JP2022032676A - Solid-state imaging device and manufacturing method of solid-state imaging device - Google Patents

Abstract

To provide a solid-state imaging device which can improve condensing efficiency even in a fine pixel, and having an excellent sensibility, and provide a manufacturing method of the solid-state imaging device.SOLUTION: A solid-state imaging device 1 according to an embodiment, comprises: a semiconductor substrate 10; a photoelectric conversion element 11; a color filter 100; a reflection reduction layer 40; and a microlens 200. The microlens 200 includes: a first microlens layer 20 arranged so as to be closest to the photoelectric conversion element 11 side; and a second microlens layer 21 formed so as to be laminated onto a lens surface of the first microlens layer 20. A refraction index of the first microlens layer 20 is within a range of 1.75 or larger and 2.2 or less, and each refraction index of the reflection reduction layer 40 and the second microlens layer 21 is lower than the refraction index of the first microlens layer 20.SELECTED DRAWING: Figure 1

Description

The present invention relates to a solid-state image sensor and a method for manufacturing a solid-state image sensor.

In recent years, solid-state image sensors such as CCD (Charge Coupled Device) image sensors and CMOS (Complementary Metal-Oxide-Semiconductor) image sensors mounted on digital cameras and the like have been increasing in pixel count and miniaturization, and are particularly fine. The pixel size is smaller than 1.1 μm × 1.1 μm.

The solid-state image sensor is colorized by providing a color filter paired with a plurality of photoelectric conversion elements. Further, the region (opening) in which the photoelectric conversion element provided in the solid-state image sensor contributes to photoelectric conversion depends on the size of the solid-state image sensor and the number of pixels. The opening is limited to about 20% or more and 50% or less with respect to the total area of the solid-state image sensor. Since a small opening directly leads to a decrease in sensitivity of the photoelectric conversion element, it is common to form a light-collecting microlens on the photoelectric conversion element in order to compensate for the decrease in sensitivity in a solid-state image pickup device. By condensing light with a microlens and guiding it to the light receiving portion of the photoelectric conversion element, the sensitivity of the solid-state image sensor can be improved.

Further, in recent years, a back-illuminated solid-state image sensor (BSI: Back Side Illumination) has been developed in order to improve the sensitivity and shading characteristics of fine pixels. In the back-illuminated solid-state image sensor, the opening can be 50% or more of the total area of the solid-state image sensor by not providing the multilayer metal wiring on the incident side of the light, and the incident light can be efficiently converted into a photoelectric conversion element. Can be captured. However, by using the backside illumination technique, the photoelectric conversion element exists on the front surface side where the light of the solid-state image sensor is incident. Therefore, the leaked light of another color filter adjacent to the color filter enters the photoelectric conversion element, the light absorbed inside the photoelectric conversion element enters the adjacent photoelectric conversion element, or the light is photoelectrically converted. Color mixing is likely to occur due to factors such as the flow of generated electrons to the circuit section of an adjacent photoelectric conversion element. As a countermeasure, a partition wall is formed between the color filters to block light by the partition wall, or the partition wall guides light as a waveguide, and a deep element separation structure is also formed between the photoelectric conversion elements. The flow of electrons and the light absorbed inside the photoelectric conversion element are separated (Patent Documents 1 and 2).

With the increase in the number of pixels and the miniaturization of solid-state image pickup devices, it is necessary to form a microlens in pairs with a photoelectric conversion element. Therefore, the size of the formed region of the microlens becomes small, and miniaturization of the microlens is required. Further, when the microlens is miniaturized as it is, the light condensing point by the microlens may shift or the light condensing ability may be inferior at the edge of the microlens. Therefore, there is a problem that the light cannot be efficiently collected in the photoelectric conversion element, such as the light hitting the partition wall portion between the color filters. In order to solve this, it is necessary to miniaturize the microlens and then increase the aspect ratio of the microlens, but it is difficult in the manufacturing process to control the shape to have a high aspect ratio while miniaturizing. Further, when the aspect ratio is increased, adjacent microlenses interfere with each other, so that there is a problem that miniaturization cannot be performed while maintaining the lens shape.

As a technique for solving such a problem, Patent Document 3 uses a material having a high refractive index as a material for a microlens to provide a flat space between microlenses (that is, a space in which a lens is not formed). A technique for improving the light capture efficiency even in a solid-state image sensor with a high pixel count or miniaturization by forming a microlens without providing the lens is disclosed.
Further, Patent Document 4 discloses a microlens in which a microlens is formed with a plurality of layers and the film thickness of each layer is adjusted to suppress uneven sensitivity. Further, Patent Document 5 shows the relationship between the refractive index of a microlens made of an inorganic material and the film thickness of the non-flattening layer formed between the microlens and the color filter, and suppresses the reduction of sensitivity characteristics. The technology to be used is disclosed.

Japanese Patent No. 6052353 International Publication No. 2017/073321 Japanese Unexamined Patent Publication No. 2005-019573 JP-A-2015-230896A Japanese Patent No. 6366101

However, if an inorganic material having a relatively high refractive index is used for the microlens, the incident light may be reflected at the interface between the microlens and the color filter or inside the microlens to reduce the light receiving efficiency.

The present invention has been made in view of the above points, and it is possible to improve the light collection efficiency even with fine pixels, and a method for manufacturing a solid-state image sensor and a solid-state image sensor having excellent sensitivity. The purpose is to provide.

The solid-state imaging device according to one aspect of the present invention corresponds to each of a semiconductor substrate, a plurality of photoelectric conversion elements provided on the semiconductor substrate and arranged in a matrix in a plan view, and the plurality of photoelectric conversion elements. The multi-color filters arranged in the lens are provided on the color filter layer two-dimensionally arranged in a preset regular pattern, the reflection reduction layer provided on the color filter layer, and the reflection reduction layer. A microlens layer having a plurality of lenses arranged corresponding to the plurality of color filters and the plurality of photoelectric conversion elements is provided, and the microlens layer is closest to the photoelectric conversion element side. The first microlens layer has a first microlens layer arranged therein and a second microlens layer formed so as to be laminated on the lens surface of the first microlens layer. The refractive index is in the range of 1.75 or more and 2.2 or less, and each refractive index of the reflection reducing layer and the second microlens layer is lower than the refractive index of the first microlens layer. It is characterized by.

In the method for manufacturing a solid-state image pickup lens according to one aspect of the present invention, a plurality of photoelectric conversion elements are arranged in a matrix in a plan view, and an element separation structure is provided between the plurality of photoelectric conversion elements on a semiconductor substrate. A step of forming a partition wall at a position surrounding the photoelectric conversion element, a step of forming a plurality of color filters at positions corresponding to the photoelectric conversion element surrounded by the partition wall, and the color filter and the partition wall. A step of forming a reflection-reducing layer composed of a resin layer or an inorganic material layer having a refractive index in the range of 1.55 or more and 1.8 or less on the upper part of the lens, and a refractive index on the upper part of the reflection-reducing layer. It is formed of a silicon nitride film or a silicon oxynitride film within the range of 1.75 or more and 2.2 or less, and is provided with a plurality of lenses at positions corresponding to the plurality of photoelectric conversion elements, and the film thickness is 100 nm or more and 400 nm or less. The step of forming the first microlens layer within the range of the above, and the silicon oxynitride film or oxidation having a refractive index lower than that of the first microlens layer on the lens surface of the first microlens layer. The process of forming the second microlens layer formed from the silicon film and
It is characterized by having.

According to the present invention, it is possible to improve the light collection efficiency even with fine pixels, and it is possible to provide a solid-state image sensor and a method for manufacturing a solid-state image sensor having excellent sensitivity.

It is a partial cross-sectional view which shows the structure of the solid-state image pickup device which concerns on 1st Embodiment of this invention. It is a partial plan view which shows the color filter arrangement of the solid-state image sensor which concerns on 1st Embodiment of this invention. It is a partial cross-sectional view which shows the structure of the modification of the solid-state image pickup device which concerns on 1st Embodiment of this invention. It is a partial cross-sectional view which shows the structure of the solid-state image pickup device which concerns on 2nd Embodiment of this invention. It is a partial plan view which shows one structural example of the microlens layer of the solid-state image pickup device which concerns on 3rd Embodiment of this invention. It is a partial plan view which shows one structural example of the solid-state image pickup device which concerns on 3rd Embodiment of this invention. It is a partial plan view which shows one structural example of the microlens layer of the solid-state image sensor formed by another method. It is a process sectional view which shows the manufacturing process of the solid-state image sensor which concerns on 1st Embodiment of this invention, and is the figure which shows up to the partition wall structure formation process. It is a process sectional view which shows the manufacturing process of the solid-state image sensor which concerns on 1st Embodiment of this invention, and is the figure which shows the 1st color filter forming process. It is a process sectional view which shows the manufacturing process of the solid-state image sensor which concerns on 1st Embodiment of this invention, and is the figure which shows the 2nd color filter forming process. It is a process sectional view which shows the manufacturing process of the solid-state image sensor which concerns on 1st Embodiment of this invention, and is the figure which shows the 3rd color filter forming process. It is a process sectional view which shows the manufacturing process of the solid-state image sensor which concerns on 1st Embodiment of this invention, and is the figure which shows the reflection reduction layer formation process. It is a process sectional view which shows the manufacturing process of the solid-state image sensor which concerns on 1st Embodiment of this invention, and is the figure which shows the microlens formation process. It is a process sectional view which shows the manufacturing process of the solid-state image sensor which concerns on 1st Embodiment of this invention, and is the figure which shows the microlens formation process. It is a process sectional view which shows the manufacturing process of the solid-state image sensor which concerns on Example 1 of this invention, and is the figure which shows the microlens flattening layer forming process. It is a process sectional view which shows the manufacturing process of the solid-state image sensor which concerns on Example 2 of this invention, and is the figure which shows the microlens formation process. It is a process sectional view which shows the manufacturing process of the solid-state image sensor which concerns on Example 3 of this invention, and is the figure which shows the microlens formation process. It is a process sectional view which shows the manufacturing process of the solid-state image sensor which concerns on Example 3 of this invention, and is the figure which shows the microlens formation process. It is a process sectional view which shows the manufacturing process of the solid-state image sensor which concerns on other embodiment of this invention, and is the figure which shows the process of forming a color filter by dry etching without making a color filter photosensitive. It is a process sectional view which shows the manufacturing process of the solid-state image sensor which concerns on other embodiment of this invention, and is the figure which shows the process of forming a color filter by dry etching without making a color filter photosensitive. It is a process sectional view which shows the manufacturing process of the solid-state image sensor which concerns on other embodiment of this invention, and is the figure which shows the process of forming a color filter by dry etching without making a color filter photosensitive.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. Here, the drawings are schematic, and the relationship between the thickness of each layer such as a color filter and the plane dimensions, the ratio of the thickness of each layer, and the like are different from the actual ones. Further, in the cross-sectional view showing the configuration of the color filter of the solid-state image sensor, the description is based on the known Bayer arrangement described later, but in the actual Bayer arrangement, the color filters of three or more colors are horizontally arranged as shown in the drawing described later. It may not be a line-up structure. Further, each embodiment shown below exemplifies a configuration for embodying the technical idea of the present invention, and the technical idea of the present invention describes the material, shape, structure, etc. of the constituent parts as follows. It is not specified by the thing. The technical idea of the present invention may be modified in various ways within the technical scope specified by the claims described in the claims.

Further, in the present embodiment, a structure having an element separation structure between the photoelectric conversion elements of the solid-state image sensor and a partition structure between the color filters will be described, but the present invention does not include these structures. It can also be used when forming a microlens for the structure of.

1. 1. First Embodiment (Structure of solid-state image sensor)
The configuration of the solid-state image sensor according to the first embodiment of the present invention will be described with reference to FIGS. 1 and 2.
As shown in FIG. 1, the solid-state image sensor 1 according to the present embodiment is located between the semiconductor substrate 10, a plurality of microlenses 200 arranged above the semiconductor substrate 10, and the semiconductor substrate 10 and the microlens 200. A provided color filter 100 and a partition wall 50 are provided, and a reflection reducing layer 40 is provided between the color filter 100 and the microlens 200. Further, a microlens flattening layer 300 is formed on the microlens 200.

The semiconductor substrate 10 is divided between a plurality of photoelectric conversion elements 11 arranged two-dimensionally, that is, in a matrix in a plan view, and the photoelectric conversion elements 11 so that electrons converted between the photoelectric conversion elements 11 do not coexist. It has a structure 12.
The cross-sectional view shown in FIG. 1 corresponds to the partial cross-sectional view taken along line I-II shown in FIG. Further, FIG. 2 corresponds to a cross-sectional view (partial plan view) taken along line III-IV shown in FIG.

The microlens 200 is formed of a plurality of microlens layers. Specifically, the microlens 200 is provided closest to the photoelectric conversion element 11 (that is, the first microlens layer 20 provided at the lowermost portion, and the second microlens formed above the first microlens layer 20). It is composed of layer 21.
The color filter 100 is composed of a plurality of color filters. Specifically, the color filter 100 is configured by arranging a first color filter 14, a second color filter 15, and a third color filter 16 in a predetermined regular pattern. That is, the color filter 100 has a rule pattern set in advance by the first color filter 14, the second color filter 15, and the third color filter 16 arranged corresponding to each of the plurality of photoelectric conversion elements 11. It is arranged two-dimensionally.

The partition wall 50 is formed between the first color filter 14, the second color filter 15, and the third color filter 16, respectively. That is, the partition walls 50 are formed between adjacent color filters. The partition wall 50 is formed of two layers, an inner partition wall 30 and a partition wall 31 that protects and covers the partition wall 30.
The reflection reduction layer 40 is formed between the color filter 100 and the microlens 200.

FIG. 1 illustrates a configuration in which the microlens flattening layer 300 is present on the upper part of the microlens 200, but the microlens flattening layer 300 may not be present depending on the configuration of the solid-state image sensor 1.
Further, although the partition wall 50 shows a two-layer structure of the partition wall 30 and the partition wall 31, it may be a one-layer structure or a two-layer or more structure depending on the structure of the solid-state image pickup device 1. Further, the partition wall 50 may have a height lower than or higher than that of the color filter 100, or may be flat. Further, when the pixel size of the solid-state image sensor 1 is large, the partition wall 50 may be omitted.

Hereinafter, in explaining the solid-state image sensor 1 according to the present embodiment, the color filter that is first formed in the manufacturing process of the color filter 100 and has the largest occupied area in the color filter 100 is defined as the first color filter 14. .. Further, the color filter formed second in the manufacturing process of the color filter 100 is defined as the second color filter 15, and the color filter formed third in the manufacturing process of the color filter 100 is defined as the third color filter 16. The same applies to other embodiments. Further, in the following description, it is assumed that the first color filter 14 is green, but the first color filter 14 may be blue or red.
Hereinafter, each component of the solid-state image sensor 1 will be described in detail.

(Photoelectric conversion element and semiconductor substrate)
As shown in FIG. 1, a plurality of photoelectric conversion elements 11 corresponding to pixel positions are two-dimensionally arranged on the semiconductor substrate 10 in the solid-state image pickup device 1 according to the present embodiment. Each of the photoelectric conversion elements 11 has a function of converting light into an electric signal. Further, on the semiconductor substrate 10 in which the plurality of photoelectric conversion elements 11 are arranged, the incident light enters the adjacent photoelectric conversion element 11 before the light incident on the solid-state image pickup element 1 is absorbed by the photoelectric conversion element 11. In addition, the element separation structure 12 is formed so that the photoelectrically converted electrons do not flow to the circuit of another photoelectric conversion element 11 adjacent to the photoelectric conversion element 11. As a method of element separation, for example, various methods such as a method of doping the semiconductor substrate 10 in the element separation region, a method of opening a gap, and a method of embedding a metal, an oxide, a nitride, a dielectric, or the like can be used. However, it is desirable to have a structure in which light and electrons do not easily flow into the circuit of another photoelectric conversion element 11 adjacent to the circuit. As a method of element separation, generally, a method of digging a semiconductor substrate 10 between the photoelectric conversion elements 11 by etching and then depositing a metal, an oxide, a dielectric, or the like is used. The semiconductor substrate 10 on which the photoelectric conversion element 11 is formed usually has a protective film formed on the outermost surface for the purpose of protecting and flattening the surface (light incident surface). The semiconductor substrate 10 is made of a material that allows visible light to pass through and can withstand a temperature of at least about 300 ° C. Here, examples of the heat-resistant material used for the semiconductor substrate 10 include oxides such as Si and SiO ₂ , nitrides such as SiN, and materials containing Si such as a mixture thereof. At this time, the photoelectric conversion element 11 is arranged at a position where the incident light can be received, according to the material of the microlens 200 described later, the height thereof, and the like.

(Color filter and bulkhead)
The first color filter 14, the second color filter 15, and the third color filter 16 (an example of the first, second, and third color filters) constituting the color filter 100 according to a predetermined arrangement pattern are incident light. It is a filter corresponding to each color (green, blue and red) that separates colors. As shown in FIG. 1, the first color filter 14, the second color filter 15, and the third color filter 16 are provided between the semiconductor substrate 10 and the reflection reduction layer 40. The first color filter 14, the second color filter 15, and the third color filter 16 are arranged in a predetermined regular pattern according to the pixel position so as to correspond to each of the plurality of photoelectric conversion elements 11. ing.

FIG. 2 shows a planar arrangement of a first color filter 14, a second color filter 15, a third color filter 16, and a partition wall 50 formed between each color filter constituting the color filter 100. It is a figure. The arrangement shown in FIG. 2 is a so-called Bayer arrangement, which is an arrangement in which a first color filter 14, a second color filter 15, and a third color filter 16 are spread.
In the present embodiment, each color filter (first color filter 14, second color filter 15, and third color filter 16) of the solid-state image sensor 1 is not necessarily limited to the Bayer arrangement, and each color filter. The color of is not limited to the three colors of red (R), green (G), and blue (B). Further, a transparent layer having an adjusted refractive index or a layer that shields visible light and transmits infrared rays (that is, a filter for infrared rays or the like) may be arranged as a part of the arrangement of the color filters 100. Further, four adjacent pixels may be arranged in the same color, and a plurality of pixels may be used so as to form a Bayer array with 16 pixels.

Each color filter constituting the color filter 100 contains a pigment (coloring agent) of a predetermined color and a thermosetting component or a photocuring component. For example, the first color filter 14 contains a green pigment as a colorant, the second color filter 15 contains a blue pigment, and the third color filter 16 contains a red pigment.

The partition wall 50 is formed between each of the plurality of color filters (first color filter 14, second color filter 15, and third color filter 16) constituting the color filter 100. In the present embodiment, the partition wall 50 provided on the side wall portion (outer periphery) of the first color filter 14 allows the first color filter 14, the second color filter 15, and the third color filter 16, respectively. Can be separated. As shown in FIG. 1, the partition wall 50 may have a two-layer structure composed of a partition wall 30 located inside (inside) the partition wall 50 and a partition wall 31 located so as to cover the outside of the partition wall 30. ..

The partition wall 30 is preferably formed of a metal material having high light-shielding property and which can be finely processed with high accuracy by etching or the like, for example, a film of tungsten (W), aluminum (Al), or copper (Cu). Further, the partition wall 30 may have a laminated structure made of a plurality of metal films, such as the lower layer of the partition wall 30 being formed of tungsten and the upper layer of the partition wall 30 being formed of titanium. Further, the partition wall 30 may be formed by using the above-mentioned metal compound instead of the above-mentioned metal simple substance.

The partition wall 31 is formed by using, for example, an oxide such as SiO ₂ or SiN, or an oxide such as SiON, a nitride, or an oxynitride so as to cover the outside of the partition wall 30. Further, the partition wall 31 may be formed with a structure capable of inducing light like an optical waveguide by forming the partition wall 31 using a material having a lower refractive index than the material constituting the color filter 100. Further, the partition wall 31 serves as a protective film for the partition wall 30, and when a metal is used for the partition wall 30, it is possible to prevent the metal from reacting (chemical reaction) and deteriorating. The partition wall 30 may be formed by using a plurality of materials such as metal and oxide, metal and nitride, or metal and oxynitride.
Further, the partition wall 31 is preferably formed of a material capable of blocking light entering the adjacent color filter 100, a material capable of reflecting light, or a material capable of forming an optical waveguide.

As shown in FIG. 2, the partition walls 30 may be formed continuously without a break, or the partition walls 31 may also be formed continuously without a break.
Further, as shown in FIG. 2, in a plan view, the width of the partition wall 50 may be narrowed or widened depending on the combination of adjacent color filters 100. For example, the width of the partition wall 50 located between the first color filter 14 and the second color filter 15 is the width of the partition wall 50 located between the first color filter 14 and the third color filter 16. It may be wider or narrower than. In this way, by narrowing the width of the partition wall 50 according to the combination of the adjacent color filters 100, the light receiving area in each pixel can be increased as needed, so that the sensitivity of the solid-state image sensor 1 can be adjusted. can do. Further, by widening the width of the partition wall 50 according to the combination of the adjacent color filters 100, the mechanical strength between each pixel can be increased as necessary, so that the strength of the solid-state image sensor 1 can be partially increased. Can be adjusted to.

The height of the partition wall 30 is preferably in the range of 30% or more and 100% or less with respect to the height of the entire color filter 100 or the height of the entire partition wall 50, and is within the range of 45% or more and 55% or less. If so, it is more preferable. When the height of the partition wall 30 is within the above numerical range, the light entering the adjacent color filter can be effectively shielded or reflected.
Further, as shown in FIG. 2, in a plan view, the width of the partition wall 30 is preferably within a range of 10% or more and 50% or less with respect to the width of the entire partition wall 50, and is within a range of 20% or more and 30% or less. If so, it is more preferable. When the width of the partition wall 30 is within the above numerical range, the light entering the adjacent color filter can be effectively shielded or reflected.

(Micro lens)
The microlens 200 in the solid-state image sensor 1 is arranged at a position corresponding to the pixel position above the semiconductor substrate 10. That is, in the microlens 200, a plurality of lenses are provided at positions corresponding to each of the plurality of photoelectric conversion elements 11. The microlens 200 can compensate for the decrease in sensitivity of the photoelectric conversion element 11 by condensing the incident light incident on the microlens 200 on each of the photoelectric conversion elements 11. As shown in FIG. 1, the microlens 200 of the present embodiment may be formed so that a plurality of microlens layers are laminated. Further, in the microlens 200, the first microlens layer 20 provided in the lower layer closest to the photoelectric conversion element 11 has a microlens shape, and the second microlens layer 21 and the second microlens layer 21 It is preferable that a plurality of microlens layers located on the upper layer are formed so as to cover the lens surface of the first microlens layer 20. Note that FIG. 1 shows a form in which the microlens 200 is formed of two layers, a first microlens layer 20 and a second microlens layer 21.

In order to improve the light collection efficiency of the microlens 200, it is preferable that the first microlens layer 20 is made of a high refractive index material. Specifically, the refractive index of the material constituting the first microlens layer 20 is preferably in the range of 1.75 or more and 2.2 or less, and preferably in the range of 1.85 or more and 2.0 or less. It is more preferable, and it is more preferable if it is in the range of 1.90 or more and 1.95 or less. In the case of a material having a refractive index of more than 2.2, the light-collecting ability of the microlens 200 is improved, but the difference in refractive index from the reflection-reducing layer 40 becomes large, so that the interface between the microlens 200 and the reflection-reducing layer 40 becomes large. The amount of reflected light may increase and the light collection efficiency may decrease. Further, when the material has a refractive index of less than 1.75, it becomes difficult to impart sufficient light-collecting ability to the microlens 200, and the light-collecting efficiency required for use may not be obtained. Hereinafter, in the present embodiment, "high refractive index" means that the refractive index is 1.75 or more, and "high refractive index material" means a material having a refractive index of 1.75 or more. ..
In the conventional structure of the microlens, an organic resin material is generally used as the material of the microlens 200, and the organic resin has a value of 1 even if the material has a relatively high refractive index. It is about 0.8 or less. Further, when an organic material having a relatively high refractive index is used, the transmittance tends to decrease.

In the case of a general camera module in which a fine solid-state image sensor 1 is used, the microlens 200 is in contact with air (air layer) having a refractive index close to 1 on the outside where light is incident. When the refractive index of the microlens 200 is 1.75 or more, the light collecting ability of the microlens 200 is improved, but the difference in the refractive index between the microlens 200 and the air (air layer) is large, so that light is reflected. It is easy to happen. In this case, the amount of light transmitted (passed) inside the microlens 200 may decrease, resulting in a decrease in light receiving sensitivity. In that case, the microlens flattening layer 300, which will be described later, may be provided.

Inorganic materials with a high refractive index and high transmission of visible light from 380 nm to 700 nm include, for example, silicon nitride (refractive index about 2.0), zirconium oxide (refractive index about 2.2), and titanium oxide (refraction). Rate about 2.49), zinc sulfide (refractive index: about 2.35), zinc oxide (refractive index: about 2.01), hafnium oxide (refractive index: about 1.91), aluminum nitride (refractive index about, 2.16), tantalum oxide (refractive index about 2.16) and the like can be considered. When forming the microlens 200, since the heat resistant temperature of the reflection reducing layer 40 and the color filter 100 provided under the microlens 200 is generally 300 ° C. or lower, a material that can be formed at this temperature or lower should be used. Is preferable. That is, the material for forming the microlens 200 can be formed at a temperature of 300 ° C. or lower, shape processing is easy by a known method such as dry etching, and any material having a low extinction coefficient, which will be described later, has a problem. However, considering the formation temperature and the reflection due to the above-mentioned difference in refractive index, it is preferable to use silicon nitride as the inorganic material. By using silicon nitride, which is transparent and exhibits a high refractive index, as an inorganic material, the first microlens layer 20 becomes a microlens having a high refractive index. Since the first microlens layer 20 exhibits a high refractive index, even if the height of the first microlens layer 20 is formed lower than the conventional one, the incident light is provided corresponding to each lens. It can be incident on each of the photoelectric conversion elements 11). Further, since the height of the first microlens layer 20 can be formed lower than the conventional one, the incident light to be incident on the adjacent pixel (photoelectric conversion element 11) is blocked by the first microlens layer 20. Therefore, it is possible to suppress a decrease in the light receiving efficiency of the solid-state image sensor 1 as a whole. Therefore, it is possible to provide the solid-state image pickup device 1 that exhibits high light-collecting characteristics without impairing the sensitivity characteristics of each color.

The refractive index of silicon nitride (Si ₃ N ₄ ) is generally about 2.0, but the ratio of silicon and nitrogen changes depending on the film forming conditions, and the refractive index changes from 1.7 to 2.2. .. Further, when silicon nitride is used as the material of the microlens 200, the transmittance needs to be high. Since the transmittance changes depending on the film thickness of the material, it can be indicated by the extinction coefficient of the optical constant as an index. The inventors have found that the extinction coefficient is important as a material used for the microlens 200. There is a difference between the refractive index and the extinction coefficient depending on the film formation conditions. Silicon nitride used in the solid-state image sensor 1 is often formed by using CVD (Chemical Vapor Deposition). The quality of the film forming conditions when forming using CVD differs depending on the control of temperature, pressure, gas type, gas flow rate, and the like. When the heat resistant temperature of the color filter 100 provided in the lower layer of the microlens 200 is 300 ° C. or lower, plasma CVD (Plasma-Plasma-) can be used in a low temperature environment using a plasma source under the above conditions. By using enhanced Chemical Vapor Deposition), silicon nitride can be formed with good quality. Under film formation conditions where the refractive index of silicon nitride is high, the extinction coefficient tends to deteriorate, and under film formation conditions where both the refractive index and extinction coefficient are excellent, the range of conditions is narrow and the stress contained in the film is high. It tends to be expensive. When the stress is high, it tends to be unsuitable for forming the microlens 200 on the color filter 100. Therefore, the silicon nitride film used for the microlens 200 (for example, the first microlens layer 20) has a refractive index in the range of 1.75 or more and 2.0 or less, and visible light having a wavelength of 380 nm or more and 700 nm or less. It is more preferable that the extinction coefficient in the range of 1.0 × 10 ^-3 or less, the refractive index is in the range of 1.85 or more and 2.0 or less, and the extinction in the wavelength range of 400 nm or more and 700 nm or less. Most preferably, the coefficient is 1.0 × 10 ^-3 or less.

The above-mentioned condition is the case where the silicon nitride film is formed, but the silicon nitride (composition formula SiON) film is formed by using N ₂ O gas containing oxygen when forming by plasma CVD. be able to. Silicon oxynitride can obtain a refractive index from silicon nitride (refractive index: 2.0) to silicon oxide (refractive index: 1.45) depending on the oxygen content, and the extinction coefficient can be easily controlled. Therefore, silicon oxynitride may be used as the material of the first microlens layer 20.

As a method for forming the first microlens layer 20, a first layer having a high refractive index is formed on the color filter 100 under various film forming conditions. As a method for forming the first layer, a method for forming the above-mentioned silicon nitride film by plasma CVD is preferable. Next, an intermediate sacrificial layer, which is a second layer for forming the lens matrix described later, is formed. The intermediate sacrificial layer is formed of a resin material, and for example, a photosensitive resin having no heat flow property (that is, a photosensitive resin having no heat flow property) is used. As a result, it is possible to prevent the photosensitive resin pattern from melting due to the heat flow, expanding the volume, and causing the adjacent lenses to come into contact with each other. As a result, it is possible to prevent the occurrence of shape collapse at the boundary portion between adjacent lenses.

Here, as the heat-flow-free photosensitive resin that can be used for forming the lens matrix, a thermoplastic resin that has a high glass transition temperature and does not lose its shape before curing by heat treatment under the conditions of 100 ° C to 220 ° C. The material is suitable. Such a photosensitive resin that does not flow heat contains a base resin having a mass average molecular weight (Mw: a value measured by gel permeation chromatography (GPC) in terms of styrene) in the range of 10,000 or more and 30,000 or less. Is preferable. More preferably, the mass average molecular weight of the base resin is in the range of 20,000 or more and 30,000 or less. When the mass average molecular weight of the base resin is 10,000 or more, both heat resistance and heat flow resistance are improved. Further, by setting the mass average molecular weight of the base resin to 30,000 or less, the solubility at the time of development does not decrease, so that the generation of residue can be suppressed.

Subsequently, a photosensitive photoresist is formed on the intermediate sacrificial layer in order to form a lens matrix. By selectively exposing and developing this photoresist, a photoresist is formed at a place where the microlens 200 is formed. Then, by heating this photoresist, the resist pattern is thermally flowed to form a lens shape.
Next, using the lens-shaped resist pattern as a mask, dry etching is performed on the entire surface of the intermediate sacrificial layer, and etching is performed at a time setting that allows the entire amount of the positive resist pattern to be scraped. To form.
Subsequently, dry etching is performed on the entire surface of the first layer using the lens matrix as a mask, and the shape of the lens matrix is transferred to the first layer to form the first microlens layer 20.
As a lens forming method for the microlens 200, a known method can be used in addition to the method using the heat flow method of the resist pattern described above. For example, by adjusting the exposure amount using a photomask having a gradation pattern, the first layer may be formed so as to have a lens shape directly after development.

At this time, the film thickness of the first microlens layer 20 in the height direction is preferably in the range of 100 nm or more and 400 nm or less. The "thickness in the height direction of the first microlens layer 20" means the first microlens layer 20 from the bottom surface of the semicircular lens (the upper surface of the reflection reduction layer 40 in the first embodiment). The height to the top of the lens. When the pitch of each photoelectric conversion element 11 (that is, the pitch of a plurality of lenses included in the first microlens layer 20) has a fine structure of 1.0 μm or less, the film thickness of the first microlens layer 20 in the height direction is , 100 nm or more and 400 nm or less is preferable, 150 nm or more and 300 nm or less is more preferable, and 180 nm or more and 220 nm or less is more preferable.

When the height of the first microlens layer 20 becomes high in a fine pixel pattern, the incident light to be incident on the adjacent pixel (photoelectric conversion element 11) is blocked by the first microlens layer 20 for solid-state imaging. There is a problem that the light receiving efficiency (light receiving sensitivity) of the element 1 is lowered. Further, if the film thickness (lens height) of the first microlens layer 20 is increased under the condition that the area of the formed portion of the microlens 200 is narrow, the film thickness of the portion corresponding to the valley portion of the lens becomes thick, and the lens shape. There is also a problem that a region is formed in which the lens collapses and the light is mixed. However, by setting the film thickness range of the first microlens layer 20 in the height direction within the above numerical range, these problems can be suppressed, and the solid-state image pickup device 1 exhibiting high light-collecting characteristics is provided. be able to.
Further, by lowering the height of the first microlens layer 20, the light collected by the photoelectric conversion element 11 increases, and the light receiving efficiency is improved. In particular, by setting the height of the first microlens layer 20 within the range of 100 nm or more and 400 nm or less, good sensitivity characteristics are obtained for all colors of green (G), blue (B), and red (R). be able to.

The second microlens layer 21 of the microlens 200 is a layer that reduces reflection by the first microlens layer 20. As described above, when the refractive index of the first microlens layer 20 is in the range of 1.75 or more and 2.2 or less, the air (air layer) having a low refractive index or the microlens flattening layer 300 is a microlens. Since it is located outside the 200, the amount of light reflected on the surface of the first microlens layer 20 increases, and the sensitivity tends to decrease. Therefore, the second microlens layer 21 of the microlens 200 composed of a plurality of layers and the microlens layer above the second microlens layer 21 control the refractive index to form an antireflection layer. That is, by providing the second microlens layer 21 having a refractive index lower than that of the first microlens layer 20 on the lens surface of the first microlens layer 20, the sensitivity characteristics of the solid-state image sensor 1 are impaired. Can be prevented. The refractive index of the second microlens layer 21 is preferably in the range of 1.4 or more and 1.75 or less, more preferably in the range of 1.4 or more and 1.60 or less, and 1.45 or more and 1 It is more preferable if it is within the range of .55 or less. If the refractive index of the second microlens layer 21 is within the above numerical range, the microlens is in contact with the air (air layer) or the microlens flattening layer 300 and the air (air layer) or the microlens flattening layer 300. Since the difference in refractive index from 200 is small, the amount of reflected incident light is reduced and the light collection efficiency is improved.

In the present embodiment, as shown in FIG. 1, when the microlens 200 has a two-layer structure, the material of the second microlens layer 21 is silicon oxide (SiO ₂ ) or an acid having a large amount of oxygen. Silicon dioxide (SiON) is suitable. That is, the material of the second microlens layer 21 preferably has a higher oxygen content than the material of the first microlens layer 20. The oxygen content in the material of the second microlens layer 21 is preferably in the range of 1.1 times or more and 10 times or less the oxygen content in the material of the first microlens layer 20. It is more preferable if it is in the range of 5 times or more and 5 times or less, and further preferable if it is in the range of 2 times or more and 4 times or less. When the oxygen content in the material of the second microlens layer 21 is within the above numerical range, the function as an antireflection layer can be reliably imparted to the second microlens layer 21.

When silicon oxide is used as the material of the second microlens layer 21, silicon oxide is used by using a coating type material such as SOG (Spin On Glass) or siloxane in addition to the vapor phase deposition method such as plasma CVD. It may be formed.
Further, any known method may be used as long as the second microlens layer 21 can be formed at a temperature of 300 ° C. or lower, which is the heat resistant temperature of the color filter 100 under the microlens 200 described above.

The film thickness of the second microlens layer 21 is preferably in the range of 5 nm or more and 200 nm or less, which is thinner than the film thickness of the first microlens layer 20, and more preferably in the range of 50 nm or more and 150 nm or less. It is preferable, and it is more preferable if it is in the range of 80 nm or more and 100 nm or less. When the height of the second microlens layer 21 becomes high in a fine pixel pattern, the incident light to be incident on the adjacent pixel (photoelectric conversion element 11) is blocked by the second microlens layer 21, and solid-state imaging is performed. There is a problem that the light receiving efficiency (light receiving sensitivity) of the element 1 is lowered. Further, if the film thickness (lens height) of the second microlens layer 21 is increased under the condition that the area of the formed portion of the microlens 200 is narrow, the film thickness of the portion corresponding to the valley portion of the lens becomes thick, and the lens shape. There is also a problem that a region is formed in which the lens collapses and the light is mixed. However, by setting the film thickness range of the second microlens layer 21 in the height direction within the above numerical range, these problems can be suppressed, and the solid-state image pickup device 1 exhibiting high light-collecting characteristics is provided. be able to.

Further, the ratio (T2 / T1) of the film thickness T2 of the second microlens layer 21 to the film thickness T1 of the first microlens layer 20 is preferably in the range of 0.125 or more and 1.0 or less. .. When the ratio of the film thickness T2 of the second microlens layer 21 to the film thickness T1 of the first microlens layer 20 is within the above numerical range, the above-mentioned problem can be suppressed and high light collection characteristics are exhibited. A solid-state image sensor 1 can be provided.

(Microlens flattening layer)
In the present embodiment, a microlens flattening layer 300 provided on the lens surface of the microlens 200 and having a flat upper surface is formed on the upper portion of the microlens 200 (in the incident direction of light). In the case of a general solid-state image sensor, the outside of the microlens 200 is an air layer, and as a camera module structure, a module structure such as a condenser lens and an infrared cut plate is formed on the upper portion thereof.

In the case of the solid-state image sensor 1 which is miniaturized so that one pixel size is 1 μm or less as shown in the present embodiment, the structure of the microlens 200 is also miniaturized, and each lens constituting the microlens 200 is also miniaturized. The distance between the lenses is close to the wavelength of light. In such a configuration, the structure of the microlens itself is likely to have an effect on incident light, such as a Fresnel lens or a diffractive lens. Therefore, it is suitable to form the microlens flattening layer 300 on the upper part of the microlens 200 even in the case of a general solid-state image sensor whose air layer is normally on the outside.

As the material of the microlens flattening layer 300, an organic material or an inorganic material can be considered, but a material in consideration of the refractive index is desirable. In the case of a general module in which the upper part of the microlens 200 is air (air layer), the material forming the microlens flattening layer 300 is located on the refractive index of air and the surface of the microlens 200. It is desirable that the material has a refractive index between that of the second microlens layer 21 and that of the second microlens layer 21. Specifically, the refractive index of the microlens flattening layer 300 is preferably in the range of 1.1 or more and 1.6 or less, and more preferably in the range of 1.20 or more and 1.45 or less. It is preferable, and it is more preferable if it is in the range of 1.30 or more and 1.40 or less. Further, the material forming the microlens flattening layer 300 is more preferably a material having a refractive index within the above numerical range and a high transmittance with respect to visible light. As a result, it is possible to suppress the light incident through the microlens flattening layer 300 from being reflected on the surface of the microlens 200, and to obtain a fine solid-state image sensor while improving the light collection characteristics and the sensitivity characteristics. Here, the above-mentioned "material having a high transmittance with respect to visible light" specifically refers to a material having a transmittance of 90% or more in a visible light region having a wavelength of 380 nm or more and 700 nm or less.

The film thickness of the microlens flattening layer 300 is not particularly limited as long as it is within the range of 100 μm or less, but a film thickness capable of flattening the microlens 200 is desirable. That is, the film thickness of the microlens flattening layer 300 does not matter as long as it can flatten the uneven shape of the microlens 200.
The material constituting the microlens flattening layer 300 may be a material having a relatively low refractive index. For example, an organic resin containing a siloxane polymer, silica, or a fluoropolymer as an inorganic polymer has a refractive index. May be a material having a refractive index of 1.6 or less, or a material having a refractive index of 1.6 or less by containing a hollow filler in an organic resin. Not limited to the above method, there is no problem as a material for the microlens flattening layer 300 as long as it is a combination of materials having high transparency and can be flattened.

(Reflection reduction layer)
The reflection reduction layer 40 is provided between the first microlens layer 20 and the color filter 100, and is a layer that reduces the reflection of incident light due to the difference in refractive index between the first microlens layer 20 and the color filter 100. .. In the present embodiment, the refractive index of the first microlens layer 20 is in the range of 1.75 or more and 2.2 or less, and the refractive index of each color filter is different for each color and is 1.4 at the wavelength transmitted through each color. It is within the range of 1.75 or less. If there is a large difference in refractive index between the color filter 100 and the first microlens layer 20, reflection of incident light is likely to occur at the interface between the color filter 100 and the first microlens layer 20, and the incident light is incident. Light may be lost and sensitivity may decrease. Therefore, the refractive index of the reflection reduction layer 40 is preferably in the range of 1.55 or more and 1.8 or less, which is a value close to the refractive indexes of the first microlens layer 20 and the color filter, and 1.65. It is more preferable if it is within the range of 1.75 or less, and further preferable if it is within the range of 1.65 or more and 1.70 or less. When the refractive index of the reflection reduction layer 40 is in the range of 1.55 or more and 1.8 or less, the difference in the refractive index between the reflection reduction layer 40 and the first microlens layer 20 becomes small, and the reflection reduction layer 40 Since the difference in refractive index between the light and the color filter 100 is small, the loss of incident light can be reduced and the light collection performance can be improved.

Further, the refractive index of the reflection reduction layer 40 is preferably higher than that of the second microlens layer 21, and specifically, 1.1 times or more the refractive index of the second microlens layer 21. It is more preferable if it is within the range of 3 times or less. The refractive index of the reflection reduction layer 40 is within the range of 1.55 or more and 1.8 or less, which is a value close to the refractive index of each of the first microlens layer 20 and the color filter, and the second microlens layer 21. When it is higher than the refractive index of, the difference in the refractive index between the layers is the smallest, so that the loss of incident light can be reduced and the light collection performance can be improved. That is, the refractive index of the reflection reduction layer 40 may be higher than the refractive index of the second microlens layer 21 and lower than the refractive index of the first microlens layer 20.

Further, the film thickness of the reflection reduction layer 40 is formed by each color filter (first color filter 14, second color filter 15, third color filter 16) forming the color filter 100 and the partition wall 50. The film thickness may be such that the steps can be flattened. By flattening the step, when the first microlens layer 20 is formed, the microlens layer can be formed without a step, and the microlens 200 with less variation can be formed.
Examples of the material of the reflection reduction layer 40 include acrylic resin, epoxy resin, polyimide resin, phenol novolac resin, polyester resin, urethane resin, melamine resin, urea resin, styrene resin and silicon resin. It is formed of a resin containing at least one of the resins and the like.

In addition to the above-mentioned organic compounds, the reflection reducing layer 40 may be composed of, for example, silicon, carbon, oxygen, hydrogen, tin, zinc, indium, aluminum, gallium, titanium, molybdenum, tungsten, niobium, tantalum, hafnium, silver and fluorine. It may be formed of a compound containing at least one of these, an oxidation compound, a nitride compound, or an acid nitride compound. As the compound of these materials, for example, ITO, ZnO, TiO ₂ , HfO ₂ , or the like can be used. Further, the reflection reduction layer 40 may be formed of a single layer or multiple layers by these materials.

In the present embodiment, as described above, the reflection reduction layer 40 may be formed by using an organic resin as the material of the reflection reduction layer 40. When silicon oxynitride is used as the material of the reflection reduction layer 40, the refractive index and the transmittance of the reflection reduction layer 40 may be adjusted by adjusting the ratio of oxygen in the silicon oxynitride. Increasing the proportion of oxygen in silicon oxynitride lowers the refractive index and improves the transmittance. The transmittance of the reflection reduction layer 40 is preferably 90% or more in the visible light region having a wavelength of 380 nm or more and 700 nm or less.

The film thickness of the reflection reduction layer 40 is preferably in the range of 10 nm or more and 300 nm or less. Further, since the reflection reduction layer 40 is preferably thin from the viewpoint of miniaturization of the solid-state image sensor 1 and suppression of color mixing, the film thickness of the reflection reduction layer 40 is more preferably within the range of 10 nm or more and 200 nm or less, more preferably 50 nm. It is more preferably in the range of 200 nm or more, and most preferably in the range of 80 nm or more and 150 nm or less. When the film thickness of the reflection reduction layer 40 is within the above numerical range, color mixing in the miniaturized solid-state image sensor 1 can be reliably suppressed, and a step generated between the color filter 100 and the partition wall 50 can be reliably suppressed. Can be flattened to.

Further, the reflection reduction layer 40 of the present embodiment also has the purpose of alleviating the difference in stress and the coefficient of thermal expansion between the color filter 100 and the first microlens layer 20, and is a thin film within a range in which this purpose can be achieved. Is preferable. It is not necessary to flatten the step generated between the color filter 100 and the partition wall 50, and only for the purpose of alleviating the difference in stress and thermal expansion coefficient between the color filter 100 and the first microlens layer 20. If so, the film thickness of the reflection reduction layer 40 may be in the range of 10 nm or more and 20 nm or less.
Further, the film thickness of the reflection reduction layer 40 may be thinner than the film thickness of the color filter 100. When the film thickness of the reflection reduction layer 40 is thinner than the film thickness of the color filter 100, the film thickness of the reflection reduction layer 40 is 0.05 times or more and 0.5 times or less the film thickness of the color filter 100. It is preferably within the range. By setting the film thickness of the reflection reduction layer 40 within the above numerical range, the solid-state image sensor 1 can be easily reduced in height.

<Modification example>
Hereinafter, a modified example of the first embodiment of the present invention will be described with reference to FIG. As shown in FIG. 3, the solid-state image sensor 1 according to the present embodiment has the microlens flattening layer 300, but the present invention is not limited to such a configuration.
For example, as shown in FIG. 3, the solid-state image sensor 1A according to the modified example of the first embodiment includes a semiconductor substrate 10, a color filter 100, a reflection reduction layer 40, a microlens 200, and a partition wall 50. ing. That is, it differs from the solid-state image sensor 1 according to the first embodiment in that it does not have the microlens flattening layer 300.
Since the semiconductor substrate 10, the color filter 100, the reflection reduction layer 40, the microlens 200, and the partition wall 50 have the same configuration as each part described in the first embodiment, the description thereof will be omitted.

<Effect of the first embodiment>
In the solid-state image sensor 1 of the present embodiment, the reflection reduction layer 40 is provided between the color filter 100 and the microlens 200. Therefore, it is possible to reduce internal reflection due to the difference in refractive index between the first microlens layer 20 and the color filter 100. Further, since the refractive index of the first microlens layer 20 is relatively high, 1.75 or more and 2.2 or less, the lens height (height from the bottom surface of the lens to the apex of the lens) is relative to 100 nm or more and 400 nm or less. Even if it is low, the lens has a high light-collecting ability. Further, the second microlens layer 21 can reduce the surface reflection of the first microlens layer 20.

Further, the reflection reduction layer 40 has the advantage of relaxing stress from other layers (for example, the color filter 100 and the first microlens layer 20) and widening the range of formation conditions of the first microlens layer 20. There is. Specifically, when using a film forming apparatus such as plasma CVD, there is an advantage that the film forming temperature can be raised. Since the stress contained in the inorganic material or the like used for forming the first microlens layer 20 tends to change depending on the film formation temperature, the stress or the like is relaxed by providing the reflection reducing layer 40 as described above. The effect is obtained. Therefore, even if the film formation temperature fluctuates slightly from the set temperature, there is an effect that a quality problem does not occur.

Further, the reflection reduction layer 40 can be flattened when a step is generated between the color filter 100 and the partition wall 50, and the microlens 200 can be suitably formed.
The reflection reduction layer 40 also functions as a stopper layer when the microlens 200 is formed by dry etching.
As a result, the light collection efficiency of the microlens 200 is improved, and a highly sensitive solid-state image sensor 1 can be formed.

In the present embodiment, the description is based on the configuration of only a normal photoelectric conversion element, but the present invention is not limited thereto. In recent years, some solid-state image pickup devices have an image plane phase difference autofocus (Phase Detection AF) formed by changing the structure of a part of the photoelectric conversion element portion for focus detection. In the image plane phase difference autofocus, the amount of deviation from the signal is obtained by using a plurality of focus detection pixels in the image sensor, and the amount of focus correction is calculated. In the case of such a structure, the arrangement may be partially changed in the image pickup device, or a plurality of pixels may be formed as one pixel. The image plane phase difference autofocus pixels have various configurations such as one pixel, two pixels, and four pixels. In that case, the color filter or microlens corresponding to the photoelectric conversion element is not necessarily the photoelectric conversion element. Is not a pair. For example, there is a configuration in which a color filter and a microlens are used for a plurality of photoelectric conversion elements, and a configuration in which a microlens is provided for a plurality of photoelectric conversion elements and a color filter. Further, the partition wall 50 is formed according to these configurations. This embodiment can be applied even in the case of a solid-state image sensor in which image plane phase difference autofocus is formed, and the microlens in that portion is not necessarily paired with the photoelectric conversion element, only the shape is slightly different. , Other parts can be the same as in the embodiment.

2. 2. Second Embodiment Next, the second embodiment of the present invention will be described with reference to FIG. As shown in FIG. 4, in the solid-state image sensor 2 according to the present embodiment, the microlens 200 has a third microlens layer 22 provided on the upper surface of the second microlens layer 21. , It is different from the solid-state image sensor 1 according to the first embodiment.
Hereinafter, the configuration of the solid-state image sensor 2 will be described differently from that of the solid-state image sensor 1 according to the first embodiment.

(Micro lens)
The microlens 200 has a first microlens layer 20, a second microlens layer 21, and a third microlens layer 22. The first microlens layer 20 and the second microlens layer 21 are the same as the first microlens layer 20 and the second microlens layer 21 of the solid-state image pickup device 1 according to the first embodiment.
The third microlens layer 22 is a layer that reduces reflection by the second microlens layer 21, and has a lower refractive index than the second microlens layer 21. The third microlens layer 22 preferably has a higher oxygen content than the second microlens layer 21. For example, the refractive index of the third microlens layer 22 may be in the range of 1.2 or more and 1.5 or less, and more preferably in the range of 1.3 or more and 1.5 or less. Within the above numerical range, the third microlens layer 22 can surely reduce the reflection by the second microlens layer 21. Further, for example, the oxygen content of the third microlens layer 22 may be in the range of 1.1 times or more and 10 times or less of the oxygen content of the second microlens layer 21. It is more preferable if it is within the range of double or less. Within the above numerical range, the third microlens layer 22 can surely reduce the reflection by the second microlens layer 21.

In the present embodiment, the microlens 200 has a configuration of three layers, but it does not matter how many layers are provided as long as the refractive index tends to decrease from the lower layer close to the photoelectric conversion element 11 to the upper layer. do not have.
The thickness of the third microlens layer 22 may be in the range of 0.5 times or more and 2 times or less of the thickness of the second microlens layer 21, and may be in the range of 0.5 times or more and 1 times or less. It is more preferable if it is inside. Within the above numerical range, it is possible to surely reduce the reflection of the incident light by the second microlens layer 21 while maintaining the low profile of the entire microlens 200.

(How to form a microlens)
The process up to the step of forming the first microlens layer 20 is the same as that of the first embodiment.
Next, the second microlens layer 21 is formed. The second microlens layer is formed so that its refractive index is in the range of 1.5 or more and 1.75 or less.
Next, the third microlens layer 22 is formed. The third microlens layer 22 is formed so that its refractive index is in the range of 1.3 or more and 1.5 or less. In this way, the refractive index is gradually lowered from the inside to the outside of the microlens 200 to reduce the difference in the refractive index from the air, thereby reducing the reflection on the surface of the microlens 200.

The second microlens layer 21 and the microlens layer above the second microlens layer 21 may be formed layer by layer or may be formed continuously. For example, silicon oxynitride formed by plasma CVD may be used to form a microlens layer above the second microlens layer 21 and the second microlens layer 21. Specifically, the above-mentioned silicon oxynitride is formed under the condition that the ratio of nitrogen is high at the initial stage, and is gradually changed to the condition where the ratio of oxygen is high in the middle, thereby forming the refractive index of silicon oxynitride. May be controlled. As a method of changing to a condition having a large proportion of oxygen on the way, for example, a method of changing the flow rate of nitrogen gas or oxygen gas can be mentioned.

Further, by increasing the film thickness of the third microlens layer 22, its refractive index and its shape can be changed. By flattening the upper surface of the third microlens layer 22, the formation of the microlens flattening layer 300, which is a subsequent step, can be omitted, and the manufacturing process of the solid-state image sensor can be facilitated. Become.
When the effect of the microlens flattening layer 300 is imparted to the microlens layer above the second microlens layer 21 and the second microlens layer 21, for example, the refractive index is 1.2 or more and 1.5 or less. A low refractive index material within the range may be used to form a microlens layer above the second microlens layer 21 and the second microlens layer 21.

<Effect of the second embodiment>
By using the second embodiment, it is possible to reduce the reflection of the incident light on the surface of the high refractive index microlens and improve the light collecting ability.

3. 3. Third Embodiment Next, a third embodiment of the present invention will be described with reference to FIGS. 5 to 7. FIG. 5 is a plan view showing a configuration example of the solid-state image sensor 3 according to the present embodiment. The solid-state image sensor 3 includes at least a semiconductor substrate 10, a color filter 100, a microlens 200A, a partition wall 50, and a reflection reduction layer 40. The solid-state image sensor 3 is different from the solid-state image sensor 1 of the first embodiment and the solid-state image sensor 2 of the second embodiment in that the microlens 200A is provided instead of the microlens 200.
Hereinafter, the microlens 200A will be described in detail. Since the semiconductor substrate 10, the color filter 100, the partition wall 50, and the reflection reduction layer 40 have the same configuration as each part described in the first embodiment and the second embodiment, the description thereof will be omitted.

(Micro lens)
Like the microlens 200, the microlens 200A is formed of a plurality of (for example, two layers) microlens layers. In the present embodiment, the microlens 200A is provided on the lens surface of the first microlens layer 20A provided on the lower layer closest to the photoelectric conversion element 11 and the second micro lens layer 20A. A case where the lens layer 21A is provided will be described. The cross-sectional structure of the first microlens layer 20A and the second microlens layer 21A is the same as that of the first microlens layer 20 and the second microlens layer 21 of the microlens 200, and FIG. 5 described later. The first microlens layer 20A and the second microlens layer 21A are not shown.

The microlens 200A has a plurality of lenses corresponding to the plurality of photoelectric conversion elements 11. As shown in FIG. 5, in the microlens 200A, a plurality of adjacent lenses are in contact with each other in a plan view. In the microlens 200A, a plurality of adjacent lenses are in linear contact with each other in a plan view, but the lenses do not necessarily have to be in linear contact with each other. Further, the microlens 200A has gaps on the corners (four corners) of the first color filter 14, the second color filter 15, and the third color filter 16 constituting the color filter 100. Each lens may have a shape that does not cover the corner portion. Further, the microlens 200A is preferably 90% or more and 100% or less if the "Fill Factor" indicating the ratio of the lens covering one pixel having the photoelectric conversion element 11 is within the range of 85% or more and 100% or less. It is more preferable if it is within the range. When the Fill Factor is in the range of 85% or more and 100% or less, the deterioration of the light collection characteristics due to the gap between adjacent lenses is suppressed, and any of the colors of green (G), blue (B) or red (R) is suppressed. Is also preferable because the light receiving sensitivity is improved.

Hereinafter, the Fill Factor will be described with reference to FIG. FIG. 6 is a plan view schematically showing the planar configuration of the four pixels and the four lenses provided on each pixel P. The Fill Factor is defined by the following equation (1).
Fill Factor [%] = {1- (a × a) / (A × A)} × 100 ・ ・ ・ (1)
Here, A indicates the pixel size, and a indicates the diagonal distance of the gap G generated in the lens, and a <A. Further, the Fill Factor indicates the ratio of the lens covering one pixel, which is 100% at the maximum.

The microlens 200A as described above can be formed by using a method for forming a microlens via lens masters 20b and 20e, which will be described later.
As shown in FIG. 7, when the microlens (200') is formed without using the lens masters 20b and 20e described later, the gap G between the lenses of the microlens 20'is large on each pixel P. Become. However, since the microlens having a high refraction coefficient collects the incident light by the lens having a high refraction coefficient as compared with the conventional microlens, the decrease in the light collecting ability is small even if the gap G between the lenses is large.

<Effect of the third embodiment>
In the present embodiment, the light collection characteristics can be improved by providing the microlens 200A in which the Fill Factor is in the range of 85% or more and 100% or less. The influence of the sensitivity characteristic due to the gap G of the microlens 200A is more affected when the lens height of the microlens 200A (first microlens layer 20) is lowered. Therefore, when the film thickness of the first microlens layer 20 is in the range of 100 nm or more and 400 nm or less, more preferably in the range of 100 nm or more and 200 nm or less, the gap G between the lenses is reduced to make it finer. Moreover, it is possible to provide a solid-state image sensor 3 with higher sensitivity.

[Example]
Hereinafter, each configuration of the solid-state image pickup device and the conventional solid-state image pickup device according to the present invention will be specifically described.
<Example 1>
In the first embodiment, the same manufacturing method as the solid-state image sensor 1 (see FIG. 1) according to the first embodiment described above will be described.
First, as shown in FIGS. 8A to 8C, the photoelectric conversion elements 11 arranged two-dimensionally are provided, and the photoelectric conversion elements 11 are separated by the element separation structure 12. A semiconductor substrate 10 was prepared (FIG. 8A). As the semiconductor substrate 10, a fine semiconductor device 11 having a pattern size of 1 μm or less was used.

(Formation of partition wall)
Next, a partition wall 50 was formed on the semiconductor substrate 10 (FIGS. 8 (b) and 8 (c)). First, a tungsten film was formed on the semiconductor substrate 10 by plasma CVD to a film thickness of 350 nm. Next, a positive resist (TDMR-AR: manufactured by Tokyo Ohka Kogyo Co., Ltd.) was spin-coated on the tungsten film at a rotation speed of 1000 rpm using a spin coater, and then prebaked at 90 ° C. for 1 minute. .. As a result, a sample coated with a photoresist, which is a photosensitive resin mask material layer (etching mask), with a film thickness of 1.5 μm was prepared. In the positive resist, which is the photosensitive resin mask material layer, the ultraviolet-irradiated portion causes a chemical reaction and dissolves in the developing solution when irradiated with ultraviolet rays.
Photolithography was performed on this sample to expose it through a photomask. As the exposure apparatus, an exposure apparatus using an i-line wavelength as a light source was used. In photolithography, the portion of the photosensitive resin mask material layer that forms the color filter was irradiated with ultraviolet light.

Next, a developing step was carried out using 2.38% by mass of TMAH (tetramethylammonium hydride) as a developing solution to form a photosensitive resin mask material layer having an opening at a place where a color filter was formed. When a positive resist is used as the photosensitive resin mask material layer, dehydration baking is often performed after development to cure the photoresist which is the photosensitive resin mask material layer. This time, a dehydration bake was carried out at a temperature of 120 ° C. The photosensitive resin mask material layer was formed with a film thickness of 1.5 μm.
Next, the tungsten film was dry-etched using the photosensitive resin mask material layer as a mask. At this time, a parallel plate type dry etching apparatus was used. Further, the etching conditions were changed in the middle so as not to affect the underlying semiconductor substrate 10, and the tungsten film was dry-etched in multiple stages.

Hereinafter, the dry etching performed in this example will be described.
First, the gas type used for dry etching will be described. In this example, etching was performed using an etching gas in which two types of SF ₆ and Ar were mixed. The gas flow rate of SF ₆ was 50 ml / min, and the gas flow rate of Ar was 100 ml / min. Further, the pressure in the chamber at this time was set to a pressure of 2 Pa, and the RF power was set to 1000 W. Using this condition, about 180 nm, which corresponds to 90% of the total film thickness of 200 nm of the tungsten film exposed from the photosensitive resin mask material layer, was etched. At this stage, the etching conditions were changed to the following.

Next, etching was performed using an etching gas in which three types of SF ₆ , O ₂ , and Ar were mixed. The gas flow rate of SF ₆ was 5 ml / min, the gas flow rate of O ₂ was 50 ml / min, the gas flow rate of Ar was 100 ml / min, and all of the tungsten film exposed from the photosensitive resin mask material layer was removed by etching.
Next, the photosensitive resin mask material layer used as the etching mask was removed. The method used at this time was a method using a solvent, and the photosensitive resin mask material layer was removed by a spray cleaning device using a stripping solution 104 (manufactured by Tokyo Ohka Kogyo Co., Ltd.). Then, ashing with oxygen plasma was performed to remove the residual photosensitive resin mask material layer. By these steps, an inner partition wall 30 having a lattice-shaped tungsten partition wall structure was formed on the semiconductor substrate 10 with a film thickness (height) of 350 nm and a width of 60 nm (FIG. 8 (b)).

Next, plasma CVD was used to form a SiO ₂ film to be the partition wall 31. As the film forming condition of SiO ₂ used at this time, the condition that a thick film is formed in the height direction with respect to the tungsten pattern (partition wall 30) is used. The height of the partition wall 50 formed by the partition wall 30 and the partition wall 31 was 600 nm (FIG. 8 (c)). When the SiO ₂ film to be the partition wall 31 was formed, the exposed surface of the semiconductor substrate 10 was covered with the SiO ₂ film constituting the partition wall 31 (FIG. 8 (c)).

(Formation of first color filter)
Next, a first color filter forming step of forming the first color filter 14 was performed (FIGS. 9 (a) and 9 (b)).
First, in order to provide the first color filter 14, a resist containing a green (G) pigment and having photosensitivity (hereinafter referred to as green resist 14a) was applied to the entire surface of the semiconductor substrate 10 on which the partition wall 50 was formed. (FIG. 9 (a)). Before applying the green resist 14a, a hydrophobic surface treatment (HMDS treatment) may be applied to the entire surface of the semiconductor substrate 10 in order to improve the adhesion.

Next, the green resist 14a was selectively exposed by photolithography and then developed to form a green filter pattern corresponding to the formation position of the first color filter 14. At this time, as the resin which is the main component of the green resist 14a, an acrylic resin having photosensitivity was used. In addition, the pigments used in the green resist 14a are each according to the color index of C.I. I. PG58, C.I. I. It was PY150 and the pigment concentration was 60% by mass. The film thickness of the green filter pattern was 600 nm.
Next, in order to firmly cure the green filter pattern, a hot plate was used to bake at 230 ° C. for 6 minutes and then cured to form the first color filter 14 (FIG. 9 (b)). After going through this heating step, even after going through steps such as the third color filter forming step, peeling of the first color filter 14 and collapse of the pattern were not confirmed.

(Formation of second color filter)
Next, a second color filter forming step of forming the second color filter 15 was performed (FIGS. 10 (a) and 10 (b)).
First, in order to provide the second color filter 15, a resist containing a blue (B) pigment and having photosensitivity (hereinafter referred to as blue resist 15a) is applied to the semiconductor substrate 10 excluding the region where the first color filter 14 is formed. It was applied to the entire surface of the above (FIG. 10 (a)). Before applying the blue resist 15a, a hydrophobic surface treatment (HMDS treatment) may be applied to the entire surface of the semiconductor substrate 10 in order to improve the adhesion.

Next, the blue resist 15a was selectively exposed by photolithography and then developed to form a blue filter pattern corresponding to the formation position of the second color filter 15. At this time, as the resin which is the main component of the blue resist 15a, an acrylic resin having photosensitivity was used. In addition, the pigments used in the blue resist 15a are each according to the color index of C.I. I. PB156, C.I. I. It was PV23, and the pigment concentration was 50% by mass. The film thickness of the blue filter pattern was 590 nm.
Next, in order to firmly cure the blue filter pattern, a second color filter 15 was formed by baking at 230 ° C. for 6 minutes using a hot plate and curing (FIG. 10 (b)). After going through this heating step, even after going through steps such as the third color filter forming step, peeling of the second color filter 15 and collapse of the pattern were not confirmed.

(Formation of a third color filter)
Next, a third color filter forming step of forming the third color filter 16 was performed (FIGS. 11 (a) and 11 (b)).
First, in order to provide the third color filter 16, a resist containing a red (R) pigment and having photosensitivity (hereinafter referred to as red resist 16a) is subjected to a first color filter 14 forming region and a second color. It was applied to the entire surface of the semiconductor substrate 10 excluding the filter 15 forming region (FIG. 11A). Before applying the red resist 16a, a hydrophobic surface treatment (HMDS treatment) may be applied to the entire surface of the semiconductor substrate 10 in order to improve the adhesion.

Next, the red resist 16a was selectively exposed by photolithography and then developed to form a red filter pattern corresponding to the formation position of the third color filter 16. At this time, as the resin which is the main component of the red resist 16a, an acrylic resin having photosensitivity was used. In addition, the pigments used in the red resist 16a are each according to the color index of C.I. I. PR254, C.I. I. It was PY139, and the pigment concentration was 60% by mass. The film thickness of the red filter pattern was 610 nm.

Next, in order to firmly cure the red filter pattern, a third color filter 16 was formed by baking at 230 ° C. for 6 minutes using a hot plate and curing (FIG. 11 (b)). At this time, the red filter pattern is surrounded by a green filter (first color filter 14) having good rectangularity and a partition wall 50, and is formed with good rectangularity. Therefore, it was confirmed that the red filter pattern was cured with good adhesion between the bottom surface of the opening and the surrounding area.
Through the above steps, a color filter 100 including a first color filter 14, a second color filter 15, and a third color filter 16 was formed.

Next, the reflection reduction layer 40 was formed on the upper surface of the color filter 100 (FIG. 12). The reflection reduction layer 40 is formed to reduce the reflection of incident light due to the difference in refractive index between the first microlens layer 20 and the color filter 100. Further, by forming the reflection reduction layer 40, the step between the color filter 100 formed of the three colors of red, blue, and green and the partition wall 50 is reduced, and the stress between the color filter 100 and the microlens 200 is relaxed. To. Further, when the microlens 200 is formed of silicon nitride, the coefficient of thermal expansion of the microlens 200 is different from the coefficient of thermal expansion of the color filter 100, but the difference in the coefficient of thermal expansion is increased by providing the reflection reduction layer 40. It also has the effect of mitigating. The reflection reduction layer 40 also has an effect as a stopper layer when the microlens 200 is formed by dry etching.

In this embodiment, the viscosity of the coating liquid containing the acrylic resin is adjusted, spin-coated at a rotation speed of 1000 rpm, and heat-treated at a temperature of 230 ° C. for 10 minutes on a hot plate to cure the resin, and the reflection reducing layer 40 is used. Formed. At this time, the film thickness of the reflection reducing layer 40 was 100 nm, and the transmittance of visible light of the reflection reducing layer 40 was 99%. The refractive index of the reflection reduction layer 40 was 1.7.

(Micro lens forming process)
Next, a microlens forming step of forming the microlens 200 was performed (FIGS. 13A (a) to 13B (d)).
A silicon nitride (SiN) film 20a was formed with a film thickness of 600 nm on the upper layer of the color filter 100 on which the reflection reduction layer 40 was formed by using plasma CVD (FIG. 13A (a)). At this time, the film formation temperature in plasma CVD was set to 200 ° C. The refractive index of the formed silicon nitride film 20a was 1.90, and the extinction coefficient in the wavelength region longer than the wavelength of 300 nm was 1.0 × 10 ^-4 or less.

Next, a resin having all of alkali solubility, photosensitivity, and thermal reflow property was applied onto the formed silicon nitride film 20a to form a photosensitive sacrificial layer. Then, the photosensitive sacrificial layer was patterned by a photolithography process using a photomask, and then heat-treated at 200 ° C. to form a lens matrix 20b which is a sacrificial layer. The lens matrix 20b was formed by providing a plurality of smooth semicircular shapes having a thickness of about 300 nm (FIG. 13A (b)).

Next, dry etching is performed using a mixed gas of CF ₄ and C ₃ F ₈ which are fluorocarbon gases, the pattern of the lens matrix 20b is transferred to the silicon nitride film 20a, and the first microlens layer 20 is formed. It was formed (FIG. 13B (c)). The etching at this time was carried out under the condition that the etching rates of the lens matrix 20b and the silicon nitride film 20a were the same and the selection ratio was 1. The dry etching time was 5 minutes. As a result, the first microlens layer 20 has a height of 300 nm from the apex of the semicircular lens of the first microlens layer 20 to the bottom surface of the semicircular lens (upper surface of the reflection reduction layer 40). Formed.

Next, on the upper layer of the first microlens layer 20, a silicon oxide (SiO ₂ ) film to be the second microlens layer 21 was formed with a film thickness of 50 nm by using plasma CVD as a film forming apparatus (FIG. 13B). (D)). The refractive index of the silicon oxide film was 1.46. As a result, the microlens 200 was formed. Here, the above-mentioned "film thickness of 50 nm" means that the film thickness at the apex of the second microlens layer 21 is 50 nm.

Next, a microlens flattening layer forming step for forming the microlens flattening layer 300 was performed (FIG. 14).
The microlens flattening layer 300 was formed using, for example, an organic resin having a refractive index of 1.21. This organic resin was rotationally coated on the semiconductor substrate 10 on which the microlens 200 was formed with a film thickness of 500 nm to flatten the surface of the semiconductor substrate 10 on which the microlens 200 was formed. The microlens flattening layer 300 is not limited to this thickness, and may be thicker than the height of the unevenness of the microlens 200 and may be thick enough to flatten the unevenness.

The solid-state image sensor 1 of the embodiment was formed by each of the above steps.
In the solid-state image sensor 1 obtained as described above, the microlens 200 is made of a material having a high refractive index, so that the light-collecting ability is improved, and the reflectance difference between the microlens 200 and the color filter is further improved. The reflection due to the light was also reduced, and the lens had good sensitivity.

<Example 2>
The second embodiment is different from the first embodiment in that the number of layers of the microlenses to be laminated is three when the microlens 200 is formed. In Example 2, each step until the first microlens layer 20 is formed is each step until the first microlens layer 20 in Example 1 is formed (FIGS. 8A to 13B (c). )). Therefore, the configuration after the formation of the first microlens layer 20 will be described below.

After forming the first microlens layer 20, plasma CVD was used as a film forming apparatus to form a silicon oxynitride (SiON) film to be the second microlens layer 21. When forming the silicon nitride (SiN) film used for forming the first microlens layer 20, SiH ₄ , NH ₃ , and N ₂ are used as the gas species, but the second micro lens layer 21 is used. When forming a silicon nitride (SiON) film used for formation, N ₂ O gas is added to the above-mentioned SiH ₄ , NH ₃ , N ₂ and mainly with N ₂ O gas and other gases. By changing the flow rate ratio of, the refractive index of the silicon nitride (SiON) film is controlled. In this embodiment, as the second microlens layer 21, a silicon nitride (SiON) film having a refractive index of 1.68 is formed so that the film thickness is 50 nm. Here, the above-mentioned "film thickness of 50 nm" means that the film thickness at the apex of the second microlens layer 21 is 50 nm.

Next, a silicon oxide (SiO ₂ ) film was formed on the second microlens layer 21 as the third microlens layer 22 by using plasma CVD (FIG. 15A). The refractive index of the silicon oxide (SiO ₂ ) film was 1.46, and the film thickness was 50 nm. Here, the above-mentioned "film thickness of 50 nm" means that the film thickness at the apex of the third microlens layer 22 is 50 nm.
Next, the microlens flattening layer 300 is formed, and the microlens flattening layer 300 is formed in the same manner as the microlens flattening layer 300 of Example 1 (FIG. 15 (b)). .. As a result, the microlens 200 was formed.

The solid-state image sensor of Example 2 was formed by each of the above steps.
In this embodiment, by forming the microlens 200 in multiple layers, there is an advantage that the reflection of the microlens 200 itself formed of the high refractive index material with respect to the incident light is further reduced and the light receiving sensitivity is further improved.

<Example 3>
Example 3 is different from Example 1 and Example 2 in that the microlens 200A is provided instead of the microlens 200. In Example 3, each step until the reflection reduction layer 40 is formed is the same as in Example 1 (FIGS. 8A to 12). Therefore, the steps after the forming step of the silicon nitride film 20a to be the first microlens layer 20 will be described below.

As shown in FIG. 16A (a), after the silicon nitride film 20a to be the first microlens layer 20 was formed, the intermediate sacrificial layer 20c was formed on the silicon nitride film 20a with a resin material.
Next, as shown in FIG. 16A (b), a mask 20d having a lens shape was formed on the intermediate sacrificial layer 20c by resist.
Subsequently, as shown in FIG. 16B (c), the intermediate sacrificial layer 20c was processed into a microlens shape by dry etching using the mask 20d as a mask to form a lens matrix 20e. At this time, the microlens shape was transferred to the intermediate sacrificial layer 20c so that the distance between the lenses was narrowed.
Subsequently, as shown in FIG. 16B (d), the silicon nitride film 20a was processed into a microlens shape by dry etching using the lens matrix 20e as a mask. As a result, the first microlens layer 20 was formed.

Finally, as in Example 1, a second microlens layer 21 was formed, and then a microlens flattening layer 300 was formed. At this time, the microlens 200 was formed so that the Fill Factor was in the range of 85% or more and 99% or less. The Fill Factor was adjusted by adjusting the shape of the mask 20d, the etching conditions of the intermediate sacrificial layer 20c and the silicon nitride film 20a. The shape of the mask 20d was adjusted by changing the photomask pattern, and the etching conditions were adjusted by changing the pressure, RF power, and gas flow rate ratio.
As described above, a plurality of adjacent lenses are in contact with each other on a line in a plan view, and gaps are formed on the corners (four corners) of the first color filter 14, the second color filter 15, and the third color filter 16. A solid-state image sensor equipped with a microlens 200 having a G was formed.

<Evaluation of solid-state image sensor>
When forming the solid-state image pickup device of each of the above-described embodiments, the height of the microlens 200 can be made variable by changing the heights of the dry-etched lens masters 20b and 20e. Further, the Fill Factor of the microlens 200 can be made variable by changing the shapes of the lens mother molds 20b and 20e. Therefore, the height of the microlens 200 and the Fill Factor were adjusted and changed to the values shown in Table 1, and the light receiving sensitivity was evaluated. Further, as a comparison, the microlens was formed by using the organic resin which is a conventional material, without applying the high refractive index material to the material of the microlens. The solid-state image sensor having the conventional structure will be described later.
As shown in Table 1, the evaluation of the solid-state image sensor was performed by using the sample No. 1-Sample No. This was performed for each of the 15 solid-state image sensors. Here, the sample No. 1-Sample No. Reference numeral 4 is a solid-state image sensor having a conventional structure, sample No. 4. 5 to sample No. Reference numeral 10 is a solid-state image sensor having the structure according to the first embodiment, sample No. Reference numeral 11 is a solid-state image sensor having the structure according to the second embodiment, sample No. 11. 12 to Sample No. 15 are solid-state image pickup devices having the structure according to the third embodiment.

<Conventional structure>
The solid-state image sensor having a conventional structure is different from the solid-state image sensor of each embodiment in that the microlens is made of an organic resin material containing an acrylic resin.
The microlens in the solid-state image sensor of the conventional structure was formed as follows. The microlens in the solid-state image pickup device having the conventional structure was formed by the same steps as in Example 1 (FIGS. 8A to 11B) up to the color filter forming step.

Next, an organic resin material containing an acrylic resin is rotationally applied to a thickness of 1.0 μm and heat-cured at 230 ° C. to form an organic resin film on the upper surfaces of the color filter 100 and the partition wall 50. did.
Next, a resin having all of alkali solubility, photosensitivity, and thermal reflow property was applied onto the formed organic resin film to form a photosensitive sacrificial layer. Then, the photosensitive sacrificial layer was patterned by a photolithography process using a photomask, and then heat-treated at 200 ° C. to form a lens matrix. Sample No. 1-Sample No. In No. 4, a lens matrix having a plurality of smooth semicircular shapes having a thickness of about 200 nm to about 500 nm was formed.

Next, dry etching is performed using a mixed gas of CF ₄ and C ₃ F ₈ which are fluorocarbon gases, and the pattern shape of the lens matrix is transferred to an organic resin film to form a resin microlens layer. Formed. The etching at this time was carried out under the conditions that the etching rates of the lens matrix and the organic resin film were the same and the selection ratio was close to 1. In this way, the sample No. 1-Sample No. In No. 4, the height from the apex of the semicircular lens of the resin-based microlens layer to the bottom surface of the semicircular lens (the upper surface of the red third color filter 16) is 200 nm to 500 nm (see Table 1), respectively. As described above, the etching time was adjusted to form a resin-based microlens layer.
Further, a silicon oxide (SiO ₂ ) film (second microlens layer) was formed on the formed resin-based microlens layer by plasma CVD.

Table 1 shows the sample Nos. 1-Sample No. The material and structure of the microlens of each solid-state image sensor of 15 are shown. The lens heights shown in Table 1 are the sample numbers. 1 to No. Reference numeral 4 indicates the lens height of the resin-based microlens layer, and the sample No. 4 is shown. 5 to sample No. Reference numeral 15 indicates the lens height of the first microlens layer.

Table 2 shows the sample Nos. 1-Sample No. The result of comparing the light-receiving sensitivity evaluation result of each solid-state image sensor of 15 is shown. The light-receiving sensitivity evaluation was performed on the sample No., which had good sensitivity with a solid-state image sensor having a conventional structure. 2 was used as a reference for the light receiving sensitivity, and the measurement was performed based on the amount of increase / decrease in the light receiving sensitivity of each sample.

Sample No. 1-Sample No. From the result of No. 4, in the conventional structure, the peak (maximum value) of the light receiving sensitivity was obtained at a microlens height of 300 nm.

The solid-state image sensor having the configuration according to the first embodiment has a structure in which the first microlens layer 20 is silicon nitride and the reflection reduction layer 40 is provided between the microlens 200 and the color filter 100. .. 5 to sample No. From the result of 10, when the film thickness of the first microlens layer 20 is within the range of 100 nm or more and 400 nm or less as compared with the microlens height of the conventional structure, better light receiving sensitivity than the conventional structure is obtained. It was shown to be. This is because the high refractive index lens improves the light collection performance, and the reflection reduction layer 40 suppresses the reflection at the interface between the color filter 100 and the high refractive index microlens (microlens 200). However, even in the microlens 200 with a high refractive index, if the lens height is less than 100 nm, the focusing performance as a lens is not sufficient, and if the lens height is 450 nm or more, the incident light that should be incident on the adjacent pixel is blocked by the microlens. It was shown that the light receiving sensitivity was slightly deteriorated. If the light receiving sensitivity is within the range of 3% reduction (-3%) with respect to the light receiving sensitivity of the solid-state image sensor having the conventional structure, the problem of the present invention can be solved.

Next, the sample No. having the same microlens height. 7 and sample No. As a result of comparison with No. 11, the sample No. of the configuration according to Example 1 was found. No. 7 and the sample No. of the configuration according to the second embodiment. In No. 11, high light-receiving sensitivity was obtained in both cases, but the sample No. 11 was obtained. The light receiving sensitivity of No. 11 is the sample No. It was slightly higher than the light receiving sensitivity of 7.
Sample No. In No. 11 (Example 2), the microlens 200 was formed in a three-layer structure, so that the sample No. 11 in which the microlens 200 was formed in a two-layer structure. The effect of reducing the reflection of light on the surface of the microlens having a higher refractive index than that of No. 7 was obtained. The light receiving sensitivity is improved as compared with 7 (Example 1). In this way, the sample No. In No. 11, although the number of process steps increases, it can be used as a structure for improving the light receiving sensitivity.

Sample No. 12-Sample No. In No. 15, since the microlens 200 was formed by the method of Example 3, the Fill Factor of the microlens 200 was as high as 85% or more and 99% or less, as compared with the case where the microlens 200 was formed by the methods of Examples 1 and 2. It was possible to obtain higher light receiving sensitivity. Therefore, it is preferable to use a solid-state image sensor having a microlens 200 in which a plurality of adjacent lenses are in contact with each other and the Fill Factor is 85% or more and 100% or less.

In Examples 1 to 3, the example in which the microlens flattening layer 300 is formed has been described, but even when the microlens flattening layer 300 is not formed for cost reduction, a high refractive index microlens having the reflection reducing layer 40 is provided. Similar to (Microlens 200), good sensitivity characteristics were confirmed.

Although the present invention has been described above with reference to each embodiment and each embodiment, the scope of the present invention is not limited to the illustrated and described exemplary embodiments, and is equivalent to that of the present invention. Also includes all embodiments that bring about a positive effect. Furthermore, the scope of the present invention is not limited to the combination of the features of the invention described in the claims, but may be formed by any desired combination of the specific features of all the disclosed features.

Further, in each of the first to third embodiments, the color filter 100 applies the material of the color filter 100 having photosensitivity to the entire surface of the semiconductor substrate 10 to selectively expose the portion where the color filter 100 is formed. The color filter pattern was formed by curing. However, the material of the color filter 100 is not limited to such a configuration. For example, as shown in FIGS. 17A to 17C, the material of the color filter (for example, the first color filter 14) is not photosensitive and is cured by heating bake using a heat-curable resin or the like (for example). 17A (a)), the photoresist 60 was coated and the pattern was exposed with a photomask (FIG. 17A (b)) to form the mask pattern 60a with the photoresist 60 (FIG. 17B (c)). A process may be used in which only the color filter at the position to be opened is removed by etching (FIG. 17B (d)) and the photoresist (mask pattern) 60a is removed by cleaning the film (FIG. 17C (e)).

Further, this dry etching process and a lithography process using a photosensitive color filter may be used in combination.
Further, in each embodiment, a known forming step may be used as the forming step of the color filter 100.
Further, in the present embodiment, the case where the partition wall 50 is formed so as to have the same height as the color filter 100 has been described, but the height of the partition wall 50 is, for example, about half the height of the color filter 100. May be good.

10 ... Semiconductor substrate 11 ... Photoelectric conversion element 12 ... Element separation structure 14 ... First color filter 15 ... Second color filter 16 ... Third color filter 20 ... 1st microlens layer 21 ... 2nd microlens layer 22 ... 3rd microlens layer 30, 31, 50 ... partition wall 40 ... reflection reduction layer 100 ... color filter 200・ ・ ・ Microlens 300 ・ ・ ・ Microlens flattening layer

Claims (17)

With a semiconductor substrate,
A plurality of photoelectric conversion elements provided on the semiconductor substrate and arranged in a matrix in a plan view, and
A plurality of color filters arranged corresponding to each of the plurality of photoelectric conversion elements are two-dimensionally arranged with a preset regular pattern, and a color filter layer.
The reflection reduction layer provided on the color filter layer and
A microlens layer provided on the reflection reduction layer and having a plurality of lenses arranged corresponding to the color filters of the plurality of colors and the plurality of photoelectric conversion elements, respectively.
Equipped with
The microlens layer is formed so as to be laminated on the lens surface of the first microlens layer and the first microlens layer arranged closest to the photoelectric conversion element side. Has a lens layer and
The refractive index of the first microlens layer is in the range of 1.75 or more and 2.2 or less.
A solid-state image sensor in which the refractive index of each of the reflection reducing layer and the second microlens layer is lower than the refractive index of the first microlens layer. The solid-state image sensor according to claim 1, wherein the refractive index of the reflection reduction layer is higher than the refractive index of the second microlens layer and lower than the refractive index of the first microlens layer. The solid-state image sensor according to claim 1 or 2, wherein the refractive index of the reflection reduction layer is in the range of 1.55 or more and 1.8 or less. The solid-state image sensor according to any one of claims 1 to 3, wherein the film thickness of the reflection reduction layer is in the range of 10 nm or more and 200 nm or less. The solid-state image sensor according to any one of claims 1 to 4, wherein the film thickness of the first microlens layer is in the range of 100 nm or more and 400 nm or less. The solid-state image sensor according to any one of claims 1 to 5, wherein the refractive index of the second microlens layer is in the range of 1.4 or more and 1.75 or less. The solid-state image sensor according to any one of claims 1 to 6, wherein the Fill Factor of the microlens layer is in the range of 85% or more and 100% or less. The solid-state imaging device according to any one of claims 1 to 7, wherein the plurality of adjacent lenses are in contact with each other in the microlens layer and have a gap on a corner portion of the color filter. The first microlens layer has an extinction coefficient of 1.0 × 10 ^-3 or less in the range of visible light having a wavelength of 380 nm or more and 700 nm or less, according to any one of claims 1 to 8. Solid-state image sensor. The solid-state image sensor according to any one of claims 1 to 9, wherein the first microlens layer is made of an inorganic material. The one according to any one of claims 1 to 10, wherein the microlens layer is formed of a nitride of silicon or an oxynitride of silicon, and the oxygen content increases as the distance from the photoelectric conversion element increases. Solid-state image sensor. The solid-state image pickup device according to any one of claims 1 to 11, further comprising a lens flattening layer provided on the lens surface of the microlens layer and formed so that the upper surface is flat. The solid-state image sensor according to claim 12, wherein the lens flattening layer is formed of a resin having a refractive index in the range of 1.1 or more and 1.6 or less. The solid-state image sensor according to any one of claims 1 to 13, wherein the film thickness of the second microlens layer is in the range of 50 nm or more and 150 nm or less. The solid-state image sensor according to any one of claims 1 to 14, wherein the film thickness of the second microlens layer is thinner than the film thickness of the first microlens layer. A step of forming a partition wall at a position surrounding the photoelectric conversion element on a semiconductor substrate in which a plurality of photoelectric conversion elements are arranged in a matrix in a plan view and an element separation structure is provided between the plurality of photoelectric conversion elements. ,
A step of forming color filters of a plurality of colors at positions corresponding to the photoelectric conversion element surrounded by the partition wall, and
A step of forming a reflection reducing layer composed of a resin layer or an inorganic material layer having a refractive index in the range of 1.55 or more and 1.8 or less on the upper part of the color filter and the partition wall.
A plurality of silicon nitride films or silicon oxynitride films having a refractive index in the range of 1.75 or more and 2.2 or less are formed on the upper part of the reflection reduction layer at positions corresponding to the plurality of photoelectric conversion elements. A step of forming a first microlens layer having a lens and having a film thickness in the range of 100 nm or more and 400 nm or less.
A step of forming a second microlens layer formed of a silicon oxynitride film or a silicon oxide film having a refractive index lower than that of the first microlens layer on the lens surface of the first microlens layer. ,
A method for manufacturing a solid-state image sensor. The step of forming the first microlens layer is
A resist pattern is formed on the resist layer by photolithography on the resist layer of a laminate in which the color filter, the reflection reduction layer, the silicon nitride film or the silicon oxynitride film, the resin layer, and the resist layer are laminated in this order. And the process to do
The process of melting the resist pattern by heat flow and forming it into a lens shape,
A step of performing dry etching on the resin layer using the resist pattern molded into the lens shape as a mask and molding the resin layer into a lens master mold.
A step of performing dry etching on the silicon nitride film or the silicon oxynitride film using the lens matrix as a mask to form the silicon nitride film or the silicon oxynitride film into a microlens.
The method for manufacturing a solid-state image sensor according to claim 16.

Images