JP2006216904A - Color solid state image sensor and method of fabricating the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obviate color mixing and a degradation of a formation accuracy of a micro lens caused by individually different thicknesses of a plurality of different color patterns that makes up a color filter. <P>SOLUTION: The color solid state image sensor 100 is provided with a plurality of photoelectric conversion elements 102 arranged on a substrate 101 to convert incident light to electric charges, a transparent film 105 that covers the substrate 101 on which the plurality of photoelectric conversion elements 102 is provided, and a color filter 106 that is made up of a plurality of color patterns formed on the transparent film 105 and having individually different thicknesses to make a color separation of the incident light. The upper surfaces of the plurality of color patterns are each flush with others. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a color solid-state imaging device and a method for manufacturing the same, and more particularly to a color filter provided in the color solid-state imaging device and a method for manufacturing the same.

  Color solid-state imaging devices using semiconductors such as CCD (Charge Coupled Device) image sensors and CMOS (Complementary Metal Oxide Semiconductor) image sensors are used in various applications such as digital cameras, digital video cameras, and mobile phones. Has been. Further, along with the spread of such devices, in addition to the demands for higher functionality and higher performance such as an increase in the number of pixels and improvement in light receiving sensitivity, there are also increasing demands for downsizing and lower prices.

  As the size and number of pixels of a color solid-state image sensor increase, the pixel size incorporated in the color solid-state image sensor will be further reduced. As the pixel size is reduced, the light receiving sensitivity, which is one of the basic performances of the color solid-state imaging device, is lowered, so that it is difficult to capture a clear image at low illuminance.

  On the other hand, along with the miniaturization of the solid-state image sensor, the miniaturization of the camera body has progressed at the same time, and the tendency to extremely miniaturize the optical system into which the image sensor is taken in, such as mounting an imaging module on a mobile phone, is rapidly increasing. It is out. As a result, the pupil position of the lens is inevitably close, and the oblique incident component of the light incident on the imaging surface increases. This is particularly noticeable at the periphery of the imaging surface.

  FIG. 4 schematically shows a cross section of a main part of a conventional color solid-state imaging device.

  The color solid-state imaging device 10 in FIG. 4 is formed using a silicon substrate 11. A plurality of photoelectric conversion elements 12 are arranged on the silicon substrate 11, and an insulating film (not shown) that covers the silicon substrate 11 and the photoelectric conversion elements 12 is formed. In addition, electrodes 13 made of, for example, polycrystalline silicon are formed at positions between the photoelectric conversion elements 12 on the silicon substrate 11. Although not shown in detail, a charge transfer portion for transferring charges generated by photoelectric conversion in the photoelectric conversion element 12 is formed by the electrode 13 or the like.

  Further, a light shielding film 14 is formed so as to cover the electrode 13, and a planarizing layer 15 made of a transparent resin or the like is formed on the photoelectric conversion element 12 and the light shielding film 14. The planarization layer 15 is formed in order to fill the unevenness caused by the formation of the electrode 13 and the like and to flatten the surface.

  On the planarization layer 15, color filters 16 are formed corresponding to the individual photoelectric conversion elements 12. Specifically, any one of a red color pattern (hereinafter referred to as an R pattern) 16R, a green color pattern (hereinafter referred to as a G pattern) 16B, and a blue color pattern (hereinafter referred to as a B pattern) is placed on each photoelectric conversion element 12. The color filter 16 is formed so as to be positioned.

  An intermediate layer 17 is formed on the color filter 16 to fill and flatten the unevenness generated on the upper surface of the color filter 16 due to the difference in thickness of each color pattern. The microlenses 18 positioned on the photoelectric conversion elements 12 are respectively formed. Here, the intermediate layer 17 is formed of a transparent resin or the like.

  Considering each pixel, as can be seen from the cross-sectional shape of the solid-state imaging device 10 in FIG. 4, the larger the incident angle of light (the greater the angle from the perpendicular to the photoelectric conversion element 12), the more the photoelectric conversion element 12 has. The amount of incident light is reduced. This is a phenomenon called a shading phenomenon, and is a cause of the occurrence of luminance distortion or the like seen in a captured image. When the shading phenomenon occurs, it becomes difficult to maintain high image quality. Accordingly, a technique for improving the light receiving sensitivity by forming the microlens 18 has been proposed.

  The color solid-state imaging device has a color filter array having a multi-layered structure such as at least a flattening layer 15, a color filter 16 and an intermediate layer 17, and further has a microlens 18 formed on the top. Thus, the color components are collected in the photoelectric conversion element 12.

  However, when such a structure is adopted, the distance between the microlens 18 and the photoelectric conversion element 12 becomes long. Therefore, when the light incident on the microlens 18 obliquely increases, the distance from the photoelectric conversion element 12 is also increased. The light collection rate decreases. This is remarkable, for example, when the aperture of the imaging lens of the camera is opened.

  Further, when the collected light approaches the opening end of the light shielding film, signal charges are likely to be mixed into the photoelectric conversion element 12 and the charge transfer unit provided in the adjacent pixels. In order to alleviate this phenomenon called smear, a color solid-state imaging device including a color microlens array in which the functions of the color filter array and the microlens are combined into one layer has been proposed.

  In the color solid-state imaging device 10 shown in FIG. 4, the photoelectric conversion elements 12 are arranged in a matrix on the silicon substrate 11. Here, the color filters 16 formed corresponding to the respective photoelectric conversion elements 12 are periodically formed with a predetermined combination of the R pattern 16R, the G pattern 16G, and the B pattern 16B as one unit in the row and column directions. Is arranged.

  In general, the R pattern 16R, the G pattern 16G, and the B pattern 16B have different film thicknesses in order to obtain predetermined spectral characteristics. For this reason, unevenness occurs on the color filter 16 formed on the planarizing layer 15. For this reason, the intermediate layer 17 is formed and planarized. Here, the intermediate layer 17 is formed, for example, by applying and curing an acrylic resin, which is a transparent resin, in several times.

  After the planarization by the intermediate layer 17 in this way, the microlens 18 is formed. For this purpose, a photosensitive resin is applied to the entire surface of the intermediate layer 17, exposed using a mask, developed, and heated to form the microlens 18 by surface tension.

  The conventional techniques described above have the following problems.

  In manufacturing a solid-state imaging device, the flatness of the intermediate layer is extremely important in the microlens formation process. However, since the color pattern of the color filter has a different film thickness for each color, a slight unevenness may occur on the surface of the intermediate layer formed for planarization. In this case, the shape (curvature) of the microlens formed on the intermediate layer changes, and a difference occurs in the light collection efficiency that should be uniform for each individual solid-state imaging device.

  FIGS. 5A to 5D are schematic views for explaining this.

  First, FIGS. 5A and 5B show a pixel including a G pattern 16G sandwiched between two B patterns 16B. Here, FIG. 5A shows a case where the surface of the intermediate layer 17 is correctly flattened and is correctly condensed with respect to the photoelectric conversion element 12G corresponding to the G pattern 16G. On the other hand, FIG. 5B shows a case where the focal point of light collection with respect to the photoelectric conversion element 12G is deviated from the photoelectric conversion element 12G, and light collection is not performed correctly. This is because the film thickness of the G pattern 16G is smaller than the film thickness of the B pattern 16B formed on both sides thereof, so that a slight depression occurs on the intermediate layer 17 and the shape of the corresponding microlens 18G is distorted. It is because it is.

  FIGS. 5C and 5D show a pixel including a G pattern 16G sandwiched between two R patterns 16R. Here, FIG. 5C shows a case where the surface of the intermediate layer 17 is accurately flattened. In this case, the condensing by the microlens 18G is correctly performed as in FIG. On the other hand, in the case of FIG. 5D, since the thickness of the G pattern 16G is larger than the thickness of the R pattern 16R formed on both sides thereof, the surface of the intermediate layer 17 is also above the B pattern 16B. There is a slight bulge in the surface. As a result, the shape of the microlens 18G is distorted, and the light is not accurately collected on the photoelectric conversion element 12G.

  Condensation abnormalities as described above are likely to occur, and when they occur, the output sensitivity of the solid-state imaging device is affected. In order to prevent this, it has been necessary to increase the number of times the resin is applied in order to improve the flatness of the intermediate layer. However, this increases the distance from the photoelectric conversion element 12 to the microlens 18 (the thickness of the layer between the photoelectric conversion element 12 and the microlens 108 increases). The shading phenomenon due to a decrease in sensitivity becomes a problem.

  For this, the intermediate layer 17 is applied in multiple layers (thickly laminated by applying many times) until it becomes sufficiently flat, and then the intermediate layer 17 is shaved by an etching process, so that the distance between the photoelectric conversion element 12 and the microlens 18 is reduced. A method of shortening has been proposed. However, in this case, the increase in the number of processes and the characteristic variation depending on the accuracy variation of the etching process are problems.

  Furthermore, even when the intermediate layer 17 is formed flat with high accuracy and the microlenses 18 are formed uniformly thereon, the color filters 16 of the respective colors are different in height after formation. There is a problem that the characteristics deteriorate due to the influence of incident light. Hereinafter, this is referred to as color mixing and will be described with reference to FIGS. 6 (a) and 6 (b).

  FIG. 6A shows a G pattern 16G and two B patterns 16B adjacent to both sides thereof. FIG. 6B shows a G pattern 16G and R patterns 16R adjacent to both sides thereof. Further, in FIGS. 6A and 6B, light 21 incident perpendicularly to the G pattern 16G and light 22 incident obliquely are shown. It is assumed that the wavelength region of incident light is a red region (wavelength 600 to 700 nm).

  In the case of FIG. 6A, since the B pattern 16B is thicker than the G pattern 16G, the oblique incident light 22 partially passes through the B pattern 16B and is absorbed. In contrast, in the case of FIG. 6B, since the R pattern 16R has a smaller film thickness than the G pattern 16G, the oblique incident light 22 similar to the case of FIG. 6A also passes through the R pattern 16R. Never do.

  For each pixel, the output should be the same if the incident light is similar. However, as described above, there is a difference in the light transmitted through the G pattern 16G between the case of FIG. 6A and the case of FIG. . As a result, a so-called line density defect occurs.

The conventional color solid-state imaging device has the above-described problems. In view of these, the object of the color solid-state imaging device of the present invention is to prevent variations in characteristics due to the presence of irregularities on the upper surface of the color filter. It is another object of the present invention to provide a method for manufacturing such a color solid-state imaging device.
JP 2002-184965 A

  In order to achieve the above object, a color solid-state imaging device according to the present invention includes a plurality of photoelectric conversion elements arranged on a substrate and converting incident light into electric charges, and a transparent substrate covering a substrate provided with the plurality of photoelectric conversion elements. A film and a color filter formed on the transparent film and having a plurality of color patterns with different thicknesses and color-separating incident light, and each upper surface of the plurality of color patterns is flush with each other Features.

  According to the color solid-state imaging device of the present invention, the color filter has a flat upper surface. In other words, the color filter is composed of a plurality of color patterns whose upper surfaces are flush with each other. For this reason, it is possible to avoid line shading defects and the like that occur in the conventional case where the upper surface height is different in a plurality of color patterns, which occurs due to the oblique incident light passing through the adjacent color patterns. Can do. For this reason, an image with high image quality can be taken.

  In addition, it is preferable that the upper surfaces of the plurality of color patterns are flush with each other because the transparent film has irregularities corresponding to different thicknesses of the plurality of color patterns.

  The color filters are formed on the transparent film and have color patterns with different thicknesses. Here, by providing unevenness on the transparent film according to the thickness of each color pattern, the combined thickness of the transparent film and each color pattern can be made constant. That is, the transparent film is thinned at the position where the thickness of the color pattern is large, and the transparent film is thickened at the position where the thickness of the color pattern is small.

  In this way, the upper surfaces of the plurality of color patterns included in the color filter can be flush with each other, and the upper surfaces of the color patterns are surely flat.

  In addition, it is preferable to further include another transparent film formed on the color filter and a plurality of microlenses formed on the other transparent film and collecting incident light on the plurality of photoelectric conversion elements, respectively.

  In the color filter of the color solid-state imaging device of the present invention, the upper surface of each color pattern is flush and the flatness of the upper surface is high. If another transparent film is further provided on such a color filter, the upper surface of the other transparent film can be easily made a surface with very high flatness. For this reason, it is easy to form microlenses with high accuracy on another transparent film, and the spectral performance of the color solid-state imaging device can be improved.

  Also, the thickness of the other transparent film may be smaller than when the upper surface of the color pattern has irregularities. For this reason, the sensitivity of the color solid-state imaging device is improved and the sensitivity to oblique incident light is also improved, so that the shading phenomenon can be suppressed.

  In order to achieve the above object, a method for manufacturing a color solid-state imaging device according to the present invention includes a step (a) of forming a plurality of photoelectric conversion elements that convert incident light into charges on a substrate, and a plurality of photoelectric conversion elements. A step (b) of forming a transparent film covering the substrate provided with a color filter, and a step (c) of forming a color filter composed of a plurality of color patterns having different thicknesses and color-separating incident light on the transparent film In step (b), the transparent film is provided with irregularities corresponding to different thicknesses of the plurality of color patterns, and in step (c), the color patterns are arranged on the irregularities so that the upper surfaces thereof are flush with each other. It is designed to form as follows.

  According to the method for manufacturing a color solid-state image pickup device of the present invention, a color solid-state image pickup having a color filter with a flat upper surface is provided by providing the transparent film with irregularities corresponding to the thicknesses of the plurality of color patterns of the color filter. An element can be formed. For this reason, the color solid-state image sensor of this invention which has effects, such as suppressing a line shading defect, can be manufactured.

  In the step (b) of forming the transparent film, a plurality of color patterns are formed by using a step of forming a material film made of a photosensitive material on the substrate and a gray scale mask capable of controlling the exposure dose. A step of exposing the material film with a plurality of exposure amounts according to the thickness of the film, and a step of forming the transparent film having irregularities by developing the exposed material film.

  In this case, the material film made of a photosensitive material is a transparent film having unevenness according to the gray gradation of the gray scale mask because the film thickness after development is determined by the difference in exposure amount. Here, by adjusting the gray scale mask that determines the exposure amount so that the unevenness of the transparent film corresponds to the thickness of each of the multiple color patterns, the top surfaces of the multiple color patterns should be aligned at the same height. Can do. As a result, the color solid-state imaging device of the present invention can be reliably manufactured.

  Here, with regard to the gray gradation of the gray scale mask, a method for obtaining a desired transmittance by dividing the light shielding film on the mask into unit cells and controlling the shape or density of the dot pattern formed in the region. (For example, refer to JP-A-2002-244273). A method of arranging regions having different optical density levels in a checkered pattern has also been proposed (see, for example, JP-A-2002-365784A).

  Further, the method further includes a step of forming another transparent film on the color filter, and a step of forming a plurality of microlenses for condensing incident light on the plurality of photoelectric conversion elements on the other transparent film, respectively. Is preferred.

  According to the method for manufacturing a color solid-state imaging device of the present invention, a color filter having a high flatness on the upper surface can be formed. For this reason, the upper surface of another transparent film formed on the color filter can be easily flattened with very high accuracy. For this reason, it is possible to easily form the microlens on another flat film with very high accuracy, and a color solid-state imaging device having excellent spectral characteristics can be manufactured.

  Furthermore, compared with the case where another transparent film is formed on a color filter having an uneven surface, the transparent film can be made thinner. At this time, in addition to the number of steps compared to a method of etching back after forming a thick transparent film to obtain a required flatness, the influence of variations in etching back processing accuracy can be avoided. For these reasons, a color solid-state imaging device with small characteristic variation can be manufactured. In addition, since the number of processes is smaller than in the prior art, cost reduction is realized.

  According to the color solid-state imaging device of the present invention, the color filter having the upper end flush with the upper surface is provided by forming the color pattern on the transparent film provided with the unevenness. For this reason, sensitivity characteristics are improved and color mixing is reduced. In addition, according to the method for manufacturing a color solid-state imaging device of the present invention, the steps of multilayer coating and subsequent etching necessary for forming another transparent film can be eliminated. For this reason, it is possible to shorten the process and to avoid variation in characteristics due to variation in etching.

  Hereinafter, a color solid-state imaging device and a manufacturing method thereof according to an embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 is a diagram schematically showing a cross-sectional configuration of the color solid-state imaging device 100 of the present embodiment, and shows a range including three photoelectric conversion elements 102 each having a color pattern of a different color as the color filter 106. ing. This will be described in detail below.

  A color solid-state imaging device 100 shown in FIG. 1 is formed using a silicon substrate 101. A plurality of photoelectric conversion elements 102 that convert incident light into electric charges are formed on the silicon substrate 101, and an insulating film (not shown) that covers the silicon substrate 101 and the photoelectric conversion elements 102 is formed. In addition, electrodes 103 made of, for example, polycrystalline silicon are formed at positions between the photoelectric conversion elements 102 on the silicon substrate 101. The electrode 103 has a function of transferring charges generated in the photoelectric conversion element 102.

  Further, a light shielding film 104 for preventing light from entering the electrode 103 is formed so as to cover the electrode 103. Further, a first transparent film 105 is formed on the photoelectric conversion element 102 and the light shielding film 104, and a color filter 106 is formed on the first transparent film 105. A second transparent film 107 made of a transparent resin is formed on the color filter 106, and microlenses 108 are provided on the second transparent film 107 corresponding to the individual photoelectric conversion elements 102. Yes.

  With such a structure, incident light to the color solid-state imaging device 100 is collected by the microlens 108, and the second transparent film 107, the color filter 106, the first transparent film 105, and the insulating film (not shown). Are sequentially entered into the photoelectric conversion element 102.

  More specifically, the color filter 106 includes a red color pattern (R pattern) 106R, a green color pattern (G pattern) 106B, and a blue color pattern (B pattern) 106B arranged in the horizontal direction. A color pattern of any one color is formed for each photoelectric conversion element 102. For this reason, only light having a specific range of wavelengths corresponding to the colors of the respective color patterns is incident on the individual photoelectric conversion elements 102.

  As described above, one color pattern and one microlens 108 are provided for one photoelectric conversion element 102. As a result, the light condensed by one micro lens 108 passes through any one of the color patterns, and is condensed on a predetermined one photoelectric conversion element 102 as light having a specific range of wavelengths. Will be.

  Further, the R pattern 106R, the G pattern 106G, and the B pattern 106B have different thicknesses. However, unevenness corresponding to such a difference in thickness is provided on the upper surface of the first transparent film 105, and the upper surfaces of the R pattern 106R, the G pattern 106G, and the B pattern 106B are aligned at the same height. ing. For this reason, the upper surface of the color filter 106 is flush.

  For example, in the case of the solid-state imaging device 100 of FIG. 1, among the three color patterns, the thickness of the B pattern 106B is the largest, and the thickness of the R pattern 106R is the smallest. The thickness of the G pattern 106G is intermediate between the other two colors. Therefore, the thickness of the first transparent film 105 is the smallest in the region where the B pattern 106B is formed, and the thickness of the first transparent film 105 is the largest in the region where the R pattern 106R is formed. It is like that. The thickness of the first transparent film 105 in the region where the G pattern 106G is formed is an intermediate thickness between the regions where the other two color patterns are formed. As a result, the top surfaces of the three color patterns are aligned at the same height, and unevenness due to the difference in film thickness of the three color patterns is prevented from occurring on the top surface of the color filter 106.

  Specifically, in the case of the conventional color solid-state imaging device 10 shown in FIG. 4, the thicknesses of the R pattern 16R, the B pattern 16B, and the G pattern 16G were, for example, 700 nm, 1200 nm, and 1000 nm in order. For this reason, on the upper surface of the color filter 16, the difference in height between the upper surface of the R pattern 16R and the upper surface of the B pattern 16B where the difference in height is greatest is 500 nm.

  On the other hand, according to the color solid-state imaging device 100 of the present embodiment, when the R pattern 106R, the B pattern 106B, and the G pattern 106G each having the same thickness as the conventional color pattern are provided, The height difference is limited to the range of processing accuracy, for example, about 40 nm.

  Further, the flatness of the surface is further enhanced by the second transparent film 107 formed on the color filter 106, and as a result, the microlens 108 is formed with extremely high accuracy. For this reason, the condensing with respect to each photoelectric conversion element 102 is performed with high accuracy, and the output sensitivity in each photoelectric conversion element 102 and the uniformity of the output sensitivity for each photoelectric conversion element 102 are improved.

  Further, since the upper surfaces of the color patterns of the respective colors are aligned at the same height, color mixing is prevented. This will be described with reference to FIGS.

  In the G pattern 16 shown in FIG. 6A in which the both sides of the conventional color filter are sandwiched by the B pattern 16, color mixing occurs as described above. This is because the oblique incident light 22 with respect to the G pattern 16 passes through the end of the adjacent B pattern 16 protruding above the G pattern 16. On the other hand, in the case of the G pattern 106G in which both sides of the solid-state imaging device 100 of this embodiment shown in FIG. 2A are sandwiched by the B pattern 106B, such color mixture does not occur. This is because the upper surface of the G pattern 106G and the upper surface of the B pattern 106B adjacent to the G pattern 106G are aligned at the same height. Note that no color mixing occurs with respect to the G pattern 106G shown in FIG. 2B in which both sides are sandwiched by the R pattern 106R.

  Further, since the thickness of the second transparent film 107 may be small, the distance between the photoelectric conversion element 102 and the microlens 108 can be shortened to suppress the shading phenomenon.

  Next, a method for manufacturing the solid-state imaging device 100 will be described with reference to FIGS.

  First, as illustrated in FIG. 3A, photoelectric conversion elements 102 are arranged and formed on a silicon substrate 101, and then insulation is omitted so as to cover the silicon substrate 101 and the photoelectric conversion elements 102. A film is formed. Next, an electrode 103 made of, for example, polycrystalline silicon is formed at a position between the photoelectric conversion elements 102 on the insulating film, and then a light shielding film 104 covering the electrode 103 is formed. Any of these may be performed by a known method.

  Further, the first transparent film 105 is formed using a photosensitive material by a method such as spin coating. This covers the photoelectric conversion element 102, the light shielding film 104, and the like.

  Next, as shown in FIG. 2B, the first transparent film 105 is exposed using a gray scale mask 110. Here, the gray scale mask 110 includes a mask pattern (R mask pattern 110R) corresponding to a region where the R pattern 106R is formed, a mask pattern (B mask pattern 110B) corresponding to a region where the B pattern 106B is formed, and G And a mask pattern (G mask pattern 110G) corresponding to a region where the pattern 106G is formed, and each mask pattern has a different gray gradation. For this reason, the first transparent film 105 is exposed with an exposure amount corresponding to each gray gradation of the pattern of the gray scale mask 110, and the film thickness is reduced according to the exposure amount by a development process performed after the exposure. The photosensitive material for forming the first transparent film 105 may be either a positive type or a negative type.

  FIG. 3C shows a process chart at the time when the development process is completed. Here, in the region where the B pattern 106B is formed, the film thickness of the first transparent film 105 is greatly reduced, and the B region recess 111B, which is the deepest recess, is formed. In the region where the G pattern 106G is formed, the film thickness of the first transparent film 105 is subsequently reduced, and the G region recess 111G, which is a recess shallower than the B region recess 111B, is formed. In the present embodiment, the thickness of the first transparent film 105 is not reduced in the region where the R pattern 106R is formed.

  In this way, an uneven pattern is formed on the first transparent film 105. In addition, about the depth of an unevenness | corrugation, it sets only as needed and is set as needed.

  Next, as illustrated in FIG. 3D, the color filter 106 including the R pattern 106 </ b> R, the B pattern 106 </ b> B, and the G pattern 106 </ b> G is formed on the first transparent film 105. At this time, since the first transparent film 105 is provided with unevenness according to the thickness of each color pattern, the upper surface of the color pattern is made to be the same height and unevenness is generated on the upper surface of the color filter 106. It can be made flush.

  Thereafter, the second transparent film 107 is formed on the color filter 106, and the microlens 108 is further formed on the second transparent film 107, whereby the solid-state imaging device 100 shown in FIG. 1 is formed.

  Here, since the color filter 106 is flat, the upper surface of the second transparent film 107 can be easily flattened with extremely high accuracy, and the second transparent film 107 is made relatively thin. Can do.

  In the conventional method, in order to reduce the film thickness, the second transparent film has been formed by etching after thick coating, but such a step is not necessary. For this reason, it is possible to eliminate characteristic variations due to accuracy variations in the etching process.

  The microlens 108 is formed by, for example, applying a photosensitive resin on the second transparent film 107 and then performing exposure and development to form a microlens pattern. To obtain a lens shape.

  As described above, according to the method for forming a color solid-state imaging device of the present embodiment, since the upper surface of the color filter 106 is flat, the thin second transparent film 107 is formed so as to be accurately flattened. Can be done. In addition, since the flatness on the second transparent film 107 is high, the microlens 108 can be formed with high accuracy.

  In the present embodiment, the case of a primary color filter composed of three color patterns of red, green, and blue has been described. However, the present invention is not limited to this. For example, a complementary color filter is formed. It can also be used in some cases. The number of colors in the color pattern is not particularly limited.

  Also, the type of solid-state imaging device such as a CMOS type and a CCD type is not particularly limited.

  In the present embodiment, all the color patterns constituting the color filter 106 are one layer. However, some or all of the color patterns may have a laminated structure composed of two or more color patterns. Even in such a case, the upper surface of the color filter 106 can be made flush by providing the first transparent film 105 with irregularities corresponding to the thickness of the laminated structure.

  According to the color solid-state imaging device and the method for manufacturing the same according to the present invention, since the upper surface of the color filter is flush, color mixing, shading phenomenon, line shading failure, and the like are suppressed, and miniaturization and increase in the number of pixels can be achieved. This is effective in advanced color solid-state imaging devices.

FIG. 1 is a diagram schematically showing a cross-sectional configuration of a solid-state imaging device according to an embodiment of the present invention. 2A to 2D are views for explaining a manufacturing process of a solid-state imaging device according to an embodiment of the present invention. FIGS. 3A and 3B are diagrams illustrating that color mixing is avoided in the color filter provided in the solid-state imaging device according to the embodiment of the present invention. FIG. 4 is a diagram schematically showing a cross-sectional configuration of a conventional solid-state imaging device. FIGS. 5A to 5D are diagrams for explaining that in a conventional solid-state imaging device, light condensing on the photoelectric conversion element by the microlens is affected by the unevenness on the upper surface of the color filter. 6A and 6B are diagrams for explaining the occurrence of color mixing in a color filter provided in a conventional solid-state imaging device.

Explanation of symbols

101 Silicon substrate 102 Photoelectric conversion element 103 Electrode 104 Light shielding film 105 First transparent film 106 Color filter 106R Red color pattern (R pattern)
106G Green color pattern (G pattern)
106B Blue color pattern (B pattern)
107 Second transparent film 108 Micro lens 110 Gray scale mask 110R Red region mask pattern (R mask pattern)
110G Green area mask pattern (G mask pattern)
110B Blue area mask pattern (B mask pattern)
111G G region recess 111B B region recess

Claims (6)

A plurality of photoelectric conversion elements arranged on a substrate and converting incident light into electric charges;
A transparent film covering the substrate provided with the plurality of photoelectric conversion elements;
A color filter composed of a plurality of color patterns with different thicknesses formed on the transparent film and separating the incident light,
A color solid-state imaging device, wherein upper surfaces of the plurality of color patterns are flush with each other. In claim 1,
A color solid-state imaging device, wherein the transparent film has irregularities corresponding to different thicknesses of the plurality of color patterns, so that upper surfaces of the plurality of color patterns are flush with each other. In claim 1 or 2,
Another transparent film formed on the color filter;
A color solid-state imaging device, further comprising: a plurality of microlenses formed on the other transparent film and collecting the incident light on the plurality of photoelectric conversion devices. A step (a) of forming a plurality of photoelectric conversion elements for converting incident light into charges on a substrate;
A step (b) of forming a transparent film covering the substrate on which the plurality of photoelectric conversion elements are provided;
Forming on the transparent film a color filter composed of a plurality of color patterns having different thicknesses and color-separating the incident light, (c),
In the step (b), unevenness corresponding to different thicknesses of the plurality of color patterns is provided on the transparent film,
In the step (c), the color pattern is formed on the unevenness so that the upper surfaces thereof are flush with each other. In claim 4,
The step (b) of forming the transparent film includes
Forming a material film made of a photosensitive material on the substrate;
A step of exposing the material film with a plurality of exposure amounts according to the thickness of the plurality of color patterns by using a gray scale mask capable of controlling an exposure dose; and
Forming the transparent film having the irregularities by developing the exposed material film, and a method for producing a solid-state imaging device. In claim 4 or 5,
Forming another transparent layer on the color filter;
A method for producing a solid-state imaging device, further comprising forming a plurality of microlenses for condensing the incident light on the plurality of photoelectric conversion elements on the other transparent layer.

Images