Single-photon avalanche diode and single-photon avalanche diode array

Abstract

A single-photon avalanche diode SPAD includes an active region (10) arranged to generate a photon-triggered avalanche current. A cover (12) is arranged on or above the active region (10). The cover (12) shields the active region (10) from incident photons. The cover (12) includes a stack of at least a first metal layer and a second metal layer (13, 14). At least one of the metal layers (13, 14), for example the first metal layer (13), includes a hole (15). The metal layers (13, 14) are arranged to define an effective hole (18) opening along the optical axis (OA) in the stack relative to the optical axis (OA). Through the effective hole (18), a portion (19) of the active region (10) is exposed to incident photons incident along the optical axis (OA). The effective hole (18) is smaller than the hole (15) arranged in the first metal layer (13).

Description

[0001] The present invention relates to a single - photon avalanche diode and a single - photon avalanche diode array.

[0002] A single - photon avalanche diode, abbreviated as SPAD, is a solid - state photodetector based on a pn - junction biased beyond the breakdown region. The bias voltage generates an electric field with a high field strength. In fact, a single photon may generate charge carriers in the depletion layer of the SPAD, and these charge carriers will trigger an avalanche current due to the impact ionization mechanism. Once triggered, after a certain time, the avalanche is quenched either actively or passively, allowing the device to be "reset" in order to detect additional photons again. After the avalanche breakdown stops, the SPAD is charged to its over - bias voltage, such as a voltage higher than the breakdown voltage. However, during this so - called dead time, the SPAD cannot detect photons.

[0003] The linear range of the SPAD is limited by the dead time because the SPAD output becomes increasingly non - linear with increasing intensity relative to the incident light intensity. To extend the linear response of the SPAD, capping structures have been proposed. For example, metal holes covering the SPAD can be designed to limit the number of incident photons. In addition, due to the high count rate of high photon counts, a high current is generated. Therefore, the attenuation of the SPAD has the additional effect of reducing current consumption.

[0004] Attenuation of the SPAD may be required under high ambient light conditions and strong signals (without strong signals, the signal may be lost in the noise). For medium - to - low ambient light conditions and weaker signals, a non - attenuated SPAD may be required. Thus, attenuated and non - attenuated SPADs can be arranged together as an array on the same chip to adapt to different light conditions. For example, the non - attenuated SPADs are deactivated under high ambient light conditions.

[0005] The maximum achievable attenuation is usually limited by the design rules of the metal layer used. For example, for a SPAD with an active area of 50μm 2 and an expected attenuation of 0.5%, the cover (such as a metal shield) can have square holes that need to be 0.5×0.5μm 2 However, in some SPAD manufacturing processes, this may violate the metal pitch rules and / or the metal enclosed area rules. In particular, for metal lines wider than the minimum metal width, the minimum pitch rule is usually much larger than the minimum pitch possible in a certain process. In addition, the maximum metal width and / or the maximum metal density are usually also limited.

[0006] The object is to provide a single - photon avalanche diode and a single - photon avalanche diode array with an extended linear response.

[0007] These objects are achieved by the subject matter of the independent claims. Further developments and embodiments are described in the dependent claims.

[0008] It should be understood that any feature described in connection with any one embodiment can be used alone or in combination with other features described hereinafter, and can also be used in combination with one or more features of any other embodiment, or in combination with any combination of any other embodiment, unless otherwise explicitly described. In addition, equivalents and modifications not described below can also be employed without departing from the scope of the single-photon avalanche diode and the array of single-photon avalanche diodes as set forth in the appended claims.

[0009] The following relates to an improved concept in the field of single-photon avalanche diodes (SPADs) and arrays of such SPADs. The concept provides a method for extending the attenuation of a SPAD by improving the cover, for example by using a combination of two or more metal layers. In fact, the actually achievable attenuation can be extended to smaller values, for example, for a typical active region, the attenuation is less than 1%.

[0010] In at least one embodiment, a single-photon avalanche diode (SPAD) includes an active region arranged to generate a photon-triggered avalanche current. A cover is arranged on or above the active region. The cover shields the active region from incident photons. The cover includes a stack of at least a first metal layer and a second metal layer.

[0011] At least one of the metal layers, such as the first metal layer, includes holes. The metal layer is arranged in the stack with respect to the optical axis to open effective holes along the optical axis. Through the effective holes, a portion of the active region is exposed to incident photons incident along the optical axis. The effective holes are smaller than the holes arranged in the first metal layer.

[0012] The proposed concept extends the actually achievable attenuation to smaller values and also extends the linear range of the SPAD. The effective holes can be established by two or more metal layers and effectively block photons incident from paths other than the optical axis to the active region. Generally, photons only enter from a certain field of view, and the effective holes can be optimized to best suit that field of view. In addition, it can be shown that such a combination of two or more metal layers is particularly insensitive to alignment errors between the metal layers. Therefore, the effect of misalignment between different metal layers has only a small impact on the achievable attenuation. In addition, the effective holes can be smaller than the minimum design rule pitch of the square that generally defines the strict requirements to be met in SPAD design. Therefore, the effective holes are largely not limited by the process.

[0013] In at least one embodiment, a metal layer, such as a first metal layer, includes more than one hole. This can result in several effective holes, each of which exposes a portion of the active region to incident photons. Each effective hole is smaller than the corresponding hole arranged in the first metal layer. Additionally, or alternatively, one or more holes may be provided on another metal layer, such as a second metal layer. Hereinafter, unless otherwise specified, the features discussed regarding a single hole, effective hole, or metal layer can be applied to several holes, effective holes, or metal layers.

[0014] In at least one embodiment, except for the portion of the active region exposed through the effective holes, the cover completely shields the active region from incident photons.

[0015] In at least one embodiment, the area of the active region exposed by the effective holes is less than 50% of the entire active region, less than 10% of the entire active region, or less than 1% of the entire active region. Thus, different degrees of attenuation can be achieved. However, the processes applied to fabricate the SPAD may limit the actually achievable attenuation and are thus not limited by the above example values.

[0016] In at least one embodiment, the metal layers are spaced apart from each other such that the first metal layer has a first distance from the active region and the second metal layer has a different second distance from the active region. For example, the metal layer having one or more holes can be located closer to or farther from the active region.

[0017] In at least one embodiment, the metal layers are interconnected by metal connectors. For example, the interconnection can be established through an interlayer dielectric or through vertical electrical connectors. For example, the electrical connectors can include through-silicon vias (TSVs).

[0018] In at least one embodiment, the second metal layer includes holes. Thus, both the first metal layer and the second metal layer can include their respective holes. In this case, the holes in the metal layers can be offset with respect to the optical axis. Alternatively, the holes in the metal layers can be aligned with respect to the optical axis.

[0019] In at least one embodiment, the optical axis is the surface normal of the active region or an inclined axis at an angle different from the normal with respect to the active region. For example, in the case where the holes in the metal layer are offset with respect to the optical axis, the axis can have an inclined angle different from the surface normal. In the case where the holes in the metal layer are aligned with respect to the optical axis, the axis can be the surface normal or parallel to the surface normal. However, the holes in different metal layers can be aligned with respect to several optical axes such that the incident photons may not travel along a straight path to reach the active region. Instead, the holes can define a path that bends once or several times according to the arrangement of the holes. This may further increase the attenuation of the incident photons.

[0020] In at least one embodiment, a second metal layer is disposed below or above a first metal layer. The arrangement of the metal layers is adjusted such that the projection of the second metal layer onto the first metal layer along the optical axis at least partially shields the holes in the first metal layer to limit the effective holes. In other words, the effective holes are defined by the holes in the first and second metal layers and their relative arrangement in the stack.

[0021] In at least one embodiment, a metal layer, such as the second metal layer, includes metal lines, metal crosses, and / or metal squares. For example, one metal layer, such as the first metal layer, is a planar layer, generally having a shape similar to the active region of the SPAD, such as a square, rectangular, or circular shape (the shape can be the same shape, having an equal or slightly larger or smaller surface area). Then the second metal layer (or other additional metal layer) can have a linear, cross-shaped, or square shape, the area of which can be smaller when compared to the planar layer of the first metal layer. For example, if only the first metal layer has holes, such as a single hole, metal lines, metal crosses, and / or metal squares can be arranged relative to the hole to define the effective hole, such that when the effective hole is considered as a projection along the optical axis, it is smaller than the hole provided in the first metal layer.

[0022] In at least one embodiment, the cover includes an additional metal layer, the metal layer of which is stacked with the first and second metal layers. For example, the concepts obtained so far can also be applied to one of these more additional metal layers including the arrangement of holes and effective holes.

[0023] In at least one embodiment, the cover and / or the active region has a circular, rectangular, or polygonal shape. Similarly, one or more of the metal layers also have a circular, rectangular, or polygonal shape. The specific shape allows several SPADs to be arranged in an array in a space-saving manner. For example, several SPADs can be arranged in a two-dimensional packaging scheme such as a circular, square, rectangular, or honeycomb structure.

[0024] In at least one embodiment, a single photon avalanche diode (SPAD) array includes a plurality of SPADs according to the above concepts. Each SPAD is arranged in a row and / or column of the array. The plurality of SPADs includes a plurality of different SPAD subsets. Each different subset has at least one SPAD, where the SPADs have different effective holes. Finally, a control unit is configured to select one or more SPADs from the subsets of SPADs according to the incident radiation intensity. In fact, individual SPADs from a given subset or the entire subset can be selected or deselected.

[0025] In at least one embodiment, the SPADs arranged in each subset have the same effective holes.

[0026] In at least one embodiment, the effective apertures of the SPADs arranged in different subsets are different.

[0027] In at least one embodiment, the control unit selects one or more SPADs from one subset or different subsets by activating or deactivating one or more SPADs or subsets. Additionally, or alternatively, all SPADs from a given subset or the entire array can always be activated without the need to be individually activated or deactivated. Then the control unit selects one or more SPADs by outputting a signal, such as an output pulse.

[0028] The principles presented above are described in more detail below with reference to the accompanying drawings, in which exemplary embodiments are shown.

[0029] In the following exemplary embodiments and drawings, each similar or identical element may be provided with the same reference numeral. However, the elements shown in the drawings and the dimensional relationships between them should not be considered to be of true scale. Instead, individual elements, such as layers, components, and regions, may be exaggerated to achieve better illustration or improved understanding.

[0030] Figure 1 An embodiment of a single photon avalanche diode is shown.

[0031] Figure 2 Another embodiment of a single photon avalanche diode is shown.

[0032] Figure 3 Another embodiment of a single photon avalanche diode is shown.

[0033] Figure 4 Another embodiment of a single photon avalanche diode is shown.

[0034] Figure 5 An embodiment of an array of single photon avalanche diodes is shown, and

[0035] Figure 6 shows another embodiment of a single photon avalanche diode.

[0036] Figure 1A cross-section of an embodiment of a single-photon avalanche diode (SPAD) is shown. The SPAD 1 includes an active region 10 disposed in a semiconductor substrate 11. For example, the active region 10 includes a pn junction that can be reverse-biased at a bias voltage exceeding the breakdown voltage of the junction. The active region 10 also includes additional electronic components, such as contact pads or depletion layers, etc., which are not shown in the figure for a simpler representation. Basically, the active region 10 is arranged to trigger a self-sustaining avalanche when a single charge carrier is injected into the junction, for example, by an incident photon. In addition, the SPAD includes means for quenching the avalanche, such as quenching the avalanche by reducing the bias voltage. The operation (including quenching) of the SPAD can be controlled by a dedicated control circuit 20 (not shown).

[0037] The SPAD 1 further includes a cover 12 disposed at a distance d1 above the active region 10. The distance d1 affects the optical characteristics of the SPAD and constitutes a design choice. For example, the value of the distance d1 can be set to any practical value that conforms to the intended application. The cover 12 includes a stack of a first metal layer 13 and a second metal layer 14. The metal layers 13, 14 are stacked on top of each other at a distance d2. The distance d2 also affects the optical characteristics of the SPAD1 and constitutes a design choice. For example, the value of the distance d2 can be set to any practical value that conforms to the intended application. Basically, the cover 12 is designed to shield the active region 10 from incident photons.

[0038] In this embodiment, the first metal layer 13 is closest to the active region 10. In addition, the first metal layer 13 includes a region having a shape and size similar to that of the active region 10. For example, this region is equal to or larger than the active region. Therefore, the first metal layer 13 disposed at a distance d1 shields the active region 10 from photons incident on the SPAD1 along the optical axis OA. For example, the optical axis OA constitutes a design choice. For example, the direction of the optical axis can be designed according to a specific field of view. In this particular embodiment, the optical axis OA is defined by the surface normal of the active region 10.

[0039] The first metal layer 13 except for the hole 15 shields the active region 10 from incident photons. The hole 15 is disposed in the metal layer 13 and can have different shapes, such as square, rectangular, circular, etc. Generally, design rules determine the metal width W1. As shown in the figure, the metal width can be defined as the width measured from the edge of the layer to the hole 15. For the metal width W1, a minimum metal spacing S1 can be defined. The minimum metal spacing S1, like the metal width W1, is usually determined by design rules and thus cannot be adjusted with all degrees of freedom.

[0040] The maximum achievable attenuation is usually limited by the design rules of the metal layers used. For example, if only the first metal layer 13 is present, it has 50μm 2The active region and a SPAD with a desired attenuation of 0.5% may have a square hole 15, which is required to be 0.5 × 0.5 μm 2 . However, in some SPAD manufacturing processes, this may violate the inter-metal spacing rule and / or the metal enclosed area rule. In particular, for metal lines that exceed the minimum metal width, the minimum spacing rule is usually much larger than the minimum spacing possible in a certain process. In addition, the maximum metal width and / or the maximum metal density are usually also restricted.

[0041] The second metal layer 14 is arranged on top of the first metal layer 13 at a distance d2. In this embodiment, the second metal layer 14 and the first metal layer 13 are interconnected by metal connectors 16. In other embodiments or additionally, an interlayer dielectric may be provided between the layers. The second metal layer 14 may have different shapes and may or may not have holes that will be further discussed below. For example, compared to the first metal layer 13, the second metal layer 14 has a similar but smaller area. In this particular embodiment, the second metal layer 14 includes a hole 17. In fact, the holes 15, 17 in the two layers are aligned or centered with respect to the optical axis OA.

[0042] The design rules of the second metal layer 14 determine the metal width W2. As shown, the metal width W2 can be defined as the width measured from the edge of layer 14 to the hole 17. For the metal width W2, a minimum metal spacing S2 can be defined. However, compared to the hole 15 in the first metal layer 13, the minimum metal spacing S2 of the second metal layer 14 can be adjusted to a smaller value. A smaller metal width W2 allows a smaller minimum metal gap S2, and thus, compared to the first metal layer 13 and the hole 15, the hole 17 can be made smaller, and ultimately, a higher degree of freedom is allowed in the design of the cover 12 while complying with the design rules determined by the SPAD manufacturing process.

[0043] Two metal layers 13, 14 are arranged together in a stack, and the holes 15, 17 of the metal layers are aligned with respect to the optical axis OA. This arrangement opens an effective hole 18 through which photons can pass along the optical axis OA and ultimately reach the active region 10. Thus, the effective region exposes a portion 19 of the active region 10 to the incident photons. The effective hole 18 is smaller than the hole 15 arranged in the first metal layer 13. In fact, in this embodiment, the effective hole 18 is determined by the hole 17 in the second metal layer 14, or in other words, by the minimum metal gap S2. The portion 19 exposed to the incident photons is determined by the holes 15, 17, that is, by the minimum metal gaps S1, S2 and the distances d1 and d2. For example, the portion 19 can be estimated by geometric optics and ray tracing originating from the hole 17. The metal connector 16 connecting the two metal layers can reduce the influence of lateral stray light. Additional metal layers can be provided and connected on top of the second metal layer through additional metal connectors 16. In this way, the effective hole can be made smaller and the attenuation can be increased.

[0044] Figure 2 Another embodiment of a single photon avalanche diode (SPAD) is shown. The SPAD 1 shown in the upper part of the figure includes two metal layers, namely the first metal layer 13 and the second metal layer 14. Similar to Figure 1 the first metal layer 13 includes a hole 15. The shape of the metal layer and the hole can be the same as Figure 1 in, or can include different shapes, such as rectangular, circular or polygonal. The first metal layer 13 is located at a distance d1 with respect to the active region 10. However, different from the Figure 1 embodiment, the second metal layer 14 is located at a closer distance d2 to the active region 10. In other words, the second metal layer 14 is closest to the active region 10. In addition, the second metal layer 14 does not have a hole, but includes a continuous shape, such as a line, a cross, a rectangle or a square.

[0045] Figure 2 The lower part of the figure shows a top view of the SPAD. The first metal layer 13 is described as a continuous plane completely covering the active region 10. Only the region under the hole 15 is exposed to the incident photons. However, the hole 15 formed in the first metal layer 13 intersects the narrow metal line of the second metal layer 14. The effective hole 18 is defined by the projection of the second metal layer 14 onto the active region 10 through the first metal layer 13. As shown, only the portions 19 above and below the metal line of the second metal layer 14 are exposed to the incident photons. In a more general consideration, the positions of the first metal layer and the second metal layer can be exchanged so that the first metal layer 13 is closest to the active region and vice versa. This applies to all embodiments discussed herein.

[0046] Figure 3Another cross-section of an embodiment of a single-photon avalanche diode (SPAD) is shown. The SPAD1 shown in the figure is based on Figure 2 an embodiment. However, the hole 15 formed in the first metal layer 13 intersects two perpendicular and narrow metal lines of the second metal layer 14. The two metal lines form a cross above or below the first metal layer 13. The effective hole 18 is defined by the projection of the second metal layer 14 through the first metal layer 13 to the active region 10. As shown, only the part 19 of the active region 10 is exposed to the incident photons. In a more general consideration, the positions of the first metal layer and the second metal layer can be exchanged only when the first metal layer 13 is closest to the active region, and vice versa.

[0047] Figure 4 Another embodiment of a single-photon avalanche diode (SPAD) is shown. Similar to Figure 2 that described, the SPAD 1 shown in the upper part of the figure includes two metal layers, namely the first metal layer 13 and the second metal layer 14. The first metal layer 13 and the second metal layer 14 can be exchanged so that the first metal layer 13 is closest to the active region, and vice versa. As shown in the lower part of the figure, the second metal layer 14 has a square shape, which, when viewed from the top, i.e., along the projection along the optical axis OA, is located inside the hole 15 arranged in the first metal layer 13. Therefore, the hole 15 is further reduced, and the stack limits the effective hole 18.

[0048] Figure 5 An embodiment of a single-photon avalanche diode array is shown. The array includes a plurality of SPADs arranged in a 3×3 matrix in a general integrated circuit 28. Nine SPADs 1 to 9 are shown. Each SPAD includes a cathode protection ring region 29 as the cathode, an anode 30, a track region 31 for connecting to the anode, and an active region 10. Only the photons incident on the active region 10 can be detected by the SPAD. The active region 10 is shielded by the anode and the track region, which in turn reduces the active region. In this example, the shape of the active region includes a square, but is not limited to this specific geometry.

[0049] Some, but not necessarily each, of the SPADs are implemented according to the above concepts. In fact, all of the SPADs 1 to 4 and 6 to 9 include a cover 12, and the cover includes at least two metal layers 13, 14. In this specific embodiment, the SPAD 5 is arranged at the center of the array and does not include the cover 12. The remaining SPADs in the array include a first metal layer 13 that completely covers the active region 10 located below the layer. Only the part 19 of the active region 10 is exposed to the incident photons through one or more holes 15 arranged in the metal layer 13.

[0050] The SPADs with the cover 12 can be divided into different subsets 21 to 27. For example, SPADs 1, 4, 7 are grouped into the first subset 21. A second metal layer 14 is present in all of these SPADs, and this metal layer includes narrow metal lines as Figure 2 shown. SPAD 2 includes two holes 15 covered by a metal cross as Figure 3 shown. SPAD 2 constitutes the second subset 22. SPAD 3 includes a hole 15 covered by a metal square as Figure 4 shown. SPAD 3 constitutes the third subset 23. Similarly, SPAD 6 includes a hole 15 covered by a metal square as Figure 4 shown. However, the hole is a bit large as it is a metal square. SPAD 6 constitutes the fourth subset 24. The middle SPAD 5 is not covered and constitutes the fifth subset 25. SPAD 7 includes two holes 15 covered by metal lines as Figure 2 shown. SPAD 7 constitutes the sixth subset 26. Finally, SPAD 9 includes a single hole 15 covered by a metal cross as Figure 3 shown. SPAD 9 constitutes the seventh subset 27.

[0051] As Figure 5 shown, the arrangement of the SPADs and the implementation of the cover are for illustrative purposes only. The number and size of the holes, as well as the shape of the holes and the second metal layer used to cover the first metal layer, can be selected according to the expected application requirements and may be limited only by the applicable design rules. In addition, the first metal layer 13 may be closest to the active region of a given SPAD, or the second metal layer 14 may be closest. Generally, the positions of the metal layers are interchangeable.

[0052] The effect of the cover 12 is that parts 19 of the active region 10 are exposed to incident photons with different attenuations. In fact, as discussed above, the respective effective holes can be reduced to values smaller than 1%. In addition to the effective holes 18 defined by the metal layer, the cover can completely cover the active region 10. By this method, the photons detected by a given SPAD are only those that pass through the effective holes and reach the active region of the SPAD.

[0053] For example, when forming the metal tracks and the anode, the metal layer can be formed. The SPADs including the cover can be fabricated by standard CMOS processes and integrated into a general integrated circuit. For example, a dielectric layer can be provided above the active region. In the next step, a metal, such as aluminum, can be deposited on the dielectric layer. The holes and the shape of a given metal layer can be defined by using a mask or by etching grooves in the electrolyte layer, which can be filled with a metal, such as copper.

[0054] As Figure 5The different coverings 12 shown result in different degrees of attenuation determined respectively by the active apertures 18. The number of photons that can be detected by a given SPAD is reduced due to the degree of attenuation determined by the corresponding covering. Consequently, the SPAD array does not need to be reset as often under varying or even high levels of illumination conditions. In fact, the likelihood of photons reaching the array during the reset phase is reduced. Such an effect is that the array misses fewer photons and linearity can even be extended to absorb the incident light level. In summary, the proposed SPAD array has a higher linear output response to light intensity.

[0055] The control unit 20 is implemented into or connected to the array. The control unit 20 selects one or more SPADs from one subset or different subsets by activating or deactivating the individual SPADs, multiple SPADs or subsets. The actual selection is influenced by the intensity of the incident radiation. This intensity can be determined by a dedicated light sensor or by the array itself.

[0056] Furthermore, or alternatively, all SPADs from a given subset or the entire array can always be activated without the need to activate or deactivate them individually.

[0057] For example, the SPAD includes means for generating a digital output pulse synchronized with the avalanche build-up, e.g. via an inverter. The leading edge of such an output pulse marks the arrival time of the detected photon. The photon-triggered avalanche current persists until the avalanche is quenched by reducing the bias voltage to or below the breakdown voltage of the pn junction. The output pulse of the SPAD can be further processed, e.g. to determine the intensity by counting the number of output pulses within a measurement time slot. Another processing involves measuring the time distribution of the output pulses, e.g. based on a time-to-digital converter. The control unit can determine which output pulses need to undergo such signal processing and ignore the other output signals by selecting or combining a set of output signals of the individual SPADs or subsets, e.g. via a multiplexer, logic such as OR, AND operations. An individual SPAD can be activated or deactivated by a dedicated quenching circuit arranged to quench the avalanche by reducing the bias voltage to or below the breakdown voltage. Such a quenching circuit can also be arranged to restore the SPAD to its operating state.

[0058] In one embodiment, the array includes 128 individual SPADs arranged in a 16×8 matrix. The 16×5 SPADs may not have a covering and thus have no attenuation. Furthermore, for example, the 16×2 SPADs have a covering with 10% attenuation and the 16×1 SPADs have a covering that attenuates the incident light to 1%. Each subset includes more than one SPAD to reduce the effect of dead pixels and further improve the dynamic range of the array.

[0059] Figure 6 shows another embodiment of a single-photon avalanche diode. A top view (A) and a cross-section along the A-A' direction (B) are shown in the figure.

[0060] The top view (A) shows a single SPAD, which can be Figure 5 a part of the array shown. The active region 10 (dashed line) is covered by the first metal layer 13, except for the hole 15. In addition, the second metal layer 14 is disposed on top of the first metal layer 13, and has a smaller area compared to the first metal layer, and has a hole 17. The holes 15, 17 in the first metal layer 13 and the second metal layer 14 define an effective hole 18, such that the region 19 of the active region is exposed to incident photons. In addition, an anode 30 and a track region 31 that establish electrical connection with the active region 10 are shown in the figure. In the top view, the anode 30 is disposed at the center of the effective hole 18.

[0061] The cross-section (B) shows a cut through the SPAD along the A-A' direction (see top view (A)). The active region 10 is disposed in the semiconductor substrate 11. In addition, the first metal layer 13 is connected to the active region through a metal connection 16. The second metal layer 14 is also connected to the first metal layer 13 through the metal connection 16. In a sense, this embodiment constitutes Figure 1 and Figure 4 a combination of... However, connecting the anode and the metal layer through metal connections can be achieved in all embodiments discussed herein.

[0062] When changing the cover from a hole that completely opens the active region to incident photons to an effective hole that blocks most of the active region from photons, the total capacitance of the anode 30 should not change significantly. However, additional capacitance may be added to the anode, and the total charge required to charge the SPAD after triggering will change, and the charging time may increase. The time delay between the incident photons and the detected electron pulse will also change. Such effects may cause different behaviors of the SPADs with different covers in the array, and thus such effects are not desirable.

[0063] These effects can be reduced by connecting the cover and the anode or by keeping the cover floating. Connecting the cover to any other potential, such as VDD, VHV, etc., may result in additional parasitic capacitance from the cover potential to the anode. The embodiment of Figure 6 is an example of connecting the cover to the anode. Here, the metal connection 16 to the anode is completed by the same metallization as a layer of the cover.

[0064] The above embodiments include an internal anode and a cathode implemented as a guard ring region surrounding the anode. Generally, the anode and the cathode can be interchanged and can be regarded as the first electrode and the second electrode, respectively. The shape and position of the electrodes can be determined according to the desired application and are not limited to the examples shown in the above embodiments.

[0065] Reference numeral

[0066] 1 to 9 SPADs

[0067] 10 Active region

[0068] 11 Semiconductor substrate

[0069] 12 Cover

[0070] 13 First metal layer

[0071] 14 Second metal layer

[0072] 15 Hole

[0073] 16 Metal connection

[0074] 17 Hole

[0075] 18 Effective hole

[0076] 19 Exposed portion (of the active region)

[0077] 20 Control unit

[0078] 21 to 27 Subset of SPADs

[0079] 28 Integrated circuit

[0080] 29 Guard ring region

[0081] 30 Anode

[0082] 31 Track region

[0083] d1 Distance

[0084] d2 Distance

[0085] OA Optical axis

[0086] S1 Metal pitch

[0087] S2 Metal pitch

[0088] W1 Metal width

[0089] W2 Metal width

Claims

1. A single-photon avalanche diode (SPAD), comprising: An active region (10) arranged to generate a photon-triggered avalanche current, A cover (12) arranged over the active region (10) and shielding the active region (10) from incident photons; Wherein: The cover (12) comprises a stack of at least a first metal layer (13) and a second metal layer (14), At least the first metal layer (13) comprises holes (15) and the second metal layer (14) comprises holes (17), The first metal layer (13) and the second metal layer (14) are arranged relative to the optical axis (OA) to open an effective hole (18) along the optical axis (OA), through which a portion (19) of the active region (10) is exposed to incident photons, The effective hole (18) is smaller than the holes (15) arranged in the first metal layer (13), and The holes (15) in the first metal layer (13) and the holes (17) in the second metal layer (14) are offset relative to the optical axis (OA).

2. The SPAD according to claim 1, wherein, Except for the portion (19) of the active region (10) exposed through the effective hole (18), the cover (12) completely shields the active region (10) from incident photons.

3. The SPAD according to claim 1 or 2, wherein The area of the active region (10) exposed by the effective hole (18) is less than 50%, 10% or 1% of the entire active region (10).

4. The SPAD according to claim 1 or 2, wherein, The first metal layer (13) and the second metal layer (14) are spaced apart from each other such that the first metal layer (13) has a first distance (d1) from the active region (10), and the second metal layer (14) has a different second distance (d1 + d2) from the active region (10).

5. The SPAD according to claim 1 or 2, wherein, The first metal layer (13) and the second metal layer (14) are interconnected by a metal connection member (16).

6. The SPAD according to claim 1 or 2, wherein, The first metal layer (13) and the second metal layer (14) are interconnected by a vertical electrical connection member.

7. The SPAD according to claim 1, wherein, The optical axis (OA) is a surface normal relative to the active region (10), or an inclined axis at a different angle from the normal relative to the active region (10).

8. The SPAD according to claim 1 or 2, wherein, The second metal layer (14) is arranged below or above the first metal layer (13) such that the projection of the second metal layer (14) onto the first metal layer (13) along the optical axis (OA) at least partially shields the holes (15) in the first metal layer (13) to define the effective hole (18).

9. The SPAD according to claim 1 or 2, wherein, The cover comprises an additional metal layer laminated with the first metal layer (13) and the second metal layer (14).

10. The SPAD according to claim 1 or 2, wherein, The cover (12) and / or the active region (10) has a circular or polygonal shape.

11. The SPAD according to claim 6, wherein, The electrical connection member is a through-silicon via (TSV).

12. The SPAD according to claim 10, wherein, The polygonal shape is a rectangular shape.

13. An array of single-photon avalanche diodes (SPADs), comprising a plurality of SPADs (1, ……, 9) as claimed in claim 1 or 2 arranged in rows and / or columns, wherein The plurality of the SPADs (1, ……, 9) includes a plurality of different SPAD subsets (21, ……, 27), each different subset including at least one SPAD (1, ……, 9) having different active apertures (18); and The control unit (20) is configured to select one or more SPADs (1, ……, 9) from the SPAD subsets (21, ……, 27) according to the incident radiation intensity.

14. The array according to claim 13, wherein, The SPADs (1, ……, 9) arranged in each subset (21, ……, 27) have the same active aperture (18).

15. The array according to claim 13 or 14, wherein, The active apertures (18) of the SPADs (1, ……, 9) arranged in different subsets (21, ……, 27) are different.

16. The array according to claim 13 or 14, wherein, The control unit (20) selects one or more SPADs by activating or deactivating the one or more SPADs (21, ……, 27), activating or deactivating one or more SPAD subsets (21, ……, 27), and / or all SPADs from a given subset or from the entire array can be always activated without being individually activated or deactivated, and the control unit (20) selects one or more SPADs by selecting an output signal.

17. The array according to claim 16, wherein, The output signal is an output pulse.

Images