of the upper N-doped layer 105, the upper surface of the left N-doped region 106, and the upper surface of the left part of the right N-doped region 107.
[0029] The bottom of the left N-doped region 106 and the right N-doped region 107 are at the same position as the bottom of the bottom N-doped layer 103.
[0030] Preferably, the lateral width of the left N-doped region 106 is smaller than the lateral width of the right N-doped region 107.
[0031] Preferably, the intermediate P+ doped layer 104 and the P+ doped pinning layer 110 are doped with boron, with a concentration greater than 1×10¹. 8 atom / cm³.
[0032] Preferably, the floating drain 112 is doped with arsenic at a concentration greater than 1 × 10¹ 8 atom / cm³.
[0033] Preferably, the gate dielectric layer 108 is silicon oxide.
[0034] In the CMOS image sensor of Example 2, the complete P+ doped pinning layer 110 on the top of the photodiode PD covers the silicon epitaxial surface, passivates the silicon surface, suppresses dark current, and can significantly reduce pixel dark current and imaging hysteresis defects; the left N doped region 106 on the side of the photodiode PD can widen the lateral range of the depletion region and make full use of the pixel photosensitive area. Example 3
[0035] A method for manufacturing a CMOS image sensor according to Embodiment 1 or Embodiment 2 includes the following steps: S1. Ion implantation is performed on the preset region of the photodiode PD in the pixel unit region of the P-doped silicon epitaxial layer 101 to form a bottom N-doped layer 103, a middle P+ doped layer 104, and an upper N-doped layer 105 stacked sequentially from bottom to top, forming a vertical three-layer N / P+ / N stacked doped structure, such as... Figure 3 As shown; S2. N-type ion implantation is performed on the silicon epitaxial layer 101 of the pixel unit region on both sides of the stacked doped structure to form a right N-doped region 107 adjacent to the right side of the stacked doped structure, as shown below. Figure 4 As shown; S3. Sequentially deposit the gate dielectric layer 108 and the polysilicon layer, then photolithographically etch the polysilicon layer to form the transport transistor (TG) polysilicon gate 109. The left part of the TG polysilicon gate 109 is located directly above the right part of the right N-doped region 107, and the right part of the TG polysilicon gate 109 is located to the right of the right N-doped region 107. Figure 6 As shown; S4. N+ doped ion implantation is performed on the silicon epitaxial layer 101 in the pixel unit region to the right of the polysilicon gate 109 of the transmission transistor TG to form an N+ doped floating drain FD, as shown. Figure 7 As shown; S5. Proceed with subsequent processes to complete the manufacturing of the CMOS image sensor.
[0036] The CMOS image sensor manufacturing method in Example 3 only adds a few ion implantation steps, is fully compatible with existing standard CMOS image sensor manufacturing processes, requires no new special equipment, has controllable mass production costs, and is compatible with various pixel process platforms such as front-illuminated and back-illuminated sensors. Example 4
[0037] In the manufacturing method of the CMOS image sensor based on Embodiment 3, in step S4, P+ doping is implanted into the upper part of the N-doped layer 105 and the upper left part of the right N-doped region 107 to form a P+ doped pinning layer 110.
[0038] Preferably, in step S2, N-type ion implantation is performed on the silicon epitaxial layer 101 of the pixel unit region on both sides of the stacked doped structure to form a left N-doped region 106 and a right N-doped region 107 adjacent to the left and right sides of the stacked doped structure, as shown below. Figure 5 As shown; In step S4, P+ doping is implanted in the upper part of the left N-doped region 106, the upper part of the upper N-doped layer 105, and the upper left part of the right N-doped region 107 to form a P+ doped pinning layer 110.
[0039] Preferably, in step S1, the thickness of the upper N-doped layer 105 is greater than the thickness of the bottom N-doped layer 103 and the thickness of the intermediate P-doped layer 104.
[0040] Preferably, the bottom of the left N-doped region 106 and the right N-doped region 107 are at the same position as the bottom of the bottom N-doped layer 103; The lateral width of the left N-doped region 106 is smaller than that of the right N-doped region 107; The intermediate P+ doped layer 104 and the P+ doped pinning layer 110 are doped with boron at a concentration greater than 1 × 10¹. 8 atom / cm³; The floating drain 112 is doped with arsenic at a concentration greater than 1 × 10¹ 8 atom / cm³.
[0041] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. A CMOS image sensor, characterized in that, Its pixel unit includes a photodiode PD, a transmission transistor TG, a reset transistor RST, a source follower transistor SF, a row select transistor SEL, and a floating drain FD; The photodiode PD adopts a vertical three-layer N / P+ / N stacked doped structure; The stacked doped structure includes a bottom N-doped layer 103, a middle P+ doped layer 104 and an upper N-doped layer 105 formed in the P-doped silicon epitaxial layer 101 of the pixel unit region, which are stacked sequentially from bottom to top. The stacked doped structure has a right N-doped region 107 adjacent to its right side. The left part of the TG polysilicon gate 109 is located directly above the right part of the right N-doped region 107, and the right part of the TG polysilicon gate 109 is located to the right of the right N-doped region 107. A gate dielectric layer 108 is formed between the transport transistor TG polysilicon gate 109, the right N-doped region 107, and the P-doped silicon epitaxial layer 101. An N+-doped floating drain FD is formed on the silicon epitaxial layer 101 of the pixel cell region to the right of the polysilicon gate 109 of the transmission tube TG.
2. The CMOS image sensor according to claim 1, characterized in that, A P+ doped pinning layer 110 is formed on the upper surface of the upper N-doped layer 105 and the upper surface of the left part of the right N-doped region 107.
3. The CMOS image sensor according to claim 2, characterized in that, The stacked doped structure has a left N-doped region 106 adjacent to it on the left side. A P+ doped pinning layer 110 is formed on the upper surface of the upper N-doped layer 105, the upper surface of the left N-doped region 106, and the upper surface of the left part of the right N-doped region 107.
4. The CMOS image sensor according to claim 3, characterized in that, The bottom of the left N-doped region 106 and the right N-doped region 107 are at the same position as the bottom of the bottom N-doped layer 103.
5. The CMOS image sensor according to claim 3, characterized in that, The lateral width of the left N-doped region 106 is smaller than that of the right N-doped region 107.
6. The CMOS image sensor according to claim 2, characterized in that, The doping of the intermediate P+ doped layer 104 and the P+ doped pinning layer 110 is boron with a concentration greater than 1 x 1018 atoms / cm3. 8 atoms / cm3; The doping of the floating drain 112 is arsenic, at a concentration greater than 1 x 1018 8 atoms / cm3.
7. The CMOS image sensor according to claim 1, characterized in that, The gate dielectric layer 108 is silicon oxide.
8. A method for manufacturing a CMOS image sensor according to claim 1, characterized in that, Includes the following steps: S1. Ion implantation is performed in the preset region of the photodiode PD in the pixel unit region of the P-doped silicon epitaxial layer 101 to form a bottom N-doped layer 103, a middle P+ doped layer 104 and an upper N-doped layer 105 stacked from bottom to top, forming a vertical three-layer N / P+ / N stacked doped structure. S2. N-type ion implantation is performed on the silicon epitaxial layer 101 of the pixel unit region on both sides of the stacked doped structure to form a right N-doped region 107 adjacent to the right side of the stacked doped structure. S3. Sequentially deposit the gate dielectric layer 108 and the polysilicon layer, and then photolithographically etch the polysilicon layer to form the transport transistor TG polysilicon gate 109. The left part of the transport transistor TG polysilicon gate 109 is located directly above the right part of the right N-doped region 107, and the right part of the transport transistor TG polysilicon gate 109 is located to the right of the right N-doped region 107. S4. N+ doped ions are implanted into the silicon epitaxial layer 101 of the pixel unit region to the right of the polysilicon gate 109 of the transmission tube TG to form an N+ doped floating drain FD. S5. Proceed with subsequent processes to complete the manufacturing of the CMOS image sensor.
9. The method for manufacturing a CMOS image sensor according to claim 8, characterized in that, In step S4, P+ doping is implanted into the upper part of the N-doped layer 105 and the upper left part of the right N-doped region 107 to form a P+ doped pinning layer 110.
10. The method for manufacturing a CMOS image sensor according to claim 8, characterized in that, In step S2, N-type ion implantation is performed on the silicon epitaxial layer 101 of the pixel unit region on the left and right sides of the stacked doped structure to form the left N-doped region 106 and the right N-doped region 107 adjacent to the left and right sides of the stacked doped structure. In step S4, P+ doping is implanted in the upper part of the left N-doped region 106, the upper part of the upper N-doped layer 105, and the upper left part of the right N-doped region 107 to form a P+ doped pinning layer 110.
11. The method for manufacturing a CMOS image sensor according to claim 10, characterized in that, The bottom of the left N-doped region 106 and the right N-doped region 107 are at the same position as the bottom of the bottom N-doped layer 103; The lateral width of the left N-doped region 106 is smaller than that of the right N-doped region 107; The intermediate P+ doped layer 104 and the P+ doped pinning layer 110 are doped with boron, with a concentration greater than 1×10¹. 8 atom / cm³; The floating drain 112 is doped with arsenic at a concentration greater than 1 × 10¹ 8 atom / cm³.
12. The method for manufacturing a CMOS image sensor according to claim 8, characterized in that, In step S1, the thickness of the upper N-doped layer 105 is greater than the thickness of the bottom N-doped layer 103 and the thickness of the intermediate P-doped layer 104.
