A multi-band SPAD light detection structure for lidar

Abstract

The application discloses a kind of multiband SPAD light detection structures for laser radar, belong to laser radar with light detector field.The light detection structure includes: P substrate layer, BN buried layer, be located on the upper end of the P substrate layer, and with the deep P well layer being located on the upper end thereof, constitute third avalanche junction;Deep P well layer, be located on the upper end of the BN buried layer, and with the second N well layer being located on the upper end close to the end portion thereof, constitute second avalanche junction;First N well layer, be located on the upper end of the deep P well layer, and with the PB layer being located on the upper end close to the end portion thereof, constitute first avalanche junction.The application is integrated three functional differentiation avalanche junction, respectively realize the accurate capture of short wavelength, medium wavelength and long wavelength photon, three junctions work together can achieve the efficient absorption of full-band photon, greatly improve the detection efficiency of device in complex light environment to multiband photon.

Description

Technical Field

[0001] This invention relates to a multi-band SPAD optical detection structure for lidar, belonging to the field of optical detectors for lidar. Background Technology

[0002] A single-photon avalanche diode (SPAD) is an avalanche photodiode operating in Geiger mode, capable of detecting a single incident photon. It boasts extremely high sensitivity and rapid response, making it widely applicable in the lidar field. In lidar, the system's accurate target identification, distance calculation, and complex environment perception capabilities hinge on the targeted detection band and high signal sensitivity of the SPAD. Among these, visible and near-infrared SPADs possess irreplaceable technological value due to their suitability for mainstream lidar applications. This is because the visible light band can accurately capture the geometric contours, surface textures, and detailed features of targets, meeting the needs for target shape discrimination in scenarios such as obstacle recognition in autonomous driving and high-precision terrain mapping. Meanwhile, the near-infrared band, with its strong penetration, low atmospheric attenuation coefficient, and excellent resistance to sunlight interference, is suitable for long-range ranging, detection in harsh weather conditions, and vegetation penetration imaging, effectively overcoming the limitations imposed by fog, haze, and rain on detection performance.

[0003] Currently, most SPADs are single-band SPADs, which can only acquire target information in a single band. If limited to the visible light band, they are difficult to achieve long-range, interference-resistant detection. If relying solely on the near-infrared band, they cannot accurately distinguish the fine shape and attributes of targets, making it difficult to meet the comprehensive requirements of lidar for detection range, recognition accuracy, and environmental adaptability. To overcome the limitations of single-band SPADs, dual-band SPADs have emerged. By integrating visible light and near-infrared dual-band detection data, they achieve the coordinated extraction of target shape features and distance information through multi-dimensional information fusion, significantly improving the target recognition accuracy and environmental adaptability of lidar in complex scenes.

[0004] See the literature "Design and Optimization of Dual-Peak Response CMOS Single-Photon Detector, Zhang Weiyu, Wang Yang, Jin Xiangliang," which introduces a dual-band SPAD. This SPAD achieves a photon detection probability of 32% at 520 nm and 12% at 840 nm, thus enabling the detection of both visible and near-infrared wavelengths. However, this dual-band SPAD still has the following shortcomings: 1. The dual-band detection scheme can only achieve selective detection of photons at two specific wavelengths. In the core application scenarios of lidar, the detection requirement of the 940 nm near-infrared band is dominant. The existing dual-band detection architecture has not yet completed targeted performance optimization for this key band, thus its detection capability is insufficient; 2. Existing SPAD detectors are based on N... +The design of the / PW junction, with its avalanche multiplication region close to the semiconductor surface, is susceptible to the influence of surface states and strong electric fields. This causes the proportion of tunneling current and surface leakage current in the total dark current to increase dramatically. These two types of currents are sensitive to temperature and pressure fluctuations and are difficult to suppress, directly increasing the device's dark count rate (DCR) and severely limiting its application in detection scenarios. 3. Existing detectors are mostly fabricated using non-standard processes, changing the doping concentration or electric field strength through customized process layers, resulting in high fabrication costs. Summary of the Invention

[0005] To address the limitations of current dual-band SPADs, such as limited detection range and insufficient multi-band detection capabilities, and also due to N... + The design of the / PW junction easily leads to problems such as increased dark count rate and high overall fabrication cost. This invention provides a multi-band SPAD optical detection structure for lidar. By constructing a three-avalanche junction architecture, it achieves efficient detection of visible light, mid-band light, and long-wavelength light in three bands. At the same time, a PB layer is introduced to reduce the dark count rate of the device, ultimately achieving a synergistic improvement in the device's multifunctionality and reliability. The technical solution is as follows: A multi-band SPAD optical detection structure for lidar, the optical detection structure comprising: P substrate layer, The BN buried layer is disposed on the upper end of the P substrate layer and together with the deep P well layer disposed on its upper end, forms a third avalanche junction; A deep P-well layer is disposed at the upper end of the BN buried layer, and together with a second N-well layer disposed near the upper end of the BN buried layer, they form a second avalanche knot. The first N-well layer is disposed on the upper end of the deep P-well layer, and together with the PB layer disposed near the upper end of the deep P-well layer, they form the first avalanche knot.

[0006] Furthermore, the upper end of the BN buried layer is provided with a deep N-well layer, which extends upward and is located between the end of the P substrate layer and the deep P-well layer.

[0007] Furthermore, a first N-well is provided at the middle of the upper end of the first N-well layer. ＋ The injection region, the upper middle part of the second N-well layer is provided with a second N-well. + The injection region, wherein a third N-well is provided at the middle of the upper end of the deep N-well layer. + Injection area.

[0008] Furthermore, a first P is provided at the upper center of the PB layer. ＋ In the injection region, a second P well is provided on both sides of the deep P well layer corresponding to the second N well layer. + Injection area, third P + Injection area.

[0009] Furthermore, the optical detection structure also includes a shallow trench isolation region (STI), the first N+ Both sides of the injection area, the second P + Both sides of the injection area, the third P + Shallow channel isolation zones (STIs) are provided on both sides of the injection area.

[0010] Furthermore, the first N + An electrically connected first electrode is provided outside the injection region, and a first lead is connected to the first electrode and led out to the first cathode port Cathode 1; the second N + A second electrode with electrical connection is provided outside the injection region, and a second lead is connected to the second electrode and led out to the second cathode port Cathode 2; the third N + An electrically connected third electrode is provided outside the injection region, and a third lead is connected to the third electrode and led out to the third cathode port Cathode 3; the first P + Injection region, the second N + Injection area, the third P + The injection area is provided with an electrically connected fourth electrode, and the fourth lead connects the fourth electrodes in series and leads them to the anode port.

[0011] Furthermore, the doping concentration of the PB layer is higher than that of the deep P-well layer.

[0012] Furthermore, the doping concentration of the PB layer The doping concentration of the deep P-well layer 103 is... .

[0013] Furthermore, the BN buried layer 101 is a lightly doped N-type region with a doping concentration of [missing information]. .

[0014] Furthermore, the optical detection structure is square or rectangular in shape, and the functional layers on the upper surface are arranged in a U-shape from the center outwards, with the corners of each U-shaped functional layer being smoothly transitioned arcs.

[0015] The beneficial effects of this invention are: This invention discloses a multi-band SPAD photodetector structure for lidar. By integrating three functionally differentiated avalanche junctions, it achieves precise capture of short-wavelength, mid-wavelength, and long-wavelength photons respectively. The three junctions work together to achieve efficient absorption of photons across the entire wavelength range, significantly improving the device's detection efficiency for multi-band photons in complex optical environments. Furthermore, it can be fabricated using standard processes, requiring only P... + Injection layer, N +The core structural layers, such as the implantation layer, N-well region, P-well region, deep P-well region, deep N-well region, PB layer and BN buried layer, can be precisely matched with the existing process system. The fabrication process does not require adjustment of mask layout, photolithography alignment accuracy and ion implantation parameters, nor does it require additional special process steps. While ensuring the integrity of the device structure design, it significantly reduces the process development and mass production costs. Meanwhile, the first avalanche junction in this application introduces a PB layer to regulate the relative position of the first avalanche junction and the shallow trench isolation region (STI), thereby moving the first avalanche junction away from the STI domain and effectively suppressing the migration of defect carriers within the STI to the avalanche multiplication region, thus reducing the device's dark count rate; simultaneously, it incorporates grounding layers for the first to third PB layers. + The injection region can significantly reduce the risk of parasitic NPN effects inside the device; in addition, the use of rounded rectangular layout design to optimize the electric field distribution of the device can effectively alleviate the problem of electric field accumulation at the edge and reduce the probability of device failure caused by electric field concentration. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a top view of the overall structure of an embodiment of the present invention; Figure 2 This is a top view of the overall structure (excluding electrodes and leads) of an embodiment of the present invention; Figure 3 This is the present invention. Figure 2 Sectional view along direction A; Figure 4 This is the present invention. Figure 2 A-direction cross-sectional view (illustration of avalanche knot); Figure 5 This is the present invention. Figure 2 A sectional view along direction A (structural schematic); Figure 6 This is the present invention. Figure 3 Electric field simulation diagram; In the figure: 100, P-substrate layer; 101, BN buried layer; 102, deep N-well layer; 103, deep P-well layer; 104, second N-well layer; 105, first N-well layer; 106, PB layer; 107, third N-well layer. + Injection area; 108, third P + Injection area; 109, second N + Injection area; 110, second P +Injection area; 111, First P + Injection region; 112, First N + Injection area; 113, Shallow trench isolation area (STI); 201, First electrode; 202, First lead; 203, Second electrode; 204, Second lead; 205, Third electrode; 206, Third lead; 207, Fourth electrode; 208, Fourth lead. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0019] Example 1 This embodiment provides a multi-band SPAD optical detection structure for lidar. Unlike existing dual-band SPADs that can only detect visible and near-infrared light, this structure utilizes the synergistic effect of three avalanche junctions to achieve the detection of different wavelengths of light (short wavelength 400~650nm; medium wavelength 650~900nm; long wavelength 900nm~1700nm), exhibiting excellent response performance in complex lighting environments and thus meeting practical application requirements. See also... Figures 1 to 5 The photodetector structure includes: P substrate 100, BN buried layer 101 is disposed on the upper end of P substrate layer 100, and a deep N well layer 102 and a deep P well layer 103 are disposed on its upper end. The deep N well layer 102 is disposed at the end of BN buried layer 101. BN buried layer 101 and deep P well layer 103 form a third avalanche junction. The third avalanche junction is a BN / DPW junction. A deep P-well layer 103 is disposed on the upper end of the BN buried layer 101, and a second N-well layer 104 is disposed near the upper end of the layer. The deep P-well layer 103 and the second N-well layer 104 form a second avalanche junction, which is an N-well / DPW junction. A first N-well layer 105 is disposed on the upper end of the deep P-well layer 103, and a PB layer 106 is disposed near its upper end. The first N-well layer 105 and the PB layer 106 form a first avalanche junction, which is an N-well / PB junction.

[0020] This application utilizes a configuration of three avalanche junctions (a first avalanche junction to a third avalanche junction) to trigger avalanche by photogenerated carriers. These functionally differentiated avalanche junctions respectively achieve precise capture of short-wavelength, medium-wavelength, and long-wavelength photons. The coordinated operation of the three avalanche junctions enables efficient absorption of photons across the entire wavelength range, significantly improving the device's detection efficiency for multi-wavelength photons in complex optical environments.

[0021] Furthermore, a first N-well is provided at the middle of the upper end of the first N-well layer 105. ＋Injection region 112, the upper middle part of the second N-well layer 104 is provided with a second N-well. + Injection region 109, a third N-well is provided at the upper middle part of the deep N-well layer 102. + Injection area 107; The first P is located at the upper center of PB layer 106. ＋ In the injection region 111, a second P well is provided on both sides of the deep P well layer 103 corresponding to the second N well layer 104. + Injection zone 110, third P + Injection area 108.

[0022] In existing technologies, dual-band SPADs suffer from parasitic effects due to the increased structural layers. The SPAD of this application, integrating a three-band design, requires additional functional layers, making the structure more complex. Electric field interference from different doped regions and avalanche regions, as well as bias voltage cross-interference, further induce parasitic effects, making them more difficult to suppress through structural design. In severe cases, this can lead to the device losing its photon detection capability. In this application, the first N... ＋ Injection zone 112, second N ＋ The injection region 109 and the deep P-well layer 103 can form a first parasitic NPN transistor; while the second N... ＋ Injection region 109, deep P-well layer 103 and third N + The injection region 107 can then form a second parasitic NPN transistor. This parasitic NPN transistor causes the device to lose its photon detection capability, therefore a corresponding structure needs to be designed to prevent the parasitic NPN from being activated. Because the second P... + Injection zone 110 and third P + The injection regions 108 are all directly connected to the low potential of the anode (Anode port) and, as heavily doped P-type semiconductor regions, possess extremely low resistivity. They form excellent ohmic contacts with the deep P-well layer 103, constituting a low-resistivity conductive path. Based on the equipotential conduction principle of semiconductor physics, this path can rapidly conduct the accumulated charge in the deep P-well layer 103, thereby forcing its potential to be stabilized at a low potential level consistent with the anode through a potential clamping effect. This suppresses the carrier accumulation effect when the deep P-well layer 103 acts as the base region of a parasitic NPN transistor, avoiding the formation of an effective current amplification path between the base region and the emitter region. Consequently, it prevents problems such as abnormally high device dark count rate, avalanche multiplication process disorder, and decreased device operating stability caused by the parasitic NPN effect, ensuring the performance reliability and consistency of the triple avalanche junction SPAD in multi-band detection processes.

[0023] Furthermore, the optical detection structure also includes a shallow trench isolation region STI 113, the first N + Both sides of injection zone 112, the second P + Both sides of injection zone 110, the third P +Shallow channel isolation regions (STIs) 113 are provided on both sides of the injection region 108. The deployment of shallow channel isolation regions (STIs) can effectively prevent edge breakdown between the anode and cathode, thereby effectively improving the reliability of the device.

[0024] Furthermore, the first N + An electrically connected first electrode 201 is provided outside the injection region 112, and a first lead 202 is connected to the first electrode 201 and led out to the first cathode port Cathode 1; the second N + An electrically connected second electrode 203 is provided outside the injection region 109, and a second lead 204 is connected to the second electrode 203 and led out to the second cathode port Cathode 2; the third N + An electrically connected third electrode 205 is provided outside the injection region 107, and a third lead 206 is connected to the third electrode 205 and led out to the third cathode port Cathode 3; the first P + Injection zone 111, second N + Injection zone 110, third P + The injection region 108 is provided with an electrically connected fourth electrode 207. The fourth lead 208 connects the fourth electrodes 207 in series and leads them out to the anode port Anode.

[0025] The first electrode 201 is connected to the first lead 202 and led out to the first cathode port Cathode1, which is the cathode port. This port Cathode1 is connected to a voltage source as the cathode of the first avalanche junction formed by the PB layer 106 and the first N-well layer 105, and inputs voltage. The second electrode 203 is connected to the second lead 204 and led out to the port Cathode2, which is the cathode port. This port is connected to a voltage source as the cathode of the second avalanche junction formed by the second N-well layer 104 and the deep P-well layer 103, and inputs voltage. The third electrode 205 is connected to the third lead 206 and led out to the port Cathode3, which is the cathode port. This port is connected to a voltage source as the cathode of the third avalanche junction formed by the BN buried layer 101 and the deep N-well layer 102, and inputs voltage. The three fourth electrodes 207 are connected to the port Anode through the fourth lead 208, which serves as the anode port and can also serve as the anode of the first to third avalanche junctions. The ports Cathode1, Cathode2, and Cathode3 can operate independently or simultaneously when powered on. When all three ports are powered on, the voltage at electrode port Cathode1 is lower than that at electrode port Cathode2, and the voltage at electrode port Cathode2 is lower than that at electrode port Cathode3, to ensure that the three avalanche junctions work together without affecting each other.

[0026] Furthermore, PB layer 106 is a highly doped P-type region (doping concentration... The deep P-well layer 103 forms a highly doped single-sided abrupt junction with the first N-well 105, i.e., the first avalanche junction. According to the avalanche breakdown principle, the highly doped region significantly compresses the width of the depletion region, causing a rapid increase in the electric field inside the depletion region under reverse bias. When the electric field strength reaches the avalanche critical electric field of silicon, only a relatively low reverse voltage is needed to trigger the carrier multiplication effect; therefore, the avalanche breakdown voltage of this junction is low. The deep P-well layer 103 is a lightly doped P-type region (doping concentration of 100%). This forms a lightly doped single-sided abrupt junction with the second N-well 104, i.e., the second avalanche junction. The low-doping characteristic widens the depletion region; under reverse bias, a higher voltage is required for the depletion region electric field to reach the avalanche threshold. Simultaneously, the probability of carrier collisional ionization in the low-doped region is lower, further increasing the threshold voltage for triggering the avalanche effect. Therefore, the avalanche breakdown voltage of this junction is significantly higher than that of the PB / N-well junction. The BN buried layer 101 is a lightly doped N-type region (doping concentration of...). The deep P-well layer 103 is a low-doped P-type region. The reverse PN junction formed by the two is a low-doped single-sided abrupt junction, namely the third avalanche junction. The depletion region of the reverse PN junction formed by the BN buried layer 101 and the deep P-well region 103 mainly extends towards the low-doped deep P-well region 103. A higher reverse voltage needs to be applied to make the electric field of the depletion region reach the avalanche critical electric field of silicon material. At the same time, the avalanche junction formed by the deep P-well region 103 and the BN buried layer 101 has a lower carrier density and a lower probability of carrier collision ionization compared to the other two avalanche junctions, which further increases the avalanche triggering threshold. Therefore, the breakdown voltage of this junction is relatively the highest.

[0027] This design employs a triple avalanche junction structure: N-well / PB, BN / DPW, and N-well / DPW. The first avalanche junction (N-well / PB) is a shallow junction structure. The PB layer's high 106 doping characteristic results in a narrow depletion region that is close to the device surface. 520nm visible light photons have high energy and shallow penetration depth, allowing for efficient absorption in the shallow depletion region. The excited electron-hole pairs rapidly avalanche multiply under the strong electric field of the junction, forming a detection signal in the visible light band. The second avalanche junction (N-well / DPW) is a medium-deep junction structure, suitable for 840nm mid-wave near-infrared photons. The penetration depth of photons in this band is such that they can penetrate to the middle layer of the device and be captured by the depletion region of the N-well / DPW junction. The charge carriers are accelerated and collide in the electric field of the junction region, triggering the avalanche effect to achieve the response in the mid-wave near-infrared band. The third avalanche junction (BN / DPW) is a deep junction structure. The low doping characteristics of the deep P-well layer 103 greatly widen the width of the depletion region and extend it to the deep layer of the device. The 940nm long-wave near-infrared photons have the greatest penetration depth and can reach the depletion region of the BN / DPW junction in the deep layer of the device. The charge carriers generated after the photons are absorbed are avalanche multiplied under the strong electric field driven by the high breakdown voltage, forming the detection signal in the long-wave near-infrared band. In this application, the BN buried layer 101 forms a carrier barrier in the vertical dimension of the device due to its heavy doping characteristics, which can effectively block the leakage current and noise carriers in the substrate (i.e., P substrate layer 100) from diffusing into the avalanche multiplication region. Meanwhile, the deep N-well layer 102 cuts off the lateral migration path of carriers between pixels in the horizontal dimension. The synergistic effect of the two can achieve all-round isolation and protection in both the vertical and horizontal dimensions. This not only greatly suppresses the problem of dark count rate increase caused by substrate leakage current, but also significantly weakens the interference of substrate noise on avalanche signals. At the same time, it can also effectively block carrier crosstalk between adjacent pixels, improve the spatial resolution and detection accuracy of the array device, and ensure the performance stability and consistency of the triple avalanche junction SPAD in complex working environments.

[0028] Furthermore, the photodetector structure of this application is a square or rectangular structure with an overall size of 20μm×20μm, with the first N + Centered on injection zone 112, from the inside out on the same planar structure, are the first shallow trench isolation zone STI 113, PB layer 106, the second shallow trench isolation zone STI 113, and the second PB layer 106. + Injection region 110, third shallow trench isolation region STI 113, second N-well layer 104, fourth trench isolation region STI 113, third P + The injection region 108, the fifth channel isolation region STI 113, and the deep N-well layer 102 are arranged in a U-shape, and the corners of each device are all smoothly transitioned arcs.

[0029] Meanwhile, this application adopts a rounded rectangular overall layout structure to replace the traditional right-angled rectangular design. The smooth transition of the arc edge effectively disperses the electric field distribution inside the device, avoids the electric field concentration effect caused by the abrupt change in geometric shape at the right-angle corner, and greatly reduces the electric field intensity in the edge area. This avoids problems such as premature avalanche breakdown and tunneling dark current surge caused by excessive local electric field, significantly reduces the risk of permanent failure caused by the accumulation of electric field at the edge, and ensures the stability and long-term reliability of the device under high reverse bias operating conditions.

[0030] refer to Figure 6 The image shows a simulation of the electric field intensity of this structure. From top to bottom, the electric field distributions of the first, second, and third avalanche junctions are shown. These three junctions are spatially layered, corresponding to the detection of short-wavelength, mid-wavelength, and long-wavelength photons, respectively. The first avalanche junction is a shallow junction structure, with a high electric field concentrated on the surface of the device, adapting to the shallow penetration characteristics of short-wavelength photons, and can efficiently absorb visible light and trigger avalanches. The second avalanche junction is a medium-deep junction structure, with the electric field distributed in the middle layer of the device, adapting to the penetration depth of mid-wavelength near-infrared photons, achieving accurate detection in the mid-wavelength band. The third avalanche junction is a deep junction structure, with a wide depletion region extending to the deep layers of the device, which can fully capture long-wavelength near-infrared photons and complete avalanche multiplication. The three junctions operate independently, with no crosstalk between their electric fields, verifying the layered detection capability of this structure for three wavelengths of photons, meeting the application requirements of multi-band fusion sensing in lidar.

[0031] In summary, in this application, the first avalanche junction, compared to the P in the literature "CMOS SPAD Based on Photo-Carrier Diffusion Achieving PDP>40% From 440 to 580 nm at 4 V Excess Bias, Chokalingam Veerappan" + In this application, due to the introduction of the PB layer 106, the first avalanche junction formed with the first N-well layer 105 is relatively deeper, thus the first avalanche junction is farther away from the first N-well layer 105. + Injection region 112 and the first P + The shallow channel isolation region STI 113 between the injection regions 111 can effectively reduce the dark count rate of SPAD caused by defects in the shallow channel isolation region STI 113, and can effectively absorb short-wavelength photons; the second avalanche junction is used to absorb medium-wavelength photons; and the third avalanche junction is used to absorb long-wavelength photons. Therefore, the three work together to achieve the absorption of multi-wavelength photons. The SPAD device integrating three avalanche junctions can achieve efficient capture of photons across the entire wavelength range, and greatly improve the detection efficiency of the device for multi-wavelength photons in complex optical environments.

[0032] 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 protection scope of the present invention.

Claims

1. A multi-band SPAD optical detection structure for lidar, characterized in that, The optical detection structure includes: P substrate layer, The BN buried layer is disposed on the upper end of the P substrate layer and together with the deep P well layer disposed on the upper end of the substrate layer, forms a third avalanche junction. A deep P-well layer is disposed at the upper end of the BN buried layer, and together with a second N-well layer disposed near the upper end of the BN buried layer, they form a second avalanche knot. The first N-well layer is disposed on the upper end of the deep P-well layer, and together with the PB layer disposed near the upper end of the deep P-well layer, they form the first avalanche knot.

2. The photodetector structure according to claim 1, characterized in that, The upper end of the BN buried layer is further provided with a deep N-well layer, which extends upward and is located between the end of the P substrate layer and the deep P-well layer.

3. The photodetector structure according to claim 1, characterized in that, The first N-well layer has a first N-well in the middle of its upper end. ＋ The injection region, the upper middle part of the second N-well layer is provided with a second N-well. + The injection region, wherein a third N-well is provided at the middle of the upper end of the deep N-well layer. + Injection area.

4. The photodetector structure according to claim 1, characterized in that, The upper middle part of the PB layer is provided with a first P ＋ In the injection region, a second P well is provided on both sides of the deep P well layer corresponding to the second N well layer. + Injection area, third P + Injection area.

5. The photodetector structure according to claim 1, characterized in that, The optical detection structure also includes a shallow trench isolation region (STI), the first N + Both sides of the injection area, the second P + Both sides of the injection area, the third P + Shallow channel isolation zones (STIs) are provided on both sides of the injection area.

6. The photodetector structure according to claim 1, characterized in that, The first N + An electrically connected first electrode is provided outside the injection region, and a first lead is connected to the first electrode and led out to the first cathode port Cathode 1; the second N + A second electrode with electrical connection is provided outside the injection region, and a second lead is connected to the second electrode and led out to the second cathode port Cathode 2; the third N + An electrically connected third electrode is provided outside the injection region, and a third lead is connected to the third electrode and led out to the third cathode port Cathode 3; the first P + Injection region, the second N + Injection area, the third P + The injection area is provided with an electrically connected fourth electrode, and the fourth lead connects the fourth electrodes in series and leads them to the anode port.

7. The photodetector structure according to claim 1, characterized in that, The doping concentration of the PB layer is higher than that of the deep P-well layer.

8. The photodetector structure according to claim 1, characterized in that, Doping concentration of the PB layer The doping concentration of the deep P-well layer 103 is... .

9. The photodetector structure according to claim 1, characterized in that... The BN buried layer 101 is a lightly doped N-type region with a doping concentration of [missing information]. .

10. The photodetector structure according to claim 1, characterized in that, The optical detection structure is square or rectangular in shape. The functional layers on the upper surface are arranged in a U-shape from the center outwards, and the corners of each U-shaped functional layer are smoothly transitioned arcs.

Images