Solid-state imaging device

Abstract

The solid-state imaging device includes: a photoelectric conversion region disposed inside the epitaxial layer; a charge holding region disposed on the light incident side with respect to the photoelectric conversion region; a dug-out region dug out from the surface of the epitaxial layer toward the photoelectric conversion region; and a light-shielding film covering the charge holding region and extending on the sidewall of the dug-out region.

Description

Technical Field

[0001] The present invention relates to a solid-state imaging device having a global shutter function. Background Art

[0002] In order not to increase the size of a solid-state imaging device and to obtain high image quality, a solid-state imaging device has a tendency to reduce the pixel size of the solid-state imaging device provided therein. In Pamphlet of International Publication No. 2017 / 131009, a solid-state imaging device is disclosed in which a columnar protrusion having a quadrilateral planar shape is provided on a transparent insulating film of a wiring layer provided in the solid-state imaging device, and incident light is concentrated and incident on a light-receiving portion due to refraction, thereby improving sensitivity and reducing optical noise. Summary of the Invention

[0003] However, in the solid-state imaging device of Pamphlet of International Publication No. 2017 / 131009, there is a problem that parasitic light sensitivity noise generated by diffracting incident light directed to the light-receiving portion due to refraction and incident on a memory portion cannot be removed.

[0004] An object of one embodiment of the present invention is to realize a solid-state imaging device capable of removing parasitic light sensitivity noise and displaying a high-quality image with high sensitivity.

[0005] In order to solve the above problems, a solid-state imaging device according to one embodiment of the present invention is a solid-state imaging device in which a plurality of pixels are arranged in a matrix on a substrate, and is characterized in that each of the plurality of pixels includes: a photoelectric conversion region configured to convert light incident on the substrate into charges and disposed inside the substrate; a charge holding region disposed inside the substrate on the light incident side of the photoelectric conversion region to hold the charges converted by the photoelectric conversion region; a dug-out region dug from the surface of the substrate on the light incident side toward the photoelectric conversion region at least to a depth corresponding to the charge holding region; and a light shielding film formed to cover the surface side of the substrate of the charge holding region and extending on the side wall of the dug-out region.

[0006] According to the solid-state imaging device of one embodiment of the present invention, parasitic light sensitivity noise can be removed and a high-quality image with high sensitivity can be displayed. Brief Description of the Drawings

[0007] Figure 1 It is a diagram showing the structure of pixels arranged in the solid-state imaging device according to the first embodiment.

[0008] Figure 2 It is a cross-sectional view of the structure of a pixel of the solid-state imaging device according to the first embodiment.

[0009] Figure 3 is a top view of a pixel of the above-described solid-state imaging device.

[0010] Figure 4 is a cross-sectional view of the structure of a pixel of the solid-state imaging device according to the comparative example.

[0011] Figure 5 is a cross-sectional view of the structure of a pixel of the solid-state imaging device according to another comparative example.

[0012] Figure 6 is a graph showing the reflectance spectrum with respect to the wavelength of the light-shielding film provided in the solid-state imaging device according to the first embodiment.

[0013] Figure 7 is a graph showing the relationship between the stress of the passivation film provided in the solid-state imaging device according to the first embodiment and the dark current.

[0014] Figure 8 is a cross-sectional view of the structure of a pixel of the solid-state imaging device according to Modification 1 of the first embodiment.

[0015] Figure 9 is a cross-sectional view of the structure of a pixel of the solid-state imaging device according to Modification 2 of the first embodiment.

[0016] Figure 10 is a cross-sectional view of the structure of a pixel of the solid-state imaging device according to Modification 3 of the first embodiment.

[0017] Figure 11 is a cross-sectional view of the structure of a pixel of the solid-state imaging device according to Modification 4 of the first embodiment.

[0018] Figure 12 is a cross-sectional view of the structure of a pixel of the solid-state imaging device according to Modification 5 of the first embodiment.

[0019] Figure 13 is a cross-sectional view of the structure of a pixel of the solid-state imaging device according to Modification 6 of the first embodiment.

[0020] Figure 14 is a cross-sectional view of the structure of a pixel of the solid-state imaging device according to Modification 7 of the first embodiment.

[0021] Figure 15 is a cross-sectional view of the structure of a pixel of the solid-state imaging device according to Modification 8 of the first embodiment.

[0022] Figure 16 is along Figure 15 a cross-sectional view taken along line AB shown.

[0023] Figure 17 It is a top view of the pixel of the above-mentioned solid-state imaging device.

[0024] Figure 18 It is a cross-sectional view of the structure of the pixel of the solid-state imaging device according to the second embodiment.

[0025] Figure 19 It is a cross-sectional view of the structure of the pixel of the solid-state imaging device according to a modification of the second embodiment. Detailed Embodiment

[0026] 〔First Embodiment〕

[0027] Hereinafter, an embodiment of the present invention will be described in detail.

[0028] Figure 1 It is a diagram showing the structure of the pixels 21 arranged in the solid-state imaging device 18 according to the first embodiment. Figure 2 It is a cross-sectional view of the structure of the pixel 21 of the solid-state imaging device 18 according to the first embodiment. Figure 3 It is a top view of the pixel 21.

[0029] The solid-state imaging device 18 includes a substrate 19. The substrate 19 includes a semiconductor substrate 1 and an epitaxial layer 2 formed on the semiconductor substrate 1.

[0030] The solid-state imaging device 18 includes: a plurality of pixels 21 arranged in a matrix on the substrate 19. Each pixel 21 includes: a photoelectric conversion region 3 for converting light incident on the substrate 19 into electric charge; a charge holding region 4 for holding the electric charge converted by the photoelectric conversion region 3; and a pixel transistor region 17 in which transistors required for converting the signal charge stored in the photoelectric conversion element formed in the photoelectric conversion region 3 into voltage, resetting the pixel, or selecting are formed.

[0031] The photoelectric conversion region 3 is disposed inside the epitaxial layer 2 to convert light incident on the substrate 19 into electric charge. The charge holding region 4 is disposed on the light incident side of the inside of the epitaxial layer 2 and closer to the light than the photoelectric conversion region 3.

[0032] Each pixel 21 includes: a dug-out region 20 that is dug out from the surface on the light incident side of the epitaxial layer 2 toward the photoelectric conversion region 3 at least to a depth corresponding to the charge holding region 4; and a light-shielding film 8 that is formed to cover the surface side of the epitaxial layer 2 of the charge holding region 4 and extends on the side wall of the dug-out region 20 in order to block light from reaching the charge holding region 4.

[0033] Each pixel 21 also includes a charge transfer gate electrode 6, and the charge transfer gate electrode 6 is used to transfer charges from the photoelectric conversion region 3 to the charge holding region 4. The charges converted from light by the photoelectric conversion region 3 are transferred to the charge holding region 4 through a charge transfer path that goes from the photoelectric conversion region 3 via the charge transfer gate electrode 6 to the charge holding region 4.

[0034] The charge transfer path refers to, in any one of the embodiments, the path from the region where the photoelectric conversion region 3 overlaps with the charge transfer gate electrode 6 (specifically, the region with the deepest potential of the photoelectric conversion region 3 when the charge transfer gate electrode 6 is turned on) to the region with the deepest potential of the charge holding region 4.

[0035] After transferring the charges to the charge holding region 4, it is the same as the operation of a so-called normal rolling shutter type CIS (CMOS Image Sensor). That is, before transferring the charges from the charge holding region 4 to the floating diffusion region (FD), the FD is reset by a reset transistor, and the charges of the FD at the time of reset are converted into a voltage by a source follower transistor. Moreover, by using a selection transistor to select a detection pixel, an analog-to-digital conversion (AD conversion) is performed on the reset signal and obtained by a latch circuit. Thereafter, immediately, the optical signal charges stored in the charge holding region 4 are transferred to the FD, and the charges transferred to the FD are converted into a voltage by a source follower transistor. Moreover, the optical signal and the reset signal are subjected to AD conversion via the turned-on selection transistor and obtained by a latch circuit. Thereafter, by using a digital circuit to subtract the reset signal from the optical signal and the reset signal, the signal generated by photoelectric conversion can be correctly obtained.

[0036] As Figure 2 and Figure 3 shown, the charge transfer gate electrode 6 overlaps both the portion with a deep potential of the photoelectric conversion region 3 and the charge holding region 4. Therefore, a charge transfer path for transferring charges from the photoelectric conversion region 3 to the charge holding region 4 is formed.

[0037] When the charge transfer gate electrode 6 is turned on, an electric field is applied under the charge transfer gate electrode 6, and the potential under the charge transfer gate electrode 6 is modulated to the deeper side. At this time, the surface of the epitaxial layer 2 is more strongly affected by the electric field, and moreover, the impurity implantation and the potential slope of the charge holding region 4 formed by the layout continue at this time, becoming a potential shape that makes the potential deeper when the charge transfer gate electrode 6 is turned off by an offset amount.

[0038] Therefore, when the charge transfer gate electrode 6 is turned on, the deepest potential region of the photoelectric conversion region 3 is formed under the region where the photoelectric conversion region 3 and the charge transfer gate electrode 6 overlap. Moreover, corresponding to the amount modulated by the charge transfer gate electrode 6 from here, the charge moves along the potential slope of the charge holding region 4 deeper than the deepest potential region of the photoelectric conversion region 3 to the deepest potential region of the charge holding region 4.

[0039] The photoelectric conversion region 3 is formed by extending from the lower side of the dug-out region 20 to the lower side of the charge holding region 4. The dug-out region 20 on the photoelectric conversion region 3 is dug to a position deeper than the charge holding region 4. At this time, the charge is transferred vertically from the photoelectric conversion region 3 below the charge holding region 4 to the charge holding region 4 directly above it. Moreover, in order to appropriately form the distance between the photoelectric conversion region 3 and the charge holding region 4 at the part where the charge is transferred vertically, a protruding portion protruding toward the charge holding region 4 is formed in the photoelectric conversion region 3. Depending on the dug-out depth of the dug-out region 20 directly above the photoelectric conversion region 3, there may sometimes be no protruding portion formed, and sometimes, on the contrary, it is necessary to maintain the distance between the photoelectric conversion region 3 and the charge holding region 4 at the part where the charge is transferred vertically. The approximate distance between the photoelectric conversion region 3 and the charge holding region 4 of this transfer part can be about 0.35 μm to 0.45 μm.

[0040] An insulating film 5 is formed on the epitaxial layer 2.

[0041] The charge transfer gate electrode 6 is formed on the insulating film 5 and extends on the side wall of the dug-out region 20 to reach the bottom of the dug-out region 20.

[0042] An antireflection film 7 for reducing the loss of light incident on the substrate 19 is formed on the charge transfer gate electrode 6 and on the dug-out region 20. A light-shielding film 8 is formed on the antireflection film 7.

[0043] The light-shielding film 8 is formed of a single or multiple film structures with a higher light reflectivity than that of a single tungsten layer.

[0044] Each pixel 21 includes: at least one wiring layer 10, which is provided on the light-shielding film 8 and has a plurality of metal wirings formed thereon; and an interlayer insulating film 9, which is formed on the light-shielding film 8.

[0045] As the semiconductor substrate 1, a known semiconductor substrate is used. For example, a silicon substrate can be used. The photoelectric conversion region 3 only needs to have a structure capable of converting incident light into charge. For example, it is composed of a photodiode. Among the multiple pixels 21 arranged in a matrix, as Figure 1 shown, charge holding regions 4 are provided between the photoelectric conversion regions 3 adjacent to each other along the row direction. For example, the charge holding region 4 is composed of an n-type layer in the epitaxial layer 2, and a potential barrier is formed between the photoelectric conversion region 3 and the charge holding region 4.

[0046] Further, a dug-out region 20 is formed above the photoelectric conversion region 3. A light-shielding film 8 is disposed on the upper side of the charge holding region 4 and the side walls of the dug-out region 20. Further, the charge transfer gate electrode 6 is configured to cover the upper surface and the side surface of the charge holding region 4 via the epitaxial layer 2 and the insulating film 5.

[0047] The solid-state imaging device 18 of the present embodiment includes: a passivation film 12 that covers the wiring layer 10 and the interlayer insulating film 9 and has a refractive index higher than that of the interlayer insulating film 9; a planarization film 13 that is provided on the passivation film 12; and a microlens 15 that is configured to cover the photoelectric conversion region 3 on the planarization film 13.

[0048] The passivation film 12 may be made of a silicon nitride-based material, and the interlayer insulating film 9 may be made of a silicon oxide-based material. In addition, the silicon nitride-based material refers to a material mainly composed of silicon nitride and may contain any impurities, etc. Further, the silicon oxide-based material refers to a material mainly composed of silicon oxide. Here, as long as it has insulating properties, the silicon oxide may contain any impurities, etc.

[0049] As Figure 2 shown, the light incident on the solid-state imaging device 18 of the present embodiment is refracted in the central axis direction of the microlens 15 by the microlens 15 and is incident on the planarization film 13 and the passivation film 12. Thereafter, the light passes through the interlayer insulating film 9 and is incident on the photoelectric conversion region 3. The light incident on the photoelectric conversion region 3 is converted into charge. The charge generated by this photoelectric conversion is transferred to the charge holding region 4 by applying a voltage to the charge transfer gate electrode 6. The charge stored in the charge holding region 4 is utilized as an image signal and forms a captured image through signal processing. In addition, since the solid-state imaging device 18 of the present embodiment adopts a global shutter method, the photoelectric conversion in the photoelectric conversion region 3 and the transfer of the charge from the photoelectric conversion region 3 to the charge holding region 4 are performed at the same time for all pixels.

[0050] During the above-mentioned photoelectric conversion, if light is directly incident on the charge holding region 4, parasitic photosensitivity noise is generated. This is because photoelectric conversion also occurs in the charge holding region 4.

[0051] The charge holding region 4 forms donors such as As and Ph in the Si-containing epitaxial layer 2 by impurity implantation, and holds charge through the pn junction capacitance mainly formed on the surface portion. That is, similar to the photodiode formed in the photoelectric conversion region 3, if light is incident on the charge holding region 4, carriers are generated by photoelectric conversion.

[0052] In addition, even if light does not directly enter the charge holding region 4, there is still a component of the charge generated by photoelectric conversion in the p-region (pixel separation region) below the charge holding region 4 that reaches the charge holding region 4 due to diffusion.

[0053] Therefore, in order to prevent light from entering the charge holding region 4, a light-shielding film 8 is provided on the upper surface and side surface of the charge transfer gate electrode 6. A metal such as tungsten is used as the structural material of the light-shielding film 8. Tungsten has a high light absorption coefficient and has high light-shielding ability even with a thin film thickness, so it is preferably the material of the light-shielding film.

[0054] Figure 4 It is a cross-sectional view of the structure of a pixel of a solid-state imaging device according to a comparative example. Figure 5 It is a cross-sectional view of the structure of a pixel of a solid-state imaging device according to another comparative example.

[0055] The dug-out region 20 is Figure 2 different from the solid-state imaging device 18 of the first embodiment described above. The epitaxial layer 2 is only dug out to a position shallower than the charge holding region 4. Moreover, the upper surface of the photoelectric conversion region 3 and the upper surface of the charge holding region 4 are arranged at the same depth within the epitaxial layer 2. The light-shielding film 8 is arranged to cover the charge transfer gate electrode 6.

[0056] Figure 4 The solid-state imaging device shown has a color filter 14 and an optical waveguide 11. The color filter 14 has the following functions: to color the incident light with red, green, and blue and convert the light of each pigment into an electrical signal in the photoelectric conversion region 3. The optical waveguide 11 has the following functions: to condense the incident light on the central axis of the pixel 21 and efficiently incident it on the photoelectric conversion region 3. Figure 5 The solid-state imaging device shown has an inner lens 16 and the above-mentioned color filter 14.

[0057] In such a Figure 4 and Figure 5 In the structure of the solid-state imaging device of the comparative example shown, the light incident on the substrate 19 not only enters the photoelectric conversion region 3 but also enters the charge holding region 4. Therefore, photoelectric conversion also occurs in the charge holding region 4. As a result, in the solid-state imaging device of the comparative example, charges other than the charges photoelectrically converted by the photoelectric conversion region 3 are stored in the charge holding region 4 that holds the signal charges before one frame, generating parasitic optical sensitivity noise.

[0058] In contrast, the solid-state imaging device 18 of the present embodiment moves the photoelectric conversion region 3 backward from the position of the charge holding region 4 toward the semiconductor substrate 1 side, and extends the photoelectric conversion region 3 along the lower surface of the charge holding region 4. A dug-out region 20 is formed in the vacant space generated by moving the photoelectric conversion region 3 backward toward the semiconductor substrate 1 side.

[0059] The side wall of the dug-out region 20 faces the side face of the charge holding region 4. The light-shielding film 8 is formed to cover the surface side of the epitaxial layer 2 of the charge holding region 4 via the antireflection film 7 and the charge transfer gate electrode 6 and extends on the side wall of the dug-out region 20.

[0060] By adopting such a structure, the light incident on the epitaxial layer 2 through the interlayer insulating film 9 is blocked by the light-shielding film 8 and cannot be incident on the charge holding region 4. As a result, no photoelectric conversion occurs in the charge holding region 4, and no parasitic light sensitivity noise is generated.

[0061] Moreover, the photoelectric conversion region 3 expands to the lower side of the charge holding region 4, so that not only can the components of the charges after photoelectric conversion in the p-region (pixel separation region) below the charge holding region 4 reach the charge holding region 4 due to diffusion be minimized, but also the saturation capacity can be improved. Furthermore, due to the disappearance of the parasitic light sensitivity noise, the solid-state imaging device 18 of the present embodiment can optimize the light condensing structure on the substrate, which is only used for improving the sensitivity, F-value characteristics (oblique incidence characteristics), and shadow suppression.

[0062] When determining the light condensing structure, the sensitivity and the parasitic light sensitivity noise are often in a trade-off relationship. For example, if the opening of the light-shielding film 8 on the photoelectric conversion region 3 is expanded, although the sensitivity is improved, the parasitic light sensitivity noise deteriorates. On the contrary, if the opening is reduced, although the sensitivity is decreased, the parasitic light sensitivity noise is improved.

[0063] In actual design of the light condensing structure, it is designed with an opening that achieves a balance between the sensitivity and the parasitic light sensitivity noise. However, in the light condensing structure according to the present embodiment, no parasitic light sensitivity noise is generated in principle. Therefore, the opening of the light-shielding film 8 can be expanded to the manufacturing limit, and thus, a light condensing structure for improving the sensitivity to infinity can be designed.

[0064] In addition, the light condensing structure can be determined without worrying about the trade-off between the sensitivity and the parasitic light sensitivity noise or the trade-off between the shadow (when shooting, light is incident on the CMOS image sensor through a lens, so the pixels at the edge of the screen usually have lower sensitivity than the pixels at the center of the screen due to the oblique light incidence, which is a phenomenon of inconsistent sensitivity) and the parasitic light sensitivity noise. Therefore, the light condensing structure of the solid-state imaging device 18 according to the present embodiment can improve the sensitivity (optical characteristics).

[0065] In the comparative example Figure 4 and Figure 5In the case of the structure shown, the portions of deep potential in the photoelectric conversion region 3 and the charge holding region 4 (the main portions where charges are stored) are both located on the outermost surface of the epitaxial layer 2 containing Si. Therefore, they have to be arranged at a certain distance apart. This is because if they are close to each other, short channeling will occur and the saturation capacity will be significantly reduced (caused by the depletion layers of the photoelectric conversion region 3 and the charge holding region 4 being connected). If they are close, not only this will happen, but also the high light spillover to the charge holding region 4 side will be significantly deteriorated. Therefore, as described above, the layout size of each pixel 21 is restricted in such a way as to prevent the photoelectric conversion region 3 from approaching the charge holding region 4. In addition, as long as it is a pixel structure having a reset transistor for discharging electrons in the pixel transistor region 17 of the pixel 21 transferred thereto, a source follower transistor for converting charges into potential, a charge discharge transistor having the function of an electronic shutter, and a selection transistor, the selection transistor of the selected pixel exists on the outermost surface of the epitaxial layer 2 containing Si. Usually, the periphery of the pixel transistor is surrounded by a shallow trench isolation (STI) and separated from the photoelectric conversion region 3 and the charge holding region 4. This is because high-concentration impurity implantation is performed in the pixel transistor. Therefore, the extension of the depletion layer is large. If a relatively large distance is not adopted, the N-type region such as the photoelectric conversion region 3 will be connected to the depletion layer, becoming a leakage source. If the STI is formed, the depletion layer has to surround the lower part of the STI. Therefore, the distance between the pixel transistor and the photoelectric conversion region 3 can be close to the width of the STI. In this way, the layout size of the photoelectric conversion region 3 can also be restricted by the pixel transistor.

[0066] In contrast, Figure 2 The main portion of charge storage (the portion of deep potential, and this portion is as close as possible to the charge holding region 4) in the photoelectric conversion region 3 formed in the solid-state imaging device 18 shown is located at a depth that is not affected by the charge holding region 4 nor by the pixel transistor. Therefore, the photoelectric conversion region 3 can be enlarged to a size where the pixel separation between the photoelectric conversion regions 3 (adjacent photodiodes) formed in adjacent pixels 21 is sufficient. Therefore, compared with the comparative example, the saturation region of the photoelectric conversion region 3 can be enlarged significantly.

[0067] Figure 6 It is a coordinate diagram showing the reflectance spectrum with respect to the wavelength of the light shielding film 8 provided in the solid-state imaging device 18 according to the first embodiment.

[0068] As a single film structure having a higher light reflectance than that of a single tungsten (W) layer for the light shielding film 8, for example, a film structure of aluminum (Al) / titanium nitride (TiN), a film structure of copper (Cu) / tantalum (Ta), a film structure of W / TiN, and a film structure of TiN / Ti / W / TiN can be cited.

[0069] Figure 6The coordinate diagram shows the results of simulating the reflectance spectra of the above-mentioned film structures. It can be seen that in the Al / TiN film structure and the Cu / Ta film structure, the reflectance is significantly higher than that of the tungsten single-layer film structure in any wavelength region. Therefore, if only the reflectance is considered, Al and Cu are more suitable as the material for the light-shielding film 8 than W. However, in order to prevent Cu from diffusing into the insulating film after forming the groove for the light-shielding film wiring, since it can only be formed by a process flow in which a stable metal or metal compound such as Ta or TaN is deposited on the insulating film, and after Cu is buried in the light-shielding film wiring groove, unnecessary portions are removed and planarized by chemical mechanical polishing (CMP), it is restricted by the manufacturing process. That is, the degree of freedom of the light-shielding film structure becomes smaller. In addition, if Cu diffuses into the insulating film and the substrate surface, it will significantly deteriorate the device characteristics such as worsening the peak of the bright spot caused by metal contamination. For these reasons, Cu is not actually used for the light-shielding film.

[0070] In addition, from the viewpoints of reliability and the like, compared with the tungsten single layer, the light-shielding film 8 preferably has a W / TiN film structure formed by depositing W on a TiN film on the insulating film. The reflectance spectrum of the W / TiN film structure is almost the same as that of the W single layer. And by forming a TiN / Ti film on the W / TiN film, the reflectance can be improved. However, as a side effect at this time, the hydrogen storage property of Ti acts in a direction that hinders the hydrogen capping effect of the interface state of the substrate, which may cause deterioration of the dark characteristics.

[0071] Figure 7 It is a coordinate diagram showing the relationship between the stress of the passivation film 12 provided in the solid-state imaging device 18 according to the first embodiment and the dark current.

[0072] Preferably, the passivation film 12 has a compressive stress of -5.0×10 -9 dyne / cm 2 or less in the following compression direction. If the film quality of the passivation film 12 changes, the refractive index also changes, but since the passivation film 12 has a compressive stress, the warpage (stress applied to the entire solid-state imaging device 18) of the image sensor (substrate 19 of the solid-state imaging device 18) can be adjusted to -5.0×10 -9 dyne / cm 2 or less.

[0073] Thereby, the side effects caused by the introduction of the optical waveguide path 11 can be covered. Compared with the materials of other peripheral components, the thermal shrinkage rate of the optical waveguide path 11 is overwhelmingly large. Therefore, the stress acting in the direction of deteriorating the dark voltage propagates to the substrate 19 containing Si. Therefore, it is important to determine the film quality of the passivation film 12 in such a way as to adjust the warpage applied to the image sensor in a direction favorable for the dark current.

[0074] If the stress of the passivation film 12 is controlled in the compressive direction, the stress in the compressive direction of the passivation film 12 propagates to the silicon on the surface of the epitaxial layer 2 including the photoelectric conversion region 3. Therefore, the bandgap of the silicon on the surface of the epitaxial layer 2 including the photoelectric conversion region 3 becomes larger. If this bandgap becomes larger, the dark current decreases exponentially.

[0075] As Figure 7 shown, the stress in the compressive direction is -5.0×10 -9 dyne / cm 2 or less and the change in the dark current is slow. Therefore, if the stress in the compressive direction of the passivation film 12 is set to -5.0×10 -9 dyne / cm 2 or less, the dark current can be reduced exponentially.

[0076] 〔Modification Example 1 of the First Embodiment〕

[0077] Figure 8 is a cross-sectional view of the structure of the pixel 21 of the solid-state imaging device 18A according to Modification Example 1 of the first embodiment.

[0078] The difference from the solid-state imaging device 18 of the first embodiment is that an optical waveguide path 11 is disposed between the passivation film 12 and the planarization film 13. For other structures and effects, they are the same as those of the solid-state imaging device 18 of the first embodiment, and thus the description is omitted.

[0079] The light incident on the microlens 15 by disposing the optical waveguide path 11 in the solid-state imaging device 18A enters the optical waveguide path 11 through the planarization film 13. In the optical waveguide path 11, the incident light is condensed on the central axis in the direction perpendicular to the substrate 19. The condensed incident light is emitted from the optical waveguide path 11 and enters the photoelectric conversion region 3 through the interlayer insulating film 9.

[0080] In this way, the optical waveguide path 11 condenses and makes the light incident on the photoelectric conversion region 3. Therefore, compared with the structure without the optical waveguide path 11 disposed, the amount of light incident on the photoelectric conversion region 3 can be increased. The solid-state imaging device 18A of the present embodiment can further improve the sensitivity.

[0081] In the structure of front side illumination (FSI), a wiring layer 10 is disposed on the light incident side, resulting in loss of a part of the light condensed by the microlens 15 being blocked by the wiring layer 10 (metal blockage). Also, in the global shutter CIS, a charge holding region 4 (storage node) has to be formed within the pixel 21. Therefore, the aperture ratio (fill factor) of the photoelectric conversion region 3 (photoelectric diode portion) with respect to the pixel area becomes smaller. As a result, the sensitivity, F-value characteristics (oblique incidence characteristics), and shadow characteristics deteriorate.

[0082] As its solution, it has an optical waveguide path structure. The optical waveguide path 11 uses a material with a refractive index higher than that of the interlayer insulating film 9 to achieve the effect of enclosing light. That is, when the light converged by the microlens 15 is incident on the entrance (upper opening) of the optical waveguide path 11, it is irradiated from the exit (lower opening) without leaking to the surroundings. Therefore, the light incident on the microlens 15 can be efficiently guided to the photoelectric conversion region 3.

[0083] In the structure of this embodiment where the area directly above the photoelectric conversion region 3 is dug out and the distance from the microlens 15 to the substrate 19 becomes farther, the optical waveguide path 11 is particularly effective. As long as the incident light is guided to the dug-out region 20 through the optical waveguide path 11, the light coming out of the optical waveguide path 11 in any direction wants to be incident on the substrate 19, but no parasitic light sensitivity noise (PLS) is caused. Therefore, the design can be carried out without considering the cone angle of the optical waveguide path 11. As a result, the degree of freedom increases, and it is easy to improve the sensitivity, F-value characteristics, and shadow.

[0084] 〔Modification Example 2 of the First Embodiment〕

[0085] Figure 9 It is a cross-sectional view of the structure of the pixel 21 of the solid-state imaging device 18B according to the modification example 2 of the first embodiment.

[0086] The difference from the solid-state imaging device 18 of the first embodiment is that a color filter 14 is arranged between the planarization film 13 and the microlens 15. For other structures and effects, they are the same as those of the solid-state imaging device 18 of the first embodiment, so the description is omitted.

[0087] By arranging the color filter 14 in the solid-state imaging device 18B, the light incident on the microlens 15 passes through the color filter 14 and only allows light of a specific wavelength to pass through, and the light other than that is absorbed by the color filter 14. Therefore, only light of a specific wavelength is incident on the photoelectric conversion region 3 and is photoelectrically converted.

[0088] In this way, the solid-state imaging device 18B of this embodiment can convert the light of red, green, and blue, which are the three primary colors of light, into electrical signals for each pixel 21 by arranging the color filter 14 that changes the wavelength of the transmitted light for each pixel 21. If the electrical signals are converted back into light after converting the light of the three colors of red, green, and blue into electrical signals, a color image can be displayed.

[0089] 〔Modification Example 3 of the First Embodiment〕

[0090] Figure 10 It is a cross-sectional view of the structure of the pixel 21 of the solid-state imaging device 18C according to the modification example 3 of the first embodiment.

[0091] It is different from the solid-state imaging device 18 of the first embodiment in that an optical waveguide 11 is disposed between the passivation film 12 and the planarization film 13, and a color filter 14 is disposed between the planarization film 13 and the microlens 15. For other structures and effects, they are the same as those of the solid-state imaging device 18 of the first embodiment, so the description is omitted.

[0092] The solid-state imaging device 18C of this embodiment has a structure that combines Modification 1 and Modification 2 of the first embodiment. By disposing the color filter 14 in the solid-state imaging device 18C, the light incident on the microlens 15 passes through the color filter 14 and only allows light of a specific wavelength to pass through, and the other light is absorbed by the color filter 14. The light of the specific wavelength that has passed through the color filter 14 passes through the planarization film 13 and is incident on the optical waveguide 11. In the optical waveguide 11, the incident light of the specific wavelength is condensed on the central axis in the direction perpendicular to the substrate 19. The condensed incident light of the specific wavelength is emitted from the optical waveguide 11, passes through the interlayer insulating film 9, and is incident on the photoelectric conversion region 3.

[0093] In this way, the solid-state imaging device 18C of this embodiment can extract only light of a specific wavelength, thereby increasing the amount of light incident on the photoelectric conversion region 3. By disposing the optical waveguide 11 and the color filter 14 in the solid-state imaging device 18C, a color image with higher sensitivity can be displayed.

[0094] 〔Modification 4 of the First Embodiment〕

[0095] Figure 11 It is a cross-sectional view of the structure of the pixel 21 of the solid-state imaging device 18D according to Modification 4 of the first embodiment.

[0096] It is different from the solid-state imaging device 18 of the first embodiment in that an internal lens 16 is disposed between the passivation film 12 and the planarization film 13. For other structures and effects, they are the same as those of the solid-state imaging device 18 of the first embodiment, so the description is omitted.

[0097] By disposing the internal lens 16 in the solid-state imaging device 18D, the incident light that has passed through the microlens 15 passes through the planarization film 13 and is incident on the internal lens 16 and is condensed toward the central axis in the direction perpendicular to the substrate 19. The condensed incident light passes through the passivation film 12 and the interlayer insulating film 9 and is incident on the photoelectric conversion region 3. By disposing the internal lens 16, the incident light from the outside can be condensed and made to be incident on the photoelectric conversion region 3.

[0098] As a result, the incident light from the outside can be efficiently converted into electrons in the photoelectric conversion region 3, so the sensitivity can be further improved.

[0099] As described above, in the FSI type global shutter CIS, deterioration of the condensing ratio due to a decrease in the filling ratio has become a problem. In addition to the above-described optical waveguide structure, an inner lens 16 can also be used to efficiently collect the light incident from the microlens 15 onto the substrate 19.

[0100] 〔Modification Example 5 of the First Embodiment〕

[0101] Figure 12 It is a cross-sectional view of the structure of the pixel 21 of the solid-state imaging device 18E according to Modification Example 5 of the First Embodiment.

[0102] The difference from the solid-state imaging device 18 of the first embodiment is that an inner lens 16 is disposed between the passivation film 12 and the planarization film 13, and a color filter 14 is disposed between the planarization film 13 and the microlens 15. For other structures and effects, they are the same as those of the solid-state imaging device 18 of the first embodiment, and thus the description is omitted.

[0103] The solid-state imaging device 18E of the present embodiment has a structure in which Modification Example 2 and Modification Example 4 of the first embodiment are combined. By disposing the color filter 14 in the solid-state imaging device 18E, the light incident on the microlens 15 passes through the color filter 14 to allow only light of a specific wavelength to pass through, and the light other than that is absorbed by the color filter 14. The light of the specific wavelength that has passed through the color filter 14 passes through the planarization film 13 and is incident on the inner lens 16. In the inner lens 16, the incident light of the specific wavelength is condensed on the central axis in the direction perpendicular to the substrate 19. The condensed incident light of the specific wavelength is emitted from the inner lens 16, passes through the interlayer insulating film 9, and is incident on the photoelectric conversion region 3.

[0104] In this way, the solid-state imaging device 18E of the present embodiment can extract only the light of a specific wavelength and increase the amount of light incident on the photoelectric conversion region 3. By disposing the inner lens 16 and the color filter 14 in the solid-state imaging device 18E, a color image with higher sensitivity can be displayed.

[0105] 〔Modification Example 6 of the First Embodiment〕

[0106] Figure 13 It is a cross-sectional view of the structure of the pixel 21 of the solid-state imaging device 18F according to Modification Example 6 of the First Embodiment.

[0107] The difference from the solid-state imaging device 18 of the first embodiment is that an optical waveguide 11 and an inner lens 16 are disposed between the passivation film 12 and the planarization film 13. For other structures and effects, they are the same as those of the solid-state imaging device 18 of the first embodiment, and thus the description is omitted.

[0108] The light incident on the microlens 15 passes through the planarization film 13 and is incident on the inner lens 16, and converges toward the central axis in the direction perpendicular to the substrate 19. The converged incident light is incident on the optical waveguide path 11. In the optical waveguide path 11, the incident light of a specific wavelength converges on the central axis in the direction perpendicular to the substrate 19. The converged incident light of the specific wavelength is emitted from the optical waveguide path 11, passes through the interlayer insulating film 9, and is incident on the photoelectric conversion region 3.

[0109] 〔Modification Example 7 of the First Embodiment〕

[0110] Figure 14 It is a cross-sectional view of the structure of the pixel 21 of the solid-state imaging device 18G according to Modification Example 7 of the first embodiment.

[0111] The difference from the solid-state imaging device 18 of the first embodiment is that the optical waveguide path 11 and the inner lens 16 are arranged between the passivation film 12 and the planarization film 13, and the color filter 14 is arranged between the planarization film 13 and the microlens 15. For other structures and effects, they are the same as those of the solid-state imaging device 18 of the first embodiment, so the description is omitted.

[0112] The light incident on the microlens 15 passes through the color filter 14 and only allows light of a specific wavelength to pass through, and the other light is absorbed by the color filter 14. The light of the specific wavelength that has passed through the color filter 14 passes through the planarization film 13 and is incident on the inner lens 16. In the inner lens 16, the incident light of the specific wavelength converges on the central axis in the direction perpendicular to the substrate 19. The converged incident light of the specific wavelength is emitted from the inner lens 16 and is incident on the optical waveguide path 11. In the optical waveguide path 11, the incident light of the specific wavelength converges on the central axis in the direction perpendicular to the substrate 19. The converged incident light of the specific wavelength is emitted from the optical waveguide path 11, passes through the interlayer insulating film 9, and is incident on the photoelectric conversion region 3.

[0113] 〔Modification Example 8 of the First Embodiment〕

[0114] Figure 15 It is a cross-sectional view of the structure of the pixel 21 of the solid-state imaging device 18H according to Modification Example 8 of the first embodiment. Figure 16 It is along Figure 15 The cross-sectional view taken along the line AB shown. Figure 17 It is a top view of the pixel 21.

[0115] The difference from the solid-state imaging device 18 of the first embodiment is that the charge transfer gate electrode 6H is buried in the epitaxial layer 2 in such a way as to cover both side surfaces of the charge holding region 4 and sandwich the charge holding region 4. For other structures and effects, they are the same as those of the solid-state imaging device 18 of the first embodiment, so the description is omitted.

[0116] The light incident on the solid-state imaging device 18H is refracted in the central axis direction of the microlens 15 by the microlens 15 and is incident on the planarization film 13 and the passivation film 12. Thereafter, this light passes through the interlayer insulating film 9 and is incident on the photoelectric conversion region 3.

[0117] The light incident on the photoelectric conversion region 3 is converted into charges. The charges generated by this photoelectric conversion are transferred to the charge holding region 4 by applying a voltage to the charge transfer gate electrode 6H.

[0118] As Figures 15 - 17 shown, the charge transfer gate electrode 6H overlaps both the portion with a deep potential of the photoelectric conversion region 3 and the charge holding region 4. Therefore, a charge transfer path for transferring charges from the photoelectric conversion region 3 to the charge holding region 4 is formed.

[0119] In this way, if the charge transfer gate electrode 6H is buried in the epitaxial layer 2 so as to cover both side surfaces of the charge holding region 4 and sandwich the charge holding region 4, even if no charge transfer gate electrode 6H is formed between the side wall of the dug-out region 20 and the charge holding region 4, charges can be transferred from the photoelectric conversion region 3 to the charge holding region 4.

[0120] Such a buried structure of the charge transfer gate electrode 6H easily forms a potential that is favorable for the transfer of charges from the photoelectric conversion region 3 to the charge holding region 4. Thus, corresponding to the case where no charge transfer gate electrode 6H exists between the side wall of the dug-out region 20 and the charge holding region 4, the opening size of the light-shielding film 8 extending on the side wall of the dug-out region 20 can be enlarged. Therefore, the light condensing rate of the photoelectric conversion region 3 is increased, and the sensitivity of the solid-state imaging device 18H is increased.

[0121] In this way, it is configured that by etching Si in the same manner as shallow trench isolation (STI) and burying the polysilicon gate electrode in the epitaxial layer 2, the charge transfer gate electrode 6H can be buried on both sides of the charge holding region 4, and a voltage can be applied to the charge holding region 4 in a manner of sandwiching it from both sides of the charge holding region 4 to assist charge transfer.

[0122] In addition, Figure 16 and Figure 17 show a structure in which the charge holding region 4 is sandwiched by the charge transfer gate electrodes 6H from both sides, but the charge transfer gate electrode 6H can also be provided only on one side of the charge holding region 4.

[0123] 〔Second Embodiment〕

[0124] Hereinafter, other embodiments of the present invention will be described. In addition, for convenience of explanation, components having the same functions as those described in the above embodiments are denoted by the same reference numerals, and their descriptions are not repeated.

[0125] Figure 18 This is a cross-sectional view of the structure of the pixel 21 of the solid-state imaging device 18I according to the second embodiment. The difference from the solid-state imaging device 18 of the first embodiment is that the depth of the dug-out area 20 directly above the photoelectric conversion area 3 is shallower than the depth of the lower end of the charge holding area 4, and the photoelectric conversion area 3 is formed only on the lower side of the dug-out area 20 and does not extend to the lower side of the charge holding area 4.

[0126] If the photoelectric conversion area 3 extends to the charge holding area 4, the photoelectric conversion area 3 is connected to the charge holding area 4. Therefore, the photoelectric conversion area 3 cannot be enlarged to the charge holding area 4, resulting in such a structure.

[0127] The charge transfer gate electrode 6 is formed on the epitaxial layer 2 so as to cover the charge holding area 4, extends on the side wall of the dug-out area 20 to reach the bottom surface, and extends on a part of the bottom surface and overlaps with the photoelectric conversion area 3 and the charge holding area 4.

[0128] When a voltage is applied to the charge transfer gate electrode 6, the charge converted by the photoelectric conversion area 3 is transferred from the photoelectric conversion area 3 through the charge transfer gate electrode 6 to the charge holding area 4.

[0129] 〔Modification of the second embodiment〕

[0130] Figure 19 This is a cross-sectional view of the structure of the pixel 21 of the solid-state imaging device 18J according to the modification of the second embodiment. The difference from the solid-state imaging device 18I of the second embodiment is that the area on the deep side of the photoelectric conversion area 3 (photoelectric diode) is expanded to the lower side of the charge holding area 4 (storage node).

[0131] The photoelectric conversion area 3 has a potential slope formed from the deep side to the shallow side, and the impurity implantation is adjusted so that the charge generated on the deep side also flows and accumulates on the shallow side. Thus, compared with the Figure 18 structure shown, the saturation capacity of the photoelectric conversion area 3 can be enlarged.

[0132] 〔Summary〕

[0133] The solid-state imaging devices 18, 18A to 18J according to Mode 1 of the present invention are solid-state imaging devices 18, 18A to 18J in which a plurality of pixels 21 are arranged in a matrix on a substrate 19. Each of the plurality of pixels 21 includes: a photoelectric conversion region 3 disposed inside the substrate 19 to convert light incident on the substrate 19 into charges; a charge holding region 4 disposed inside the substrate 19 on the light incident side of the photoelectric conversion region 3 to hold the charges converted by the photoelectric conversion region 3; a dug-out region 20 dug from the surface of the substrate 19 on the light incident side toward the photoelectric conversion region 3 at least to a depth corresponding to the charge holding region 4; and a light-shielding film 8 formed to cover the surface side of the substrate 19 of the charge holding region 4 and extend on the side wall of the dug-out region 20.

[0134] According to the above structure, the charge holding region for holding the charges converted by the photoelectric conversion region is disposed inside the substrate on the light incident side of the photoelectric conversion region. Moreover, the dug-out region dug from the surface of the substrate on the light incident side is disposed toward the photoelectric conversion region at least to a depth corresponding to the charge holding region. In addition, the light-shielding film for blocking light from reaching the charge holding region is formed to cover the surface side of the substrate of the charge holding region and extend on the side wall of the dug-out region.

[0135] Therefore, not only the upper surface of the charge holding region but also the side surface of the charge holding region is covered by the light-shielding film. As a result, a solid-state imaging device can be provided that does not generate, in principle, parasitic light sensitivity noise caused by light leaking due to diffraction of light incident on the substrate in the charge holding region.

[0136] In the above Mode 1, the solid-state imaging devices 18, 18A to 18J according to Mode 2 of the present invention preferably further include charge transfer gate electrodes 6, 6H for transferring the charges from the photoelectric conversion region 3 to the charge holding region 4, and the charges are transferred to the charge holding region 4 through a charge transfer path that passes from the photoelectric conversion region 3 through the charge transfer gate electrodes 6, 6H to the charge holding region 4.

[0137] According to the above structure, by applying a voltage to the charge transfer gate electrode, the charges converted by the light incident on the substrate can be transferred from the photoelectric conversion region to the charge holding region.

[0138] In the solid-state imaging devices 18, 18A to 18H, 18J according to Mode 3 of the present invention, it is preferable that the photoelectric conversion region 3 is formed to extend from below the dug-out region 20 to below the charge holding region 4.

[0139] According to the above structure, the saturation capacity of the photoelectric conversion region can be enlarged.

[0140] In the solid-state imaging device 18H according to Mode 4 of the present invention, it is preferable that the charge transfer gate electrode 6H is buried in the substrate (epitaxial layer 2) in such a manner as to cover the side surface of the charge holding region 4.

[0141] According to the above structure, it is not necessary to provide a charge transfer gate electrode on the side of the dug-out region of the charge holding region, and the opening of the light shielding film can be enlarged. Therefore, the light condensing rate of the photoelectric conversion region can be improved, and the sensitivity of the solid-state imaging device can be increased.

[0142] In the solid-state imaging devices 18, 18A to 18J according to Mode 5 of the present invention, it is preferable that the light shielding film 8 is formed of a single or a plurality of film structures having a higher light reflectance than that of a single tungsten layer.

[0143] According to the above structure, the reflectance of the light toward the charge holding region due to the light shielding film can be increased. Therefore, the light shielding performance of the light shielding film can be improved, and at the same time, the light toward the photoelectric conversion region can be increased, so that the sensitivity can be improved.

[0144] The solid-state imaging devices 18A, 18C, 18F, 18G according to Mode 6 of the present invention preferably further include: an interlayer insulating film 9 formed on the light shielding film 8; and an optical waveguide path 11 formed on the interlayer insulating film 9 by a material having a refractive index different from that of the interlayer insulating film 9 in order to guide the light to the photoelectric conversion region 3.

[0145] According to the above structure, the light incident on the substrate can be efficiently guided to the photoelectric conversion region disposed below the dug-out region through the optical waveguide path.

[0146] The solid-state imaging devices 18D, 18E, 18F, 18G according to Mode 7 of the present invention preferably further include: an interlayer insulating film 9 formed on the light shielding film 8; and an inner lens 16 formed on the interlayer insulating film 9 in order to condense the light so as to guide the light to the photoelectric conversion region 3, and the inner lens 16 is formed of the same material as the material constituting the interlayer insulating film 9 or at least one or more materials having a refractive index different from that of the material constituting the interlayer insulating film 9.

[0147] According to the above structure, the light incident on the substrate can be efficiently guided to the photoelectric conversion region disposed below the dug-out region through the inner lens.

[0148] The solid-state imaging devices 18A, 18C, 18D, 18E, 18F, 18G according to Mode 8 of the present invention preferably further include a passivation film 12 formed between the interlayer insulating film 9 and the optical waveguide 11 or the inner lens 16, and the passivation film 12 has a compressive stress in the following direction of -5.0×10 -9 dyne / cm 2 or less.

[0149] According to the above structure, the dark current of the solid-state imaging device can be exponentially reduced.

[0150] The present invention is not limited to the above-described embodiments, and various modifications can be made within the scope shown in the claims. Embodiments obtained by appropriately combining technical solutions separately disclosed in different embodiments are also included in the technical scope of the present invention. Moreover, by combining the technical solutions separately disclosed in each embodiment, new technical features can be formed.

Claims

1. A solid-state imaging device in which a plurality of pixels are arranged in a matrix on a substrate, wherein the solid-state imaging device is characterized in that: Each of the plurality of pixels includes: A photoelectric conversion region disposed inside the substrate to convert light incident on the substrate into charges; A charge holding region disposed inside the substrate on the light incident side of the photoelectric conversion region to hold the charges converted by the photoelectric conversion region; A dug-out region dug from the surface of the substrate on the light incident side toward the photoelectric conversion region to at least a depth corresponding to the charge holding region; And A light-shielding film formed to cover the surface side of the substrate of the charge holding region and extending on the side walls of the dug-out region, The photoelectric conversion region is formed to extend from below the dug-out region to below the charge holding region and has a protruding portion protruding toward the charge holding region.

2. The solid-state imaging device according to claim 1, wherein: It further includes a charge transfer gate electrode for transferring the charges from the photoelectric conversion region to the charge holding region, The charges are transferred to the charge holding region through a charge transfer path that passes from the photoelectric conversion region through the charge transfer gate electrode to the charge holding region.

3. The solid-state imaging device according to claim 1 or 2, wherein: The charge transfer gate electrode is buried in the substrate so as to cover the side surface of the charge holding region.

4. The solid-state imaging device according to claim 1 or 2, wherein: The light-shielding film is formed of a single or multiple film structures having a higher light reflectivity than that of a single layer of tungsten.

5. The solid-state imaging device according to claim 1 or 2, characterized in that, It further includes: An interlayer insulating film formed on the light-shielding film; and An optical waveguide formed on the interlayer insulating film by a material having a refractive index different from that of the interlayer insulating film to guide the light to the photoelectric conversion region.

6. The solid-state imaging device according to claim 1 or 2, characterized in that, It further includes: An interlayer insulating film formed on the light-shielding film; and An inner lens formed on the interlayer insulating film to condense the light so as to guide the light to the photoelectric conversion region, The inner lens is formed of the same material as the material constituting the interlayer insulating film or at least one or more materials having a refractive index different from that of the material constituting the interlayer insulating film.

7. The solid-state imaging device according to claim 5, wherein: It further includes a passivation film formed between the interlayer insulating film and the optical waveguide, The passivation film has a compressive stress in the compression direction of -5.0×10 -9 dyne / cm 2 or less.

8. The solid-state imaging device according to claim 6, wherein: It further includes a passivation film formed between the interlayer insulating film and the inner lens, The passivation film has a compressive stress in the compression direction of -5.0×10 -9 dyne / cm 2 or less.

Images