Solid-state imaging device and manufacturing method

Abstract

A solid-state imaging element according to the present invention comprising a semiconductor substrate having a plurality of light-receiving elements in a matrix, and a plurality of microlenses formed to respectively correspond to the plurality of light-receiving elements, wherein a protrusion made of photosensitive resin is formed between or at a boundary of adjacent microlenses.

Description

[Technical Field]

[0001] The present invention relates to a solid-state imaging device equipped with a microlens, and more particularly to a solid-state imaging device used in a range image sensor that is being developed both domestically and internationally, and a method for manufacturing the same. This application claims priority from Japanese Patent Application No. 2020-147599, filed September 2, 2020, the contents of which are incorporated by reference. [Background technology]

[0002] Conventionally, methods for measuring three-dimensional distances using two-dimensional image sensors have been used, such as with smartphones, with dual or triple cameras, to measure distances using techniques such as triangulation and defocusing. On the other hand, distance image sensors capable of three-dimensional distance measurement are image sensors that can capture images containing distance information to the target object. They obtain distance information by irradiating the target with infrared light and detecting the time it takes for the reflected light from the target to be received (TOF: Time of Flight). These distance image sensors use infrared light for distance measurement and a separate image sensor for color images.

[0003] In recent years, three-dimensional image sensors capable of measuring distances on a two-dimensional pixel array have been developed for distance image sensors, and are attracting attention as next-generation image sensors. However, three-dimensional image sensors capable of measuring distances require space for time measurement in TOF, which reduces the aperture ratio of the light-receiving elements that acquire image information. This necessitates the use of microlenses to improve sensitivity.

[0004] For example, in the case of a hemispherical microlens used for this purpose, with a diameter of 20 μm, which is roughly the same as the pixel size, the lens height must be 10 μm to 15 μm. As such, the lens diameter required is orders of magnitude larger than that of the microlenses used in conventional solid-state imaging devices, but as the pixel size increases, the height of the microlens also tends to increase.

[0005] Regarding the manufacturing method of microlenses, since permanent resist capable of forming microlenses with a thickness of 10 μm or more by thermal flow is not available, microlenses are formed by an etch-back method. In the etch-back method, a layer that will become the microlens is first formed, and then a resist pattern is formed on top of it, and the resist pattern is then formed into a lens shape by thermal flow. Next, the lens-shaped resist pattern is dry-etched to transfer the lens shape to the layer that will become the microlens (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 2004-200360 Summary of the Invention [Problem to be solved by the invention]

[0007] However, when a microlens with a diameter of 20 μm and a height of 10 μm is formed by the etch-back method, cracks may occur in the valleys between adjacent microlenses. This is thought to be because the microlens is significantly thicker and has a larger diameter than conventional microlenses, causing stress due to thermal contraction of the microlens to concentrate in the valleys. Another presumed cause is the lens's large curvature.

[0008] When cracks occur between the microlenses in this way, the light-collecting characteristics of the lenses deteriorate, and the sensitivity of the image sensor decreases.

[0009] In order to solve the above problems, an object of the present invention is to provide a solid-state imaging device having microlenses in which cracks do not occur in the valleys where the microlenses overlap. [Means for solving the problem]

[0010] A first aspect of the present invention is a solid-state imaging device having a semiconductor substrate on which a plurality of light receiving elements are arranged in a matrix, and a plurality of microlenses formed corresponding to the plurality of light receiving elements, respectively. In this solid-state imaging device, adjacent microlenses and the connecting region has: A protrusion made of photosensitive resin is formed. The protrusions have a thickness equal to or less than that of the microlenses, and are covered with the same resin as the microlenses.

[0011] A second aspect of the present invention is a method for manufacturing the solid-state imaging device according to the first aspect. This manufacturing method includes the steps of forming a protrusion made of photosensitive resin on a semiconductor substrate, forming a transparent resin layer on the semiconductor substrate to cover the protrusion, and forming a plurality of microlenses using the transparent resin layer. [Effects of the Invention]

[0012] According to the solid-state imaging device of the present invention, protrusions are formed in the areas where adjacent microlenses are connected (valleys of the microlens array) to increase the volume of the resin that makes up the valleys of the microlenses. Therefore, even if the force generated by heat or the like during formation by the etch-back method is concentrated in the valleys, the resin layer in the cross section of the valleys is increased, so the stress, which is the force per unit cross-sectional area, is reduced, and it is possible to suppress the occurrence of cracks in the valleys.

[0013] According to the second aspect of the present invention, it is possible to provide the solid-state imaging device according to the first aspect of the present invention. [Brief explanation of the drawings]

[0014] [Figure 1] 1A and 1B are diagrams illustrating the configuration of a solid-state imaging device of the present invention, in which (a) is a top view illustrating a part of the solid-state imaging device, and (b) is a cross-sectional view taken along the AA′ cutting line in (a). [Figure 2] 3A to 3C are diagrams illustrating a method for manufacturing a solid-state imaging device according to the present invention. [Figure 3] FIG. 10 is a diagram showing an overlapping region of microlenses. [Figure 4] 10A and 10B are diagrams showing other examples of protrusions. [Figure 5] FIG. 10 is a schematic cross-sectional view showing another example of a solid-state imaging device according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0015] <Solid-state imaging element> The solid-state imaging device of the present invention will be described with reference to FIGS. The solid-state imaging element 10 of the present invention comprises a semiconductor substrate 1 for a time-of-flight distance image sensor, which includes a silicon wafer 5 having a plurality of light receiving elements 4 arranged in a matrix, and microlenses 2 formed corresponding to the position of each light receiving element 4 on the semiconductor substrate 1. The microlenses 2 are formed in the form of a microlens array, with the peripheral portions between adjacent microlenses 2 connected together.

[0016] In the solid-state imaging device 10 of the present invention, the thickness of the microlenses 2 is equal to or greater than half the lens pitch of the microlens array. In a time-of-flight distance image sensor, in addition to the light receiving element, space for TOF measurement is required. This reduces the aperture ratio. To compensate for the reduced aperture ratio, it is necessary to form microlenses. For example, if the lens pitch is a hemispherical lens with a 20 μm diameter, which is roughly the same as the diameter of the microlens, the lens height (thickness) needs to be 10 μm to 15 μm.

[0017] In the connecting region where adjacent microlenses 2 are connected (referring to the lowered region between microlenses 2, or the region where the valley of the microlens array is formed), a protrusion 3 is formed to increase the volume of the resin that forms the valley of the microlens 2.

[0018] Here, we will explain the connecting region of the microlens 2. When the peripheral portions of two adjacent, substantially hemispherical microlenses overlap and connect to each other, the thickness of the resin forming the valleys of the microlenses 2 in the connecting region varies depending on the state of overlap. For example, when two adjacent microlenses are in contact (not overlapping) at their peripheries, there is no overlap of the resin forming the microlenses 2 in the connecting region. When the peripheries of two adjacent microlenses 2 overlap shallowly in the connecting region, the resin layer (hereinafter simply referred to as the resin layer) forming the peripheries of the microlenses 2 is thinner than when they overlap deeply. On the other hand, when adjacent microlenses 2 overlap deeply, the resin layer in the connecting region is thicker. In any of the three cases described above, the resin layer in the connecting region is thinner than the peak position (vertex position of the microlens 2) of the microlens 2, and therefore is the lowest region of the microlens 2. Note that the thickness of the resin layer in the connecting region varies depending on factors such as variations in the positional accuracy of the microlens 2.

[0019] Therefore, depending on the overlap state of two adjacent microlenses, the thickness of the resin layer formed by the overlap of the resin of the microlenses 2 in the connecting region varies. The first aspect of the present invention is applicable to either state.

[0020] Furthermore, in the connection region, the thickness of the resin layer changes depending on whether there is nothing between the microlens 2 and the semiconductor substrate 1 or whether a protrusion 3 is formed between the microlens 2 and the semiconductor substrate 1 (hereinafter also referred to as the base) as shown in Figure 1. In the former case, a resin layer is formed with a thickness determined by the overlapping state of two adjacent microlenses. In the latter case, the thickness of the protruding portion 3 formed on the semiconductor substrate 1 is added to the resin layer having the thickness of the former case, so that the resin layer formed is thicker in the latter case.

[0021] For the above reasons, when the protrusions 3 are formed on the base of the connecting region as in the first embodiment of the present invention, the thickness of the resin layer in the valley portion can be increased, which increases the cross-sectional area of ​​the resin layer and reduces stress, thereby suppressing the occurrence of cracks due to thermal distortion, etc., even when forming large microlenses such as microlenses for time-of-flight.

[0022] Furthermore, by using a resin for the protrusions 3 that has a higher refractive index than the resin material for the microlenses 2, it is possible to increase the sensitivity of the solid-state imaging element, since stray light that would otherwise not be incident on the light receiving elements 4 can be reused by being reflected by the protrusions 3. In this case, it is preferable that the difference in refractive index between the resin material for the microlenses 2 and the resin material for the protrusions 3 exceeds 0.1.

[0023] <Method of manufacturing a solid-state imaging device> Next, a method for manufacturing a solid-state imaging device according to the second embodiment of the present invention will be described with reference to FIGS. The method for manufacturing a microlens in the second aspect of the present invention is a method for manufacturing a microlens having a microlens array in which microlenses 2 are formed corresponding to the position of each light receiving element 4 of a semiconductor substrate 1 for a time-of-flight type distance image sensor, the semiconductor substrate 1 having a plurality of light receiving elements 4 arranged in a matrix, and the microlenses 2 are connected at their peripheral portions.

[0024] The method for manufacturing a microlens of the present invention includes the following steps: Step 1 (FIG. 2(b)) of forming protrusions 3 on a semiconductor substrate 1 (FIG. 2(a)) in at least a portion of a connecting region where adjacent microlenses 2 are connected; Step 2 (FIG. 2(c)) of forming a transparent resin layer 6 that will become the microlenses 2 on top of the protrusions 3; Step 3 (FIG. 2(d)) of forming a photosensitive resist layer 7 on the transparent resin layer 6; Step 4 (FIG. 2(e)) of exposing and developing the photosensitive resist layer 7 using a predetermined photomask to form a resist pattern 8; Step 5 (FIG. 2(f)) of converting the resist pattern 8 into a lens shape by a thermal flow treatment in which the resist pattern 8 is heated to a softening temperature or higher; and Step 6 (FIG. 2(g)) of forming a lens shape in the transparent resin layer 6 by dry etching using the resist pattern 9 converted into a lens shape as an etching mask. In particular, the second aspect includes Steps 1 and 2.

[0025] Furthermore, in the method for manufacturing a microlens of the present invention, after step 3 (FIG. 2(d)) of forming a photosensitive resist layer 7 on a transparent resin layer 6, the photosensitive resist layer 7 may be exposed to light using a gray-tone photomask (gray-tone mask) and developed to form a lens shape (FIG. 2(f)) without performing a thermal flow treatment on the developed resist pattern 8 shown in FIG. 2(e).

[0026] (Process 1) Step 1 is a step of forming protrusions 3 in a matrix shape in plan view on the semiconductor substrate 1 so that each light receiving element 4 of the semiconductor substrate 1 is positioned approximately at the center, passing through at least a portion of the connecting region where the valleys of adjacent microlenses 2 are formed (FIG. 2(b)). The protrusions 3 need only be formed in at least the connecting region of the microlenses 2. The protrusions 3 do not have to be in a matrix shape in plan view, and may be formed intermittently in the regions corresponding to the valleys of the microlenses 2.

[0027] 1(a), the protrusions 3 are matrix-shaped protrusions 3 with each light receiving element 4 of the semiconductor substrate 1 arranged at the center. The protrusions 3 can be formed by applying a photosensitive resin onto the semiconductor substrate 1, drying it, and then exposing and developing it using a photomask with a predetermined exposure pattern.

[0028] The method for applying the photosensitive resin does not need to be particularly limited. Various application methods can be used, including spin coating, which is used in semiconductor manufacturing processes, as well as die coating, roll coating, and screen printing. The method can be selected appropriately taking into consideration the size of the substrate and the required film thickness uniformity.

[0029] The preferred method for drying the photosensitive resin applied to the semiconductor substrate 1 is to dry it in heated clean air using a clean oven used in the semiconductor manufacturing process, but there is no need to limit the drying method as long as it is possible to dry the photosensitive resin without foreign matter adhering to it.

[0030] The method of exposing and developing the photosensitive resin can be carried out using a projection type exposure device and development device used in the semiconductor manufacturing process, but there is no need to be limited to this, and any exposure method and development method that can form the required pattern can be used.

[0031] The protrusions 3 may also be part of a matrix of protrusions 3. In that case, as shown in Fig. 1(b), the protrusions 3 may be formed so as to be disposed in the portions where the microlenses 2 between adjacent microlenses 2 overlap and connect with each other. The protrusions 3 may not be formed in other portions.

[0032] At least, if the protrusions 3 are formed so as to be disposed in the connecting region, the cross-sectional area of ​​that portion increases. By increasing the cross-sectional area, it is possible to reduce the stress, which is the force per unit cross-sectional area, in the portion where the microlenses 2 are connected between adjacent microlenses 2. It is preferable that the protrusion 3 is arranged so as to pass through the center of the connecting region. The center of the connecting region is where the bottom of the valleys formed by adjacent microlenses 2 is formed, which can further enhance the stress relaxation effect. In other words, it is preferable that the protrusion 3 is arranged in the region where the bottom of the valleys of adjacent microlenses 2 is formed. It is even more preferable that the protrusions 3 fill the entire connecting region. In other words, it is preferable that the protrusions 3 are formed so as to cover the entire valleys of adjacent microlenses 2. It is thought that the lowest parts of the valleys are most susceptible to stress, but by covering the periphery of the lowest parts with the protrusions 3, it is possible to further alleviate the stress per unit cross-sectional area.

[0033] (Process 2) Step 2 is a step of forming a transparent resin layer 6 that will become the microlenses 2 from above the protrusions 3 (FIG. 2(c)). The coating method and drying method used in step 1 can be used.

[0034] (Step 3) Step 3 is a step of forming a photosensitive resist layer 7 on the transparent resin layer 6 (FIG. 2(d)). The coating method and drying method used in step 1 can be used.

[0035] (Step 4) Step 4 is a step of forming a resist pattern 8 by exposing and developing the photosensitive resist layer 7 using a predetermined photomask (FIG. 2(e)). The exposure and development methods used in step 1 can be used. The resist pattern 8 is arranged in the area where the microlens 2 is to be formed.

[0036] (Step 5) Step 5 is a step of changing the resist pattern 8 into a lens shape by a thermal flow treatment in which the resist pattern 8 is heated to a temperature above the softening temperature (Fig. 2(f)). The resist pattern 8 formed in step 4 is heated to a temperature above the softening point of the resin material that makes up the resist pattern 8, causing thermal flow, and the resist pattern 8 can be changed into a lens shape as shown in Fig. 2(f).

[0037] If the thickness (height) of the resist pattern 9 changed into a lens shape is a and the thickness of the transparent resin layer 6 is b, then for example, the transparent resin layer 6 having a thickness of b is also formed on the protrusion 3.

[0038] In step 4, a lens-shaped resist pattern as shown in Figures 2(d) to 2(f) may be formed by exposure and development using a gray-tone mask. In this case, the step (step 5) of performing a thermal flow treatment by heating the resist pattern 8 shown in Figure 2(e) to a temperature above its softening temperature can be omitted.

[0039] (Step 6) Step 6 is a step of forming a lens shape in the transparent resin layer 6 by dry etching using the resist pattern (resist pattern after thermal flow) 9 that has been changed into a lens shape as an etching mask (FIG. 2(g)).

[0040] As shown in Figure 2(f), if the thickness of the resist pattern 9 changed into a lens shape is a and the thickness of the transparent resin layer 6 is b, then, as shown in Figure 2(g), dry etching can be stopped just when a microlens 2 having a thickness of b of the transparent resin layer 6 is formed. It is preferable to use materials for the resist pattern 9 and the transparent resin layer 6 that have the same rate at which the resin is etched away by dry etching. Such a case can be exemplified by the case where both are made of the same resin, but they may also be made of different resins.

[0041] That is, the thickness (height) of the microlens 2 is b, and in this state, the transparent resin remains on the protruding portion 3 with a thickness of ba.

[0042] Therefore, in the valleys between the microlenses 2, protrusions 3 and a transparent resin layer with a thickness of ba are formed.

[0043] Here, the transparent resin with a thickness ba is the same as the thickness of the transparent resin formed in the valley portions when no protrusions 3 are formed in the valley portions. Therefore, when no protrusions 3 are formed in the valley portions, the thickness of the transparent resin formed in the valley portions between the microlenses 2 is ba, but when protrusions 3 are formed in the valley portions, the thickness of the resin layer formed in the valley portions is the sum of the thickness ba of the transparent resin and the thickness of the protrusions 3. Therefore, the thickness of the resin layer in the valley portion increases, which reduces stress in the valley portion and suppresses the occurrence of cracks.

[0044] Other aspects of the protrusions will be described. For example, as shown in Fig. 3, when the diameter of a microlens 2 is larger than one side of a pixel region that is square in plan view, an overlapping region occurs between adjacent microlenses 2. The multiple protrusions 3A shown in Fig. 4 are formed to have a shape similar to this overlapping region in plan view. Therefore, there are no protrusions 3A at the corners of the pixel region (on both sides of the microlens 2 in the diagonal direction of the pixel region), and unlike the protrusions 3 described above, they are not continuous in a matrix form.

[0045] The inventors' investigations confirmed that the frequency of cracks increases when the resin thickness in the connecting region is 2 μm or less. On the other hand, it was also confirmed that cracks do not occur in corners where stress is applied differently, even if the thickness is 2 μm or less. Therefore, even if the protrusions 3A are provided only in the overlapping region, by increasing the thickness of the resin layer in the connecting region, it is possible to effectively suppress the occurrence of cracks and prevent cracks from occurring in the corners.

[0046] According to the inventor's investigation, when the thickness of the resin in the connecting region exceeds 2 μm, the frequency of cracks drops significantly, and when the thickness is 5.5 μm or more, the frequency of cracks becomes almost zero. From this point of view, it can be said that the thickness of the protrusion 3A is preferably 1 μm or more and 5 μm or less.

[0047] When the protrusions 3A are provided, the resin thickness at the corners is reduced. This allows for a larger difference in height between the valleys between the microlenses and the tops of the microlenses in the diagonal direction of the pixel area, making it easier to increase the curvature of the microlenses in the diagonal direction. This has the advantage of making it easier to improve the light collection efficiency in the diagonal direction.

[0048] Furthermore, the protrusion 3A has an advantage even in an embodiment where there is no overlapping area between the microlenses. In step 6, dry etching is performed until the transparent resin layer 6 on the protrusions 3A is removed, resulting in a structure in which the protrusions 3A are exposed between the microlenses 2, as shown in FIG. 5. This configuration completely prevents cracks from occurring in the connecting regions, while the protrusions 3A prevent the silicon wafer 5 from being exposed and protect it. When the transparent resin layer is formed in step 2, due to factors such as surface tension, it tends to be slightly thicker at the corners than on the protrusions 3A. Therefore, it is possible to remove only the transparent resin layer 6 on the protrusions 3A by dry etching, while leaving the transparent resin layer 6 on the corners to protect the silicon wafer 5.

[0049] In this embodiment, the microlenses are made of a non-photosensitive resin, while the protrusions are made of a photosensitive resin, so they are made of different materials. Making the refractive indexes of the two equal allows the microlenses and the protrusions to be optically formed into a uniform resin layer, which has the advantage of making it easier to stabilize the optical characteristics of the solid-state imaging device. From the perspective of forming the microlenses and the protrusions into a uniform resin layer, it is preferable that the difference in refractive index between the two be within 0.1.

[0050] The crack suppression according to the present invention is particularly effective when the diameter of the microlens 2 (the maximum dimension in the radial direction when the microlens 2 is not a perfect circle in plan view) is 15 μm or more. [Explanation of symbols]

[0051] 1. Semiconductor substrate (for time-of-flight distance image sensors) 2 Microlenses 3, 3A protrusion 4 Photodetector 5. Silicon wafers 6 Transparent resin layer 7 Photosensitive resist layer 8 Resist pattern 9. Resist pattern changed into a lens shape (or resist pattern after thermal flow treatment) 10 Solid-state imaging element a) Thickness (height) of the resist pattern changed into a lens shape b Thickness of the transparent resin layer

Claims

1. A solid-state imaging device having a semiconductor substrate on which a plurality of light receiving elements are arranged in a matrix, and a plurality of microlenses formed corresponding to the plurality of light receiving elements, a connecting region in which adjacent microlenses are connected to each other; a protrusion made of a photosensitive resin is formed in the connecting region, The protrusion has a thickness equal to or less than the thickness of the microlens and is covered with the same resin as the microlens. Solid-state imaging element.

2. the difference between the refractive index of the resin constituting the protrusion and the refractive index of the resin constituting the microlens is within 0.1; The solid-state imaging device according to claim 1 .

3. The refractive index of the resin constituting the protrusion is higher than that of the resin constituting the microlens. The solid-state imaging device according to claim 1 .

4. A method for manufacturing a solid-state imaging device according to any one of claims 1 to 3, comprising: forming a protrusion made of a photosensitive resin on the semiconductor substrate; forming a transparent resin layer on the semiconductor substrate to cover the protruding portion; forming a plurality of microlenses using the transparent resin layer; Equipped with A method for manufacturing a solid-state imaging device.

5. forming the plurality of microlenses by dry etching; The method for manufacturing a solid-state imaging device according to claim 4 .

Images