JP2011061133A - Semiconductor image sensor and method of manufacturing the same - Google Patents

Abstract

【Task】
In a pixel configuration using only total reflection of light, the angle of the total reflection surface is an important design factor, and it is difficult to achieve compatibility with other characteristics of the semiconductor image sensor. Further, total reflection is not realized depending on the incident angle of light, and light sensitivity is lowered and resolution is deteriorated due to light leaking to adjacent pixels.
[Solution]
A recess is formed in the upper part of the photosensitive element 11, and an optical material 19 is formed by filling the inside with an insulating material. The optical element 19 is composed of a first material and a second material, and the second material is the main component in the central region of the optical element, and the first material is the main component in the peripheral region in the optical element. Thus, the refractive index distribution is realized. After the incident light 25 passes through the color filter 20, it is refracted in the optical element or totally reflected by the side wall and then reaches the photosensitive element 11.
[Selection] Figure 1

Description

  The present invention relates to a pixel configuration of a semiconductor image sensor.

  Semiconductor image sensors have become increasingly widespread in many image input devices due to advances in the number of pixels and miniaturization due to advances in semiconductor technology. In particular, digital cameras and mobile phones have rapidly expanded their application as image recording and scanners. In these application fields, there is a strong demand for further increase in the number of pixels and miniaturization of the semiconductor image sensor, and various improvements in the pixel configuration have been made. Among these, a configuration in which a condensing microlens is mounted on each pixel is widely used. As an example of such a microlens, a transparent resin is applied to the entire surface of the wafer, leaving only the photosensitive element region of the pixel, and then deforming the resin in a high temperature atmosphere to form a convex lens shape. Due to the light condensing action of the microlens, it is possible to guide light incident outside the photosensitive element region in the pixel to the photosensitive element, and the sensitivity of the image sensor is improved. With the increase in the number of pixels and the reduction in size, the area ratio of the photosensitive element region to the pixel area (so-called aperture ratio) tends to be small, and mounting of microlenses has been frequently used as an effective design technique.

  However, since the surface of the image sensor is not flat when the microlens is mounted, dust and dirt are liable to adhere, and special attention is required in the assembly process. Moreover, even if the microlens is mounted, the incident state of light is different between the pixel in the central part of the sensor and the pixel in the peripheral part, so that unevenness in brightness and photosensitivity occurs in the imaging screen.

  In order to solve such a problem, the following cited patent document 1 proposes a pixel configuration method that uses total reflection of light as an alternative to a microlens.

  FIG. 18 shows a pixel configuration example described in Patent Document 1. In the figure, the incident light beam that has passed through the color filter is reflected by the total reflection surface 1 and reaches the light receiving unit 2. By selecting materials for the transparent material portion 3 and the low refractive index portion 4 as appropriate, a total reflection phenomenon using a refractive index difference is realized.

JP-A-5-235313

  In Patent Document 1, it is specifically exemplified that the total reflection surface is constituted by an interface between “a solid material such as silicon oxide and silicon nitride” and “a gas material such as vacuum and air”. However, it is difficult to implement these combinations with an actual semiconductor image sensor. In particular, in the configuration cited in FIG. 18, the portion 4 is formed in a hollow state, and it is difficult to optically flatten the surface of the total reflection surface 1 (surface on the 4 side).

  Furthermore, cited patent document 1 shows that the angle formed by the total reflection surface 1 and the vertical direction is “26.5 degrees” in order to realize total reflection. However, in the design of a semiconductor image sensor, it is not common to set this angle independently of other structural factors. The above-mentioned angles are related to many design requirements, such as limitations on vertical and horizontal dimensions caused by known semiconductor manufacturing processes, the magnitude of the output electric signal with respect to incident light, and the area ratio of the light receiving portion occupied in the pixel. It is determined. That is, if the angle of “26.5 degrees” cannot be realized, the effect of the cited patent will be reduced.

  As described in the previous section, when the angle of the total reflection surface is larger than “26.5 degrees”, total reflection due to the difference in refractive index does not occur, and a part of the incident light is 4 Enter into the area. This light component induces deterioration of the characteristics of the semiconductor image sensor such as a decrease in photosensitivity due to failure to contribute to photoelectric conversion in the corresponding pixel and a decrease in resolution due to photoelectric conversion in adjacent pixels.

  A semiconductor image sensor provided with a transparent optical element for condensing light above each of the arranged photosensitive elements, wherein the refractive index of the central region of the optical element occupies outside the central region of the optical element Higher than the refractive index of the peripheral region.

  The optical element is composed of at least two kinds of transparent materials having different refractive indexes, and the content of the material that increases the refractive index is decreased and the refractive index is decreased as it goes from the central region to the peripheral region. Increase material content.

  A refractive index in a part of the optical element is determined by a composition ratio of the at least two kinds of transparent materials in a part of the optical element.

Note that the refractive index at an arbitrary location in the optical element depends on the configuration of the transparent material at that location. For example, since the central region of the optical element is mainly composed of Si ₃ N ₄ or the like, the refractive index in this region is substantially equal to the refractive index (= 2.0) of Si ₃ N ₄ . For example, since the peripheral region of the optical element is composed mainly of SiO ₂ or the like, the refractive index in this region is approximately equal to the refractive index of SiO ₂ (= 1.5). Furthermore, for example, since Si ₃ N ₄ and SiO ₂ are mixed in an intermediate region between the central region and the peripheral region of the optical element, the refractive index thereof has an intermediate value.

In general, when a transparent material having a refractive index higher than this refractive index is mixed with a transparent material having a certain refractive index, the refractive index of the mixed transparent material can be increased. Conversely, when a transparent material having a low refractive index is mixed, the refractive index of the mixed transparent material can be lowered. For example, when Si ₃ N ₄ (high refractive index) is mixed with SiO ₂ (low refractive index), Si ₃ N ₄ becomes a “material that increases the refractive index”. Conversely, when SiO ₂ is mixed with Si ₃ N ₄ , SiO ₂ becomes a “material that lowers the refractive index”.

  In the case where the optical element is composed of two kinds of materials, a plurality of pixels including photosensitive elements are arranged, and a transparent light-collecting function provided in an insulating layer above the photosensitive elements is provided. A semiconductor image sensor comprising an optical element, wherein the optical element is composed of a first material having a first refractive index and a second material having a second refractive index, and a central region of the optical element In other words, it can be said that the second material is the main component, and the peripheral region away from the central region in the optical element is the first material. In addition, the paragraph 0010 described above can be realized by configuring the first refractive index of the first material to be smaller than the second refractive index of the second material.

  In the case where the optical element is composed of two types of materials, the optical element composed of the first material and the second material is moved from the central region toward the peripheral region. The paragraph 0011 described above can be realized by configuring so that the content ratio of the second material monotonously decreases and the content ratio of the first material monotonously increases.

  Note that, as a configuration example of the optical element, when the first element and the second material are used, the total content of these two materials is 100%. For example, when the content of the first material is 20% at a specific location in the optical element, the content of the second material is 80%. In the second specific location from the specific location toward the central region of the optical element, the content of the first material is, for example, 10%, and the content of the second material is 90%. Become.

  In the above paragraph 0016, “monotonic” means, for example, “the change in the content ratio is one direction of increasing as the spatial position goes to the central region, and the content rate once increased decreases again. It means "There is no". However, in an actual semiconductor process, it is inevitable that slight fluctuations occur in the process conditions and the quality of the vapor growth film during the manufacturing process. For this reason, when viewed from a micro perspective, it is not “monotonous” but increases or decreases as a whole while repeating minute increases and decreases. The term “monotonic” used in this specification may include a minute increase / decrease due to such fluctuation.

  As the optical element moves from the central region toward the peripheral region, the refractive index of the optical element is decreased stepwise by increasing the content of the low refractive index material stepwise.

  In a semiconductor image sensor including a plurality of pixels including a photosensitive element and a transparent optical element having a light collecting function provided in an insulating layer above the photosensitive element, as a configuration example of the optical element A plurality of layers may be spatially arranged from a central region of the optical element to a peripheral region away from the central region. In this case, the content ratio of the material in each layer constituting the plurality of layers is changed for each of the layers, and the refractive index of the layer in the central region is set so that the refractive index of the layer in the peripheral region is It is comprised so that it may become larger than a refractive index. In this case, the refractive index decreases stepwise (stepwise) at the boundary between the layers constituting the plurality of layers as it goes from the central region to the peripheral region.

The “layer” in paragraph 0020 above will be outlined in more detail. The optical element in the pixel has a plurality of layers from the central region to the peripheral region, and the specific layer is composed of at least two or more materials as described above. The content ratio of each material in the specific layer is determined for each layer, and the content ratio is such that the refractive index increases as the layer is closer to the central region. For example, when the three cases of SiO ₂ , Si ₃ N ₄ , and Ta ₂ O ₅ are described as the material,
(1) A layer located in the center of the optical element:
Content ratio of SiO ₂ and Si ₃ N ₄ = 0%
Ta ₂ O ₅ content = 100%
(2) A layer located in the central region slightly away from the center of the optical element
(For example, it is a layer located outside (1) described above):
Content ratio of SiO ₂ and Ta ₂ O ₅ = 0%
Si ₃ N ₄ content = 100%
(3) Layer located between the central region and the peripheral region:
Content ratio of SiO ₂ and Si ₃ N ₄ = 50%
Ta ₂ O ₅ content = 0%
(4) Layer located on the outermost periphery of the optical element in the peripheral region:
100% content of SiO ₂
Content ratio of Si ₃ N ₄ and Ta ₂ O ₅ = 0%
Examples of such contents are given.

  The paragraph 0020 described above describes that the optical element is composed of a plurality of layers having a finite thickness. In this case, it is shown that the content ratio of the material in each layer is constant, and the content ratio changes stepwise at the boundary with the adjacent layer, that is, the refractive index changes stepwise. Yes.

  As the optical element moves from the central region to the peripheral region, the optical element continuously increases the content of the low refractive index material so that the refractive index continuously decreases.

  The paragraph 0023 described above corresponds to the case where the optical element is composed of a plurality of layers having an infinitesimal thickness. In such a case, the number of “layers” becomes infinite, and it becomes difficult to clearly define the boundaries of “layers”. That is, the above-described content rate does not indicate a value in a specific “layer”, but a content value at a location (a spatial position in the optical element) where this “layer” exists. For this reason, the content rate changes continuously from the central region toward the peripheral region without rapidly changing.

In addition, as a transparent material which comprises the said optical element, various materials, such as an inorganic material, an organic material, and a hybrid material, can be utilized in combination. For example, as an inorganic material, ITO (indium tin oxide, Sn / In ₂ O ₃ ), ATO (antimony tin oxide, Sb / SnO ₂ ), zinc oxide (ZnO), titanium oxide, which are well known as transparent conductive materials, are used. (TiO ₂ ), tantalum oxide (Ta ₂ O ₅ ), zirconia (ZrO ₂ ), silica (SiO ₂ ), alumina (Al ₂ O ₃ ), barium titanate (BaTiO ₃ ), aluminum hydroxide (AlO (OH)), etc. There is. It is known that some of these substances may be used in combination with a technique for ensuring transparency by reducing the particle size and dispersing. In addition to the above metal oxides, oxides such as silicon oxide (SiO ₂ ) and nitrogen oxide (Si ₃ N ₄ ), fluorine compounds such as magnesium fluoride (MgF ₂ ), lead glass, various types There is also glass. Examples of the organic material include a fluorine-based polymer (fluorine resin) and a silicone-based polymer (silicone resin).

The optical element is characterized by being made of two or more transparent materials such as silicon oxide, silicon nitride, and tantalum oxide. In the case where the optical element is composed of two kinds of materials, the first material may be a material containing silicon oxide, and the second material may be a material containing silicon nitride. It is done. It is known that substances having different component ratios exist in “silicon oxide” and “silicon nitride”. In general, “silicon oxide” as a material used in the semiconductor field contains a large amount of SiO _2, but “silicon nitride” is represented by a substance represented by SiNx, typically Si ₃ N ₄ . Although these materials should be correctly called “silicon nitride-based materials”, they are described as “Si ₃ N ₄ ” or “silicon nitride” in this specification for convenience.

  The paragraph 0011 and the like described above indicate that the material of the optical element is not single and uniform, but “graded material” in which the material changes from the periphery toward the center. For example, when the optical element is composed of silicon oxide and silicon nitride, the silicon nitride ratio is high in the central region of the optical element, while the silicon oxide ratio is high in the peripheral region in the optical element. It is done. When this state is expressed by a refractive index, it corresponds to “the central region of the optical element has a higher refractive index than the peripheral region”.

  A semiconductor image sensor having a transparent optical element having a light collecting function, in which a plurality of pixels including a photosensitive element are arranged and provided in an insulating layer above the photosensitive element, the size of the optical element being The light incident side (upper side) may be large and the photosensitive element side may be small. Furthermore, it is preferable that the size of the optical element on the light incident side is substantially equal to the size of the pixel of the semiconductor image sensor. Further, it is preferable that the size of the optical element on the photosensitive element side is substantially equal to or smaller than the size of the photosensitive element. For example, the representative shape of the optical element is an inverted square frustum.

  The optical element is provided with a layer that reflects light on a sidewall in contact with an insulating layer of the semiconductor image sensor in which the optical element is disposed.

  The light reflecting layer may be provided on the entire side wall. Further, it may be provided only on the upper part of the side wall (upper part on the side where light enters the image sensor).

  A step of providing a recess in the insulating layer above the photosensitive element, and a vapor phase growth step of filling the recess with at least two types of transparent materials having different refractive indexes to form the optical element. In the phase growth process, a semiconductor image sensor manufacturing method is adopted in which the vapor phase growth conditions are changed over the period from the initial stage of the process to the end of the process.

  Since the side wall of the optical element is formed along the depression provided in the insulating layer, the surface of the side wall becomes optically flat. Since this manufacturing process is composed of a technique frequently used in the semiconductor field, the optical element can be easily manufactured. Therefore, since there is no unevenness on the side wall, the desired light collecting effect can be exhibited without inhibiting the total reflection of light.

  The refractive index distribution of the optical element is not uniform, and the refractive index in the central region is manufactured to be larger than the refractive index in the peripheral region in the optical element. For this reason, light refraction occurs in the optical element, and the light is directed to a region having a higher refractive index. For this reason, it becomes possible to increase the light component toward the central region in the optical element before the light reaches the side wall without total reflection at the side wall of the optical element. For this reason, in the conventional example, the restriction on the angle of the side wall, which is an important pixel design factor and at the same time, is reduced, and the degree of freedom in pixel design can be greatly improved. As a result, the thickness of a plurality of insulating layers, the shape of the optical element, and the like can be determined based on improvements in the electrical characteristics and photoelectric conversion characteristics of the semiconductor image sensor.

  According to the pixel configuration of the semiconductor image sensor according to the present invention, there is an advantage that a microlens is not required and the surface of the image sensor can be flattened with a material of an optical element having a condensing function. Such flattening is advantageous in that surface contamination of the image sensor can be avoided and mechanical cleaning means can be employed.

  A condensing function similar to that of a microlens is achieved by configuring the optical element by filling a light-transmitting material into a recess provided in an insulating layer above the photosensitive element constituting the semiconductor image sensor. Realized. Further, the light collecting function is improved by spatially distributing materials having different refractive indexes as the optical element.

  Even incident light having a large incident angle at which total reflection at the side wall cannot be expected can be reflected into the optical element by the reflection layer provided on the side wall. As a result, there is no light passing through the side wall, and it is possible to prevent characteristic deterioration such as a decrease in photosensitivity and a decrease in resolution.

  In order to realize the refractive index distribution, a manufacturing method in which the growth conditions are changed from the initial stage of the vapor phase growth to the end of the process by using the vapor phase growth process can be used.

It is a figure which shows the pixel structure of a semiconductor image sensor. (First embodiment) It is a figure which shows the refractive index distribution in an optical element. It is a model figure for performing the light trace analysis in an optical element (cylindrical shape). It is a figure which shows the light trace in an optical element (cylindrical shape). (Stepwise refractive index distribution) It is a figure which shows the light trace in an optical element (cylindrical shape). (Linear refractive index distribution) FIG. 3 is a model diagram for analyzing light traces in an optical element (vertical inverted truncated cone shape). It is a figure which outlines the manufacturing process of a semiconductor image sensor. It is a figure which outlines the manufacturing process of a semiconductor image sensor. (Second Embodiment) It is a figure which outlines the manufacturing process of a semiconductor image sensor. (Third embodiment) It is a figure which outlines the manufacturing process of a semiconductor image sensor. (There is a reflective layer on the side wall) (Fourth Embodiment) It is a figure which shows the pixel structure of a semiconductor image sensor. (Fifth embodiment) It is a figure which shows the pixel structure of a semiconductor image sensor. (There is a microlens on the bottom of the optical element.) (Sixth embodiment) It is a figure which shows the manufacturing process of FIG. It is a figure which shows the pixel structure of a semiconductor image sensor. (There is a microlens in the upper opening of the optical element) (Seventh embodiment) It is a figure which shows the structure of a semiconductor image sensor. (Eighth embodiment) It is a figure which shows the structure of a semiconductor image sensor. (Ninth embodiment) It is a figure which shows the structure of a semiconductor image sensor. (Tenth embodiment) It is a figure which shows the structure of the conventional semiconductor image sensor.

  Hereinafter, the present invention will be described in detail based on each embodiment of the present invention shown in the drawings.

<First Embodiment>
FIG. 1 is a diagram illustrating an example of a pixel configuration as a first embodiment of the present invention. In the figure, the aspect ratio of the figure does not necessarily match the actual image sensor, and merely shows the configuration and operation of the pixels conceptually. In the figure, 10 is a semiconductor substrate such as Si, 11 is a photosensitive element such as a photodiode, and 12 is for connecting the photoelectric conversion signal from 11 to an external circuit and simultaneously charging 11 to a reference potential at the same time. 13 is a gate electrode for electrically connecting 11 and 12, 14 is a transistor composed of 11 to 13, 15 is a first insulating layer, 16 is a drive for the image sensor and It is a wiring layer necessary for signal processing, and also functions as a light shielding layer for preventing incident light from being irradiated outside the photosensitive element region. Reference numeral 17 denotes a second insulating layer made of an oxide film, a resin, or the like laminated on the surface of the above 16 and has a function of flattening the surface of the semiconductor image sensor. Further, 18 is a third insulating layer laminated on the second insulating layer. A recess is formed in a part of the first insulating layer 15, the second insulating layer 17, and the third insulating layer 18 so as to be positioned above the photosensitive element 11, and an insulating material or a resin material is formed in the inside. Filled to form an optical element 19. That is, 19 is composed of an upper opening facing the incident light, a bottom surface that is a boundary with the photosensitive element side, and a side wall. As illustrated in the figure, the cross section of the optical element 19 does not have a uniform cross-sectional area in the vertical direction of the figure, but is formed so that the cross-sectional area increases toward the top. Further, it is desirable that the bottom surface of 19 is not in direct contact with the photosensitive element 11 and is in contact with a part of the first insulating layer 15, but this is not restrictive.

  In the present embodiment, incident light exemplified by 25 passes through the color filter 20, then enters the upper opening of the optical element 19, reaches the bottom surface, passes through the bottom surface, and enters the photosensitive element 11. In this optical element, as will be described later, the light reaches the bottom surface by being totally reflected once or a plurality of times at a place where the refractive index changes, or totally reflected by the side wall. That is, all the light amounts incident on the upper opening of the optical element are efficiently condensed and guided to the photosensitive element. That is, the optical element 19 has a function similar to that of a microlens that has been frequently used in the past. A fourth insulating layer 21 as a protective film is laminated on the color filter 20 and the optical element 19 to flatten the surface of the semiconductor image sensor. In the configuration of FIG. 1, compared with the conventional example, the top surface of the pixel, that is, the surface of the semiconductor image sensor is in principle flat, thereby overcoming various problems such as contamination and dust adhesion that tend to occur in the conventional configuration. Will be.

  In the embodiment shown in FIG. 1, the semiconductor image sensor for capturing a color image is exemplified. However, the present embodiment is not limited to color, and includes an image sensor for capturing black and white images. In the case of a monochrome image sensor, the color filter 20 is not necessary, and the upper opening of the optical element 19 is in direct contact with the fourth insulating layer 21. Furthermore, the fourth insulating layer 21 is not necessary when the surface of the third insulating layer 18 and the upper opening of the optical element 19 are formed to have a flattened surface that is substantially coplanar. There is also.

  In FIG. 1, the thickness of the third insulating layer 18 will be described. In the pixel of the semiconductor image sensor, since the planar spread (area) of the photosensitive element 11 is smaller than the pixel area, all incident light represented by 25 can be efficiently condensed and concentrated on the photosensitive element. , Essential for increasing photosensitivity. In FIG. 1, the shape of the optical element 19 is set so that the area becomes larger as it goes upward, and the upper opening of the optical element becomes larger. In the figure, reference numerals 22 and 23 denote pixel arrangement pitches (corresponding to pixel areas). Since the position of the photosensitive element 11 within the pixel does not necessarily coincide with the position of the geometric pixel center, the planar relative positions of 22 and 23 are shifted, but the dimensions are the same. As conceptually shown in the figure, the size of the upper opening of the optical element 19 can be made substantially equal to the pixel pitch 23 by setting the thickness of the third insulating layer 18 large. In such a configuration, the entire pixel surface can receive incident light equivalently. That is, the thickness of the third insulating layer 18 is an important design factor for making the upper opening of the optical element 19 equal to the pixel pitch. Of course, even if the thickness of the third insulating layer 18 is small or the angle of the side wall of the optical element is increased (the inclination of the side wall is made closer to the horizontal direction), the upper opening can be made equal to the pixel pitch. Such a configuration is not preferable because a defect that makes it difficult to expect a total reflection effect on the side surface described later occurs.

<Shape of optical element>
In the embodiment shown in FIG. 1, only the cross-sectional shape of the optical element 19 in the vertical direction is illustrated. For the planar shape of the optical element in the horizontal direction (the shape when it is cut)
(1) The bottom surface, which is the boundary with the photosensitive element side, is approximately equal to the size of the photosensitive element and preferably has the same shape as the photosensitive element, but this is not restrictive. The shape of the photosensitive element is usually a rectangle (including a square) in many cases, but the bottom surface may be a shape other than this rectangle, for example, a polygon or a circle.
(2) The upper opening facing the incident light is approximately the same as the size of the pixel of the semiconductor image sensor and preferably has the same shape as the pixel, but this is not restrictive. The shape of the pixel is usually a rectangle (including a square) in many cases, but the upper opening may be a shape other than the rectangle, for example, a polygon or a circle. However,
In order to make all the light incident on the pixel contribute to photoelectric conversion, it is preferable that the upper opening has the same shape and the same size as the pixel.
(3) Although it is preferable that the intermediate part (part with a side wall) of an optical element is a shape similar to the said upper opening, it is not this limitation. Usually, the shape is such that the upper opening and the bottom surface are connected, for example, a quadrangular frustum (the upper surface has a larger area than the lower surface).

  The manufacturing process of the optical element 19 shown in FIG. 1 will be described later, but its refractive index is not uniform and is spatially distributed in the cross section of 19. FIG. 2A to FIG. 2C show examples of refractive index distributions in the cross section of the portion indicated by the one-dot chain line 27 in FIG. In FIG. 6A, the refractive index distribution is a stepped step 28, and in FIG. 5B, the refractive index distribution is a continuous straight line 29, and FIG. Then, the refractive index distribution is a continuous curve 30. In any case, the refractive index is large in the central region of the optical element 19 and the refractive index is small in the peripheral region. Further, the refractive index distribution shown in the figure is characterized by increasing monotonously toward the center.

  FIG. 2 (a) shows a stepwise refractive index distribution that changes stepwise. In other words, this corresponds to the case where a plurality of layers exist from the central region to the peripheral region of the optical element, and each layer has a specific refractive index. On the other hand, FIG. 2B shows a refractive index distribution that is linear and continuously changing. This distribution corresponds to the case in FIG. 2A in which the layer thickness is extremely small and the number of layers is extremely large. That is, it can be considered that the step difference of the staircase in FIG. 2A is reduced, and the refractive index distribution is linear and continuous.

<Refractive index distribution in optical element (cylindrical shape)>
FIG. 3 is a diagram for explaining the light collecting function of the optical element having the refractive index distribution described in the previous section. FIG. 2A is a conceptual structural diagram when it is assumed that the optical element has a cylindrical shape with a circular cross section. Concentric figures are drawn from the central part to the outer peripheral part, and it has a layered structure. Furthermore, the case where the refractive index changes stepwise at each circular boundary, that is, the stepwise refractive index distribution illustrated in FIG. 2A is shown. FIG. 3B shows a further simplified model. That is, in the same figure (a), it cut | disconnected longitudinally from the center part to the outer peripheral part, and is simplified to the structure where the some board was laminated | stacked in layer form. In this case, the back side of FIG. 5B corresponds to the center portion of the optical element, and the near side corresponds to the outer peripheral portion of the optical element. Further, FIG. 4C shows a conceptual diagram of the simplified model of FIG. In FIG. 3C, the rightmost layer corresponds to the central part (refractive index is maximum) of the optical element, and the leftmost layer corresponds to the outer peripheral part (refractive index is minimum) of the optical element. Further, in FIG. 3C, the lower side corresponds to the photosensitive element side, and the upper side corresponds to the surface side of the semiconductor image sensor (the above-described upper opening, the side on which light from the imaging lens is incident). FIG. 2C shows a state in which light is incident at a constant incident angle from the upper right of the figure. Below, the light trace which this light ray draws inside the optical element which has a layered structure is illustrated as a simulation result.

<Light trace in optical element (cylindrical shape-1)>
FIG. 4 shows the trace of incident light within the optical element described above. The horizontal axis of the figure shows the distance from the center of the optical element (distance corresponds to 0 micrometer) to the outer periphery (distance corresponds to 6 micrometers). The vertical axis indicates the depth of the optical element from the upper opening. The figure also shows that the refractive index decreases stepwise step by step from the center toward the outer periphery every 1 micrometer. That is, the six-layer structure has a refractive index of 2.0 (corresponding to Si ₃ N ₄ ) at the center and a refractive index of 1.5 (corresponding to SiO ₂ ) at the outer periphery. As described above, this figure assumes a configuration in which the optical element has a cylindrical shape (6 layer structure) with a diameter of 12 micrometers and a height of 10 micrometers, and the refractive index is distributed stepwise. . Further, in the figure, the incident angle (see FIG. 3C) of the light incident on the top of the optical element is used as a parameter. FIG. 4A shows a case where light is incident on the center of the optical element, and FIG. 5B shows a case where light is incident on a position 2 micrometers away from the center of the optical element toward the outer peripheral side.

In FIG. 4A, when the incident angle is 10 degrees to 40 degrees, total reflection is performed at a specific boundary where the refractive index changes. For example, (1) At an incident angle of 10 degrees, the light is totally reflected at a distance of 1 micrometer and does not enter the region having a refractive index of 1.9, but is reflected to the region having a refractive index of 2.0.
(2) When the incident angle is 20 degrees, the light is totally reflected at a distance of 2 micrometers, does not enter the region having a refractive index of 1.8, and is reflected to the region having a refractive index of 1.9.
(3) At an incident angle of 30 degrees, the light is totally reflected at a distance of 3 micrometers, does not enter the region having a refractive index of 1.7, and is reflected to the region having a refractive index of 1.8.
(4) At an incident angle of 40 degrees, the light is totally reflected at a distance of 5 micrometers, does not enter the region having a refractive index of 1.5, and is reflected to the region having a refractive index of 1.6.
(5) At an incident angle of 50 degrees, depending on the refractive index of the material constituting the outside of the optical element, total reflection may occur at a distance of 6 micrometers.
(6) In the case of an incident angle of 60 degrees, depending on the refractive index of the substance constituting the outside of the optical element,
There may be cases where light escapes outward from the optical element.
In FIG. 4B, total reflection is performed at a specific boundary where the refractive index changes when the incident angle is 10 degrees to 30 degrees. For example, (1) At an incident angle of 10 degrees, the light is totally reflected at a distance of 3 micrometers, does not enter the region having a refractive index of 1.7, and is reflected to the region having a refractive index of 1.8.
(2) At an incident angle of 20 degrees, total reflection occurs at a distance of 4 micrometers (in the depth direction, 13 micrometer), and it does not enter a region with a refractive index of 1.6, and a refractive index of 1.7.
It is reflected to the area.
(3) At an incident angle of 30 degrees, the light is totally reflected at a distance of 5 micrometers, does not enter the region having a refractive index of 1.5, and is reflected to the region having a refractive index of 1.6.
(4) At an incident angle of 40 degrees, depending on the refractive index of the material constituting the outside of the optical element, total reflection may occur at a distance of 6 micrometers.
(5) When the incident angles are 50 degrees and 60 degrees, depending on the refractive index of the material constituting the outside of the optical element, light may escape from the optical element to the outside.
As shown in FIGS. 4A and 4B, it can be seen that when the incident angle increases, light may escape outward from the side wall corresponding to the outer periphery of the optical element. In order to prevent the passage of this light, a reflective layer is provided on the side wall (at a distance of 6 micrometers) of the optical element, and all the light is propagated in the optical element by reflecting the light with this reflective layer. Can do. With such a configuration, all the incident light can be confined in the optical element by the total reflection at the specific boundary and the reflection at the side wall in the optical element having the refractive index distribution. Light can be guided from the bottom surface of the optical element to the photosensitive element.

<Light trace in optical element (cylindrical shape) -2>
FIG. 5 shows another example of the light trace of the incident light within the optical element described above. The horizontal axis of the figure shows the distance from the center of the optical element (distance corresponds to 0 micrometer) to the outer periphery (distance corresponds to 6 micrometers). The vertical axis represents the depth from the top (upper opening) of the optical element. This figure shows a refractive index distribution as exemplified in FIG. 2B, that is, a case where the refractive index decreases linearly from the center toward the outer periphery. In the figure, the refractive index at the center is 2.0 (corresponding to Si ₃ N ₄ ), and the refractive index at the outer periphery is 1.4 (assuming that it corresponds to SiO ₂ ). In the figure, the incident angle of the light incident on the top of the optical element (see FIG. 3C) is used as a parameter. FIG. 4A shows a case where light is incident on the center of the optical element, and FIG. 5B shows a case where light is incident on a position 2 micrometers away from the center of the optical element toward the outer peripheral side.

  FIG. 5 shows a light trace simulation result similar to FIG. However, since it is assumed that the refractive index distribution is linearly continuous, the light trace is a continuous curve, and the state of total reflection is also shown. However, strictly speaking, total reflection does not occur, and the light trace travels vertically downward from the position where the light trace is vertical on the drawing. In the analysis, for convenience of calculation, the calculation is performed by dividing by 0.1 micrometers, and the occurrence of total reflection is observed because there is local fluctuation in the refractive index distribution even in an actual semiconductor image sensor. Become. Also in this figure, since the same interpretation as described in the above paragraph 0047 can be made, details are omitted.

  Furthermore, it is clear that similar results can be obtained even with a refractive index distribution that changes in a continuous curve as shown in FIG.

<Light trace in optical element (vertical inverted truncated cone shape)>
In FIG. 3 to FIG. 5, the light trace results are shown assuming that the optical element is cylindrical. The shape of the optical element is not limited to this, and may be another shape. For example, it may be a bucket shape (an upside down truncated cone) in which the optical element has an upward cross-sectional area. FIG. 6A shows a cross section of a bucket-shaped optical element. As will be described later, when an optical element is manufactured, a transparent insulating film is sequentially deposited from the side wall side. For this reason, in the case of the stepwise refractive index distribution as shown in FIG. 2A, the shape is as shown in FIG. FIG. 5B shows a simplified view of only a portion separated by a broken line in FIG. That is, the simulation results shown in FIGS. 4 to 5 can be applied to the bucket-shaped optical element described above. However, the sum of the incident light angle indicated by 50 and the side wall angle indicated by 51 in FIG. 6A corresponds to the incident angle in FIG. That is, the angle as the parameter in FIG. 4 needs to be read as “the value obtained by subtracting the side wall angle from the display angle in FIG. 4 is the angle corresponding to the actual incident light [50 in FIG. 6A]”. is there. For example, if the side wall angle is 20 degrees, the light trace of “30 degrees” in FIG. 4A corresponds to 10 degrees in FIG.

<The light trace in FIG. 4 corresponds to FIG. 6>
As described in the previous section, when the side wall angle is 20 degrees, when FIG. 4 is read corresponding to FIG.
(1) Incident light to the center of the optical element
1) Incident light from 0 degrees to 20 degrees is totally reflected at the surface where the refractive index changes in the optical element
Be done
2) Incident light of 30 degrees or more may leak from the outer periphery of the optical element. (2) Incident light at a location 2 micrometers away from the center of the optical element.
1) Incident light from 0 to 10 degrees is totally reflected at the surface where the refractive index changes in the optical element
Be done
2) It can be said that incident light of 20 degrees or more may leak from the outer peripheral portion of the optical element. The angle of incident light in this section is described in the coordinate system of FIG. 6A (corresponding to 50 in FIG. 6A). Further, for light leaking from the outer peripheral portion of the optical element, if a reflective layer is provided on the side wall of the optical element and reflected by the reflective layer into the optical element, light leakage can be prevented. .

  The analysis results shown in FIG. 4 to FIG. 6 indicate that light having a much larger incident angle can be efficiently collected as compared with the case of “Patent Document 1” cited in this specification. From another point of view, the degree of freedom in design in terms of the height of the optical element, the side wall angle, and the like is increased, and it can be said that the effectiveness of the present invention is high.

  From the simulation results shown in FIG. 4 and FIG. 5, according to the present invention, when the incident light beam is directed to the center of the optical element by the refractive index distribution and total reflection cannot be expected at the outer periphery of the optical element, The reflective layer allows incident light to be directed to the center of the optical element. As a result, it is possible to realize a semiconductor image sensor capable of efficient photoelectric conversion with respect to light rays having a wide range of incident angles and without being restricted by the size and shape of the optical element.

<Optical Element Manufacturing Process-1>
FIG. 7 is a diagram outlining a manufacturing process of a semiconductor image sensor including the optical element. In the figure, the same reference numerals as those in FIG. 1 denote the same components. In FIG. 2A, a photosensitive element 11 and a diffusion layer 12 composed of a photodiode or the like are formed on a semiconductor substrate 10, and then a first insulating layer 15, a gate electrode 13, and a light are formed using a well-known technique. A wiring layer 16 having a shielding function is formed. In the same figure (a), the case where the wiring layer 16 is a single layer is illustrated, but it may be composed of a plurality of layers. Subsequently, a second insulating layer 17 is formed on the entire surface including the wiring layer 16. Further, a third insulating layer 18 is laminated on 17. The constituent materials 17 and 18 may be a transparent resin or an insulating material such as an oxide film or a nitride film. The surfaces of the second insulating layer 17 and the third insulating layer 18 are generally flat, but need not be flat. In FIG. 2B, a recess 61 is formed above the photosensitive element 11 using a mask layer 60 made of a patterned photoresist, metal, insulating material, or the like. For this formation, chemical etching or the like is mainly used. Specific etching methods include wet etching with hydrofluoric acid as the main component, dry etching using plasma and reactive gas, and anisotropy with almost no change in cross-sectional shape in the depth direction as a kind of dry etching. Etching and isotropic etching with a change in cross-sectional shape in the depth direction can be mentioned. Moreover, you may combine a some etching technique for formation of the hollow 61. FIG.

  Note that the mask layer 60 is not necessary after the formation of the depression 61 and may be removed. If the mask layer 60 is opaque to incident light, it can be left as it can be used as a light shielding layer. Further, as another example of the opaque light shielding layer, a light shielding layer made of carbon, metal, or opaque resin is provided between the 60 and the 18, and a part of the light shielding layer is formed prior to the formation of the recess 61. May be removed by etching or the like.

  In FIG. 7C, the entire surface including the inside of the depression 61 is filled and covered with a transparent insulating material or resin material to form an optical element 62 and a surface layer 63. Since the surface layer 63 is an unnecessary element, it may be removed by chemical means such as etching or mechanical means such as polishing. However, if the surfaces of 62 and 63 are flat, they are not removed. Also good. If the step of removing 63 is employed, only the optical element 62 is formed inside the recess 61. Note that the optical element 62 needs to have optical transparency with respect to the light wavelength of the image detected by the photosensitive element. For example, a semiconductor image sensor for color imaging has a transparent insulating material or resin material in the visible light wavelength range, a semiconductor image sensor for black and white imaging has a transparent material in the near infrared wavelength range and the visible light wavelength range, In a semiconductor image sensor for infrared imaging, a transparent material is selected in the infrared wavelength region. A process for filling the recess 61 with these materials will be described later. In the following drawings, a configuration in which the resin of the surface layer 63 is removed is shown. In FIG. 7D, the color filter 20 is provided on the optical element 62 by a known technique. Further, the fourth insulating layer 21 is laminated on the surface of the color filter 20, and the surface of the entire configuration is flattened.

  In FIG. 7B, as for the cross-sectional shape of the depression 61, it is desirable that the area of the lower part (bottom face) is smaller than the upper part (upper opening) of the depression as shown in the figure. This shape becomes an important factor for the reflection of the incident light 25 shown in FIG. Regarding the difference in cross-sectional area, the size and shape of the pixel and the photosensitive element, and the thicknesses of the first insulating layer 15, the second insulating layer 17, and the third insulating layer 18 are taken into consideration. Determined.

<Optical Element Manufacturing Process-2>
In this section, the process of filling the optical element 62 shown in FIG. 7C and further realizing the distributed refractive index exemplified in FIGS. 2A to 2C will be outlined. The distributed refractive index, which is a feature of the present invention, is realized by depositing an insulating material while continuously changing atmospheric conditions such as gas, pressure, and power in a known vapor phase growth process. Specifically, at the initial stage of the vapor phase growth process, only SiO ₂ is deposited on the outer peripheral portion of the optical element. Then, a mixed film of SiO ₂ —Si ₃ N ₄ is deposited on the intermediate portion of the optical element by increasing the nitrogen concentration in the atmosphere and decreasing the oxygen concentration over time. In the final step, Si ₃ N ₄ is deposited in the center of the optical element by reducing the oxygen concentration to zero or extremely. By such a continuously variable method of atmospheric conditions, it is possible to realize the refractive index of the peripheral region of the optical element from approximately 1.45 of SiO ₂ to approximately 2.0 of Si ₃ N _{4 in} the central region. It becomes. Alternatively, a control method in which the atmosphere in the vapor phase growth process contains more Si with time can be applied. In the above description, an example in which the atmospheric condition is continuously changed is described. However, the refractive index is changed stepwise as shown in FIG. 2A by changing the atmospheric condition at regular intervals. It is also possible to realize the distribution.

The manufacturing conditions for the process are determined by the manufacturing equipment used, the atmospheric conditions during vapor phase growth, and the like. As an example, if the growth conditions of SiO ₂ are given, the power supplied to the upper electrode in the high frequency band = 300 W, the power supplied to the lower electrode = 200 W, the ambient temperature = 300 degrees Celsius, the vacuum = 70 Pascal, O ₂ ( The flow rate of the reaction gas is 1000 cubic centimeters per minute, N2 (reactive gas) is 100 cubic centimeters per minute, and TEOS (reactive gas) is 50 cubic centimeters per minute. As another example, if the growth conditions of Si ₃ N ₄ are given, the power supply in the high frequency band = 100 W, the ambient temperature = 300 degrees Celsius, the degree of vacuum = 130 Pascal, the flow rate of Si ₃ H 4 (reactive gas) = per minute 5 cubic centimeters, NH3 (reactive gas) = 100 cubic centimeters per minute, and the film formation speed under these conditions is about 3 angstroms per second.

In the previous section, the constituent material of the optical element is a combination of SiO ₂ and Si ₃ N ₄ , but the present invention is not limited to this combination. For example, other known materials such as Ta ₂ O ₅ can be used. Furthermore, a stepwise refractive index distribution as shown in FIG. 2A can be achieved by changing the constituent material of the optical element at each stage. For example, only SiO _2, mixed film of SiO ₂ and _Si 3 _{N 4,} only _Si 3 _{N 4,} and the combination of order as the only _Ta 2 _{O 5} is also possible.

<Second Embodiment>
FIG. 8 shows a second embodiment of the present invention and shows another method of forming the recess. In the figure, the same reference numerals as those in FIG. 1 denote the same components. In the figure, a recess 71 is formed above the photosensitive element 11 using a mask layer 70 made of a patterned photoresist, metal, insulating material, or the like. Typical examples of such forming means include wet etching and isotropic RIE (reactive ion etching). This means is characterized in that etching is performed to the lower part of the end of the mask layer 70, and the dimension of the uppermost portion of the 71 is larger than the patterning opening of the mask layer 70. Although not shown, an optical element is formed in the recess 71 in the subsequent process.

<Third Embodiment>
FIG. 9 shows a third embodiment of the present invention, which shows another method of forming the recess. In the figure, the same reference numerals as those in FIG. 1 denote the same components. In the figure, a depression 81 is formed above the photosensitive element 11 using a mask layer 80 made of a patterned photoresist, metal, insulating material, or the like. The depression 81 is formed in two steps, a first stage and a second stage. In the first stage, a vertical recess having a shape substantially equal to the shape of the patterning opening of the mask layer 80 is formed by anisotropic RIE or the like. Subsequently, in the second stage, etching proceeds in the lateral direction to the lower portion of the end portion of the mask layer 80 by isotropic RIE. Through these two steps, a plug-shaped depression as shown in the figure is realized. In the first-stage anisotropic RIE, the cross-sectional shape of the recess can be changed by changing the etching conditions. For this reason, as illustrated in FIG. 9, the cross-sectional area is gradually increased from the lower part (bottom face) to the upper part (upper opening) of the depression, or a vertical side wall shape is formed with substantially the same cross-sectional area. It is also possible. The present embodiment is characterized in that a plurality of etching means are combined to form a recess. Although not shown, an optical element is formed in the recess 81 in the subsequent process.

<Fourth Embodiment>
FIG. 10 shows a fourth embodiment of the present invention and shows another method for forming the recess. In the figure, the same reference numerals as those in FIG. 1 denote the same components. In the figure, a recess 91 is formed above the photosensitive element 11 by using a patterned photoresist, a mask layer 90 made of metal, an insulating material, or the like, and a side wall of the recess is made of a metal thin film or the like. A reflective layer 92 is provided. The reflective layer 92 is provided by a known means such as vapor deposition. In general vapor deposition means, a reflective layer may be formed up to the bottom of the recess 91, but it is necessary to remove only the reflective layer at the bottom using a technique such as RIE. On the other hand, metal particles may fly from an oblique direction using an improved vapor deposition apparatus, and a reflective layer may be formed only on the side wall. Although not shown, an optical element is formed in the recess 91 in the subsequent process. The present embodiment has an advantage that incident light can be efficiently reflected and light can be condensed on the photosensitive element 11 even when it is not possible to use the total reflection of light on the side wall of the optical element.

<Modification of Fourth Embodiment>
In the embodiment shown in FIG. 10, the reflective layer 92 is provided on the entire side wall of the recess 91. However, referring to the calculated light traces (FIGS. 4 and 5), it can be seen that reflection on the side wall by the reflective layer occurs above the side wall. In other words, the light directed downward of the side wall in the optical element is folded at a portion where the refractive index changes in the optical element before reaching the side wall, and is directed toward the central region of the optical element. For this reason, even if the reflective layer 92 is formed only in the upper part of the side wall, it does not affect the light collecting effect. In this case, the “deposition from an oblique direction” method described in the latter half of the previous section can be adopted. That is, the vapor deposition from the oblique direction has an advantage that the manufacturing process is simplified because the reflective layer is not formed at the bottom of the depression.

<Fifth Embodiment>
FIG. 11 is a fifth embodiment of the present invention and shows another configuration example of the pixel of the semiconductor image sensor. In the figure, the same reference numerals as those in FIG. 1 denote the same components. In the figure, the aspect ratio of the figure does not necessarily match the actual image sensor, and merely shows the configuration and operation of the pixels conceptually. In the figure, reference numeral 100 denotes a color filter provided on the second insulating layer 17. After forming this color filter, the third insulating layer 18 is laminated, and then the optical element 101 is formed. After the incident light 102 enters the optical element, the color is separated by the color filter 100 and reaches the photosensitive element 11. Unlike the configuration shown in FIG. 1, in this configuration, the bottom of the optical element is located away from the photosensitive element 11, but by providing a refractive index distribution in the optical element 101, the condensing function of incident light is reduced. Achieved.

<Sixth Embodiment>
FIG. 12 is a sixth embodiment of the present invention and shows another configuration example of the pixel of the semiconductor image sensor. In the figure, the same reference numerals as those in FIG. 1 denote the same components. In the figure, the aspect ratio of the figure does not necessarily match the actual image sensor, and merely shows the configuration and operation of the pixels conceptually. The configuration shown in the figure is characterized in that a microlens 110 is provided below the optical element 111. 110 can be formed of a resin material or an insulating material, but a material having a refractive index greater than a refractive index of 17 is selected. The incident light 112 is totally reflected at the specific boundary as described above due to the refractive index distribution in the optical element or reaches the bottom of the optical element due to reflection on the side wall, and then is refracted by the microlens 110 and collected. The light reaches the photosensitive element 11 after being illuminated. In this configuration, the microlens plays an auxiliary role in the light collecting function. This microlens prevents light that has passed through the bottom surface of the optical element from propagating in the lateral direction, which can contribute to improvement in photoelectric conversion efficiency. In addition, there is an advantage that the degree of freedom of design in the pixel configuration is further expanded.

<Manufacturing process of micro lens>
FIG. 13 is a diagram outlining the manufacturing method of FIG. 12, and the same reference numerals as those in FIG. 1 denote the same components. The structure shown in FIG. 5A created by the above-described process is subjected to dry or wet isotropic etching using a patterned mask layer 120 as shown in FIG. A concave lens-shaped recess 121 is formed. After this forming step, as shown in FIG. 5C, a resin material or an insulating material having a refractive index larger than the refractive index of the second insulating layer 17 is coated on the entire surface of the structure, and the lens layer 122 is formed. The Subsequently, as shown in FIG. 4D, the lens layer 122 is partially removed by a process called etch back, and is planarized so as to have substantially the same plane as 120, thereby forming the microlens 123. The The shape of 123 in the figure is different from the shape of 110 in FIG. 12, but the condensing function of incident light can be regarded as the same. In FIG. 4D, the mask layer 120 is not removed, but there is no need for removal as long as the surfaces of 120 and 123 are planarized. Further, the 120 does not only function as a mask during etching, but also has other functions, for example, a light shielding function made of a material having a low transmittance with respect to incident light. The mask layer 120 is preferably not removed. FIG. 11 (e) shows a manufacturing example for realizing the same shape as 110 in FIG. 12, unlike FIG. 12 (d). That is, 122 is provided after the mask layer 120 in FIG. 6B is removed, and the top surface is planarized by an etch back process to form 123.

<Seventh Embodiment>
FIG. 14 is a seventh embodiment of the present invention and shows another configuration example of the pixel of the semiconductor image sensor. In the figure, the same reference numerals as those in FIG. 1 denote the same components. In the figure, the aspect ratio of the figure does not necessarily match the actual image sensor, and merely shows the configuration and operation of the pixels conceptually. In the figure, reference numeral 131 denotes a microlens embedded in the upper opening surface of the optical element 130, which is formed in a manner similar to the process outlined in FIG. The condition is that the refractive index of 131 is larger than the refractive index of the optical element 130, and a material that satisfies the condition is selected. Also in this embodiment, the lens layer 131 functions as an auxiliary microlens and contributes to an increase in the collection efficiency of incident light.

<Eighth Embodiment>
FIG. 15 shows an eighth embodiment of the present invention. The same reference numerals as those in FIG. 1 denote the same components. This figure shows a case where three pixels illustrated in FIG. 1 are arranged in the horizontal direction to form an image sensor portion 140. 1 illustrates the pixel structure illustrated in FIG. 1, that is, the structure in which the color filter 20 is disposed above the optical element, the structures illustrated in FIGS. 11 to 14 may be used. A cover glass 141 is adhered and laminated on 140. As described above, in the present invention, since the surface of the semiconductor image sensor is essentially flat, the cover glass 141 can be adhered to the surface of the image sensor portion 140 as shown in FIG. This close contact structure prevents the semiconductor image sensor from being broken because the cover glass 141 functions as a reinforcing material against sudden application of force. Further, since the surface of the image sensor portion 140 can be mechanically cleaned before the cover glass 141 is laminated, it is advantageous from the viewpoint of preventing contamination.

  In this embodiment, the term “cover glass” is used. However, if it is described more strictly, it is “a flat plate having at least the characteristic of being transparent in a specific light wavelength region”. That is, the cover glass may have a plurality of functions other than the surface coating. For example, when a semiconductor image sensor is applied to a visible image capturing digital camera, an infrared cutoff filter function that optically removes near-infrared light that affects the imaging effect can be cited. In the case of monochrome imaging, the near-infrared light energy contained in the incident light may be actively used to increase the apparent light sensitivity. In such a case, the cover glass needs to be transparent to the wavelength range from visible light to near infrared light. In addition, in the case where the semiconductor image sensor is applied to an infrared imaging device whose purpose is thermal image detection, an infrared filter that removes visible light affecting the imaging effect can be cited. In these application examples, a filter layer is formed on the cover glass itself or on the front or back surface of the cover glass. As the structure of the filter layer, there are a method of directly forming an interference filter in which multiple layers of metal and dielectric thin films are laminated on the glass surface, and a method of stretching a plastic film laminated in multiple stages and stretching it on the glass surface. Another example of the function added to the cover glass is a color filter. In this case, a color filter array equal to the pixel pitch is formed on the front or back surface of the cover glass. That is, in FIG. 15, there is an example in which the color filter 20 is not formed as a part of the image sensor portion 140 but is formed as a part of the cover glass 141. In such a configuration, the color filter array may be directly formed on the cover glass surface by a well-known method, or the color filter array is formed on a thin glass plate different from the cover glass and then attached to the cover glass. Also good. The term “cover glass” in the present specification includes these functional examples exemplified in this section, and both are included in the present invention.

<Ninth Embodiment>
FIG. 16 shows a ninth embodiment of the present invention, which is an example of function addition to the cover glass outlined in the previous section. In the figure, the same numbers as those in FIG. 15 indicate the same components. In this embodiment, unlike the embodiment of FIG. 15, the color filter 20 is not mounted. In the figure, reference numeral 151 denotes a cover glass including a fluorescent film in a part thereof. This fluorescent film converts incident ultraviolet light 152 into visible light and emits it, and the visible light is converted into an image signal by the image sensor portion 140. Since the visible light radiation from the fluorescent film is not directional, the configuration of the optical elements arranged in 140 is preferably a configuration having a reflective layer on the side wall of the optical element as illustrated in FIG. Not as long. Further, in the present embodiment, an example in which the incident light is ultraviolet light is shown. However, the incident light is not limited to this, and an application example in which radiation light such as X-rays is incident has the configuration of 151. It can be applied by changing.

<Tenth Embodiment>
FIG. 17 shows a tenth embodiment of the present invention. The same reference numerals as those in FIG. 15 denote the same components. 1 illustrates the pixel structure illustrated in FIG. 1, that is, the structure in which the color filter is disposed on the optical element, the structures illustrated in FIGS. 11 to 14 may be used. In the present embodiment, the cover glass 161 includes convex lenses 162 arranged corresponding to the pixels. In this configuration, 162 functions as an auxiliary microlens, and the light collecting effect on the photodiode 11 can be further enhanced. Various known techniques can be used for the method of forming 162. For example, after processing the lower surface of the cover glass 161 into a depression on the concave lens, the material to be 162 is embedded. Here, in order to exhibit the function of a convex lens, it is necessary to select a material having a refractive index 162 larger than that of 161. However, unlike the conventional example, the convex lens 162 is only required to have an auxiliary function, and therefore, restrictions on its shape and characteristics are moderate. Also in this embodiment, since the surface of the image sensor portion 140 can be brought into close contact with the lower surface of the cover glass 161, various problems in the conventional microlens mounting example are solved.

  In the embodiment illustrated in FIGS. 15 to 17, the structure called a cover glass is shown as a flat plate having a uniform thickness, but this is not restrictive. That is, only a region corresponding to the pixel of the image sensor portion 140 is formed of a transparent material, and a region including an electronic circuit such as a drive system or a signal processing system is a metal, and the transparent material is hermetically sealed to the metal. There may also be a configuration. Such a configuration is particularly suitable for applications that require high reliability.

  In this specification, as shown in FIG. 1, a pixel configuration of an image sensor considered as a standard configuration is described. However, there are various image sensor configurations, and the present invention can be applied to all of them. For example, a so-called back-illuminated image sensor for increasing sensitivity, a so-called amplifying image sensor in which each pixel has an amplifying function or the photodiode itself has an amplifying function for increasing sensitivity. Configuration, CCD image sensor that reads signals from pixels via horizontal and vertical registers, and the image sensor itself has a laminated structure, and each layer has an imaging function, signal processing function, memory function, and input / output There are three-dimensional configurations in which control functions are assigned. All the configuration examples described here are included in the present invention.

  The pixel configuration of the semiconductor image sensor described in detail in this specification does not require a condensing microlens. In addition to application to digital cameras and mobile phones, this configuration is widely applied to infrared image sensors, optical semiconductor elements, semiconductor devices for optical signal processing, and even displays that emit light. it can.

DESCRIPTION OF SYMBOLS 1 Total reflection surface 2 Light-receiving part 3 Transparent material part 4 Low-refractive-index part 10 Semiconductor substrate 11 Photosensitive element 12 Diffusion layer 13 Gate electrode 14 Transistor 15 1st insulating layer 16 Wiring layer 17 2nd insulating layer 18 3rd insulation Layers 19, 62, 101, 111, 130 Optical element 20, 100 Color filter 21 Fourth insulating layer 22, 23 Pixel pitch 25, 102, 112 Incident light 27 Partial section 28 Stepwise refractive index distribution 29 Linear Refractive index distribution 30 Continuous refractive index distribution 50 Incident angle 51 Side wall angle 60, 70, 80, 90, 120 Mask layer 61, 71, 81, 91 Recess 63 Surface layer 92 Reflective layer 110, 123, 131 Microlens 121 Recess 122 Lens layer 140 Image sensor portion 141, 151, 161 Cover glass 152 UV 162 convex lens

Claims (6)

A semiconductor image sensor provided with a transparent optical element for condensing above each of the arranged photosensitive elements,
The semiconductor image sensor according to claim 1, wherein a refractive index of a central region of the optical element is higher than a refractive index of a peripheral region occupying outside the central region of the optical element. The optical element is composed of at least two kinds of transparent materials having different refractive indexes,
As going from the central region to the peripheral region,
2. The semiconductor image sensor according to claim 1, wherein the content ratio of the material for increasing the refractive index is decreased and the content ratio of the material for decreasing the refractive index is increased. The refractive index in a part of the optical element is
3. The semiconductor image sensor according to claim 1, wherein a part of the optical element includes a case where the optical element is determined by a composition ratio of the at least two kinds of transparent materials. As going from the central region to the peripheral region,
The optical element is
By gradually increasing the content of the low refractive index material,
The semiconductor image sensor according to claim 2, wherein the refractive index is lowered stepwise. As going from the central region to the peripheral region,
The optical element is
By continuously increasing the content of the low refractive index material,
The semiconductor image sensor according to claim 2, wherein the refractive index is continuously lowered. 6. The semiconductor according to claim 1, wherein the optical element is provided with a light reflecting layer on a side wall in contact with an insulating layer of the semiconductor image sensor in which the optical element is disposed. Image sensor.