Light detection device

Abstract

To provide a light detection device that makes it possible to improve performance.SOLUTION: A light detection device comprises: a semiconductor substrate having a first face which light enters and a second face located on the side opposite the first face; and a plurality of sensor pixels disposed on the semiconductor substrate. The plurality of sensor pixels includes a first pixel. The first pixel has a first irregular part provided on the first face and a first photoelectric conversion part capable of photoelectrically converting light incident from the first face and generating electric charges. The first irregular part extends in a first direction in a plan view from the normal direction of the first face. When a cross section of the first irregular part cut by a plane orthogonal to the first direction is assumed to be a first cross section, the first cross section has an asymmetrical structure that the lateral side faces are not symmetrical.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present disclosure relates to a light detection device. [Background technology]

[0002] In imaging devices such as back-illuminated CMOS (Complementary Metal Oxide Semiconductor) image sensors, a concave-convex structure (e.g., a moth-eye structure) may be provided on the incident surface of a semiconductor substrate (see Patent Document 1). The concave-convex structure may, for example, tilt the incident light to extend the optical path length, thereby improving the photoelectric conversion efficiency Qe (Quantum Efficiency).

[0003] Also, an imaging device is known that has a structure in which a photodiode and a memory are stacked in the light incidence direction, and a light-shielding film is disposed between the photodiode and the memory (see, for example, Patent Document 2). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] International Publication No. 2017 / 126329 [Patent Document 2] International Publication No. 2021 / 117648 Summary of the Invention [Problem to be solved by the invention]

[0005] When incident light passes through a photodiode and enters an unintended area (e.g., a memory), photoelectric conversion occurs in the unintended area, which may generate an unintended charge signal. This unintended charge signal may degrade the performance of the imaging device.

[0006] The present disclosure has been made in consideration of the above circumstances, and has an object to provide a photodetector capable of suppressing degradation in performance. [Means for solving the problem]

[0007] A photodetector according to an embodiment of the present disclosure includes a semiconductor substrate having a first surface on which light is incident and a second surface located opposite to the first surface, and a plurality of sensor pixels provided on the semiconductor substrate. The plurality of sensor pixels include a first pixel. The first pixel has a first uneven portion provided on the first surface, and a first photoelectric conversion portion capable of generating an electric charge by photoelectrically converting light incident from the first surface. The first uneven portion extends in a first direction in a plan view from a normal direction to the first surface. When a cross section of the first uneven portion cut by a plane perpendicular to the first direction is defined as a first cross section, the first cross section has an asymmetric structure in which left and right side surfaces are not symmetrical.

[0008] According to this, the light (incident light) incident on the first concave-convex portion can be refracted and can travel obliquely with respect to the first surface. This can extend the optical path length in the first photoelectric conversion portion, thereby improving the photoelectric conversion efficiency Qe. In addition, since the concave-convex portion has an asymmetric structure, it becomes easy to direct the traveling direction of the incident light in a specific direction within the semiconductor substrate, and it is possible to prevent the incident light from being incident on an unintended region. This can prevent the generation of unintended signal charges, making it possible to reduce PLS (Parastic Light Sensitibity). This can prevent a decrease in the performance of the photodetector. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a block diagram illustrating an example of the functional configuration of an imaging device according to an embodiment of the present disclosure. [Diagram 2] FIG. 2 is an equivalent circuit diagram showing an example of the configuration of the readout circuit. [Diagram 3] FIG. 3 shows a planar layout of a pixel region having two pixels in the X direction and four pixels in the Y direction. [Figure 4] FIG. 4 is a diagram illustrating a first example of a cross-sectional configuration of an imaging device according to an embodiment of the present disclosure. [Diagram 5] FIG. 5 is a diagram illustrating a second cross-sectional configuration example of an imaging device according to an embodiment of the present disclosure. [Figure 6] 6A to 6C are cross-sectional views showing a process sequence of a method for forming a concave-convex portion suitable for an imaging device according to an embodiment of the present disclosure. [Figure 7A] FIG. 7A is a plan view diagrammatically illustrating the anisotropy of alkaline wet etching of a Si(111) substrate. [Figure 7B] FIG. 7B is a plan view that illustrates the anisotropy of alkaline wet etching of a Si(111) substrate. [Figure 8] FIG. 8 is a diagram showing an example of the cross-sectional shape of the protrusions of the concave-convex portion. [Figure 9] FIG. 9 is a diagram showing a layout example (part 1) of trenches and uneven portions formed on a Si(111) substrate. [Figure 10] FIG. 10 is a diagram showing a layout example (part 2) of trenches and uneven portions formed on a Si(111) substrate. [Figure 11] FIG. 11 is a diagram showing a layout example (part 3) of trenches and uneven portions formed on a Si(111) substrate. [Figure 12] FIG. 12 is a plan view (part 1) showing variations in layout examples (minimum unit) of trenches and uneven portions formed on a Si(111) substrate. [Figure 13] FIG. 13 is a plan view (part 2) showing variations in layout examples (minimum unit) of trenches and uneven portions formed on a Si(111) substrate. [Figure 14] FIG. 14 is a plan view (part 3) showing variations in layout examples (minimum unit) of trenches and uneven portions formed on a Si(111) substrate. [Figure 15] FIG. 15 is a plan view illustrating a configuration example of a color filter according to an embodiment of the present disclosure. [Figure 16] FIG. 16 is a plan view showing an example of a mixed arrangement of concave and convex portions according to an embodiment of the present disclosure. [Figure 17]FIG. 17 is a cross-sectional view showing a configuration of an imaging device according to a modified example of the embodiment of the present disclosure. [Figure 18] FIG. 18 is a cross-sectional view showing a configuration of an imaging device according to a modified example of the embodiment of the present disclosure. [Figure 19] FIG. 19 is a cross-sectional view showing a configuration of an imaging device according to a modified example of the embodiment of the present disclosure. [Figure 20] FIG. 20 is a cross-sectional view showing a configuration of an imaging device according to a modified example of the embodiment of the present disclosure. [Figure 21] FIG. 21 is a cross-sectional view showing a configuration of an imaging device according to a modified example of the embodiment of the present disclosure. [Figure 22] FIG. 22 is a cross-sectional view showing a configuration of an imaging device according to a modified example of the embodiment of the present disclosure. [Diagram 23] FIG. 23 is a cross-sectional view showing a configuration of an imaging device according to a modified example of the embodiment of the present disclosure. [Figure 24] FIG. 24 is a cross-sectional view showing a configuration of an imaging device according to a modified example of the embodiment of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, an embodiment of the present disclosure will be described with reference to the drawings. In the description of the drawings referred to in the following description, the same or similar parts are denoted with the same or similar reference numerals. However, it should be noted that the drawings are schematic, and the relationship between thickness and planar dimensions, the thickness ratio of each layer, etc., differ from the actual ones. Therefore, the specific thickness and dimensions should be determined with reference to the following description. In addition, it goes without saying that the drawings include parts with different dimensional relationships and ratios.

[0011] The definitions of directions such as up and down in the following description are merely for the convenience of explanation and do not limit the technical ideas of the present disclosure. For example, if an object is rotated 90 degrees and observed, up and down are converted to left and right and read, and if it is rotated 180 degrees and observed, up and down are obviously read inverted.

[0012] In the following description, the thickness direction of the semiconductor substrate 11 means the direction from one of the front surface 11A and the back surface 11B of the semiconductor substrate to the other. Also, a plan view means a view from the normal direction of the back surface 11B of the semiconductor substrate 11.

[0013] <Embodiment> Hereinafter, an embodiment of the present disclosure will be described in detail. An imaging device will be described as an example of the "photodetection device" of the present disclosure. The imaging device is, for example, a global shutter type back-illuminated image sensor using a CMOS (Complementary Metal Oxide Semiconductor) image sensor or the like. The imaging device receives light from a subject for each pixel, performs photoelectric conversion, and generates a pixel signal, which is an electrical signal.

[0014] The global shutter method is a method in which exposure of all pixels starts and ends simultaneously. Here, all pixels refer to all pixels that form a valid image, excluding dummy pixels that do not contribute to image formation. Also, as long as image distortion and exposure time differences are small enough that they do not cause problems, they do not necessarily have to be simultaneous. For example, the global shutter method also includes a case in which the operation of performing simultaneous exposure in units of multiple rows (several dozen rows, etc.) is repeated while shifting the rows in units of multiple rows in the row direction. The global shutter method also includes a case in which simultaneous exposure is performed only on a portion of the pixel area.

[0015] A back-illuminated image sensor is an image sensor in which a photoelectric conversion unit such as a photodiode that receives light from a subject and converts it into an electrical signal is disposed for each pixel between a light receiving surface on which light from a subject is incident and a wiring layer on which wiring such as transistors that drive each pixel is provided. Note that the present disclosure may also be applicable to image sensors of imaging methods other than CMOS image sensors.

[0016] (Example of overall configuration of imaging device) Fig. 1 is a block diagram showing an example of the overall configuration of an imaging device 101 according to an embodiment of the present disclosure. The imaging device 101 in Fig. 1 is formed on a semiconductor substrate 11, and therefore is, more accurately, a solid-state imaging device, but will be simply referred to as the imaging device 101 below. The imaging device 101 in Fig. 1 includes a pixel array section 111 in which a plurality of sensor pixels 121 that perform photoelectric conversion are arranged in a matrix, i.e., in a two-dimensional plane. The sensor pixel 121 is an example of a "pixel" in the present disclosure. The pixel signal photoelectrically converted in the pixel array section 111 is read out via a readout circuit.

[0017] The imaging device 101 includes, for example, a pixel array section 111, a vertical driving section 112, a ramp wave module 113, a column signal processing section 114, a clock module 115, a data storage section 116, a horizontal driving section 117, a system control section 118, and a signal processing section 119.

[0018] The imaging device 101 is composed of a single or multiple semiconductor substrates 11. For example, the imaging device 101 can be composed by electrically connecting, to the semiconductor substrate 11 on which the pixel array section 111 is formed, another semiconductor substrate 11 on which the vertical driving section 112, the ramp wave module 113, the column signal processing section 114, the clock module 115, the data storage section 116, the horizontal driving section 117, the system control section 118, the signal processing section 119, and the like are formed, by Cu-Cu bonding or the like.

[0019] The pixel array section 111 has a plurality of sensor pixels 121 including photoelectric conversion elements that generate and accumulate electric charges according to the amount of light incident from a subject. The sensor pixels 121 are arranged in both the horizontal direction (row direction) and the vertical direction (column direction) as shown in Fig. 1. In the pixel array section 111, a pixel drive line 122 is wired along the row direction for each pixel row made up of sensor pixels 121 arranged in a row in the row direction, and a vertical signal line 123 is wired along the column direction for each pixel column made up of sensor pixels 121 arranged in a row in the column direction.

[0020] The vertical driving unit 112 includes a shift register, an address decoder, etc. The vertical driving unit 112 supplies signals and the like to the plurality of sensor pixels 121 via the plurality of pixel driving lines 122, thereby driving all of the plurality of sensor pixels 121 in the pixel array unit 111 simultaneously or driving them on a pixel row basis.

[0021] The ramp module 113 generates a ramp signal used for A / D (Analog / Digital) conversion of pixel signals and supplies it to a column signal processing unit 114. The column signal processing unit 114 is composed of, for example, a shift register, an address decoder, etc., and performs noise removal processing, correlated double sampling processing, A / D conversion processing, etc. to generate pixel signals. The column signal processing unit 114 supplies the generated pixel signals to a signal processing unit 119.

[0022] The clock module 115 supplies each unit of the image capture device 101 with a clock signal for operation.

[0023] The horizontal driving unit 117 sequentially selects the unit circuits corresponding to the pixel columns of the column signal processing unit 114. Through the selective scanning by the horizontal driving unit 117, the pixel signals that have been signal-processed for each unit circuit in the column signal processing unit 114 are output to the signal processing unit 119 in sequence.

[0024] The system control unit 118 is composed of a timing generator that generates various timing signals, etc. The system control unit 118 controls the driving of the vertical driving unit 112, the ramp wave module 113, the column signal processing unit 114, the clock module 115, and the horizontal driving unit 117 based on the timing signals generated by the timing generator.

[0025] The signal processing unit 119 performs signal processing such as arithmetic processing on the pixel signals supplied from the column signal processing unit 114 while temporarily storing data in the data storage unit 116 as necessary, and outputs an image signal made up of each pixel signal.

[0026] (Example of configuration of readout circuit and pixel area) Fig. 2 is an equivalent circuit diagram showing an example of the configuration of the readout circuit 120. Fig. 3 is a planar layout diagram showing an example of the configuration of a part of a pixel region in the pixel array section 111. Fig. 3 shows a planar layout of a pixel region with two pixels in the X direction and four pixels in the Y direction.

[0027] 2 and 3, the readout circuit 120 has four transfer transistors TRZ, TRY, TRX, TRG, a discharge transistor OFG, a reset transistor RST, an amplification transistor AMP, and a selection transistor SEL. These transistors are N-type MOS transistors. In the layout example shown in FIG. 3, the reset transistor RST, the amplification transistor AMP, and the selection transistor SEL are formed on and bonded to a semiconductor substrate separate from the semiconductor substrate 11 on which the pixel array section 111 is arranged. For this reason, the reset transistor RST, the amplification transistor AMP, and the selection transistor SEL are not shown in the planar layout of FIG. 3.

[0028] The following mainly describes an example in which a photodiode PD is used as the photoelectric conversion unit 51. The transfer transistor TRZ is connected to the photodiode PD in the sensor pixel 121, and transfers the charge (pixel signal) photoelectrically converted by the photodiode PD to the transfer transistor TRY. The transfer transistor TRZ is assumed to be a vertical transistor, and has a vertical gate electrode.

[0029] The transfer transistor TRY transfers the charge transferred from the transfer transistor TRZ to the transfer transistor TRX. The transfer transistors TRY and TRX may be replaced with one transfer transistor. A charge holding unit (MEM) 54 is connected to the transfer transistors TRY and TRX. The potential of the charge holding unit (MEM) 54 is controlled by a control signal applied to each gate electrode of the transfer transistors TRY and TRX.

[0030] For example, when the transfer transistors TRY, TRX are turned on, the potential of the charge holding unit (MEM) 54 becomes deeper, and when the transfer transistors TRY, TRX are turned off, the potential of the charge holding unit (MEM) 54 becomes shallower. Then, for example, when the transfer transistors TRZ, TRY, TRX are turned on, the charge stored in the photodiode PD is transferred to the charge holding unit (MEM) 54 via the transfer transistors TRZ, TRY, TRX. The drain of the transfer transistor TRX is electrically connected to the source of the transfer transistor TRG, and the gates of the transfer transistors TRY, TRX are connected to the pixel drive line.

[0031] The charge holding unit (MEM) 54 is an area that temporarily holds the charge accumulated in the photodiode PD in order to realize a global shutter function. The charge holding unit (MEM) 54 holds the charge transferred from the photodiode PD.

[0032] The transfer transistor TRG is connected between the transfer transistor TRX and the floating diffusion FD, and transfers the charge held in the charge holding unit (MEM) 54 to the floating diffusion FD in response to a control signal applied to the gate electrode. For example, when the transfer transistor TRX is turned off and the transfer transistor TRG is turned on, the charge held in the charge holding unit (MEM) 54 is transferred to the floating diffusion FD. The drain of the transfer transistor TRG is electrically connected to the floating diffusion FD, and the gate of the transfer transistor TRG is connected to the pixel drive line.

[0033] The floating diffusion FD is a floating diffusion region that temporarily holds the charge output from the photodiode PD via the transfer transistor TRG. For example, a reset transistor RST is connected to the floating diffusion FD, and a vertical signal line VSL is connected to the floating diffusion FD via an amplification transistor AMP and a selection transistor SEL.

[0034] The emission transistor OFG initializes (resets) the photodiode PD in response to a control signal applied to its gate electrode. The drain of the emission transistor OFG is connected to a power supply line VDD, and the source is connected between the transfer transistors TRZ and TRY.

[0035] For example, when the transfer transistor TRZ and the discharge transistor OFG are turned on, the potential of the photodiode PD is reset to the potential level of the power supply line VDD. That is, the photodiode PD is initialized. In addition, the discharge transistor OFG forms an overflow path between the transfer transistor TRZ and the power supply line VDD, for example, and discharges the charge overflowing from the photodiode PD to the power supply line VDD.

[0036] The reset transistor RST initializes (resets) each region from the charge holding unit (MEM) 54 to the floating diffusion FD in response to a control signal applied to the gate electrode. The drain of the reset transistor RST is connected to a power supply line VDD, and the source is connected to the floating diffusion FD. For example, when the transfer transistor TRG and the reset transistor RST are turned on, the potentials of the charge holding unit (MEM) 54 and the floating diffusion FD are reset to the potential level of the power supply line VDD. That is, by turning on the reset transistor RST, the charge holding unit (MEM) 54 and the floating diffusion FD are initialized.

[0037] The amplifier transistor AMP has a gate electrode connected to the floating diffusion FD and a drain connected to a power supply line VDD, and serves as an input section of a source follower circuit that reads out charges obtained by photoelectric conversion in the photodiode PD. That is, the amplifier transistor AMP has a source connected to the vertical signal line VSL via the selection transistor SEL, and thus constitutes a source follower circuit together with a constant current source connected to one end of the vertical signal line VSL.

[0038] The selection transistor SEL is connected between the source of the amplification transistor AMP and the vertical signal line VSL, and a control signal is supplied to the gate electrode of the selection transistor SEL as a selection signal. When the control signal is turned on, the selection transistor SEL is turned on, and the sensor pixel 121 connected to the selection transistor SEL is turned on. When the sensor pixel 121 is turned on, the pixel signal output from the amplification transistor AMP is read out to the column signal processing circuit 22 via the vertical signal line VSL.

[0039] 3, the transfer transistors TRG, TRX, TRY, TRZ and the discharge transistor OFG in the readout circuit 120 of one sensor pixel 121 are arranged in order in the Y direction. The arrangement of the transistors in two sensor pixels 121 adjacent to each other in the Y direction is symmetrical with respect to the pixel boundary in the Y direction. The arrangement of the transistors in the readout circuits 120 for two sensor pixels 121 adjacent to each other in the X direction is alternately reversed and the same.

[0040] A charge holding unit (MEM) 54 is disposed at a position facing the transfer transistors TRG, TRX, and TRY (below in FIG. 4 described later). In addition, in FIG. 4 described later, the photodiode PD in one sensor pixel 121 is disposed so as to straddle below the transfer transistors TRG, TRX, and TRY of that sensor pixel 121 and below the discharge transistor OFG and transfer transistors TRZ, and TRY of the sensor pixel 121 adjacent in the X direction.

[0041] The planar layout of each transistor in the readout circuit 120 is not limited to the aspect shown in Fig. 3. If the arrangement of each transistor in the readout circuit 120 is changed, the locations of the photodiode PD and the charge storage unit (MEM) 54 arranged below them will also change.

[0042] (Example of cross-sectional configuration of an imaging device) (1) Configuration example 1 FIG. 4 is a diagram showing a cross-sectional configuration example 1 of an imaging device 101 according to an embodiment of the present disclosure. FIG. 4 shows a cross section obtained by cutting the planar layout diagram shown in FIG. 3 along line AA. The symbols "P" and "N" in FIG. 4 represent a P-type semiconductor region and an N-type semiconductor region, respectively. In addition, + and - attached to P and N mean that the semiconductor region has a relatively high or low impurity concentration, respectively, compared to a semiconductor region not marked with + or -. However, even if the semiconductor regions are marked with the same P and P (or the same N and N), it does not mean that the impurity concentrations of the respective semiconductor regions are strictly the same. This also applies to the subsequent drawings.

[0043] As shown in FIG. 4, the imaging device 101 includes a semiconductor substrate 11, a photoelectric conversion section 51 (an example of a “first photoelectric conversion section” in the present disclosure), a charge holding section (MEM) 54, a charge transfer section 50, a vertical gate electrode 52V which is a vertical electrode of the transfer transistor TRZ, a first light-shielding section 12, and a second light-shielding section 13.

[0044] The semiconductor substrate 11 is, for example, a single crystal silicon substrate having a crystal orientation with a plane index (111). The front surface 11A and the back surface 11B of the semiconductor substrate 11 are each a (111) plane. Hereinafter, the semiconductor substrate 11 may also be referred to as a Si(111) substrate. One of the reasons for using a Si(111) substrate is that, as will be described later, a process of alkaline wet etching in a direction along the crystal plane is included. Here, the plane index (111) is intended to include crystal orientations that are opposite to any direction in three-dimensional directions, such as (-111), (1-11), and (11-1).

[0045] In addition, the imaging device 101 includes inter-pixel isolation portions 20 and 30, an etching stopper 17, a transparent insulating film 56, a color filter CF, and a light-receiving lens LNS. In this specification, one principal surface of the semiconductor substrate 11 on which the light-receiving lens LNS is arranged is referred to as a back surface 11B or a light-receiving surface, and one principal surface on which the readout circuit 120 is arranged is referred to as a front surface 11A.

[0046] The photoelectric conversion unit 51 in the semiconductor substrate 11 has, for example, an N-type semiconductor region 51A, an N-type semiconductor region 51B, and a P-type semiconductor region 51C in this order from the position closest to the back surface 11B. The light incident on the back surface 11B is photoelectrically converted in the N-type semiconductor region 51A to generate charges, and then the charges are accumulated in the N-type semiconductor region 51B. The boundary between the N-type semiconductor region 51A and the N-type semiconductor region 51B is not necessarily clear, and it is sufficient that the N-type impurity concentration gradually increases from the N-type semiconductor region 51A to the N-type semiconductor region 51B. In addition, a P+ type semiconductor region having a higher P-type impurity concentration than the P-type semiconductor region 51C may be provided between the N-type semiconductor region and the P-type semiconductor region 51C. In this way, the layer configuration of the photoelectric conversion unit 51 formed in the semiconductor substrate 11 is not necessarily limited to that shown in FIG. 4.

[0047] The first light-shielding portion 12 is a member that functions to prevent light from entering the charge-holding portion (MEM) 54, and is provided to surround the charge-holding portion (MEM) 54. Specifically, the first light-shielding portion 12 has a two-layer structure made up of an inner layer portion 12A and an outer layer portion 12B that surrounds the inner layer portion 12A. The inner layer portion 12A is made of a material that includes at least one of a light-shielding elemental metal, a metal alloy, a metal nitride, and a metal silicide.

[0048] For example, the material of the inner layer portion 12A may be Al (aluminum), Cu (copper), Co (cobalt), W (tungsten), Ti (titanium), Ta (tantalum), Ni (nickel), Mo (molybdenum), Cr (chromium), Ir (iridium), platinum iridium, TiN (titanium nitride), or a tungsten silicon compound. Among them, Al (aluminum) is the most optically preferable material. The inner layer portion 12A may be made of graphite or an organic material. The outer layer portion 12B is made of an insulating material such as silicon oxide (SiO2). The outer layer portion 12B ensures electrical insulation between the inner layer portion 12A and the semiconductor substrate 11.

[0049] The second light-shielding portion 13 is disposed closer to the back surface 11B of the semiconductor substrate 11 than the first light-shielding portion 12. The second light-shielding portion 13 has a vertical light-shielding portion 13V extending in the thickness direction of the semiconductor substrate 11 (which is also the depth direction from the back surface 12B; the Z-axis direction) and a horizontal light-shielding portion 13H extending in the horizontal direction of the semiconductor substrate 11. The vertical light-shielding portion 13V also serves as the inter-pixel separation portion 30.

[0050] As shown in FIG. 4, the cross-sectional shape of the second light-shielding portion 13 is a T-shape formed by the vertical light-shielding portion 13V and the horizontal light-shielding portion 13H. The horizontal light-shielding portion 13H is connected to the end of the vertical light-shielding portion 13V on the front surface 11A side. The horizontal light-shielding portion 13H is disposed at a position overlapping with the vertical gate electrode 52V in the thickness direction (Z-axis direction). This prevents light incident from the rear surface 11B side of the semiconductor substrate 11 from being directly incident on the vertical gate electrode 52V by being shielded by the horizontal light-shielding portion 13H. The second light-shielding portion 13 is made of a material having excellent light reflection characteristics (e.g., Al) or a material having excellent light absorption characteristics (e.g., W).

[0051] The inter-pixel isolation parts 20 and 30 are provided along the boundaries between pixels. The inter-pixel isolation parts 30 correspond to the vertical light shielding parts 13V of the second light shielding part 13 described above. The inter-pixel isolation parts 20 and 30 are wall-like members that extend in the depth (Z-axis) direction along the boundaries between adjacent sensor pixels 121 and surround each photoelectric conversion part 51. The inter-pixel isolation parts 20 and 30 can electrically isolate the adjacent sensor pixels 121 from each other.

[0052] The inter-pixel isolation section 20 is made of an insulating material such as silicon oxide (SiO2). Alternatively, like the vertical light shielding section 13V, the inter-pixel isolation section 20 is made of a material with excellent light reflection characteristics (e.g., Al) or a material with excellent light absorption characteristics (e.g., W). This allows the inter-pixel isolation section 20 to prevent light from being incident from one adjacent sensor pixel 121 to the other adjacent sensor pixel 121, thereby preventing color mixing.

[0053] 4, inter-pixel separation portion 20 has only a vertical light-shielding portion, but inter-pixel separation portion 20 may have a vertical light-shielding portion and a horizontal light-shielding portion. The cross-sectional shape of inter-pixel separation portion 20 may be various shapes, such as a T-shape or a cross shape.

[0054] Both of the inter-pixel separation portions 20, 30 (vertical light shielding portions 13V) can prevent light incident on each sensor pixel 121 from the back surface 11B side of the semiconductor substrate 11 from leaking out to adjacent sensor pixels 121, thereby reducing inter-pixel crosstalk.

[0055] The gate electrodes of the transfer transistors TRZ, TRY, TRX, and TRG in the readout circuit 120 and the discharge transistor OFG are all provided on the front surface 11A side of the semiconductor substrate 11 via an insulating layer 18. The charge holding unit (MEM) 54, which is an N-type semiconductor region, is provided in a P-type semiconductor region 51C in the semiconductor substrate 11. More specifically, the charge holding unit (MEM) 54 is disposed between the front surface 11A of the semiconductor substrate 11 and the horizontal light shielding unit 12H of the first light shielding unit 12. As shown in FIG. 4, the first light shielding unit 12 surrounds the periphery of the charge holding unit (MEM) 54 so that light from the back surface 11B side is not incident on the charge holding unit (MEM) 54.

[0056] The transfer transistor TRZ has a horizontal gate electrode 52H arranged in the horizontal plane direction of the semiconductor substrate 11, and a vertical gate electrode 52V extending in the thickness direction (Z-axis direction) of the semiconductor substrate 11. The deepest position of the vertical gate electrode 52V is, for example, in the N-type semiconductor region 52A. In the example of FIG. 4, an example is shown in which each sensor pixel 121 has two vertical gate electrodes 52V, but there is no limit to the number of vertical gate electrodes 52V, and one or more may be used. The transfer transistor TRZ transfers the charge photoelectrically converted by the photoelectric conversion unit 51 to the transfer transistor TRY via the vertical gate electrode 52V.

[0057] As shown in FIG. 4, a planarization film 55 is provided between the rear surface 11B and the color filter CF. The planarization film 55 is provided along the rear surface 11B of the semiconductor substrate 11. The planarization film 55 connects the rear surface 11B and the color filter CF in close contact. The planarization film 55 is made of, for example, an organic material such as a resin or an insulating film such as silicon oxide (SiO2). The refractive index of the planarization film 55 is lower than that of the semiconductor substrate 11. If the refractive index of the semiconductor substrate 11 is a and the refractive index of the planarization film 55 is b, a>b.

[0058] 4, a transparent insulating film 56 may be interposed between the rear surface 11B and the planarization film 55. The transparent insulating film 56 is provided along the uneven portion 60. The transparent insulating film 56 may be, for example, silicon oxide (SiO2), silicon nitride (SiN), silicon oxynitride (SiON), hafnium oxide (HfO2), aluminum oxide (Al2O3), zirconium oxide (ZrO2), tantalum oxide (Ta2O5), titanium oxide (TiO2), lanthanum oxide (La2O3), praseodymium oxide (Pr2O3), cerium oxide (CeO2), neodymium oxide (Nd ... The transparent insulating film 56 is made of, for example, fluorine oxide (Pm2O3), samarium oxide (Sm2O3), europium oxide (Eu2O3), gadolinium oxide (Gd2O3), terbium oxide (Tb2O3), dysprosium oxide (Dy2O3), holmium oxide (Ho2O3), thulium oxide (Tm2O3), ytterbium oxide (Yb2O3), lutetium oxide (Lu2O3), yttrium oxide (Y2O3), resin, etc. The transparent insulating film 56 may be made of one or more of these materials stacked together.

[0059] A color filter CF is disposed on the surface of the planarizing film 55, and a light-receiving lens LNS is disposed on the surface of the color filter CF. The color filter CF and the light-receiving lens LNS are provided for each pixel.

[0060] (Uneven part on the back) As shown in FIG. 4, the semiconductor substrate 11 has an uneven portion 60 (an example of a "first uneven portion" of the present disclosure) provided on the rear surface 11B. In the example shown in FIG. 4, each of the multiple sensor pixels 121 has multiple uneven portions 60. The uneven portion 60 can mitigate a sudden change in refractive index at the interface, which is one of the causes of light reflection, and reduce the influence of reflected light. In other words, the refractive index changes gradually along the direction in which light is incident, thereby reducing light reflection.

[0061] The uneven portion 60 extends in a first direction (for example, the Y-axis direction in FIG. 4) in a plan view from the rear surface 11B. The depth d60 of the uneven portion 60 is the same among the multiple sensor pixels 121. The depth of the uneven portion 60 is indicated by the distance from the apex of the convex portion of the uneven portion 60 to the bottom surface of the concave portion in the thickness direction of the semiconductor substrate 11 (the Z-axis direction).

[0062] If a cross section of the uneven portion 60 cut in a plane (XZ plane) perpendicular to the first direction (Y-axis direction) is taken as a first cross section, this first cross section has an asymmetric structure in which the left and right side surfaces are not symmetric. For example, the uneven portion 60 includes a convex portion 61 and a concave portion 62. In addition, side surfaces 63 and 64 (one example of the "left and right side surfaces" of the present disclosure) are provided between the convex portion 61 and the concave portion 62. In a cross section of the uneven portion 60 cut in the XZ plane, the left and right side surfaces 63 and 64 are inclined with respect to the thickness direction (Z-axis direction) of the semiconductor substrate 11. In a cross section of the uneven portion 60 cut in the XZ plane, the left and right side surfaces 63 and 64 are not symmetric. The inclination angles with respect to the thickness direction (Z-axis direction) of the semiconductor substrate 11 are different between the side surface 63 and the side surface 64.

[0063] For example, each of the sensor pixels 121 has a plurality of concave-convex portions 60. The plurality of concave-convex portions 60 are aligned in a second direction (X-axis direction) intersecting the first direction in a plan view.

[0064] As shown in Fig. 4, incident light L that passes through the rear surface 11B is scattered by the uneven portion 60. For example, incident light L that is perpendicularly incident on the rear surface 11B passes through the uneven portion 60, is refracted, and travels obliquely within the N-type semiconductor region 51A. In this way, the traveling direction of the incident light L changes obliquely. This makes it possible to extend the optical path length.

[0065] 4, the incident light L passing through the uneven portion 60 is reflected by the vertical light shielding portion 13V and the horizontal light shielding portion 13H of the second light shielding portion 13, and the first light shielding portion 12. By increasing the number of reflection points of the incident light L, the optical path length can be further extended. The first light shielding portion 12 and the second light shielding portion 13 can confine the incident light L within the N-type semiconductor region 51A to easily cause diffuse reflection, and the optical path length can be extended. As a result, the incident light L is easily converted into electric charges in the N-type semiconductor region 51A, and the photoelectric conversion efficiency Qe can be improved.

[0066] Furthermore, since the uneven portion 60 has an asymmetric structure as described above, it becomes easy to refract (bend) the incident light L incident on the rear surface 11B toward a specific direction (for example, toward the horizontal light shielding portion 13H). This extends the optical path length of the incident light L and prevents the incident light L from directly entering the charge holding portion (MEM) 54 or the floating diffusion FD. Since it is possible to suppress the generation of unintentional signal charges in the charge holding portion (MEM) 54 or the floating diffusion FD, it becomes possible to reduce PLS (Parastic Light Sensitibity). This effect of reducing PLS becomes more noticeable in the sensor pixel 121 that detects light of a red (R) component that is long wavelength and difficult to attenuate.

[0067] (2) Configuration example 2 FIG. 5 is a diagram showing a cross-sectional configuration example 2 of the imaging device 101 according to the embodiment of the present disclosure. FIG. 5 is drawn upside down compared to FIG. 4. In the cross-sectional configuration example 2 shown in FIG. 5, the vertical light shielding portion 12V of the first light shielding portion 12 also serves as the inter-pixel separation portion 20. As shown in FIG. 5, the vertical light shielding portion 12V is embedded, for example, in a trench provided from the rear surface 11B toward the front surface 11A of the semiconductor substrate 11 via an insulating film (for example, the transparent insulating film 56 shown in FIG. 4). The vertical light shielding portion 12V extends in the depth (Z-axis) direction along the boundary position between adjacent sensor pixels 121.

[0068] The cross-sectional shape of the first light-shielding portion 12 is a T-shape formed by the vertical light-shielding portion 12V and the horizontal light-shielding portion 12H. The horizontal light-shielding portion 12H is connected to the end of the vertical light-shielding portion 12V on the front surface 11A side. The horizontal light-shielding portion 12H extends from one side of the vertical light-shielding portion 12V (i.e., the inter-pixel separation portion 20) toward the center of the sensor pixel 121, and is disposed between the rear surface 11B, which is the light-receiving surface, and the charge storage portion (MEM) 54. The horizontal light-shielding portion 12H is disposed at a position overlapping with the charge storage portion (MEM) 54 in the thickness direction (Z-axis direction). As a result, the light incident from the rear surface 11B side of the semiconductor substrate 11 is shielded by the horizontal light-shielding portion 12H, and is prevented from directly entering the charge storage portion (MEM) 54. The first light-shielding portion 12 is made of, for example, a material having excellent light reflection characteristics (for example, Al) or a material having excellent light absorption characteristics (for example, W).

[0069] The horizontal light shielding portion 13H of the second light shielding portion 13 also extends from the vertical light shielding portion 13V (i.e., the inter-pixel separation portion 30) toward the center of the sensor pixel 121, and is disposed between the back surface 11B, which is the light receiving surface, and the vertical gate electrode 52V of the transfer transistor TRZ. The horizontal light shielding portion 13H is disposed at a position overlapping with the vertical gate electrode 52V in the thickness direction (Z-axis direction). As a result, light incident from the back surface 11B side of the semiconductor substrate 11 is shielded by the horizontal light shielding portion 13H, and is prevented from directly entering the vertical gate electrode 52V. The second light shielding portion 13 is made of, for example, a material having excellent light reflection characteristics (for example, Al) or a material having excellent light absorption characteristics (for example, W).

[0070] 5, horizontal light shielding portion 13H of second light shielding portion 13 is located closer to back surface 13B than horizontal light shielding portion 12H of first light shielding portion 12. End portion 12He of horizontal light shielding portion 12H on the pixel center side and end portion 13He of horizontal light shielding portion 13H on the pixel center side are disposed at positions overlapping each other in the thickness direction of semiconductor substrate 11 (Z-axis direction).

[0071] 5 shows a case where the vertical light shielding portion 12V has a form of DTI (Deep Trench Isolation) that does not penetrate the semiconductor substrate 11, but this is merely an example. The vertical light shielding portion 12V may penetrate between the back surface 11B and the front surface 11A of the semiconductor substrate 11.

[0072] In the second cross-sectional configuration example, like the first cross-sectional configuration example, it is possible to improve the photoelectric conversion efficiency Qe and reduce the PLS.

[0073] In the cross-sectional configuration example 2 shown in FIG. 5, the cross-sectional shapes of the uneven portions 60 are different between adjacent sensor pixels 121. For example, the sensor pixels 121 include a sensor pixel 121(R) (an example of a "first pixel" in the present disclosure) that detects light of a red (R) component, a sensor pixel 121(G) (an example of a "second pixel" in the present disclosure) that detects light of a green (G) component, and a sensor pixel 121(G) (not shown) that detects light of a blue (B) component. The sensor pixel 121(R) that detects red light has a red color filter CF(R) on the back surface 11B side, the sensor pixel 121(G) that detects green light has a green color filter CF(G) on the back surface 11B side, and the sensor pixel 121(B) that detects blue light has a blue color filter CF(B) (not shown) on the back surface 11B side. Hereinafter, the sensor pixel that detects red light is also referred to as a "red sensor pixel". The sensor pixel that detects green light is also referred to as a "green sensor pixel". A sensor pixel that detects blue light is also called a "blue sensor pixel."

[0074] The red sensor pixel 121(R) has an uneven portion 60(R) (an example of a "first uneven portion" in the present disclosure) provided on the rear surface 11B of the semiconductor substrate 11. The green sensor pixel 121(G) has an uneven portion 60(G) (an example of a "second uneven portion" in the present disclosure) provided on the rear surface 11B of the semiconductor substrate 11. The uneven portions 60(R), 60G each extend in a first direction (Y-axis direction) when viewed in a plan view from the rear surface 11B side.

[0075] As shown in Fig. 5, a cross section (an example of a "first cross section" in the present disclosure) of the uneven portion 60(R) of the red sensor pixel 121(R) cut in an XZ plane perpendicular to the Y-axis direction has an asymmetric structure in which the left and right side surfaces are not symmetrical, similar to the uneven portion 60 shown in Fig. 4. In the uneven portion 60(R), the inclination of the side surface with respect to the thickness direction (Z-axis direction) of the semiconductor substrate 11 is larger for the side surface 63 (an example of a "first side surface" in the present disclosure) than for the side surface 64 (the "second side surface" in the present disclosure). In a plan view from the Z-axis direction, the side surface 63 has a larger area than the side surface 64. The horizontal light-shielding portion 13H of the second light-shielding portion 13 is located on the optical path of light that is incident on the side surface 63 and refracted.

[0076] A cross section of the uneven portion 60(G) of the green sensor pixel 121(B) taken along the XZ plane (an example of a "second cross section" in this disclosure) also has an asymmetric structure in which the left and right sides are not symmetrical, similar to the uneven portion 60 shown in Figure 4.

[0077] 5, the shape of a cross section obtained by cutting the uneven portion 60(R) of the red sensor pixel 121(R) along the XZ plane is different from the shape of a cross section obtained by cutting the uneven portion 60(G) of the green sensor pixel 121(B) along the XZ plane. That is, the uneven portions 60 of adjacent sensor pixels 121 also have asymmetric structures.

[0078] For example, the green sensor pixel 121(B) has a plurality of uneven portions 60(G). The plurality of uneven portions 60(G) are arranged in the second direction (X-axis direction). The uneven portions 60(G) are arranged more densely in the X-axis direction than the uneven portions 60(R). More specifically, the side surfaces 63, 64 of the uneven portion 60(R) and the side surfaces 63, 64 of the uneven portion 60(G) are inclined with respect to the thickness direction (Z-axis direction) of the semiconductor substrate 11. The side surface 63 of the uneven portion 60(G) has a larger inclination angle with respect to the Z-axis direction than the side surface 63 of the uneven portion 60(R). Similarly, the side surface 64 of the uneven portion 60(G) has a larger inclination angle with respect to the Z-axis direction than the side surface 64 of the uneven portion 60(R). This allows the incident light L to be scattered further in the green sensor pixel (G) than in the red sensor pixel 121(R).

[0079] Compared to red light, green light has a short wavelength and is easily attenuated, and tends to have difficulty reaching the charge storage unit (MEM) 54 located deep from the rear surface 11B (light receiving surface). Therefore, in the green sensor pixel 121(G), the uneven portion 60(G) is densely arranged to prioritize scattering the incident light L over guiding the incident light L to the horizontal light shielding portion 13H. In the green sensor pixel 121(G), the incident light L is scattered to further extend the optical path length while being guided to the horizontal light shielding portion 13H. This makes it possible to further improve the photoelectric conversion efficiency Qe in the green sensor pixel 121(G) while keeping the PLS low.

[0080] Although not shown in Fig. 5, the blue sensor pixel 121(B) may also have an uneven portion 60(B) (not shown) provided on the rear surface 11B of the semiconductor substrate 11. For example, the uneven portion 60(B) extends in the first direction (Y-axis direction) in a plan view from the rear surface 11B side. The uneven portion 60(B) has an asymmetric structure in which the left and right side surfaces are not symmetrical, similar to the uneven portion 60 shown in Fig. 4.

[0081] In this case, it is preferable that the concave-convex portions 60(B) are arranged more densely in the X-axis direction than the concave-convex portions 60(G). This allows the incident light L to be further scattered in the blue sensor pixel (G), making it possible to further improve the photoelectric conversion efficiency Qe while keeping the PLS low.

[0082] (Method of forming uneven portion) Next, a method for forming the uneven portion 60 will be described. FIG. 6 is a cross-sectional view showing a process sequence of a method for forming the uneven portion 60 suitable for the imaging device 101 according to an embodiment of the present disclosure. In step ST1 of FIG. 6, a semiconductor substrate 11 on which a photoelectric conversion section 51 (see FIG. 5) (not shown) and the like are formed is prepared. Next, a hard mask (not shown) is formed on the back surface 11B (upper surface in FIG. 6) of the semiconductor substrate 11. The hard mask is made of an insulating material such as a silicon nitride film (SiN film) or a silicon oxide film (SiO2 film).

[0083] Next, as shown in step ST2 of FIG. 6, the back surface 11B of the semiconductor substrate 11 is partially and shallowly dug down by dry etching using a hard mask to form a trench T. For example, the trench T is <110> Form it long in the direction.

[0084] 6, a part of the Si(111) surface constituting the rear surface 11B of the semiconductor substrate 11 is removed by wet etching using an alkaline aqueous solution (hereinafter also referred to as alkaline wet etching). As the alkaline aqueous solution, inorganic solutions such as KOH, NaOH, and CsOH can be used, and organic solutions such as EDP (ethylenediaminepyrocatechol aqueous solution), N2H4 (hydrazine), NH4OH (ammonium hydroxide), and TMAH (tetramethylammonium hydroxide) can be used.

[0085] Here, we perform anisotropic etching, which utilizes the property that the etching rate varies depending on the surface orientation of Si(111). Specifically, for a Si(111) substrate, <111> With respect to the etching rate in the direction, i.e., the direction with three Si back bonds <110> In alkaline wet etching, the etching rate of the (111) plane is lower than that of the (110) and (112) planes.

[0086] 7A and 7B are plan views that show the anisotropy of alkaline wet etching of a Si(111) substrate. <112> 7B shows a trench T long in the direction of the <110> For example, as shown in the plan view of FIG. 7A, a trench is formed on the (111) surface of a Si(111) substrate. <112> A trench T elongated in the direction is formed by dry etching. When the trench T is subjected to alkaline wet etching, etching proceeds from the side surface of the trench T as shown by the arrow in Fig. 7A, and for example, a diamond-shaped recess 62 in a plan view is formed.

[0087] As shown in the plan view of FIG. 7B, the (111) surface of the Si(111) substrate is <110> A long and narrow trench T is formed in the direction by dry etching. When this trench T is subjected to alkaline wet etching, etching proceeds from the side surface of the trench T as shown by the arrow in Fig. 7B, and for example, a recess 62 having a long and narrow rectangular shape in a plan view is formed.

[0088] In step ST3 of FIG. 6, the trench T is formed due to the anisotropy of alkaline wet etching with respect to Si. <111> With respect to direction <110> The etching rate is high in the direction from the top to the bottom. As a result, the trench T is widened and the left and right side surfaces are formed at an angle to the back surface 11B. In addition, in this alkaline wet etching, the corners Tc at the opening ends of the trench T are etched from both the top and bottom directions as shown by the arrows in step ST3, and are therefore removed before other parts. As a result, as shown in step ST4 of FIG. 6, an uneven portion 60 having an asymmetric structure in which the left and right side surfaces 63, 64 are not symmetrical is formed on the back surface 11B of the semiconductor substrate 11. For example, the side surface 63 is a (112) surface, and the side surface 64 is a (111) surface.

[0089] (Examples of convex shapes) 6 shows a case where the cross-sectional shape of the protrusion 61 is a triangle, but the cross-sectional shape of the protrusion 61 is not limited to a triangle. The cross-sectional shape of the protrusion 61 may be any shape, provided that the left and right side surfaces 63, 64 are not symmetrical and have an asymmetrical structure.

[0090] FIG. 8 is a diagram showing an example of the cross-sectional shape of the protrusion 61. As shown in FIG. 8, the cross-sectional shape of the protrusion 61 may be a triangle, a quadrilateral (rectangle), a polygonal (polygon), or a non-linear shape. Non-linear 1 in FIG. 8 illustrates a case where the side surface 63 is linear and the side surface 64 is non-linear (curved). Non-linear 2 in FIG. 8 illustrates a case where both the side surfaces 63 and 64 are non-linear (curved). Each of the protrusions 61 shown in FIG. 8 has an asymmetric structure in which the left and right side surfaces 63 and 64 are not symmetric. The cross-sectional shape of the protrusion 61 can be controlled by, for example, the width of the trench T and the processing time of the alkaline wet etching.

[0091] (Example of layout of trench and uneven parts) FIG. 9 is a diagram showing a layout example (part 1) of trenches T and unevenness 60 formed in a Si(111) substrate. The left side of FIG. 9 shows the shape in plan view (hereinafter, planar shape) and the shape in cross section (hereinafter, cross section) of trenches T formed by dry etching. The cross section obtained by cutting the plan view in the upper left of FIG. 9 along line A9-A9' corresponds to the cross section in the lower left of FIG. 9. The right side of FIG. 9 shows the plan shape and cross section of unevenness 60 formed by performing alkaline wet etching on trenches T shown on the left. The cross section obtained by cutting the plan view in the upper right of FIG. 9 along line B9-B9' corresponds to the cross section in the lower right of FIG. 9.

[0092] As shown in Figure 9, <110> A long trench T in the direction is formed by dry etching, and then alkaline wet etching is performed. <110> The uneven portion 60 having an asymmetric structure that is long in the direction is formed. For example, <112> By forming the concave and convex portions 60 in the direction, it is possible to form a plurality of concave and convex portions 60 in a line and space pattern.

[0093] FIG. 10 is a diagram showing a layout example (part 2) of trenches T and unevenness 60 formed in a Si(111) substrate. The left side of FIG. 10 shows the planar shape and cross-sectional shape of trenches T formed by dry etching. The cross-section taken along line A10-A10' of the plan view in the upper left of FIG. 10 corresponds to the cross-sectional view in the lower left of FIG. 10. The right side of FIG. 10 shows the planar shape and cross-sectional shape of unevenness 60 formed by performing alkaline wet etching on trenches T shown on the left side. The cross-section taken along line B10-B10' of the plan view in the upper right of FIG. 10 corresponds to the cross-sectional view in the lower right of FIG. 10.

[0094] As shown in Figure 10, <112> When a trench T elongated in the direction is formed by dry etching and then alkaline wet etching is performed, an uneven portion 60 having an asymmetric structure with a rhombus in plan view is formed. For example, <110> By forming the grooves 60 in a line in different directions, the diamond-shaped projections 60 can be arranged side by side.

[0095] FIG. 11 is a diagram showing a layout example (part 3) of trenches T and unevenness 60 formed in a Si(111) substrate. The left side of FIG. 11 shows the planar shape and cross-sectional shape of trenches T formed by dry etching. The cross-section taken along line A11-A11' of the plan view in the upper left of FIG. 11 corresponds to the cross-section in the lower left of FIG. 11. The right side of FIG. 11 shows the planar shape and cross-sectional shape of unevenness 60 formed by performing alkaline wet etching on trenches T shown on the left. The cross-section taken along line B11-B11' of the plan view in the upper right of FIG. 11 corresponds to the cross-section in the lower right of FIG. 11.

[0096] As shown in Figure 11, <110> a trench elongated in the direction of the <112> A cross-shaped trench T, which is a cross between a trench elongated in the direction and a trench extending in the direction of the substrate in a plan view, is formed by dry etching. After that, when alkaline wet etching is performed, etching proceeds from the side surface of the trench T, and an uneven portion 60 having an asymmetric structure with a hexagonal planar shape is formed. For example, when the cross-shaped trench T is <110> By forming the recesses 60 in different directions, the recesses 60 can be arranged side by side.

[0097] 9 to 11, the multiple uneven portions 60 extend in at least one direction in a plan view from the back surface side of the semiconductor substrate 11, and are arranged side by side in another direction perpendicular to the one direction. Next, multiple layout variations in a plan view of the uneven portions 60 will be further shown.

[0098] 12 to 14 are plan views (parts 1 to 3) showing variations of layout examples (minimum unit) of trenches T and uneven portions 60 formed in a Si(111) substrate. The uneven portion 60 shown in Figs. 12 and 13 is the minimum unit of the layout, and is one of a plurality of uneven portions 60 arranged in a line-and-space pattern, for example. Note that, although the layouts shown in Figs. 12 and 13 show trenches T formed by dry etching, in the actual uneven portion 60, this trench T may be integrated with a recess formed by alkaline wet etching, and may be indistinguishable in appearance.

[0099] 12 and 13, and in FIG. 14 described later, sub-numbers -1 to -9 are added to the reference numeral 60 in order to distinguish between the layouts of the concave-convex portion 60. When there is no need to distinguish between the layouts, the sub-numbers are omitted.

[0100] As shown in FIGS. 12 and 13, in an embodiment of the present disclosure, <110> a trench elongated in the direction of the <112> The uneven portion 60 arranged in a line-and-space pattern may be formed by using a trench T having a shape combining a trench elongated in the vertical direction. The trench T is formed by dry etching. After the formation of the trench T, Si is etched from the side surface of the trench T by alkaline wet etching, forming a recess 62 and side surfaces 63 and 64.

[0101] In each of the uneven portions 60-1 to 60-4 shown in Fig. 12, the side surfaces 63, 64 are formed in a straight line shape in a plan view by alkaline wet etching. In each of the uneven portions 60-5 to 60-8 shown in Fig. 13, the side surfaces 63, 64 are formed in a broken line shape or a curved shape in a plan view by alkaline wet etching.

[0102] The planar shape of the side surfaces 63 and 64 can be controlled by the processing time of the alkaline wet etching. For example, the uneven portion 60-5 shown in FIG. 13 can be formed by shortening the processing time of the alkaline wet etching compared to the uneven portion 60-1 shown in FIG. 12. Furthermore, the uneven portion 60-6 shown in FIG. 13 can be formed by shortening the processing time of the alkaline wet etching compared to the uneven portion 60-4 shown in FIG. 12. The uneven portion 60-8 shown in FIG. 13 can be formed by shortening the processing time of the alkaline wet etching compared to the uneven portion 60-7 shown in FIG. 13.

[0103] In an embodiment of the present disclosure, for example, <110> Direction and <112> A trench T may be formed in the (111) surface of the Si(111) substrate in a direction intersecting the direction. <110> Direction and <112> Two trenches T extending in directions intersecting both the vertical and horizontal directions are formed in the (111) surface of the Si(111) substrate in a mutually intersecting manner, and then alkaline wet etching is performed. This method also makes it possible to form the uneven portion 60-9 having an asymmetric structure. The planar shape of the uneven portion 60-9 shown in FIG. 14 is, for example, a hexagon, but the planar shape of the uneven portion 60-9 may be formed into a shape other than a hexagon by adjusting the processing time of the alkaline wet etching.

[0104] (Color filter) Fig. 15 is a plan view showing a configuration example of a color filter CF according to an embodiment of the present disclosure. As shown in the plan view on the left side of Fig. 15, the color filter CF has three colors, for example, red (R), green (G), and blue (B), and has a configuration in which these are arranged in a Bayer pattern. Alternatively, as shown in the plan view in the center of Fig. 15, the color filter CF may have a pixel arrangement called Quad Bayer Coding in which four adjacent sensor pixels 121 in a 2x2 arrangement are configured with color filters of the same color (Modification 1). Alternatively, as shown in the plan view on the left side of Fig. 15, the color filter CF may have three colors, cyan (C), magenta (M), and yellow (Y), and has a configuration in which these are arranged in a Bayer pattern.

[0105] 15 is used, an asymmetric uneven portion 60 is provided on the rear surface 11B of the semiconductor substrate 11 facing the color filter CF, thereby achieving the effects of the above-described embodiment.

[0106] (Mixed arrangement of concave and convex parts) Fig. 16 is a plan view showing an example of a mixed arrangement of concave-convex portions 60 according to an embodiment of the present disclosure. As shown in the plan view on the left side of Fig. 16, in an imaging device 101, concave-convex portions 60 having an asymmetric structure may be arranged only in some of the sensor pixels 121 (e.g., the red sensor pixel 121(R)). Also, as shown in the plan view in the center of Fig. 16, concave-convex portions 60 having an asymmetric structure may be arranged in all of the sensor pixels 121 (e.g., the red sensor pixel 121(R), the green sensor pixel 121(G), and the blue sensor pixel 121(B)).

[0107] Alternatively, as shown in the plan view on the left side of Fig. 16, the uneven portion 60 having an asymmetric structure may be provided in some of the sensor pixels 121 (e.g., the red sensor pixel 121(R)), and the uneven portion 70 having a different shape from the uneven portion 60 having the asymmetric structure may be provided in the other sensor pixels 121 (e.g., the green sensor pixel 121(G) and the blue sensor pixel 121(B)). Examples of the uneven portion 70 include an uneven portion having a symmetrical structure with left and right side surfaces arranged symmetrically, and an uneven portion having a hexagonal planar shape as shown in the figure.

[0108] 16, in the red sensor pixel 121(R), the presence of the uneven portion 60 can improve the photoelectric conversion efficiency Qe and reduce the PLS. In the green sensor pixel 121(G) and the blue sensor pixel 121(B), the presence of the uneven portion 70 can improve the photoelectric conversion efficiency Qe.

[0109] (Effects of the embodiment) As described above, the imaging device 101 according to the embodiment of the present disclosure includes a semiconductor substrate 11 having a back surface 11B on which light is incident and a front surface 11A located opposite to the back surface 11B, and a plurality of sensor pixels 121 provided on the semiconductor substrate 11. The plurality of sensor pixels 121 include a red sensor pixel 121(R). The red sensor pixel 121(R) has an uneven portion 60 provided on the back surface 11B and a photoelectric conversion portion 51 (photodiode PD) capable of photoelectrically converting light incident from the back surface 11B to generate an electric charge. The uneven portion 60 extends in a first direction (for example, the Y-axis direction in FIGS. 4 and 5) in a plan view from the normal direction of the back surface 11B. If a cross section of the uneven portion 60 cut on a plane (for example, an XZ plane) perpendicular to the first direction is taken as a first cross section, the first cross section has an asymmetric structure in which the left and right side surfaces are not symmetrical.

[0110] According to this, the imaging device 101 can refract the light (incident light) incident on the concave-convex portion 60 and make the light travel obliquely with respect to the rear surface 11B. This can extend the optical path length in the first photoelectric conversion portion, thereby improving the photoelectric conversion efficiency Qe.

[0111] Furthermore, since the uneven portion 60 has an asymmetric structure, it becomes easy to direct the traveling direction of incident light in a specific direction within the semiconductor substrate 11, and it is possible to prevent the incident light from entering an unintended region (for example, the charge storage portion (MEM) 54, the floating diffusion FD, etc.). This makes it possible to prevent the generation of unintended signal charges, thereby making it possible to reduce PLS.

[0112] The imaging device 101 can achieve both an improvement in the photoelectric conversion efficiency Qe and a reduction in PLS, and therefore can suppress degradation in performance.

[0113] <Modification> (depth of uneven part) In the above embodiment, it has been described that the depth d60 of the uneven portion 60 is the same among the multiple sensor pixels 121. However, the embodiment of the present disclosure is not limited to this. The depth d60 of the uneven portion 60 may be different among the multiple sensor pixels 121.

[0114] Fig. 17 is a cross-sectional view showing a configuration of an imaging device 101A according to a modified example of the embodiment of the present disclosure. In the imaging device 101B shown in Fig. 17, the depth d60(R) of the concave-convex portion 60(R) provided in the red sensor pixel 121(R) and the depth d60(G) of the concave-convex portion 60(G) provided in the green sensor pixel 121(G) are different from each other. For example, the depth d60(R) is deeper than the depth d60(G) (d60(R)>d60(G)).

[0115] When forming the uneven portions 60(R) and 60(G), the trenches T of the uneven portion 60(R) and the trenches T of the uneven portion 60(G) are formed separately by dry etching to have different depths. Then, by simultaneously performing alkaline wet etching, the uneven portions 60(R) and 60(G) having different depths can be formed.

[0116] Even with this configuration, the same effects as those of the above embodiment can be achieved by each of the concave and convex portions 60(R), 60(G) having an asymmetric structure in which the left and right side surfaces are not symmetrical.

[0117] (Number of uneven parts) In the embodiment of the present disclosure, the number of concave-convex portions included in each of the plurality of sensor pixels may be one or more. Also, the concave-convex portion may be provided only in some of the plurality of sensor pixels.

[0118] 18 to 21 are cross-sectional views showing configurations of imaging devices 101B to 101E according to modified examples of the embodiment of the present disclosure. The imaging devices 101B and 101C shown in FIGS. 18 and 19 are examples in which one sensor pixel 121 is provided with one uneven portion 60. The imaging device 101D shown in FIG. 20 is an example in which, among the multiple sensor pixels 121, the green sensor pixel 121(G) is provided with multiple uneven portions 60, and the red sensor pixel 121(R) is not provided with the uneven portion 60. The imaging device 101E shown in FIG. 21 is an example in which one sensor pixel 121 (for example, two adjacent sensor pixels 121(R) and 121(G)) is provided with one uneven portion 60.

[0119] In any of the imaging devices 101B to 101E, the uneven portion 60, the uneven portion 60(G) each have an asymmetric structure in which the left and right side surfaces 63, 64 are not symmetrical, thereby achieving the same effects as the above embodiment.

[0120] (Arrangement of light receiving lens) 22 and 23 are cross-sectional views showing configurations of imaging devices 101F and 101G according to modified examples of the embodiment of the present disclosure. In the embodiment of the present disclosure, as in the imaging device 101F shown in FIG. 22, one light receiving lens LNS may be arranged for a plurality of sensor pixels 121 (for example, four adjacent sensor pixels 121 of 2×2, or two adjacent sensor pixels 121 of 2×1). Also, as in the imaging device 101G shown in FIG. 23, a plurality of light receiving lenses LNS (for example, four adjacent light receiving lenses LNS of 2×2, or two adjacent light receiving lenses LNS of 2×1) may be arranged for a single sensor pixel 121.

[0121] In either of the aspects shown in FIG. 22 and FIG. 23, in the sensor pixel 121 having the concave-convex portion 60 with an asymmetric structure, it is possible to achieve both an improvement in the photoelectric conversion efficiency Qe and a reduction in the PLS.

[0122] (Arrangement of light-shielding parts) In the above embodiment, it has been described that horizontal light shielding portion 12H of first light shielding portion 12 is arranged symmetrically with respect to vertical light shielding portion 12V as shown in Fig. 5. Also, it has been described that horizontal light shielding portion 13H of second light shielding portion 13 is arranged symmetrically with respect to vertical light shielding portion 13V as shown in Fig. 5. However, the embodiment of the present disclosure is not limited thereto.

[0123] Fig. 24 is a cross-sectional view showing a configuration of an imaging device 101H according to a modified example of the embodiment of the present disclosure. For example, as shown in Fig. 24, the charge holding units (MEMs) 54 do not have to be arranged symmetrically in adjacent sensor pixels 121. In this case, the horizontal light shielding unit 12H of the first light shielding unit 12 does not have to be arranged symmetrically like the charge holding units (MEMs) 54. The horizontal light shielding unit 12H is arranged between the rear surface 11B, which is the light receiving surface, and the charge holding units (MEMs) 54, thereby making it possible to suppress the incidence of light on the charge holding units (MEMs) 54.

[0124] In addition, the vertical gate electrodes 52V of the transfer transistors TRZ of the adjacent sensor pixels 121 may not be arranged symmetrically. In this case, the horizontal light shielding portion 13H of the second light shielding portion 13 may not be arranged symmetrically, like the vertical gate electrodes 52V. The horizontal light shielding portion 13H is arranged between the rear surface 11B, which is the light receiving surface, and the vertical gate electrodes 52V, thereby suppressing the incidence of light on the vertical gate electrodes 52V.

[0125] In the imaging device 101H shown in Fig. 24, the shape of the uneven portion 60 is set so that the incident light L hits the horizontal light shielding portion 13H not only in the red sensor pixel 121(R) but also in the green sensor pixel 121(G). For example, when the thickness of the semiconductor substrate 11 becomes thinner and the green light can reach the charge holding portion (MEM) 54 more easily, the structure shown in Fig. 24 can be adopted to effectively prevent the green light from reaching the charge holding portion (MEM) 54. The structure shown in Fig. 24 is a suitable example for making the semiconductor substrate 11 thinner.

[0126] 24, an inter-pixel isolation portion 20 may be provided on the front surface 11A side of the semiconductor substrate 11, and an inter-pixel isolation portion 30 may be provided on the back surface 11B side of the semiconductor substrate 11, and the inter-pixel isolation portions 20, 30 may be disposed to face each other in the thickness direction (e.g., the Z-axis direction) of the semiconductor substrate 11. Even in this embodiment, a horizontal light-shielding portion 12H may be provided integrally with the inter-pixel isolation portion 20. A horizontal light-shielding portion 13H may be provided integrally with the inter-pixel isolation portion 30.

[0127] (Type of semiconductor substrate) In the above embodiment, a Si(111) substrate is used as the semiconductor substrate 11. However, in the embodiment of the present disclosure, the type of the semiconductor substrate 11 is not limited to this. For example, the semiconductor substrate 11 may be a Si(111) substrate having a (112) surface on the front surface 11A and the back surface 11B. <112> A substrate may be used, or a Si substrate having a (110) surface on the front side 11A and the back side 11B. <110> Even in this configuration, by providing the uneven portion 60 having an asymmetric structure on the light receiving surface of the semiconductor substrate 11, the same effects as those of the above embodiment can be achieved.

[0128] <Other embodiments> As described above, the present disclosure has been described by the embodiments and modifications, but the descriptions and drawings forming a part of this disclosure should not be understood as limiting the present disclosure. Various alternative embodiments, examples and operating techniques will become apparent to those skilled in the art from this disclosure.

[0129] For example, the material constituting the first light-shielding part 12 and the material constituting the second light-shielding part 13 may be appropriately selected and used. Specifically, when priority is given to improving the photoelectric conversion efficiency Qe, the first light-shielding part 12 and the second light-shielding part 13 are each made of a material with excellent reflectivity (e.g., Al). When priority is given to reducing PLS, the first light-shielding part 12 and the second light-shielding part 13 are each made of a material with excellent light absorption characteristics (e.g., W). When a good balance is to be achieved between improving the photoelectric conversion efficiency Qe and reducing the PLS, the first light-shielding part 12 is made of a material with excellent light absorption characteristics (e.g., W), and the second light-shielding part 13 is made of a material with excellent light reflection characteristics (e.g., Al). This makes it possible to efficiently obtain the desired effect.

[0130] In the above embodiment, the imaging devices 101 to 101H having a global shutter function are exemplified as the "photodetection device" of the present disclosure. However, the "photodetection device" of the present disclosure may also be applied to an imaging device that does not have a global shutter function. In that case, it is preferable that the floating diffusion FD is shielded from light by, for example, a first light shielding portion. Alternatively, the "photodetection device" of the present disclosure may be applied to, for example, a distance measuring device having a SPAD (single photon avalanche diode) pixel. In that case, it is preferable that the PN junction of the SPAD is shielded from light by, for example, a first light shielding portion. In this way, the technology of the present disclosure may be applied to an imaging device that does not have a global shutter function or a distance measuring device having a SPAD pixel, and even in that case, it is possible to suppress the generation of unintentional charge signals and suppress deterioration of performance.

[0131] Thus, the present technology naturally includes various embodiments not described herein. At least one of various omissions, substitutions, and modifications of components can be made without departing from the spirit of the above-mentioned embodiments and modifications. In addition, the effects described in this specification are merely examples and are not limited to the present invention, and other effects may also be present.

[0132] The present disclosure can also be configured as follows. (1) A semiconductor substrate having a first surface onto which light is incident and a second surface located on the opposite side to the first surface; a plurality of sensor pixels provided on the semiconductor substrate; the plurality of sensor pixels includes a first pixel; The first pixel is A first uneven portion provided on the first surface; a first photoelectric conversion unit capable of generating electric charges by photoelectrically converting light incident from the first surface; the first concave-convex portion extends in a first direction in a plan view from a normal direction of the first surface, When a cross section of the first uneven portion is cut along a plane perpendicular to the first direction, the cross section is defined as a first cross section. The photodetector, wherein the first cross section has an asymmetric structure in which the left and right sides are not symmetrical. (2) The light detection device according to (1), wherein the left and right side surfaces of the first uneven portion are each inclined with respect to a thickness direction from the first surface to the second surface of the semiconductor substrate. (3) the first pixel has a plurality of the first concave-convex portions, The light detection device according to (1) or (2), wherein the plurality of first concave-convex portions are aligned in a second direction intersecting the first direction in the plan view. (4) an inter-pixel separator provided on the semiconductor substrate and separating adjacent pixels from each other among the plurality of sensor pixels; The first pixel is a first charge holding section that is provided on the opposite side of the first surface with the first photoelectric conversion section in between and is capable of holding charges transferred from the first photoelectric conversion section; The photodetection device described in any one of (1) to (3), further comprising a first light-shielding portion provided between the first surface and the first charge retention portion and extending from the inter-pixel separation portion toward the center of the first pixel. (5) The first pixel is a first transfer transistor that is provided on the opposite side of the first surface with the first photoelectric conversion unit therebetween and that transfers charges generated by the first photoelectric conversion unit; a second light-shielding portion provided between the first surface and the first transfer transistor and extending from the inter-pixel isolation portion toward a center portion of the first pixel, The light detection device according to (4), wherein the second light-shielding portion is located closer to the first surface than the first light-shielding portion. (6) The photodetector described in (5), wherein the end portion on the center side of the first light-shielding portion and the end portion on the center side of the second light-shielding portion are positioned so as to overlap each other in the thickness direction from the first surface to the second surface of the semiconductor substrate. (7) When one of the left and right side surfaces of the first uneven portion is defined as a first side surface and the other is defined as a second side surface, the first side surface has a greater inclination than the second side surface in a thickness direction from the first surface to the second surface of the semiconductor substrate; The first side surface has a larger area than the second side surface in the plan view, The photodetector according to (6), wherein the second light-shielding portion is located on an optical path of light that is incident on the first side surface and refracted. (8) an inter-pixel separator provided on the semiconductor substrate and separating adjacent pixels from each other among the plurality of sensor pixels; The plurality of sensor pixels include a second pixel disposed adjacent to the first pixel via the inter-pixel separation portion, and into which light of a different color than that of the first pixel is incident on the first surface; The second pixel is A second uneven portion provided on the first surface; a second photoelectric conversion unit capable of generating electric charges by photoelectrically converting light incident from the first surface, The second uneven portion extends in the first direction in the plan view, When a cross section of the second uneven portion cut by the plane perpendicular to the first direction is defined as a second cross section, The photodetector according to any one of (1) to (7), wherein the second cross section has an asymmetric structure in which left and right side surfaces are not symmetrical. (9) The light detection device according to (8), wherein the first cross section and the second cross section have different shapes. (10) The photodetector according to (8) or (9), wherein the left and right side surfaces of the second uneven portion are each inclined with respect to a thickness direction from the first surface to the second surface of the semiconductor substrate. (11) the second pixel has a plurality of the second concave-convex portions, The light detection device according to any one of (8) to (10), wherein the second concave-convex portions are aligned in a second direction intersecting the first direction in the plan view. (12) The light detection device according to (11), wherein the second concave-convex portions are arranged more densely in the second direction than the first concave-convex portions. (13) The photodetector according to any one of (1) to (12), wherein the first surface of the semiconductor substrate is a (111) surface. [Explanation of symbols]

[0133] 11 Semiconductor substrate 11A surface 11B Back side 12 First light shielding section 12H, 13H Horizontal light shielding area 12V, 13V vertical light shield 13 Second light shielding section 17 Etching stopper 18 Insulating layer 20, 30 Inter-pixel separation section 22 Column signal processing circuit 50 Charge transfer section 51 Photoelectric conversion unit 51A, 52A N-type semiconductor region 51B N-type semiconductor region 51C P-type semiconductor region 52H Horizontal gate electrode 52V Vertical gate electrode 54 Charge retention unit (MEM) 55 Planarization film 56 Transparent insulating film 60, 60-1, 60-2, 60-3, 60-4, 60-5, 60-6, 60-7, 60-8, 60-9, 60(B), 60(G), 60(R), 70 Uneven part 61 Convex 62 Recess 63, 64 Side 101, 101A, 101B, 101C, 101D, 101E, 101F, 101G, 101H Imaging device 111 Pixel array section 112 Vertical drive unit 113 Ramp Wave Module 114 Column signal processing section 115 Clock Module 116 Data Storage Unit 117 Horizontal drive unit 118 System Control Unit 119 Signal Processing Section 120 Readout circuit 121 sensor pixels 121(R) Red sensor pixel 121(G) Green sensor pixel 121(B) Blue sensor pixel 122 pixel drive line 123 Vertical signal line AMP Amplification transistor CF Color Filter FD Floating Diffusion L Incident light LNS receiving lens OFG emission transistor PD Photodiode RST Reset transistor SEL Select transistor T-Trench Tc Corner VDD power line VSL vertical signal line

Claims

1. a semiconductor substrate having a first surface onto which light is incident and a second surface located opposite to the first surface; a plurality of sensor pixels provided on the semiconductor substrate; the plurality of sensor pixels includes a first pixel; The first pixel is A first uneven portion provided on the first surface; a first photoelectric conversion unit capable of generating charges by photoelectrically converting light incident from the first surface; The first uneven portion extends in a first direction in a plan view from a normal direction of the first surface, When a cross section of the first uneven portion is cut along a plane perpendicular to the first direction, the cross section is defined as a first cross section. The photodetector, wherein the first cross section has an asymmetric structure in which the left and right sides are not symmetrical.

2. The light detection device according to claim 1 , wherein left and right side surfaces of the first uneven portion are inclined with respect to a thickness direction from the first surface to the second surface of the semiconductor substrate.

3. the first pixel has a plurality of the first concave-convex portions, The light detection device according to claim 1 , wherein the first concave-convex portions are aligned in a second direction intersecting the first direction in the plan view.

4. an inter-pixel separator provided on the semiconductor substrate and separating adjacent pixels from each other among the plurality of sensor pixels; The first pixel is a first charge holding unit that is provided on the opposite side of the first surface with the first photoelectric conversion unit therebetween and is capable of holding charges transferred from the first photoelectric conversion unit; 2. The photodetection device according to claim 1, further comprising a first light-shielding portion provided between the first surface and the first charge retention portion and extending from the inter-pixel isolation portion toward a center portion of the first pixel.

5. The first pixel is a first transfer transistor that is provided on the opposite side of the first surface with the first photoelectric conversion unit therebetween and transfers charges generated by the first photoelectric conversion unit; a second light-shielding portion provided between the first surface and the first transfer transistor and extending from the inter-pixel isolation portion toward a center portion of the first pixel, The light detection device according to claim 4 , wherein the second light-shielding portion is located closer to the first surface than the first light-shielding portion.

6. The photodetector according to claim 5 , wherein an end portion of the first light-shielding portion on the side of the center and an end portion of the second light-shielding portion on the side of the center are positioned so as to overlap each other in a thickness direction from the first surface to the second surface of the semiconductor substrate.

7. When one of the left and right side surfaces of the first uneven portion is defined as a first side surface and the other is defined as a second side surface, the first side surface has a greater inclination than the second side surface with respect to a thickness direction from the first surface to the second surface of the semiconductor substrate; The first side surface has a larger area than the second side surface in the plan view, The light detection device according to claim 6 , wherein the second light blocking portion is located on an optical path of light that is incident on the first side surface and refracted.

8. an inter-pixel separator provided on the semiconductor substrate and separating adjacent pixels from each other among the plurality of sensor pixels; The plurality of sensor pixels include a second pixel disposed adjacent to the first pixel via the inter-pixel separation portion, and into which light of a different color than that of the first pixel is incident on the first surface; The second pixel is A second uneven portion provided on the first surface; a second photoelectric conversion unit capable of generating charges by photoelectrically converting light incident from the first surface, The second uneven portion extends in the first direction in the plan view, When a cross section of the second uneven portion cut by the plane perpendicular to the first direction is defined as a second cross section, The light detection device according to claim 1 , wherein the second cross section has an asymmetric structure in which left and right sides are not symmetrical.

9. The light detection device according to claim 8 , wherein the first cross section and the second cross section have different shapes.

10. The light detection device according to claim 8 , wherein left and right side surfaces of the second uneven portion are inclined with respect to a thickness direction from the first surface to the second surface of the semiconductor substrate.

11. the second pixel has a plurality of the second concave-convex portions, The light detection device according to claim 8 , wherein the second concave-convex portions are aligned in a second direction intersecting the first direction in the plan view.

12. The light detection device according to claim 11 , wherein the second concave-convex portions are arranged more densely in the second direction than the first concave-convex portions.

13. 2. The photodetector device of claim 1, wherein the first surface of the semiconductor substrate is a (111) surface.

Images