Solid-state imaging device

Abstract

To provide a solid-state imaging device that can effectively reduce a variation in spectral response over a wide wavelength range.SOLUTION: A solid-state imaging device 1 includes a semiconductor substrate 20, which has a main surface 20a on which a plurality of light-sensitive areas 3 is provided and an insulating film 30, which is provided on the main surface 20a of the semiconductor substrate 20. A plurality of irregularities R is formed on a surface (main surface 30b) of the insulating film 30 on the opposite side of the main surface 20a of the semiconductor substrate 20. In the light-sensitive areas 3, there are multiple height differences of the irregularities R.SELECTED DRAWING: Figure 9

Description

[Technical field]

[0001] The present disclosure relates to a solid-state imaging device. [Background technology]

[0002] In solid-state imaging devices constituting image sensors such as CMOS, there are some that are sensitive to a wide wavelength range including, for example, the ultraviolet range to the near infrared range. In solid-state imaging devices sensitive to the ultraviolet range, an insulating film such as a BPSG (Boro-phospho silicate glass) film is formed as a protective film in the photosensitive region to suppress deterioration of the element due to ultraviolet light. In order for the insulating film to fully function as a protective film, for example, a thickness of about 1 μm is required. In this case, it is considered that interference of incident light occurs between the upper surface of the insulating film and the main surface of the semiconductor substrate, and the spectral sensitivity varies with respect to the wavelength of the incident light. To address this issue, for example, in the solid-state imaging device described in Patent Document 1, a base pattern of periodic convex portions is formed on the surface of the light receiving element, thereby forming one or more concaves and convexes having a height difference of about one wavelength or more of incident light on the surface of the insulating film. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 6-125068 Summary of the Invention [Problem to be solved by the invention]

[0004] In the solid-state imaging device described in the above-mentioned Patent Document 1, unevenness is formed on the surface of the insulating film, and the interference of the incident light in the insulating film is reduced, thereby reducing the variation in the spectral sensitivity to the wavelength of the incident light. However, in this conventional solid-state imaging device, the unevenness pattern appearing on the surface of the insulating film is uniform, so the effect of reducing the interference of the incident light in the insulating film is limited. Therefore, further ingenuity is required to effectively reduce the variation in the spectral sensitivity in a wide wavelength range including, for example, the ultraviolet range to the near-infrared range.

[0005] The present disclosure has been made to solve the above-mentioned problems, and has an object to provide a solid-state imaging device that can effectively reduce variations in spectral sensitivity over a wide wavelength range. [Means for solving the problem]

[0006] A solid-state imaging device according to one aspect of the present disclosure comprises a semiconductor substrate having a main surface on which a plurality of photosensitive regions are provided, and an insulating film provided on the main surface of the semiconductor substrate, wherein a plurality of projections and recesses are formed on the surface of the insulating film opposite the main surface of the semiconductor substrate, and the photosensitive regions have a plurality of height differences between the projections and recesses.

[0007] In this solid-state imaging device, in the photosensitive region, the multiple unevennesses provided in the insulating film have multiple height differences. Due to the presence of multiple height differences, when incident light enters the photosensitive region, multiple interferences with different optical path lengths occur in the insulating film. As a result, the periods of variation in the spectral sensitivity to the wavelength of the incident light cancel each other out, making it possible to reduce the sensitivity variation in a wide wavelength region including, for example, the ultraviolet region to the near-infrared region.

[0008] In the photosensitive region, when the height difference of the unevenness is measured at intervals of 0.01 μm, the maximum occurrence frequency of the surface height of the insulating film based on the top of the insulating film may be 5% or less. In this case, the unevenness of the unevenness of the insulating film will be various, and the effect of canceling out the periods of the variations in the spectral sensitivity to the wavelength of the incident light will be enhanced. Therefore, it is possible to more effectively reduce the sensitivity variations in a wide wavelength region including, for example, the ultraviolet region to the near infrared region.

[0009] In the photosensitive region, when the height difference of the unevenness is measured at intervals of 0.01 μm, the standard deviation of the occurrence frequency of the surface height of the insulating film based on the top of the insulating film may be 1% or less. In this case, the unevenness of the unevenness of the insulating film will be various, and the effect of canceling out the periods of the variations in the spectral sensitivity to the wavelength of the incident light will be enhanced. Therefore, it is possible to more effectively reduce the sensitivity variations in a wide wavelength region including, for example, the ultraviolet region to the near infrared region.

[0010] In the photosensitive region, the height difference between the top and bottom of the unevenness may be 0.5 μm or more and 0.9 μm or less. When this height difference is 0.5 μm or more, the optical path length difference of the incident light in the insulating film becomes sufficiently large, so that the effect of canceling out the periods of the variations in the spectral sensitivity to the wavelength of the incident light is further enhanced. Furthermore, when this height difference is 0.9 μm or less, the depth of the recess is not excessive, and it is possible to suppress the absolute value of the spectral sensitivity from changing due to the refraction and reflection of the incident light on the surface of the insulating film.

[0011] The main surface of the semiconductor substrate may be a flat surface, in which case the shape of the main surface of the semiconductor substrate is simplified, thereby reducing the cost of the solid-state imaging device. Effect of the Invention

[0012] According to the present disclosure, it is possible to effectively reduce the variation in spectral sensitivity over a wide wavelength range. [Brief description of the drawings]

[0013] [Figure 1] FIG. 1 is a plan view showing a schematic configuration of a solid-state imaging device. [Diagram 2] 2 is a cross-sectional view taken along line II-II in FIG. [Diagram 3] FIG. 2 is a plan view showing a concave-convex pattern of an insulating film. [Figure 4] FIG. 4 is an enlarged view of a main part of FIG. [Diagram 5] 5A is a cross-sectional view taken along line C1-C2 in FIG. 4, and FIG. 5B is a cross-sectional view taken along line C1-C3 in FIG. [Figure 6] 5 is a schematic cross-sectional view showing how projections and recesses are formed in an insulating layer. FIG. [Figure 7] 11 is a schematic cross-sectional view showing the relationship between heat treatment and the spacing between projections and recesses. FIG. [Figure 8] 11 is a schematic cross-sectional view showing the relationship between heat treatment and the spacing between projections and recesses. FIG. [Figure 9] 5 is a graph showing a profile of height differences between projections and recesses of the pattern shown in FIG. 4. [Figure 10] 5 is a graph showing the frequency of occurrence of the surface height of an insulating film in the pattern shown in FIG. [Figure 11] 5 is a graph showing the spectral sensitivity characteristics in the ultraviolet to near infrared range of a solid-state imaging device having the concaves and convexes of the pattern shown in FIG. [Figure 12] 5 is a graph showing an enlarged view of the spectral sensitivity characteristics in the ultraviolet region of a solid-state imaging device having the concaves and convexes of the pattern shown in FIG. [Figure 13] FIG. 11 is a plan view showing a concave-convex pattern of an insulating film according to a comparative example. [Figure 14] FIG. 14 is an enlarged view of a main part of FIG. [Figure 15] 15 is a cross-sectional view taken along line D1-D2 in FIG. 14. [Figure 16] 15 is a graph showing a profile of height differences between projections and recesses of the pattern shown in FIG. 14. [Figure 17] 15 is a graph showing the frequency of occurrence of the surface height of an insulating film in the pattern shown in FIG. 14. [Figure 18]15 is a graph showing the spectral sensitivity characteristics in the ultraviolet to near infrared range of a solid-state imaging device having the concaves and convexes of the pattern shown in FIG. [Figure 19] 15 is a graph showing an enlarged view of the spectral sensitivity characteristics in the ultraviolet region of a solid-state imaging device having the concaves and convexes of the pattern shown in FIG. 14. [Figure 20] 1 is a graph showing the spectral sensitivity characteristics in the ultraviolet to near infrared range of a solid-state imaging device having an insulating film with no irregularities. [Figure 21] 11 is a graph showing an enlarged view of the spectral sensitivity characteristics in the ultraviolet region of a solid-state imaging device having an insulating film with no irregularities. [Figure 22] 1 is a graph showing the relationship between the maximum frequency of occurrence of the surface height of an insulating film and peaks and valleys of the spectral sensitivity. [Figure 23] 1 is a graph showing a shift in the peaks and valleys of spectral sensitivity. [Figure 24] 1 is a graph showing the relationship between the standard deviation of the occurrence frequency of the surface height of an insulating film and the peaks and valleys of the spectral sensitivity. [Diagram 25] FIG. 13 is a diagram showing the behavior of peaks and valleys of spectral sensitivity when the height difference between the highest and lowest points of the unevenness is changed. [Figure 26] 4 is a schematic cross-sectional view showing the change in unevenness before and after heat treatment. FIG. [Figure 27] 13 is an enlarged view of a main part showing a modified example of the concave and convex pattern of the insulating film. FIG. [Figure 28] FIG. 13 is an enlarged view of a main portion showing another modified example of the concave-convex pattern of the insulating film. [Figure 29] FIG. 13 is an enlarged view of a main portion showing another modified example of the concave-convex pattern of the insulating film. [Diagram 30] FIG. 13 is an enlarged view of a main portion showing another modified example of the concave-convex pattern of the insulating film. [Diagram 31] FIG. 13 is an enlarged view of a main portion showing another modified example of the concave-convex pattern of the insulating film. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Hereinafter, preferred embodiments of a solid-state imaging device according to one aspect of the present disclosure will be described in detail with reference to the drawings. [Schematic configuration of solid-state imaging device]

[0015] Fig. 1 is a plan view showing a schematic configuration of a solid-state imaging device. Fig. 2 is a cross-sectional view taken along line II-II in Fig. 1. As shown in Fig. 1, the solid-state imaging device 1 includes a plurality of photosensitive regions 3, a plurality of transfer gate sections 5, a plurality of anti-blooming gate sections 7, a plurality of anti-blooming drain sections 9, and a shift register section 11. These components are formed on a main surface 20a of a semiconductor substrate 20. The solid-state imaging device 1 of this embodiment is, for example, a front-illuminated image sensor, and one photosensitive region 3 constitutes one pixel.

[0016] Each photosensitive region 3 generates charges according to the intensity of the incident light in response to the incidence of light. That is, the photosensitive region 3 functions as a photoelectric conversion unit. In this embodiment, the planar shape of the photosensitive region 3 is a rectangle formed by two long sides and two short sides. The multiple photosensitive regions 3 are aligned along a second direction (direction along the short side direction of the photosensitive region 3) perpendicular to a first direction along the long side direction of the photosensitive region 3, and are arranged in an array in a one-dimensional direction. The shape of the photosensitive region 3 is not limited to the approximately rectangular shape described above, and may be various shapes.

[0017] Each transfer gate section 5 corresponds to a respective photosensitive region 3 and is disposed on one short side of the planar shape of the photosensitive region 3. That is, the multiple transfer gate sections 5 are lined up in the second direction on one short side of the planar shape of the photosensitive region 3. The transfer gate sections 5 acquire charges generated in the photosensitive region 3 and transfer the acquired charges in the first direction as signal charges. An isolation region 13 is disposed between adjacent transfer gate sections 5. The isolation region 13 provides electrical isolation between the transfer gate sections 5.

[0018] Each anti-blooming gate portion 7 corresponds to a photosensitive region 3, and is disposed on the other short side of the planar shape of the photosensitive region 3. That is, the multiple anti-blooming gate portions 7 are lined up in the second direction on the other short side of the planar shape of the photosensitive region 3. The anti-blooming gate portion 7 acquires charges generated in the photosensitive region 3, and transfers the acquired charges in the first direction as unwanted charges. The isolation region 13 is disposed between adjacent anti-blooming gate portions 7. The isolation region 13 provides electrical isolation between the anti-blooming gate portions 7.

[0019] Each anti-blooming drain section 9 corresponds to a corresponding one of the anti-blooming gate sections 7, and is disposed adjacent to the anti-blooming gate section 7 in the first direction. That is, the anti-blooming drain sections 9 are lined up in the second direction on the other short side of the planar shape of the photosensitive region 3. The anti-blooming drain sections 9 are connected to a predetermined fixed potential, and discharge unnecessary charges transferred from the corresponding anti-blooming gate section 7.

[0020] The shift register units 11 correspond to the multiple transfer gate units 5, respectively, and are arranged adjacent to the transfer gate units 5 in the first direction. That is, the multiple shift register units 11 are lined up in the second direction on the other short side of the planar shape of the photosensitive region 3. The shift register units 11 receive the signal charges transferred from the transfer gate units 5, transfer them in the second direction, and sequentially output them to the read amplifier unit 15. The signal charges output from the shift register units 11 are converted into voltages by the read amplifier unit 15, and are output to the outside of the solid-state imaging device 1 as voltages for each of the photosensitive regions 3 arranged in the second direction.

[0021] A light-shielding film LS is disposed in the area excluding the multiple photosensitive regions 3. In this embodiment, the light-shielding film LS is disposed so as to cover the transfer gate section 5, the anti-blooming gate section 7, the anti-blooming drain section 9, and the shift register section 11. The light-shielding film LS prevents light from entering these areas, and can prevent the generation of unnecessary charges due to light entering these areas.

[0022] In the photosensitive region 3, as shown in FIG. 2, an insulating film 30 is provided on a main surface 20a of a semiconductor substrate 20. The semiconductor substrate 20 has a main surface 20a and a main surface 20b that face each other. In this embodiment, the main surface 20a and the main surface 20b are both flat surfaces, and the main surface 20a is a light incident surface of the semiconductor substrate 20. The semiconductor substrate 20 includes, from the main surface 20b side, a p-type semiconductor region 21, a p-type semiconductor region 22, an n+ type semiconductor region 23, a p+ type semiconductor region 24, and an oxide film 25. In this embodiment, the semiconductor substrate 20 is made of Si. When the semiconductor substrate 20 is made of Si, a Group 3 element such as B is used as the p-type impurity, and a Group 5 element such as P or As is used as the n-type impurity.

[0023] The oxide film 25 is, for example, a silicon oxide film. This oxide film 25 functions as a gate oxide film of a MOS transistor in the transfer gate portion 5. The oxide film 25 also has a role of preventing components from the insulating film 30 from penetrating into the semiconductor substrate 20 in the photosensitive region 3. For example, when the insulating film 30 is a BPSG film as described later, the oxide film 25 prevents B (boron) and P (phosphorus) from the BPSG film from penetrating into the semiconductor substrate 20.

[0024] In addition, when the oxide film 25 is provided on the semiconductor substrate 20, the surface of the oxide film 25 can be regarded as the main surface 20a of the semiconductor substrate 20. When the oxide film 25 is not provided on the semiconductor substrate 20, the surface of the p+ type semiconductor region 24 becomes the main surface 20a of the semiconductor substrate 20. The semiconductor regions constituting the semiconductor substrate 20 are not limited to the configuration shown in FIG. 2. For example, the p+ type semiconductor region 24 may not be provided, and the n+ type semiconductor region 23 may be located on the p- type semiconductor region 22, and the oxide film 25 may be formed directly on the n+ type semiconductor region 23.

[0025] The insulating film 30 has a main surface 30a and a main surface 30b facing each other. The main surface 30a faces the main surface 20a of the semiconductor substrate 20, and the main surface 30b faces the opposite side to the main surface 20a of the semiconductor substrate 20. The insulating film 30 can be formed by, for example, deposition. The main surface 30a follows the shape of the main surface 20a of the semiconductor substrate 20 and is a flat surface. On the other hand, a plurality of irregularities R are formed on the main surface 30b. The irregularities R are formed in a gently continuous shape by forming irregularities with a rectangular cross section by etching or mechanical polishing, and then flowing a part of the irregularities with a rectangular cross section by heat treatment. The insulating film 30 is, for example, a BPSG (Boro-phospho silicate glass) film, and also functions as an anti-reflection film (AR film).

[0026] The thickness of the insulating film 30 varies from one portion to another due to the multiple projections and recesses R, but is, for example, 0.5 μm to 3 μm. The lower limit of the thickness of the insulating film 30 is determined in consideration of the function as a protective film, that is, sufficient ultraviolet light resistance. The upper limit of the thickness of the insulating film 30 is determined in consideration of process limitations (e.g., ease of forming contact holes), etc. [Details of the unevenness of the insulating film]

[0027] FIG. 3 is a plan view showing the uneven pattern of the insulating film. FIG. 4 is an enlarged view of the main part. FIG. 3 shows the photosensitive region 3 for three pixels, and the first direction shown in FIG. 1 corresponds to the Y direction, the second direction corresponds to the X direction, and the thickness direction of the semiconductor substrate 20 shown in FIG. 2 corresponds to the Z direction. In FIG. 3 and FIG. 4, the region of the unevenness R corresponding to the convex portion 31 is shown as a solid color, and the region corresponding to the concave portion 32 is shown as a dotted color. When the height position of the top 31a of the convex portion 31 is 1 and the height position of the bottom 32a of the concave portion 32 is 0, the region corresponding to the convex portion 31 is a region whose height position is 0.5 or more and 1 or less, and the region corresponding to the concave portion 32 is a region whose height position is 0 or more and less than 0.5.

[0028] 3 and 4, in the photosensitive region 3, a plurality of irregularities R are formed on the main surface 30b side of the insulating film 30 by sinusoidally undulating the main surface 30b of the insulating film 30. In the example of Fig. 3 and Fig. 4, recesses 32 having a substantially square shape in a plan view are arranged in a matrix in the in-plane direction of the main surface 30b of the insulating film 30, and the remaining portions are protrusions 31. In each of the X direction and the Y direction, one side of the recesses 32 is about 3.0 µm, and the pitch between the recesses 32, 32 is about 3.0 µm.

[0029] In the photosensitive region 3, there are a plurality of height differences of the unevenness R. For example, when looking at a cross section (see FIG. 4) from the center C1 of one recess 32 to the center C2 of the next recess 32 in the Y direction, as shown in FIG. 5(a), the thickness T1 of the insulating film 30 at the bottom 32a of the recess 32 is about 0.85 μm, and the height difference of the unevenness R (the thickness from the bottom 32a of the recess 32 to the top 31a of the protrusion 31) T2 is about 0.7 μm. The maximum thickness (=T1+T2) of the insulating film 30 is about 1.55 μm.

[0030] Also, for example, when looking at a cross section (see FIG. 4) from the center C1 of one recess 32 to the center C3 of the next recess 32 adjacent in the X and Y directions, as shown in FIG. 5(b), the thickness T1 of the insulating film 30 at the bottom 32a of the recess 32 is about 0.85 μm, and the height difference T2 of the unevenness R (the thickness from the bottom 32a of the recess 32 to the top 31a of the protrusion 31) is about 0.9 μm. The maximum thickness (= T1 + T2) of the insulating film 30 is about 1.75 μm.

[0031] Furthermore, the height difference F between the top and bottom of the unevenness R (see FIG. 9) is 0.5 μm or more and 0.9 μm or less. The top refers to the top 31a of the multiple protrusions 31 that has the highest height from the main surface 20a of the semiconductor substrate 20, and the bottom refers to the bottom 32a of the multiple recesses 32 that has the lowest height from the main surface 20a of the semiconductor substrate 20.

[0032] The unevenness R of the insulating film 30 is formed by, for example, dry etching and heat treatment. First, as shown in FIG. 6(a), a resist 33 is formed by photolithography on the main surface 30b of the insulating film 30 at positions corresponding to the formation positions of the convex portions 31. Next, as shown in FIG. 6(b), the portions of the insulating film 30 where the resist 33 is not formed are dry etched to form the concave portions 32. After the concave portions 32 are formed, the resist 33 is removed. Then, the insulating film 30 is heat-treated to cause a portion of the unevenness R having a rectangular cross section to flow, thereby forming gently continuous unevenness R on the main surface 30b side of the insulating film 30 as shown in FIG. 6(c).

[0033] According to the above manufacturing method, the amount of flow of the irregularities R during heat treatment is constant, so that the greater the pitch between the recesses 32, 32 before heat treatment, the higher the height position of the apex 31a of the protrusions 31, and the smaller the pitch between the recesses 32, 32 before heat treatment, the lower the height position of the apex 31a of the protrusions 31. In other words, the height difference T2 of the recesses 32 can be adjusted by the pitch between the recesses 32, 32 before heat treatment.

[0034] In the example of Fig. 7(a), before the heat treatment, the width W1 of the protrusion 31 and the width W2 of the recess 32 are both 3.0 μm, and the thickness T2 from the bottom 32a of the recess 32 to the top 31a of the protrusion 31 is 1.4 μm. In this case, after the heat treatment, as shown in Fig. 7(b), the width W1 of the protrusion 31 and the width W2 of the recess 32 are both maintained at 3.0 μm, and the height difference T2 of the recess 32 is 0.7 μm. On the other hand, in the example of Fig. 8(a), before the heat treatment, the width W1 of the protrusion 31 and the width W2 of the recess 32 are both 4.2 μm, and the thickness T2 from the bottom 32a of the recess 32 to the top 31a of the protrusion 31 is 1.4 μm. In this case, after the heat treatment, as shown in FIG. 8(b), the width W1 of the protrusions 31 and the width W2 of the recesses 32 are both maintained at 4.2 μm, and the height difference T2 of the recesses 32 is 0.9 μm.

[0035] 9 is a graph showing a height difference profile of the unevenness of the pattern shown in FIG. 4. In the profile in this figure, the horizontal axis represents the position in the X direction, and the vertical axis represents the height position of the insulating film 30. On the vertical axis, the height position of the top 31a of the convex portion 31 is set as the reference (=0 μm). This height difference profile of the unevenness R was obtained by scanning a depth gauge along a scanning line G1 set on the diagonal line of the photosensitive region 3 for one pixel. By scanning the depth gauge along the scanning line G1, a height difference profile equivalent to that obtained when the scanning line is slid obliquely over one concave portion 32 and the surrounding convex portions 31 to measure the depth can be obtained.

[0036] As shown in FIG. 9, in the pattern of the unevenness R shown in FIG. 4, the height position of the top 31a of the convex portion 31 is 0 μm at the center of the scanning direction, gradually decreases from the center to the end of the scanning direction, and is −0.2 μm at both ends of the scanning direction. The height position of the bottom 32a of the concave portion 32 is −0.2 μm at the center of the scanning direction, gradually decreases from the center to the end of the scanning direction, and is −0.9 μm at both ends of the scanning direction. From the result of FIG. 9, it can be seen that the difference in height position between the top position and the bottom position of one amplitude in the height difference profile is small at the center of the scanning direction and gradually increases from the center to the end. That is, it can be considered that there are multiple height differences of the unevenness R of the main surface 30b of the insulating film 30 in the photosensitive region 3.

[0037] In the photosensitive region 3, when the height difference of the unevenness R is measured at intervals of 0.01 μm, the maximum value of the occurrence frequency of the surface height of the insulating film 30 based on the top of the insulating film 30 is 5% or less. FIG. 10 is a graph showing the occurrence frequency of the surface height of the insulating film in the pattern shown in FIG. 4. In the figure, the horizontal axis shows the surface height of the insulating film, and the vertical axis shows the occurrence frequency. The depth resolution when calculating the occurrence frequency is set to 0.01 μm. As shown in the figure, in the unevenness R pattern shown in FIG. 4, the surface height of the insulating film is widely distributed in the range from 0 μm to -0.9 μm, and the occurrence frequency peaks near -0.2 μm. The occurrence frequency is about 4% even near the peak position of -0.2 μm, and is about 1.5% to 4% in the ranges from -0.2 μm to 0 μm and from -0.2 μm to -0.3 μm, and is 1% or less in the range from -0.4 μm to -0.9 μm. From this result, it can be seen that in the unevenness R of the pattern shown in FIG. 4, the height difference of the unevenness is not biased to a specific value, but rather includes height differences of various values ​​more or less uniformly. [Effect of unevenness of insulating film]

[0038] In the solid-state imaging device 1 having the above-mentioned configuration, since the unevenness R provided on the main surface 30b side of the insulating film 30 in the photosensitive region 3 has a plurality of height differences, when the incident light enters the photosensitive region 3, a plurality of interferences with different optical path lengths are generated in the insulating film 30. As a result, the periods of variation in the spectral sensitivity to the wavelength of the incident light cancel each other out, and it becomes possible to reduce the sensitivity variation in a wide wavelength region including, for example, the ultraviolet region to the near-infrared region.

[0039] In addition, in the solid-state imaging device 1, when the height difference of the unevenness R in the photosensitive region 3 is measured at intervals of 0.01 μm, the maximum occurrence frequency of the surface height of the insulating film based on the top of the insulating film 30 is 5% or less. This results in a wide variety of height differences of the unevenness R in the insulating film 30, enhancing the effect of canceling out the periods of variation in the spectral sensitivity to the wavelength of incident light. Therefore, it becomes possible to more effectively reduce the sensitivity variation in a wide wavelength range including, for example, the ultraviolet range to the near-infrared range.

[0040] In the solid-state imaging device 1, the height difference F between the top and bottom of the unevenness R in the photosensitive region 3 is 0.5 μm or more and 0.9 μm or less. Since the height difference F is 0.5 μm or more, the optical path length difference of the incident light in the insulating film 30 becomes sufficiently large, and the effect of the periods of the variations in the spectral sensitivity to the wavelength of the incident light canceling each other out is further enhanced. Furthermore, since the height difference F is 0.9 μm or less, the depth of the recesses 32 is not excessive, and it is possible to suppress the absolute value of the spectral sensitivity from changing due to the refraction and reflection of the incident light on the surface of the insulating film 30.

[0041] In the solid-state imaging device 1, the main surface 20a of the semiconductor substrate 20 on which the insulating film 30 is provided is a flat surface. This simplifies the shape of the main surface 20a of the semiconductor substrate 20, thereby reducing the cost of the solid-state imaging device 1.

[0042] 11 and 12 are graphs showing the spectral sensitivity characteristics of a solid-state imaging device having the unevenness of the pattern shown in FIG. 4. FIG. 11 shows the spectral sensitivity characteristics in a wide wavelength range including the ultraviolet region to the near infrared region, and FIG. 12 shows the spectral sensitivity characteristics in the ultraviolet region in an enlarged manner. In FIG. 11 and FIG. 12, the horizontal axis shows wavelength, and the vertical axis shows sensitivity, and the spectral sensitivity characteristics of four samples having the same configuration as the above embodiment are plotted in an overlapping manner. As shown in FIG. 11 and FIG. 12, in the example, the periods of the variation in the spectral sensitivity with respect to the wavelength of the incident light cancel each other out, so that no peaks or valleys are generated in the spectral sensitivity profile, and it can be confirmed that the variation in the spectral sensitivity can be effectively reduced in a wide wavelength range including the ultraviolet region to the near infrared region. It can also be confirmed that the variation in the spectral sensitivity characteristics between the samples is very small.

[0043] On the other hand, Fig. 13 is a plan view showing a pattern of projections and recesses of an insulating film according to a comparative example. Fig. 14 is an enlarged view of a main part thereof. In the example of Figs. 13 and 14, the projections and recesses R have convex portions 31 extending in a strip shape in the X direction and concave portions 32 extending in a strip shape in the Y direction in a plan view, and are configured by arranging the projections 31 and the recesses 32 alternately in the Y direction. The width of the projections 31 and the width of the recesses 32 in the Y direction are both 3.0 μm.

[0044] In this comparative example, for example, when looking at a cross section (see FIG. 14) from the center D1 of one recess 32 to the center D2 of the next recess 32 adjacent in the Y direction, the thickness T1 of the insulating film 30 at the bottom 32a of the recess 32 is about 0.85 μm, and the height difference T2 of the unevenness R (the thickness from the bottom 32a of the recess 32 to the top 31a of the protrusion 31) is about 0.9 μm, as shown in FIG. 15. The maximum thickness (=T1+T2) of the insulating film 30 is about 1.75 μm.

[0045] Fig. 16 is a graph showing the height difference profile of the unevenness of the pattern shown in Fig. 14. This height difference profile of the unevenness R was acquired by scanning a depth gauge along a scanning line G2 (see Fig. 13) set in the Y direction of a photosensitive region for one pixel. In the pattern of the unevenness R shown in Fig. 14, the height position of the top 31a of the convex portion 31 is 0 μm near the center of the convex portion 31 in the Y direction, and the height position of the bottom 32a of the concave portion 32 is -1 μm near the center of the concave portion 32 in the Y direction. From the results in Fig. 16, it can be seen that the height difference of the unevenness R is uniform in the pattern shown in Fig. 14.

[0046] FIG. 17 is a graph showing the occurrence frequency of the surface height of the insulating film of the pattern shown in FIG. 14. In the same figure, as in the case of FIG. 10, the depth resolution for calculating the occurrence frequency is set to 0.01 μm. As shown in the figure, in the pattern of the unevenness R shown in FIG. 14, the occurrence frequency of the surface height of the insulating film is concentrated in the range from 0 μm to -0.1 μm, and the occurrence frequency peak located near 0 μm reaches about 7%. From this result, it can be seen that in the unevenness R of the pattern shown in FIG. 14, the height difference of the unevenness is biased to a specific value.

[0047] 18 and 19 are graphs showing the spectral sensitivity characteristics of a solid-state imaging device having the unevenness of the pattern shown in FIG. 14. FIG. 18 shows the spectral sensitivity characteristics in a wide wavelength range including the ultraviolet region to the near infrared region, and FIG. 19 shows the spectral sensitivity characteristics in the ultraviolet region in an enlarged manner. In FIG. 18 and FIG. 19, similarly to the cases of FIG. 11 and FIG. 12, the spectral sensitivity characteristics of four samples having the same configuration as the comparative example are plotted in an overlapping manner. As shown in FIG. 18 and FIG. 19, in the comparative example, it is difficult for the periodic variations in the spectral sensitivity to the wavelength of incident light to cancel each other out, and it is found that peaks and valleys occur in the spectral sensitivity profile compared to the example. It is also found that the variation in the spectral sensitivity characteristics between the samples is larger compared to the example.

[0048] 20 and 21 are graphs showing the spectral sensitivity characteristics of a solid-state imaging device according to a reference example. In this reference example, the spectral sensitivity characteristics of four samples having an insulating film with no irregularities are plotted in an overlapping manner. As shown in Fig. 20 and Fig. 21, in the reference example, the effect of the periods of the variations in the spectral sensitivity with respect to the wavelength of the incident light cancel each other out is not achieved, and large peaks and valleys are generated in the spectral sensitivity profile compared to the comparative example, and the variation in the spectral sensitivity characteristics between the samples is even larger. [Considerations regarding unevenness of insulating film]

[0049] In the above-described embodiment, the maximum value of the frequency of occurrence of the surface height of the insulating film 30 based on the top of the insulating film 30 is 5% or less. In calculating the maximum value of the frequency of occurrence of the surface height of the insulating film 30, the relationship between the maximum value of the frequency of occurrence of the surface height of the insulating film 30 and the peaks and valleys of the spectral sensitivity was examined. Here, the peaks and valleys of the spectral sensitivity were determined as the average absolute value of the amount of variation in the spectral sensitivity for each wavelength over the entire wavelength range (200 nm to 1000 nm). The absolute value of the amount of variation in the spectral sensitivity for each wavelength is calculated as ΔS / S×100(%), where S is the average value of the spectral sensitivity at a certain wavelength, and ΔS is the difference between S and the maximum or minimum value of the spectral sensitivity at a certain wavelength.

[0050] FIG. 22 is a graph showing the relationship between the maximum occurrence frequency of the surface height of the insulating film and the peaks and valleys of the spectral sensitivity. In the figure, the horizontal axis shows the maximum occurrence frequency of the surface height of the insulating film, and the vertical axis shows the peaks and valleys of the spectral sensitivity. As shown in FIG. 22, there is a certain correlation between the maximum occurrence frequency of the surface height of the insulating film and the peaks and valleys of the spectral sensitivity, and it can be seen that the smaller the maximum occurrence frequency of the surface height of the insulating film, the smaller the peaks and valleys of the spectral sensitivity. From the results of FIG. 22, in the range where the maximum occurrence frequency of the surface height of the insulating film is 5% or less, the peaks and valleys of the spectral sensitivity are suppressed to 1% or less. Therefore, it can be confirmed that the maximum occurrence frequency of the surface height of the insulating film being 5% or less is a significant condition for reducing the variation in the spectral sensitivity.

[0051] The reason why the depth resolution for calculating the occurrence frequency of the surface height of the insulating film is set to 0.01 μm is because the peaks and valleys of the spectral sensitivity shift when the height difference of the unevenness is changed. FIG. 23 is a graph showing the shift of the peaks and valleys of the spectral sensitivity. As shown in the figure, the peaks and valleys of the spectral sensitivity are inverted when the waveform of the spectral sensitivity shifts by 1 / 2 a period. For example, at a wavelength of 200 nm, when the height difference of the unevenness changes by 0.04 μm, the waveform of the spectral sensitivity shifts by 1 / 2 a period accordingly. Therefore, by measuring the depth of the unevenness with a resolution of 0.01 μm, which is 1 / 4 of 0.04 μm, it is possible to calculate the occurrence frequency of the phenomenon in which the peaks and valleys of the spectral sensitivity are inverted.

[0052] The wavelength of 200 nm is near the lower limit of the measurement wavelength band of a general light receiving sensor. At wavelengths longer than this, the height difference of the unevenness required to shift the spectral sensitivity waveform by 1 / 2 period becomes larger. Therefore, if the depth resolution for calculating the occurrence frequency is specified based on a wavelength of 200 nm, a solid-state imaging device targeting incident light on the longer wavelength side will have sufficient resolution.

[0053] Furthermore, when the height difference of the unevenness R in the photosensitive region 3 is measured at intervals of 0.01 μm, the standard deviation of the occurrence frequency of the surface height of the insulating film 30 based on the top of the insulating film 30 may be 1% or less. In this case, too, there will be many types of height difference of the unevenness R in the insulating film 30, and the effect of the periods of variation in the spectral sensitivity to the wavelength of incident light canceling each other out is enhanced. Therefore, it is possible to more effectively reduce the sensitivity variation in a wide wavelength range including, for example, the ultraviolet range to the near infrared range.

[0054] FIG. 24 is a graph showing the relationship between the standard deviation of the occurrence frequency of the surface height of the insulating film and the peaks and valleys of the spectral sensitivity. In the figure, the horizontal axis shows the standard deviation of the occurrence frequency of the surface height of the insulating film, and the vertical axis shows the peaks and valleys of the spectral sensitivity. As shown in FIG. 24, there is a certain correlation between the standard deviation of the occurrence frequency of the surface height of the insulating film and the peaks and valleys of the spectral sensitivity, and it can be seen that the smaller the standard deviation of the occurrence frequency of the surface height of the insulating film, the smaller the peaks and valleys of the spectral sensitivity. From the results of FIG. 24, in the range where the standard deviation of the occurrence frequency of the surface height of the insulating film is 1% or less, the peaks and valleys of the spectral sensitivity are suppressed to 1% or less. Therefore, it can be confirmed that the standard deviation of the occurrence frequency of the surface height of the insulating film being 1% or less is a significant condition for reducing the variation in the spectral sensitivity.

[0055] In the above-mentioned embodiment, the height difference F (see FIG. 9) between the top and bottom of the unevenness R is 0.5 μm or more and 0.9 μm or less. In this regard, FIG. 25 shows the behavior of the peaks and valleys of the spectral sensitivity when the height difference F is changed. As shown in the figure, when the height difference F is 0.1 μm, the optical path length difference of the incident light in the insulating film is small, so the shift amount of the spectral sensitivity based on each optical path length difference is small. Therefore, it is considered that the effect of the periodicity of the variation in the spectral sensitivity canceling each other out occurs only in a limited wavelength range. On the other hand, when the height difference F is 0.5 μm and 0.9 μm, the optical path length difference of the incident light in the insulating film is sufficiently large, so the shift amount of the spectral sensitivity based on each optical path length difference is large. Therefore, the effect of the periodicity of the variation in the spectral sensitivity for the wavelength of the incident light canceling each other out is improved, and it is possible to further reduce the peaks and valleys of the spectral sensitivity.

[0056] When the height difference F exceeds 0.9 μm, the optical path length difference of the incident light in the insulating film becomes larger, but on the other hand, the depth of the recess becomes excessive, and it is considered that the absolute value of the spectral sensitivity changes due to refraction and reflection of the incident light on the surface of the insulating film. Therefore, when the height difference F is in the range of 0.5 μm to 0.9 μm, it is possible to reduce the sensitivity variation in a wide wavelength range including, for example, the ultraviolet region to the near infrared region without causing the problem of the change in the absolute value of the spectral sensitivity.

[0057] The interval between the projections and recesses R before the heat treatment is preferably 2 μm to 4 μm. As shown in FIG. 26, when the interval between the projections and recesses R before the heat treatment is 1.0 μm (here, the width of the projections 31 and the width of the recesses 32 are both 1.0 μm), the amount of the insulating film 30 flowing from the projections 31 to the recesses 32 during the heat treatment becomes excessive, and the main surface 30b of the insulating film 30 after the heat treatment tends to be flattened. When the interval between the projections and recesses R before the heat treatment is 3.0 μm (here, the width of the projections 31 and the width of the recesses 32 are both 3.0 μm), the amount of the insulating film 30 flowing from the projections 31 to the recesses 32 during the heat treatment becomes appropriate, and the main surface 30b of the insulating film 30 after the heat treatment corrugates in a sinusoidal wave shape, so that the projections and recesses R as shown in FIG. 3 and FIG. 4 are formed.

[0058] On the other hand, when the interval between the projections and recesses R before the heat treatment is 5.0 μm (here, the width of the projections 31 and the width of the recesses 32 are both 5.0 μm), the amount of flow of the insulating film 30 from the projections 31 to the recesses 32 during the heat treatment is insufficient, so that although the insulating film 30 after the heat treatment has projections and recesses, flat portions tend to remain at the tops 31a of the projections 31 and the bottoms 32a of the recesses 32. In this case, it is considered that the occurrence frequency of each height difference of the projections and recesses R tends to be biased to a specific value. Therefore, as described above, by setting the interval between the projections and recesses R before the heat treatment to 2 μm to 4 μm, it becomes easy to have various height differences of the projections and recesses R in the insulating film 30. Note that the amount of flow of the insulating film 30 from the projections 31 to the recesses 32 during the heat treatment also changes depending on the temperature during the heat treatment and the impurity concentration in the insulating film 30, but as long as the heat treatment uses a standard manufacturing process for solid-state imaging devices, it is considered that the appropriate range of the interval between the projections and recesses R can be generalized. [Modification of the unevenness of the insulating film]

[0059] Various modifications can be applied to the unevenness R of the insulating film 30. For example, in Fig. 3 and Fig. 4, the recesses 32 having a substantially square shape in a plan view are arranged in a matrix in the in-plane direction of the main surface 30b of the insulating film 30, but as shown in Fig. 27(a), the recesses 32 having a substantially circular shape in a plan view may be arranged in a matrix in the in-plane direction of the main surface 30b of the insulating film 30. As shown in Fig. 27(b), the recesses 32 having a substantially equilateral triangular shape in a plan view may be arranged in a matrix in the in-plane direction of the main surface 30b of the insulating film 30, and as shown in Fig. 27(c), the recesses 32 having a substantially equilateral hexagonal shape in a plan view may be arranged in a matrix in the in-plane direction of the main surface 30b of the insulating film 30.

[0060] 28(a), the positions of the projections 31 and the recesses 32 may be reversed with respect to the configurations of FIGS. 3 and 4. That is, the projections 31 having a substantially square shape in plan view may be arranged in a matrix in the in-plane direction of the main surface 30b of the insulating film 30. In this case, if the pitch between the projections 31, 31 is large, the amount of flow of the insulating film 30 during heat treatment may be insufficient. Therefore, as shown in FIG. 28(b), it is also preferable to reduce the pitch between the projections 31, 31 while maintaining the area of ​​the projections 31, and ensure a sufficient area of ​​the projections 31 relative to the photosensitive region 3.

[0061] In addition, the pattern of the unevenness R of the insulating film 30 does not necessarily have to be a two-dimensional pattern, and unevenness R of a one-dimensional pattern can also be adopted. For example, in the example of Fig. 29(a), the unevenness R has convex parts 31 extending in a strip shape in the X direction in a plan view and concave parts 32 extending in a strip shape in the Y direction, and is configured by arranging the convex parts 31 and the concave parts 32 alternately in the Y direction. In the example of Fig. 29(a), unlike the comparative examples shown in Figs. 13 and 14, pairs of convex parts 31 and concave parts 32 with a width of 3.0 µm in the Y direction and pairs of convex parts 31 and concave parts 32 with a width of 2.5 µm in the Y direction are arranged alternately in the Y direction.

[0062] Furthermore, as shown in Fig. 29(b), in the unevenness R of the one-dimensional pattern, the widths in the Y direction of the protrusions 31 and recesses 32 may be changed along the X direction. In the example of Fig. 29(b), the width in the Y direction of the protrusions 31 gradually increases from 2.5 µm to 3.5 µm in the X direction, and conversely, the width in the Y direction of the recesses 32 gradually decreases from 3.5 µm to 2.5 µm in the X direction.

[0063] Even in these configurations, since there are multiple height differences in the unevenness R provided on the main surface 30b side of the insulating film 30, multiple interferences with different optical path lengths occur within the insulating film 30 when incident light enters the photosensitive region 3. This causes the periods of variation in the spectral sensitivity to the wavelength of the incident light to cancel each other out, making it possible to reduce the sensitivity variation in a wide wavelength region including, for example, the ultraviolet region to the near-infrared region.

[0064] Also, as shown in FIG. 30, the height positions of the bottoms 32a of the recesses 32 may be different in the unevenness R of the one-dimensional pattern. In the example of FIG. 30, the widths of the protrusions 31 and the recesses 32 in the Y direction are both 3.0 μm. Also, the recesses A whose bottoms 32a are relatively low and the recesses 32B whose bottoms 32a are relatively high are arranged alternately on either side of the protrusions 31. In this case, the height positions of the recesses 32 are not uniform, so that even if the protrusions 31 and the recesses 32 are arranged at a uniform pitch in the Y direction, it is possible to make the unevenness R have a plurality of height differences. The recesses 32A and 32B whose bottoms 32a are different in height positions can be easily manufactured by, for example, changing the amount of dry etching of the insulating film 30 before heat treatment. In the embodiment of FIG. 29 and FIG. 30, it is possible to suppress crosstalk to adjacent pixels caused by refraction at the interface between the protrusions 31 and the recesses 32. Furthermore, even if the pitch in the X direction in the photosensitive region 3 becomes small, the unevenness R can be easily formed.

[0065] In the examples of Fig. 29 and Fig. 30, the one-dimensional pattern of unevenness is arranged in the Y direction, but as shown in Fig. 31, the one-dimensional pattern of unevenness R may be arranged in the X direction. In the example of Fig. 31, the unevenness R is formed by an unevenness row 35A in which the convex portions 31 and the concave portions 32 having a width of 3.0 μm in the X direction are arranged in the X direction, an unevenness row 35B in which the convex portions 31 and the concave portions 32 having a width of 2.5 μm in the X direction are arranged in the X direction, an unevenness row 35C in which the convex portions 31 and the concave portions 32 having a width of 2.0 μm in the X direction are arranged in the X direction, and an unevenness row 35D in which the convex portions 31 and the concave portions 32 having a width of 1.5 μm in the X direction are arranged in the X direction. In the Y direction, the unevenness rows 35A to 35D are arranged in this order from both end sides of the photosensitive region 3, and the unevenness rows 35D are adjacent to each other in the center. Between each of the unevenness rows 35A to 35D, the convex portions 31 are arranged with a width of 1 μm.

[0066] Even in these configurations, since there are multiple height differences in the unevenness R provided on the main surface 30b side of the insulating film 30, multiple interferences with different optical path lengths occur within the insulating film 30 when incident light enters the photosensitive region 3. This causes the periods of variation in the spectral sensitivity to the wavelength of the incident light to cancel each other out, making it possible to reduce the sensitivity variation in a wide wavelength region including, for example, the ultraviolet region to the near-infrared region. [Explanation of symbols]

[0067] 1... solid-state imaging device, 3... photosensitive region, 20... semiconductor substrate, 20a... main surface, 30... insulating film, 30b... main surface, R... unevenness.

Claims

1. a semiconductor substrate having a main surface on which a plurality of photosensitive regions are provided; an insulating film provided on the main surface of the semiconductor substrate; a plurality of gently continuing projections and recesses are formed on a surface of the insulating film opposite to a main surface of the semiconductor substrate; In the photosensitive region, the unevenness has a plurality of height differences, A solid-state imaging device in which, when the height difference of the unevenness is measured at intervals of 0.01 μm in the photosensitive region, the maximum frequency of occurrence of the surface height of the insulating film based on the topmost part of the insulating film is 5% or less.

2. a semiconductor substrate having a main surface on which a plurality of photosensitive regions are provided; an insulating film provided on the main surface of the semiconductor substrate; a plurality of gently continuing projections and recesses are formed on a surface of the insulating film opposite to a main surface of the semiconductor substrate; In the photosensitive region, the unevenness has a plurality of height differences, A solid-state imaging device in which, when the height difference of the unevenness is measured at intervals of 0.01 μm in the photosensitive region, the standard deviation of the occurrence frequency of the surface height of the insulating film based on the topmost part of the insulating film is 1% or less.

3. 3. The solid-state imaging device according to claim 1, wherein the height difference between the top and bottom of the irregularities in the photosensitive region is 0.5 [mu]m or more and 0.9 [mu]m or less.

4. 4. The solid-state imaging device according to claim 1, wherein the main surface of the semiconductor substrate is a flat surface.

Images