Single photon avalanche diode array, receiving sensor and laser radar

Abstract

The invention belongs to the technical field of optical devices, and provides a single photon avalanche diode array, a receiving sensor and a laser radar, at least two SPAD units are arranged in an array, a microlens converges incident light to the corresponding SPAD unit, a back metal grid connects the SPAD unit and a corresponding external electrode, a first dielectric layer is arranged between the microlens and the SPAD unit, and a second dielectric layer is arranged between the microlens and the SPAD unit. The second dielectric layer is arranged between the SPAD units and the front metal wiring layer, and the front metal wiring layer is electrically connected with the corresponding SPAD units through contact metal wires. The deep groove isolation columns are arranged between the adjacent SPAD units, and the metal filling structures are arranged between the deep groove isolation columns and the back metal grids, so that self-excitation photons generated when the SPAD units are excited by incident light can be isolated, the self-excitation photons are prevented from entering the adjacent SPAD units through grid gaps, the probability of photon crosstalk is reduced, and the optical crosstalk of the device is reduced.

Description

[0001] This application is a divisional application. The original application has the application number 202311637529.X and the application date is November 30, 2023. The entire contents of the original application are incorporated herein by reference. Technical Field

[0002] This application belongs to the field of optical device technology, and in particular relates to a single-photon avalanche diode array, a receiving sensor, and a lidar. Background Technology

[0003] Single-photon avalanche diode arrays (SPADs) have been widely used in laser detection and ranging systems (Lidar), autonomous driving, fluorescence lifetime imaging, quantum communication, and biomedical imaging due to their advantages such as high sensitivity detection capability for intensities down to single photons, high time resolution, and strong anti-interference capability. These SPADs, consisting of at least two connected in parallel as a single-point silicon photomultiplier tube (SiPM), are used as a single point. With advancements in silicon manufacturing technology, the design of SPAD arrays using silicon as the absorption and avalanche material has become a hot topic in industry. Furthermore, by utilizing various CMOS manufacturing technologies, SPADs can be easily integrated with different quenching circuits and readout circuits, thus offering significant technological advantages.

[0004] Single-photon avalanche diode arrays can be combined in circuits to form two typical application forms. The first is the silicon photomultiplier tube (SiPM), which is composed of at least two single-photon avalanche diode array cells connected in parallel. The whole array outputs a single signal. Since the pulse signals generated by different photons can be superimposed, the SiPM can respond to different light intensities and output a single point signal. The other typical form is the SPAD array, where each SPAD cell is a separate pixel. The light intensity detected by each pixel can be output separately, which has the ability to detect photons in parallel. Therefore, it can be used for direct imaging and has excellent imaging performance in some low-light environments, such as biomedical microscopy, fluorescence lifetime microscopy, and Flash radar.

[0005] However, optical crosstalk remains a critical and difficult problem to solve for both SiPM and SPAD arrays. For SiPM, crosstalk photons can excite SPAD cells that do not normally receive incident photons. A high probability of optical crosstalk can lead to an overestimation of the received signal energy, causing problems such as incorrect calculation of object reflectivity and increased distance measurement errors. For SPAD arrays, phenomena such as image dilation and abnormal output of entire rows or columns can occur. Summary of the Invention

[0006] To address the aforementioned technical problems, this application provides a single-photon avalanche diode array and a silicon photomultiplier tube, aiming to solve the problem in current single-photon avalanche diode arrays where crosstalk photons cause SPAD units without incident photons to be excited, resulting in a large probability of optical crosstalk.

[0007] The first aspect of this application provides a single-photon avalanche diode array, the single-photon avalanche diode array comprising: At least two SPAD units; At least two microlenses are provided, each corresponding to one of the at least two SPAD units, and the microlenses are used to focus the incident light onto the corresponding SPAD unit. At least two front-side metal trace layers, and each of the at least two front-side metal trace layers corresponds to at least two of the SPAD cells; A metal mesh on the back is used to connect the SPAD unit to the corresponding external electrode; A first dielectric layer is disposed between the microlens and the SPAD unit, and the first dielectric layer is used to reduce the reflection of the incident light; A second dielectric layer is disposed between the SPAD unit and the front metal trace layer; wherein the front metal trace layer is electrically connected to the corresponding SPAD unit through contact metal wires; Deep groove isolation columns are disposed between adjacent SPAD units to isolate adjacent SPAD units; A metal-filled structure is disposed between the deep groove isolation pillar and the back metal mesh to suppress self-excited photons generated by the incident light from the SPAD unit from entering the adjacent SPAD unit through the mesh gaps.

[0008] In one embodiment, the vertical cross-section of the back metal mesh is an inverted trapezoid.

[0009] In one embodiment, the vertical cross-section of the back metal mesh is a multi-layer stepped structure, wherein the width of each step in the multi-layer stepped structure gradually increases.

[0010] In one embodiment, the vertical cross-section of the back metal mesh is an arc-shaped structure, used to reflect photons illuminating its surface to the SPAD unit.

[0011] In one embodiment, an intermetallic dielectric layer is provided between adjacent front metal trace layers, and an etch stop layer is provided between the intermetallic dielectric layer and the deep trench isolation pillar. The deep trench isolation pillar extends into the etch stop layer and divides the second dielectric layer into at least two dielectric units, which correspond to the SPAD unit.

[0012] In one embodiment, an intermetallic dielectric layer is provided between adjacent front metal trace layers, and an etching stop layer is provided between the intermetallic dielectric layer and the deep trench isolation pillar. The length of the portion of the deep trench isolation pillar extending into the second dielectric layer is greater than half the thickness of the second dielectric layer.

[0013] In one embodiment, the SPAD unit includes a first doped region and a second doped region stacked on the second dielectric layer, wherein the first doped region and the second doped region form a PN junction.

[0014] In one embodiment, the deep trench isolation column is made of tungsten metal.

[0015] In one embodiment, the metal filler layer has a ring-shaped structure.

[0016] A second aspect of this application also provides a silicon photomultiplier tube, the silicon photomultiplier tube comprising a single-photon avalanche diode array as described in any of the preceding embodiments.

[0017] A third aspect of this application also provides a receiving sensor, the receiving sensor comprising a single-photon avalanche diode array as described in any embodiment of the first aspect.

[0018] A fourth aspect of this application also provides a receiving sensor, the receiving sensor including a silicon photomultiplier tube as provided in the second aspect.

[0019] A fifth aspect of this application also provides a lidar, which includes a transmitting sensor and a receiving sensor according to the third and fourth aspects. The transmitting sensor is used to transmit a detection laser, and the receiving sensor is used to receive the echo of the detection laser and obtain detection information of a target object based on the echo.

[0020] The beneficial effects of this application's embodiments are as follows: At least two SPAD units are arranged in an array. A microlens focuses incident light onto the corresponding SPAD unit. A back metal mesh connects the SPAD unit to the corresponding external electrode. A first dielectric layer is disposed between the microlens and the SPAD unit, and a second dielectric layer is disposed between the SPAD unit and the front metal trace layer. The front metal trace layer is electrically connected to the corresponding SPAD unit through contact metal wires. By setting deep trench isolation pillars between adjacent SPAD units and setting a metal filling structure between the deep trench isolation pillars and the back metal mesh, self-excited photons generated by the SPAD unit under the excitation of incident light can be isolated, preventing self-excited photons from entering adjacent SPAD units through the mesh gaps, reducing the leakage of self-excited photons, lowering the probability of photon crosstalk, and reducing the optical crosstalk of the device. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the single-photon avalanche diode array provided in the embodiments of this application. Figure 1 ; Figure 2 This is a top view schematic diagram of the single-photon avalanche diode array provided in the embodiments of this application; Figure 3 This is a schematic diagram of the cross-section AA of the single-photon avalanche diode array provided in the embodiments of this application; Figure 4 This is a schematic diagram of the cross-section BB of the single-photon avalanche diode array provided in the embodiments of this application; Figure 5 This is a schematic diagram of the single-photon avalanche diode array provided in the embodiments of this application. Figure 2 ; Figure 6 This is a schematic diagram of the single-photon avalanche diode array provided in the embodiments of this application. Figure 3 ; Figure 7 This is a schematic diagram of the single-photon avalanche diode array provided in the embodiments of this application. Figure 4 ; Figure 8 This is a schematic diagram of the single-photon avalanche diode array provided in the embodiments of this application. Figure 5 ; Figure 9 This is a schematic diagram of the single-photon avalanche diode array provided in the embodiments of this application. Figure 6 ; Figure 10 This is a schematic diagram of the single-photon avalanche diode array provided in the embodiments of this application. Figure 7 ; Figure 11 This is a schematic diagram of the single-photon avalanche diode array provided in the embodiments of this application. Figure 8 ; Figure 12 A schematic diagram of self-excited crosstalk of a one-dimensional linear array provided in an embodiment of this application; Figure 13 A schematic diagram of two-dimensional array self-excited crosstalk provided in an embodiment of this application; Figure 14 A schematic diagram of the crosstalk optical field between the conventional structure and the metal-filled structure extension scheme provided in the embodiments of this application; Figure 15 A schematic diagram of the crosstalk optical field between the conventional structure and the deep trench isolation column extension scheme provided in the embodiments of this application; Figure 16 The embodiments of this application provide a closed-loop scheme that simultaneously employs deep trench isolation column extension and metal filling structure (such as...). Figure 9 A schematic diagram of the device structure with self-excited photon crosstalk during the process; Figure 17 The comprehensive solution provided for the embodiments of this application (such as...) Figure 11 A schematic diagram illustrating the improvement of self-excited crosstalk in the device structure when ( ). Detailed Implementation

[0022] To make the technical problems, technical solutions, and beneficial effects to be solved by this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and are not intended to limit the scope of this application.

[0023] It should be noted that when a component is referred to as being "fixed to" or "set on" another component, it can be directly on or indirectly on that other component. When a component is referred to as being "connected to" another component, it can be directly connected to or indirectly connected to that other component.

[0024] It should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0025] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or at least two of that feature. In the description of this application, "at least two" means one or more, unless otherwise explicitly specified.

[0026] Single-photon avalanche diode arrays (SPADs) have been widely used in laser detection and ranging systems (Lidar), autonomous driving, fluorescence lifetime imaging, quantum communication, and biomedical imaging due to their advantages such as high sensitivity detection capability for single photons with low intensity, high time resolution capability, and strong anti-interference capability. Silicon photomultiplier tubes (SiPMs), which consist of at least two SPADs connected in parallel as a single point, are used.

[0027] To reduce crosstalk characteristics in devices, optimization can be performed at the device, package, and system application levels. However, application-level crosstalk improvement schemes only avoid the impact of crosstalk on signal readout, at the cost of complex scanning methods and circuit implementations, and do not solve the optical crosstalk problem. Package-level crosstalk improvement schemes are similar. Device-level crosstalk is the most fundamental, yet the available methods are limited. While FSI devices (front-illuminated receivers) have better crosstalk characteristics, their low PDE limits their application, and the industry still uses back-side illumination (BSI) devices with higher PDEs. Introducing deep trench isolation pillars (DTI) can greatly reduce optical crosstalk, but its effect is mainly on reducing crosstalk to adjacent SPADs; distant SPADs can still be affected by crosstalk through other paths. In particular, for SiPM devices, it may cause chain-like avalanche triggering, resulting in crosstalk affecting a large area of ​​the device. To address the aforementioned technical problems, this application provides a single-photon avalanche diode array, see [link to relevant documentation]. Figure 1 As shown, the single-photon avalanche diode array in this embodiment includes: at least two SPAD units 500, at least two microlenses 100, at least two front metal trace layers 130, a back metal mesh 200, a first dielectric layer 310, a second dielectric layer 320, a deep trench isolation pillar 400, and a metal filling structure 140.

[0028] In this embodiment, at least two SPAD units 500 are arranged adjacent to each other, and at least two microlenses 100 correspond one-to-one with each of the at least two SPAD units 500. The microlenses 100 are used to focus incident light onto the corresponding SPAD unit 500. At least two front metal trace layers 130 correspond one-to-one with each of the at least two SPAD units 500. The back metal mesh 200 is used to connect the SPAD unit 500 to the corresponding external electrode. A first dielectric layer 310 is disposed between the microlens 100 and the SPAD unit 500, and the first dielectric layer 310 is used to reduce the reflection of incident light. A second dielectric layer 320 is disposed between the SPAD unit 500 and the front metal trace layer 130. The front metal trace layer 130 is electrically connected to the corresponding SPAD unit 500 through a contact metal wire 900. A deep groove isolation pillar 400 is disposed between adjacent SPAD units 500, and the deep groove isolation pillar 400 is used to isolate adjacent SPAD units 500; a metal filling structure 140 is disposed between the deep groove isolation pillar 400 and the back metal mesh 200, and the metal filling structure 140 is used to suppress self-excited photons generated by the SPAD unit 500 under the excitation of incident light from entering the adjacent SPAD unit 500 through the mesh gap.

[0029] In this embodiment, the microlens 100 acts as a light convergent, focusing the incident light onto the corresponding SPAD unit 500. The back metal mesh 200 serves to lead one of the electrodes of the SPAD unit 500 out from the back. By setting deep trench isolation pillars 400 between adjacent SPAD units 500 and a metal filling structure 140 between the deep trench isolation pillars 400 and the back metal mesh 200, the deep trench isolation pillars 400 can electrically isolate adjacent SPAD units 500. Self-excited photons generated by the SPAD unit 500 when excited by incident light can be reflected when they hit the deep trench isolation pillars 400, increasing absorption and preventing the self-excited photons from interfering with adjacent SPAD units 500. This is an important way to improve crosstalk. The metal-filled structure 140 can isolate self-excited photons generated by the SPAD unit 500 under the excitation of incident light, reduce the crosstalk of self-excited photons from the gap between the back metal mesh 200 and the deep trench isolation pillar 400 to the adjacent SPAD unit 500, reduce the overflow of self-excited photons, reduce the probability of photon crosstalk, and reduce the optical crosstalk of the device.

[0030] In one embodiment, when there are multiple SPAD units 500, the multiple SPAD units 500 are arranged in an array. For example, in the case of 4 SPAD units 500, the 4 SPAD units 500 are arranged in an array.

[0031] Figure 2 This is a top view of a SPAD unit 500. At least two SPAD units 500 form a SPAD array. Each SPAD unit 500 is provided with a corresponding microlens 100. The metal filling structure 140 is the metal connecting the deep groove isolation pillar 400 and the back metal mesh 200. Figure 3 for Figure 2 A schematic diagram of the cross-section at position AA of the component. Figure 4 for Figure 2 A schematic diagram of the cross-section at position BB of the component. (See diagram below.) Figure 4 As shown, in the BB section, the presence of the metal-filled structure 140 prevents crosstalk photons from entering the adjacent SPAD unit 500 through the dielectric gap between the back metal mesh 200 and the deep trench isolation pillar 400. However, in the AA section where the metal-filled structure 140 is not present (e.g....), crosstalk photons cannot enter the adjacent SPAD unit 500. Figure 3 As shown in the figure, a considerable amount of unabsorbed light or self-generated photons will enter the adjacent SPAD unit 500 through the gap, causing crosstalk.

[0032] In one embodiment, the metal-filled structure 140 is a ring-shaped structure, such as... Figure 5As shown, the shape of the metal filling structure 140 is the same as that of the back metal mesh 200. The metal filling structure 140 can completely fill the gap between the back metal mesh 200 and the deep groove isolation pillar 400 around the SPAD unit 500, thus avoiding the gaps such as... Figure 2 The gap appearing at position AA of the mid-section is also filled by the metal filling structure 140. This part of the crosstalk will be completely suppressed, and the photons can be reflected back into the Si material of the substrate for reabsorption. It can also improve the photon detection efficiency (PDE) to a certain extent, reduce the crosstalk of self-excited photons from the gap between the back metal mesh 200 and the deep trench isolation pillar 400 to the adjacent SPAD unit 500, reduce the overflow of self-excited photons, reduce the probability of photon crosstalk, and reduce the optical crosstalk of the device.

[0033] In some embodiments, the deep trench isolation post 400 can be made of tungsten metal, and a dielectric material is filled around the deep trench isolation post 400. The presence of the deep trench isolation post 400 will reflect crosstalk photons, greatly reducing the probability of optical crosstalk.

[0034] In some embodiments, the width of the metal filling structure 140 is smaller than the width of the deep groove isolation post 400.

[0035] In some embodiments, by selecting appropriate first dielectric layer 310 and second dielectric layer 320, and designing different dielectric layer refractive indices and film thicknesses, the first dielectric layer 310 and second dielectric layer 320 can form an optical filter to absorb or reflect photons of different wavelengths in self-excited photons, thereby reducing crosstalk in the device.

[0036] In some embodiments, in array-packaged SiPM devices, trench structures can be etched into the package structure to reduce crosstalk between SiPMs.

[0037] In one embodiment, the SPAD unit 500 includes a first doped region 600 and a second doped region 700 stacked on a second dielectric layer, wherein the first doped region 600 and the second doped region 700 form a PN junction.

[0038] In some embodiments, the first doped region 600 and the second doped region 700 are two doped regions of different polarities, and a PN junction is formed between the first doped region 600 and the second doped region 700. The amplification region 800 is the region with the strongest electric field of the PN junction. The amplification region 800 is mainly used for the generation and amplification of avalanches. The depletion region is the space charge region generated by the device under a certain voltage. Electrons or holes generated in it can drift into the amplification region 800 through the electric field to generate avalanches.

[0039] In one embodiment, the vertical cross-section of the back metal mesh 200 is an inverted trapezoid.

[0040] The path of optical crosstalk is as follows Figure 3 As shown, because the back metal mesh 200 has a trapezoidal shape, photons reflected from the sidewalls of the back metal mesh 200 have a high probability of entering other SPAD units. Furthermore, this crosstalk may propagate to distant locations due to the small reflection from the lens, causing crosstalk to remote devices and further exacerbating the avalanche chain reaction of crosstalk. Figure 6 As shown, in this embodiment, an inverted trapezoidal back metal mesh 200 is used. At this time, the interface of the back metal mesh 200 tilted inward will reflect the light incident on it back into its own device, reducing photon overflow and lowering the probability of crosstalk of self-excited photons.

[0041] In one embodiment, see Figure 7 As shown, the vertical cross-section of the back metal mesh 200 is a multi-layer stepped structure, and the width of each step in the multi-layer stepped structure gradually increases.

[0042] In this embodiment, the vertical cross-section of the back metal mesh 200 is a multi-layer stepped structure, and the width of the multi-layer stepped structure increases step by step, causing the back metal mesh 200 to tilt inward, reflecting the light incident on it back into its own device, reducing photon leakage and lowering the probability of crosstalk of self-excited photons.

[0043] In one embodiment, the width of the multi-tiered stepped structure increases progressively, and the width of the multi-tiered stepped structure is set as an arithmetic sequence.

[0044] In one embodiment, see Figure 8 As shown, the vertical cross-section of the back metal mesh 200 is an arc-shaped structure, which is used to reflect photons illuminating its surface to the SPAD unit 500.

[0045] In this embodiment, by setting the vertical cross-section of the back metal mesh 200 to an arc-shaped structure, the interface between the back metal mesh 200 and the microlens 100 can form a total internal reflection interface tilted inward to the SPAD unit 500, which reflects the photons illuminating its surface to the SPAD unit 500 (i.e., the area of ​​the SPAD unit 500 itself), reducing the overflow of self-excited photons and lowering the crosstalk probability of self-excited photons.

[0046] In one embodiment, see Figure 8 As shown, an intermetallic dielectric layer 120 is provided between adjacent front metal trace layers 130, and an etch stop layer 110 is provided between the intermetallic dielectric layer 120 and the deep trench isolation pillar 400. The deep trench isolation pillar 400 extends into the etch stop layer 110 and divides the second dielectric layer 320 into at least two dielectric units. The at least two dielectric units correspond to the SPAD unit 500 and can block crosstalk light reflected by the front metal trace layer 130 and the etch stop layer 110.

[0047] In one embodiment, since the first doped region 600 and the second doped region 700 have a certain thickness, and the deep trench isolation pillar 400 extends into the second dielectric layer 320, the crosstalk light reflected by the front metal trace layer 130 and the etch stop layer 110 can be blocked when the length of the portion of the deep trench isolation pillar 400 extending into the second dielectric layer 320 is greater than half the thickness of the second dielectric layer 320.

[0048] In practical applications, the length of the portion of the deep trench isolation post 400 that extends into the second medium layer 320 can be set according to the width of the deep trench isolation post 400.

[0049] In one embodiment, the thickness of the portion of the deep trench isolation post 400 extending into the second dielectric layer 320 may be less than the thickness of the second dielectric layer 320.

[0050] In one embodiment, in the single-photon avalanche diode array, the metal filling structure 140 is a ring structure that can completely fill the gap between the back metal mesh 200 and the deep trench isolation pillars 400 around the SPAD unit 500, thus avoiding the gaps such as... Figure 2 The gap appearing at position AA of the mid-section will completely suppress this part of the crosstalk. Combined with the deep trench isolation pillar 400 extending into the etching stop layer 110, a crosstalk prevention scheme of deep trench isolation extension and metal filling structure ring is formed, such as... Figure 9 As shown.

[0051] In one embodiment, in a single-photon avalanche diode array, such as Figure 10 As shown, the back metal mesh 200 has an inverted trapezoidal shape. The inwardly tilted interface of the back metal mesh 200 reflects incident light back into the device, reducing photon leakage and lowering the probability of crosstalk from self-excited photons. Simultaneously, the deep trench isolation pillars 400 extend into the etch stop layer 110, further reducing the probability of photon crosstalk from the bottom. The combination of deep trench isolation extension and the inverted trapezoidal back metal mesh 200 forms an anti-crosstalk scheme that effectively reduces interference from crosstalk photons to adjacent SPAD units 500.

[0052] In one embodiment, based on the metal filling structure 140 being a ring structure and the deep trench isolation pillars 400 extending into the etching stop layer 110, setting the cross-section of the back metal mesh to an inverted trapezoidal shape can form a crosstalk prevention combination scheme of deep trench isolation extension, metal filling structure ring, and inverted trapezoidal back metal mesh, such as... Figure 10 As shown.

[0053] In the device structure of the BSI described above, the deep trench isolation pillar 400 can block most of the crosstalk photons, but there are still a few paths that can obviously cause crosstalk that have not been resolved, making it impossible to further reduce crosstalk.

[0054] like Figure 1 As shown, the deep trench isolation pillar 400 rests above the second dielectric layer 320. In this case, crosstalk photons can be reflected into the adjacent SPAD unit 500 via the front metal trace layer 130, or reflected into the adjacent SPAD unit 500 via the interface formed by the etch stop layer 110 and the second dielectric layer 320. To reduce the probability of photons interfering from the bottom, in this embodiment, the deep trench isolation pillar 400 extends downwards to the etch stop layer to block crosstalk light reflected by the front metal trace layer 130 and the etch stop layer 110.

[0055] In some embodiments, the etch stop layer 110 may be a silicon carbon nitride (SiCN) material.

[0056] In one embodiment, the deep trench isolation column 400 is made of tungsten metal.

[0057] In this embodiment, when the self-excited photons generated by the SPAD unit 500 after being illuminated propagate within the device, the light incident on the deep trench isolation pillars 400 around it will be repeatedly reflected, increasing absorption and preventing photons from interfering with adjacent units. Therefore, setting deep trench isolation pillars 400 around the SPAD unit 500 is also an important way to improve crosstalk.

[0058] In one embodiment, the back metal mesh 200 is made of aluminum.

[0059] In some embodiments, multiple layers of metal traces may be formed in the front metal trace layer 130 to lead out another electrode of the SPAD unit 500 to the front electrode, and the corresponding electrical signal may also be output through the front metal trace layer 130.

[0060] To illustrate the improvement effect and level of optical crosstalk of the single-photon avalanche diode array in the above embodiments, the optical crosstalk of the single-photon avalanche diode array is simulated and verified below.

[0061] The formula for calculating the optical crosstalk probability OXT is: ;(Formula 1) Wherein, G(r) is the carrier velocity generated after crosstalk light from the SPAD unit 500 at the absorption center of the adjacent SPAD unit 500; ATP(r) is the carrier avalanche probability in the depletion region; Gain is the number of carriers released each time an avalanche occurs in the SPAD unit 500, i.e., the gain; PPS(λ) is the avalanche photon generation spectrum, which physically represents the distribution of the average number of photons released per carrier per avalanche with respect to wavelength.

[0062] In the calculation, G(r) can be obtained through optical simulation. Random photons with a spectral width of 0.4-1 μm are placed in the central SPAD unit 500. The intensity of each photon is set according to PPS(λ), referencing the spontaneous spectrum of avalanche photons. The absorbed light power within adjacent SPAD units 500 is then monitored and converted into carrier generation rate. ATP(r) can be calculated using the McIntyre model based on the electric field magnitude of the SPAD unit 500, determining the probability of avalanche triggering at different locations. The gain is set to 0.8 M in the calculation. During simulation, due to the randomness of the self-excited photons' position and radiation direction, this application sets the self-excited photons to be generated at three different locations (left, center, and right of the device). Simultaneously, three dipole polarization directions (X, Y, Z) are used for the photon radiation direction. The average of nine simulation results is taken as the OXT radiation probability.

[0063] When calculating the overall crosstalk of a one-dimensional long array of SPADs, it is difficult to directly perform optical simulations for a large number of SPADs. Therefore, this application uses a 4-spads model to approximate the one-dimensional long array. Figure 11 As shown. For ease of understanding, let's first consider the crosstalk situation with three devices, specifically the crosstalk situation experienced by SPAD2. After SPAD0 generates self-excited photons, it has a certain probability of triggering an avalanche in SPAD1. The avalanche of SPAD1 also has a probability of generating photons that further trigger an avalanche in SPAD2. Therefore, the crosstalk of SPAD2 can be generated directly from SPAD0 (probability χ2), or it can be generated through the SPAD0→SPAD1→SPAD2 link. Since these two paths can occur almost simultaneously, the probability of SPAD2 receiving optical crosstalk is the sum of the probabilities of the two links, minus the probability of simultaneous occurrence. This can be expressed by Formula 2, where χ1 is the probability of direct crosstalk between two adjacent SPADs: OXT3-spad= χ1 +(χ2 +χ1·χ1- χ2·χ1·χ1); (Formula 2) Furthermore, this simulation model is extended to 4-spads, and its crosstalk chain can be referenced. Figure 10 As shown, therefore, under the 4-spads model, the crosstalk of the one-dimensional array of SPAD cells 500 can be calculated by Equation 3: OXT4-spad=χ1 +(χ2 +χ1•χ1- χ2•χ1•χ1)+χ3+ 2•χ1•χ2+ χ1^3- 2•χ1•χ2•χ3-χ3•χ1^3- χ1^2•χ2^2-2•χ1^3•χ2+ χ1^2•χ2^2•χ3 + 2•χ1^3•χ2•χ3 + χ1^3•χ2^2 - χ1^3•χ2^2•χ3; (Formula 3) In Formula 3, χi (i=1, 2, 3) represents the probability that SPAD0 directly generates crosstalk to the i-th SPAD unit 500, which can be obtained by combining optical simulation with Formula 1. Thus, the optical crosstalk model of a one-dimensional long linear array SPAD is obtained.

[0064] Building upon this, this application employs a similar method to extend it to 2D arrays, referencing... Figure 13 As shown, the crosstalk received by SPAD0 comes from eight surrounding directions. Due to symmetry, only the crosstalk ρ1 in the horizontal direction and the crosstalk ρ2 in the diagonal direction need to be simulated and calculated.

[0065] This application uses an approximate method to simulate an infinitely large 2D array, specifically employing a one-dimensional 4-spads model to obtain an approximate horizontal crosstalk ρ1, which is then used for equivalent calculation. Figure 11 The crosstalk between SPAD1 and SPAD2 and SPAD0 is calculated; then, the crosstalk magnitude of the SPADs in the diagonal direction is simulated to obtain ρ2, which is equivalent to... Figure 12 Crosstalk from SPAD3 to SPAD0.

[0066] At this point, SPAD0 experiences two crosstalk paths: SPAD1→SPAD0 and SPAD3→SPAD1→SPAD0. The model ignores some theoretically possible but practically unlikely paths, such as SPAD3→SPAD1→SPAD2→SPAD0. Therefore, the crosstalk model for the entire 2D array is given by Equation 4: OXT -2D =4ρ1+4ρ2+4(ρ1·ρ2-ρ1·ρ1·ρ2); (Formula 4) Based on the crosstalk improvement effect in this application, sequential simulation verification was performed, and the crosstalk improvement effect was compared with the crosstalk performance of traditional device structure simulation, as shown in Table 1.

[0067] Table 1

[0068] The simulation data in Table 1 clearly shows a significant improvement in crosstalk performance. When optimizing the deep trench isolation pillar 400 process alone, the crosstalk probability decreased significantly in both the horizontal and diagonal directions, ultimately reducing the total OXT to 60% of its original value. The optimization scheme using the metal-filled structure 140 closed-loop alone also achieved the same improvement. Simultaneously employing both the deep trench isolation extension scheme and the metal-filled structure 140 closed-loop scheme further reduced crosstalk performance, particularly the χ1 probability, i.e., the significant suppression of crosstalk to the nearest neighbor SPAD unit 500, reducing the OXT to 46% of its original value. Furthermore, the comprehensive scheme proposed in this application (such as...) further improved the performance. Figure 11 As shown), OXT decreased to 36% of its original value, with the most significant improvement observed in the nearest SPAD unit 500.

[0069] Figure 14-17 This more intuitively demonstrates the optical crosstalk reduction effect of the proposed solution. Figure 14 The diagram shows the crosstalk situation of the self-excited light in the traditional structure and the crosstalk situation of the metal-filled extended structure 140. Figure 14 The leftmost SPAD unit 500 avalanche generated self-excited photons. Figure 14 The image shows a photon position located to the left of the device. Comparing the two images, it's clear that the 140° closed-loop metal-filled structure (such as...) Figure 1 The proposed scheme significantly suppressed crosstalk light in this path. Figure 15 The crosstalk situation of the traditional structure and the deep trench isolation extension structure is shown. When the deep trench isolation extends to the etch stop layer, it can be seen that the crosstalk light at the bottom is also significantly suppressed.

[0070] Figure 16 To simultaneously employ a deep trench isolation extension and a metal-filled structure 140 closed-loop scheme (such as...) Figure 9 When examining the crosstalk of self-excited photons, it is easy to see that the crosstalk light directly from below the back metal mesh 200 and the bottom of the deep groove isolation pillar 400 has been greatly reduced, and the main crosstalk comes from the lens part. Figure 17 To adopt a comprehensive improvement plan (such as...) Figure 11 In the case of [missing information], the inverted trapezoidal shape of the back metal mesh 200 reflects a large portion of the crosstalk light back into its own SPAD cells, significantly suppressing optical crosstalk to neighboring devices and thus reducing the overall optical crosstalk of the two-dimensional array. Specifically, the gain used in the model calculations of this scheme is relatively small, and the value of the self-excited photon emission spectrum is also relatively small. Therefore, for the actual fabricated device, the improvement in optical crosstalk performance will be more significant when the incident light intensity and overvoltage are increased.

[0071] This application also provides a silicon photomultiplier tube, which includes a single-photon avalanche diode array as described in any of the above embodiments.

[0072] The beneficial effects of this embodiment are as follows: Incident light is focused onto the corresponding SPAD unit via a microlens. A back metal mesh connects the SPAD unit to the corresponding external electrode. A first dielectric layer is disposed between the microlens and the SPAD unit, and a second dielectric layer is disposed between the SPAD unit and the front metal trace layer. The front metal trace layer is electrically connected to the corresponding SPAD unit via contact metal wires. By setting deep trench isolation pillars between adjacent SPAD units and a metal filling structure between the deep trench isolation pillars and the back metal mesh, self-excited photons generated by the SPAD unit under incident light excitation can be isolated, preventing self-excited photons from entering adjacent SPAD units through mesh gaps. This reduces the leakage of self-excited photons, lowers the probability of photon crosstalk, and reduces optical crosstalk of the device.

[0073] This application embodiment also includes a receiving sensor, which may include one or more of the above-described single-photon avalanche diode arrays or one or more of the above-described silicon photomultiplier tubes.

[0074] This application embodiment also includes a lidar, which includes a transmitting sensor and the receiving sensor described above. The transmitting sensor is used to emit a detection laser, and the receiving sensor is used to receive the echo of the detection laser and obtain the target object based on the echo.

[0075] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional device areas and modules is used as an example. In practical applications, the above functions can be assigned to different functional device areas and modules as needed, that is, the internal structure of the device can be divided into different functional device areas or modules to complete all or part of the functions described above. In the embodiments, the functional device areas and modules can be integrated into one device, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0076] Furthermore, the specific names of each functional device area and module are merely for easy differentiation and are not intended to limit the scope of protection of this application. The specific working processes of the units and modules in the above system can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0077] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0078] In addition, the functional device regions in the various embodiments of this application can be integrated into one device, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0079] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A single-photon avalanche diode array, characterized in that, The single-photon avalanche diode array includes: At least two SPAD units; At least two microlenses are provided, each corresponding to one of the at least two SPAD units, and the microlenses are used to focus the incident light onto the corresponding SPAD unit. At least two front-side metal trace layers, and each of the at least two front-side metal trace layers corresponds to at least two of the SPAD cells; A metal mesh on the back is used to connect the SPAD unit to the corresponding external electrode; A first dielectric layer is disposed between the microlens and the SPAD unit, and the first dielectric layer is used to reduce the reflection of the incident light; A second dielectric layer is disposed between the SPAD unit and the front metal trace layer; wherein the front metal trace layer is electrically connected to the corresponding SPAD unit through contact metal wires; Deep groove isolation columns are disposed between adjacent SPAD units to isolate adjacent SPAD units; A metal-filled structure is disposed between the deep groove isolation pillar and the back metal mesh to suppress self-excited photons generated by the incident light from the SPAD unit from entering the adjacent SPAD unit through the mesh gaps; The metal filling structure is a ring structure that completely fills the gap between the back metal mesh and the deep groove isolation pillars around the SPAD unit.

2. The single-photon avalanche diode array as described in claim 1, characterized in that, The vertical cross-section of the back metal mesh is an inverted trapezoid.

3. The single-photon avalanche diode array as described in claim 1, characterized in that, The vertical cross-section of the back metal mesh is a multi-layered stepped structure, and the width of each step in the multi-layered stepped structure gradually increases.

4. The single-photon avalanche diode array as described in claim 1, characterized in that, The vertical cross-section of the back metal mesh is an arc-shaped structure, which is used to reflect photons illuminating its surface to the SPAD unit.

5. The single-photon avalanche diode array as described in any one of claims 1-4, characterized in that, An intermetallic dielectric layer is provided between adjacent front metal trace layers, and an etching stop layer is provided between the intermetallic dielectric layer and the deep trench isolation pillar. The length of the portion of the deep trench isolation pillar extending into the second dielectric layer is greater than half the thickness of the second dielectric layer.

6. The single-photon avalanche diode array as described in claim 5, characterized in that, The etching stop layer is silicon carbon nitride (SiCN).

7. The single-photon avalanche diode array as described in any one of claims 1-4, characterized in that, The SPAD unit includes a first doped region and a second doped region stacked on the second dielectric layer, wherein the first doped region and the second doped region form a PN junction.

8. The single-photon avalanche diode array as described in any one of claims 1-4, characterized in that, The deep trench isolation column is made of tungsten metal.

9. A receiving sensor, characterized in that, The receiving sensor includes a single-photon avalanche diode array as described in any one of claims 1-8.

10. A lidar, characterized in that, The lidar includes a transmitting sensor and a receiving sensor as described in claim 9; The emission sensor is used to emit a detection laser; The receiving sensor is used to receive the echo of the detection laser and obtain the detection information of the target object based on the echo.

Images