JP2011022432A - Multilayer film optical filter, solid imaging element, imaging apparatus, display device, and communication device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a multilayer film optical filter which is advantageous in terms of improving the transmitting efficiency while assuring condensing effect. <P>SOLUTION: The multilayer film optical filter 10 is constituted by alternately laminating a first dielectric layer 12 and a second dielectric layer 14 having different refractive indices and constituted to include a multilayer film which transmits a light having a specified wavelength selectively. In all the first dielectric layers 12 and the second dielectric layers 14, the optical film thickness of each of the first dielectric layer 12 and the second dielectric layer 14 adjacent to each other in the lamination direction is formed so as to be 1/4 of the wavelength λ of the light to be selectively transmitted. At least a portion in the thickness direction of the multilayer film, that is a direction in which the first and the second dielectric layers 12, 14 are laminated, is formed as a spherical portion 20 where each of the first and the second dielectric layers 12, 14 is formed along a spherical surface exhibiting a projected shape in the same direction in the direction in which the first and the second dielectric layers 12, 14 are laminated. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a multilayer optical filter, a solid-state imaging device, an imaging device, a display device, and a communication device.

Opportunities to use imaging devices such as digital still cameras and video recorders that capture and image a subject with a two-dimensional solid-state imaging device are increasing.
Currently, there are CCD (Charge Coupled Device) type and CMOS (Complementary Metal Oxide Semiconductor) type solid state imaging devices as mainstream solid state imaging devices.
In these image pickup devices, each pixel collects an optical signal from a subject and converts it into a signal charge, and the charge is taken out of the device as digital data or analog data and imaged.
These solid-state imaging devices are generally sensitive to specific electromagnetic wave wavelengths. For example, a solid-state imaging device based on silicon has sensitivity to wavelengths shorter than near infrared rays (˜1.1 μm).
In the case of X-rays, which are electromagnetic waves having higher energy than visible wavelengths, a plurality of charges are generated in a cascade by one photon, so that the silicon-based solid-state imaging device has energy resolution (wavelength resolution).
On the other hand, silicon has no energy resolution (wavelength resolution) with respect to the electromagnetic wave wavelength in the visible wavelength to near-infrared region where the wavelength is several hundred nm to 1.1 μm. ) Cannot be determined whether a large amount of light has accumulated.
As described above, the CCD solid-state imaging device and the CMOS solid-state imaging device have high sensitivity in the visible wavelength range, but cannot distinguish color information such as red, green, and blue light like human eyes. .

Therefore, in general color imaging devices, the following method is often used in order to acquire a color image.
(1) An on-chip color filter that selectively transmits a specific color component for each pixel of a two-dimensional element is provided, and the light intensity in a plurality of colors (wavelengths) in an adjacent pixel group of about 2 × 2 pixels (vertical and horizontal two pixels). Information is acquired and restored as a color image by demosaic processing in the subsequent stage.
(2) After separating the light into wavelength components through a spectral element such as a dichroic prism, the light intensity data of each color component is acquired by a plurality of (for example, three) monochrome sensors, and color images are formed by subsequent signal processing. .

In particular, in the digital still camera for consumer use and the camera module for mobile phone, the method (1) is generally adopted.
That is, transmissive color filters of three colors R (red), G (green), and B (blue) are arranged in a staggered pattern for a 2 × 2 pixel unit, and light of each color component in each pixel. It is common to take a technique of detecting the intensity and forming a color image by demosaic processing.
When a plurality of, for example, three RGB imaging elements are used for one imaging device, the system (2) is used for broadcasting applications and the like because the system becomes large and complicated and expensive. Often used in relatively expensive camcorders.

  As a combination of color filters, the following filters may be used in addition to the RGB color filters. For example, a white filter (no filter), a complementary color system (cyan, magenta, yellow), a near-infrared filter, an ultraviolet filter, or a filter that further categorizes the visible wavelength range and acquires each wavelength component can be used.

In color filters used in such solid-state imaging devices, organic materials such as pigments and dyes are often used.
However, the binding energy of molecules containing carbon and hydrogen, which are the constituent elements of these color filters, is about the same as that of ultraviolet energy, and when irradiated with high energy light for a long time, carbon bonds and bonds between carbon and hydrogen are broken. There is a case.
Therefore, when used outdoors for a long time exposed to sunlight including ultraviolet rays, when used in an environment where ultraviolet rays are particularly strong (for example, highland, snowy field, sea), or exposed to high temperature and humidity for a long time. It has been pointed out that the following inconvenience occurs. That is, a change occurs in the transmission characteristics of the color filter, resulting in deterioration of the color reproduction characteristics of the captured image (see Non-Patent Document 1).

On the other hand, color filters made of materials other than organic materials include the following.
The first example is a thin film filter that utilizes the wavelength dependence of the absorption coefficient of an inorganic substance.
As an example, there is an amorphous silicon thin film (see Patent Document 1).
Since the absorption coefficient in the visible wavelength region of amorphous silicon becomes smaller as the wavelength increases, when considering a thin film having a specific thickness, an electromagnetic wave having a wavelength = 0.7 μm is more easily transmitted than an electromagnetic wave having a wavelength = 0.5 μm.
Therefore, a long pass filter that preferentially transmits long wavelengths can be easily realized by controlling the film thickness.
Since amorphous silicon is an inorganic material, it can overcome the shortcomings of organic color filters, such as deterioration due to ultraviolet rays, but on the other hand, it is difficult to realize a short-pass filter that transmits only short wavelengths, and RGB color separation requires subsequent arithmetic processing. It becomes necessary and the problem remains in terms of color reproducibility.

The second example is a multilayer optical filter (photonic filter) using a photonic crystal (see Patent Documents 2 and 3).
The multilayer optical filter is configured by periodically laminating optical materials having different refractive indexes. Then, by adjusting the interval between the periodically laminated optical materials, multiple reflection can be selectively generated with respect to the electromagnetic wave having a specific wavelength, and the electromagnetic wave having an arbitrary wavelength and bandwidth can be taken out.
Therefore, it can be said that the multilayer optical filter is an excellent optical element in that an arbitrary wavelength can be extracted by adjusting the physical size.
In this case, the thickness of each of the laminated optical materials having different refractive indexes needs to be λ / 4n (n is the refractive index of the optical material) of the wavelength to be transmitted. In other words, it is necessary to set the optical film thickness of the laminated optical materials having different refractive indexes to λ / 4 of the wavelength to be transmitted.
Here, the optical film thickness is a value indicated by the product nd of the refractive index n of the optical material and the physical film thickness d of the optical material.

The multilayer optical filter has a high degree of freedom in design and a chemically stable structure, and thus has superior characteristics as compared with an organic filter.
However, there are problems. The multilayer optical filter laminates multilayer films along a plane parallel to a two-dimensional plane formed by the solid-state imaging device.
On the other hand, in many solid-state imaging devices, in order to improve the light collection efficiency, each pixel has an on-chip micro convex lens (refer to Patent Document 4) or a light collecting device (on-chip lens) made of a subwavelength structure. (See Patent Documents 5 and 6).

Tokushui WO06 / 028128 Special reproduction WO05 / 013369 JP 2007-103401 A Patent No. 02600250 Patent No. 3547665 JP 2006-351972

IEEE Electron Device Letters, Vol.27, No.6, June 2006, p457-459

In the above-described conventional multilayer optical filter, light collection by an on-chip lens becomes a problem.
That is, the light that has entered the on-chip lens is collected by the lens at the center of each pixel. In other words, when considering vertically incident light, light is incident perpendicularly to the vicinity of the center of the on-chip lens, while the light in each lens peripheral region has a wavefront inclined with respect to the center of the pixel.
Therefore, when viewed from obliquely incident light, the interval between the multilayer films is not strictly the film thickness of λ / 4n of the optical material, but becomes a value larger than λ / 4n.
As a result, light cannot interfere efficiently, and as a result, there is a disadvantage that the transmission efficiency of the multilayer optical filter is lowered.
The present invention has been made in view of such circumstances, and an object of the present invention is to provide a multilayer optical filter that is advantageous in improving the transmission efficiency while ensuring the light collecting effect.
Another object of the present invention is to provide a signal display device and an information communication device using such a multilayer optical filter.
Another object of the present invention is to provide a solid-state imaging device using such a multilayer optical filter and an imaging device using such a solid-state imaging device.

The multilayer optical filter of the present invention includes a multilayer film that is configured by alternately laminating first dielectric layers and second dielectric layers having different refractive indexes and selectively transmits light of a specific wavelength. The optical film thickness of each of the first dielectric and the second dielectric adjacent to each other in the stacking direction is formed to be ¼ of the wavelength of the selectively transmitted light. , At least part of the thickness direction of the multilayer film, which is the direction in which the first dielectric layer and the second dielectric layer are laminated, is the first dielectric layer and the second dielectric layer, respectively. Is formed as a spherical portion formed along a spherical surface having a convex shape in the same direction in the direction of stacking.
The solid-state imaging device of the present invention includes a multilayer optical filter and a photoelectric conversion unit that photoelectrically converts light transmitted through the multilayer optical filter, and the multilayer optical filter includes a first dielectric having a different refractive index. The first dielectric is formed by alternately laminating layers and second dielectric layers, includes a multilayer film that selectively transmits light of a specific wavelength, and is adjacent in the direction of lamination And the second dielectric layer is formed so that the optical film thickness thereof is ¼ of the wavelength of the selectively transmitted light, and the first dielectric layer and the second dielectric layer are laminated. At least part of the thickness direction of the multilayer film, which is a direction in which the first dielectric layer and the second dielectric layer are projected in the same direction in the direction in which they are laminated, is a spherical surface. Shaped as a spherical part formed along It is.
The imaging apparatus of the present invention includes an imaging unit having a solid-state imaging device, a control unit that controls the imaging unit, and an operation unit that operates the imaging unit, and the solid-state imaging device includes a multilayer optical filter. And a photoelectric conversion unit that photoelectrically converts light that has passed through the multilayer optical filter, wherein the multilayer optical filter includes first and second dielectric layers having different refractive indexes that are alternately stacked. Each optical film thickness of each of the first dielectric and the second dielectric adjacent to each other in the stacking direction is configured to include a multilayer film that selectively transmits light of a specific wavelength. , Which is formed to be ¼ of the wavelength of the selectively transmitted light, and in the thickness direction of the multilayer film, which is a direction in which the first dielectric layer and the second dielectric layer are stacked. At least in part, the first dielectric layer and the second dielectric layer Each conductor layer is formed as a spherical portion formed along a spherical surface which exhibits a convex in the same direction in the direction in which a stack thereof.
Further, the display device of the present invention selectively transmits a luminance image generating unit that generates a luminance image whose luminance is modulated for each pixel, and light that forms the luminance image for each wavelength that is preset for each pixel. A filter array, wherein the filter array is constituted by a multilayer optical filter, and the multilayer optical filter is formed by alternately laminating first dielectric layers and second dielectric layers having different refractive indexes. The optical film thickness of each of the first dielectric and the second dielectric adjacent to each other in the stacking direction is configured to include a multilayer film that selectively transmits light having a specific wavelength. At least part of the thickness direction of the multilayer film, which is formed to be ¼ of the wavelength of light to be transmitted and is a direction in which the first dielectric layer and the second dielectric layer are laminated Is the first dielectric layer Each of the fine said second dielectric layer is formed as a spherical portion formed along a spherical surface which exhibits a convex in the same direction in the direction in which a stack thereof.
The communication apparatus of the present invention includes at least one of an optical signal transmission unit that transmits an optical signal and an optical signal detection unit that receives an optical signal, and the at least one of the optical signal transmission unit or the optical signal detection unit includes: A filter that selectively transmits light having a specific wavelength in the light included in the optical signal, the filter being formed of a multilayer optical filter, and the multilayer optical filter having a refractive index different from the first The first dielectric layer and the second dielectric layer are alternately stacked, include a multilayer film that selectively transmits light of a specific wavelength, and are adjacent to each other in the stacking direction. Each of the dielectric film and the second dielectric film is formed to have an optical film thickness that is ¼ of the wavelength of the selectively transmitted light, and the first dielectric layer, the second dielectric layer, Is the direction in which the layers are stacked At least part of the thickness direction of the layer film is formed along a spherical surface in which each of the first dielectric layer and the second dielectric layer has a convex shape in the same direction in the stacking direction. It is formed as a spherical portion.

According to the present invention, at least a part of the multilayer film in the thickness direction is along a spherical surface in which each of the first dielectric layer and the second dielectric layer has a convex shape in the same direction in the stacking direction. It is formed as a spherical portion formed in this way.
Therefore, the light path difference when passing through the multilayer film hardly occurs regardless of the light passing through the vicinity of the center of the multilayer film and the light passing through the periphery of the multilayer film. In this case, the distance between the multilayer films substantially coincides with the optical film thickness.
Therefore, each light is efficiently interfered in the multilayer film, which is advantageous in improving the transmission efficiency while ensuring the light collecting effect of the multilayer film optical filter.

It is sectional drawing which shows the basic structure of the conventional multilayer optical filter. FIG. 2A is a schematic cross-sectional view of one pixel of a solid-state imaging device provided with an organic color filter 202, and FIG. 2B is a schematic cross-sectional view of one pixel of a solid-state imaging device provided with a general multilayer optical filter 205. is there. (A) is a schematic cross-sectional view of one pixel of a solid-state imaging device provided with a conventional multilayer optical filter as a comparative example, and (B) is a solid-state imaging device 1 provided with the multilayer optical filter of the first embodiment. It is a cross-sectional schematic diagram of a pixel. It is a diagram which shows the transmission characteristic when light with a visible wavelength of 400 nm to 700 nm is vertically incident on the multilayer optical filter as a TE mode. It is a diagram which shows the transmission characteristic when light with a visible wavelength of 400 nm to 700 nm is vertically incident on a multilayer optical filter as a TM mode. (A) is sectional drawing which shows the structure of the multilayer optical filter 10 of 1st Embodiment, FIG.6 (B) is sectional drawing which shows the structure of the multilayer optical filter 10 of 2nd Embodiment. It is sectional drawing which shows the structure of the multilayer optical filter 10 of 3rd Embodiment. It is sectional drawing which shows the structure of the multilayer optical filter 10 of 4th Embodiment. It is sectional drawing which shows the structure of the multilayer optical filter 10 in the modification of 4th Embodiment. It is sectional drawing which shows the structure of the multilayer optical filter 10 of 5th Embodiment. FIG. 11 is a cross-sectional view showing the configuration of the multilayer optical filter 10 in a modification of the fifth embodiment. It is sectional drawing of the CMOS type solid-state image sensor in which the multilayer optical filter 10 in 6th Embodiment was mounted. It is sectional drawing of the CMOS type solid-state image sensor in which the multilayer optical filter 10 in 7th Embodiment was mounted. It is sectional drawing of the CMOS type solid-state image sensor in which the multilayer film optical filter 10 in 8th Embodiment was mounted. It is sectional drawing of the multilayer optical filter 10 in 9th Embodiment. It is sectional drawing of the multilayer optical filter 10 in 10th Embodiment. (A) is sectional drawing which shows the CMOS type solid-state image sensor comprised similarly to 6th Embodiment (FIG. 12), (B) is the multilayer optical filter 10 of 11th Embodiment mounted. FIG. 5C is a cross-sectional view of a CMOS solid-state image sensor, and FIG. 6C is a cross-sectional view of a CMOS solid-state image sensor on which an optical filter 34 composed of an inorganic thin film filter 1205 (or an organic filter 1205) is mounted. It is sectional drawing which shows the structure of the multilayer film optical filter 10 of 12th Embodiment. It is a perspective view which shows the structure of each sub wavelength structure in FIG. It is sectional drawing of the two-dimensional solid-state image sensor which provided both the multilayer filter 10 and the polarizing filter 50 in 13th Embodiment. FIG. 21A is an explanatory diagram showing a general RGB filter arrangement structure in which transmitted light is separated into three wavelength bands of R, G, and B by the multilayer optical filter 10, and FIG. 21B is a wavelength in this RGB filter. It is a diagram which shows the relationship between a transmittance | permeability. (A) is explanatory drawing which shows the arrangement structure of the filter which isolate | separates transmitted light into seven wavelength bands with the multilayer optical filter 10, (B) is a diagram which shows the relationship between the wavelength and transmittance | permeability in this filter. It is explanatory drawing which shows the outline of a bias sputtering device. It is a block diagram which shows the structure of the imaging device 400 in 16th Embodiment. It is a block diagram which shows the structure of the electronic device 70 which has the display apparatus 60 in 17th Embodiment. It is a block diagram which shows the structure of the communication apparatus 90 in 18th Embodiment.

(First embodiment)
Next, a first embodiment of the present invention will be described with reference to the drawings.
For convenience of explanation, first, the basic structure of a conventional multilayer optical filter will be described with reference to FIGS. 1 and 2, and then the multilayer optical filter of the first embodiment will be described.
As shown in FIG. 1, reference numeral 101 is an upper reflector, and reference numeral 103 is a lower reflector.
Light enters from above and transmits downward.
Reference numeral 102 denotes a spacer layer.
Each of the upper reflector 101 and the lower reflector 103 is a multilayer film composed of four layers, that is, a total of eight layers. The white portion indicates the low refractive index portion 2 and the shaded portion indicates the high refractive index portion 4.
The number of layers is not limited to 8 and may be any number such as 6 layers, 10 layers, and 12 layers.
The optical film thickness of each of the layers 2 and 4 included in each of the upper reflector 101 and the lower reflector 103 is λ / 4 of the incident electromagnetic wave wavelength to be transmitted through the multilayer optical filter. The physical film thickness of each of 4 is λ / 4n (where n is the refractive index of the medium).
When the light transmitted through the multilayer optical filter has a visible wavelength (550 nm), the physical film thickness of each of the layers 2 and 4 is 137.5 [nm] / n.
In other words, when the refractive index of the low refractive index medium is n1 and the refractive index of the high refractive index medium is n2, the physical film thickness of each of the layers 2 and 4 is 137.5 / n1 [nm], 137.5 / n2 [nm].
If the physical film thicknesses of the layers 2 and 4 are d1 and d2, the relationship of d1 · n1 = d2 · n2 = optical film thickness is established between them.
Here, the optical film thickness is an index obtained by multiplying the film thickness d of each of the layers 2 and 4 (dielectric layers) by the refractive index n.

As an example, when the low refractive index layer 10 is made of SiO ₂ , the refractive index in the visible wavelength region is 1.45, so the physical film thickness is preferably about 94.8 [nm].
If the high refractive index layer 12 is made of silicon nitride (Si ₃ N ₄ ), the refractive index in the same wavelength region is 2.05, so the physical film thickness is preferably about 67.1 [nm].
Further, when the high refractive index layer 12 is made of titanium dioxide (TiO ₂ ), the refractive index in the same wavelength region is 2.5, so the film thickness is preferably about 55 [nm].
The larger the ratio of the refractive index between the low refractive index medium and the high refractive index medium, the wider the reflection / transmission characteristics and bandwidth of the multilayer optical filter.
Reference numeral 102 denotes a spacer layer, and the transmission wavelength can be adjusted by adjusting the thickness of this layer. The medium constituting the spacer layer 102 may be a third medium having the same refractive index as the low refractive index medium, the same as the high refractive index medium, or a refractive index different from both media.

FIG. 2 is a schematic cross-sectional view of one pixel of the solid-state imaging device.
FIG. 2A shows a pixel provided with an organic color filter 202, and FIG. 2B shows a pixel provided with a general multilayer optical filter 205.
Reference numerals 201 and 204 denote on-chip convex lenses, which are typical condensing elements for efficiently collecting light on the photosensors 203 and 208.
Reference numeral 202 denotes a color filter made of an organic material, which is generally a layer containing a nanosize pigment that transmits R, G, B light and absorbs other light, for example.
Reference numeral 205 denotes a multilayer optical filter.
The multilayer optical filter 205 is a structure in which a set of periodic structures 206 of a high refractive index portion indicated by oblique lines and a low refractive index portion indicated by whitening are set as one set.
Here, each optical film thickness constituting one period of the periodic structure 206 is 1/4 of the incident electromagnetic wave wavelength λ, that is, one period corresponds to a half wavelength in the optical film thickness.
The multilayer optical filter 205 is composed of an upper reflector and a lower reflector, a spacer layer 207 is inserted between them, and by changing the thickness of the spacer layer 207, a desired electromagnetic wave is transmitted, A bandpass filter that does not transmit other wavelengths is realized.
The multilayer optical filter 205 may be made of any material as long as it is transparent in the electromagnetic wave band to be transmitted, and can be easily realized with an inorganic material. Therefore, the multilayer optical filter is resistant to external stress such as ultraviolet rays and is chemically stable. Is superior in that it can be realized.

FIG. 3A is a schematic cross-sectional view of one pixel of a solid-state image sensor provided with a conventional multilayer optical filter as a comparative example.
FIG. 3A schematically shows how incident light A, B, and C are converged by the on-chip condensing element 6. Reference numeral 8 denotes a solid-state image sensor.
The multilayer film has a structure in which the low refractive index layer 2 and the high refractive index layer 4 are laminated at an interval corresponding to the optical film thickness of λ / 4, and the optical thickness of the spacer layer 207 between the multilayer films. Determines the wavelength to be transmitted.
As shown in FIG. 3A, even when the physical film thickness of each of the layers 2 and 4 of the multilayer film is strictly set to λ / 4n (n is the refractive index of the medium), there are the following disadvantages. .
That is, in the light A that passes through the vicinity of the center of the on-chip convex lens 6 and the light B and C that pass through the periphery, the light A and the light B and C are equivalent to the light paths of the light B and C that are obliquely incident. An optical path difference will occur.
Therefore, for the light A, the distance between the multilayer films substantially matches the optical film thickness, but for the lights B and C, the distance between the multilayer films deviates from the optical film thickness.
As a result, the lights A, B, and C do not interfere efficiently in the multilayer film, which is disadvantageous in ensuring the transmission efficiency of the multilayer optical filter.

FIG. 3B is a schematic cross-sectional view of one pixel of the solid-state imaging device provided with the multilayer optical filter of the first embodiment.
As shown in FIG. 3B, the multilayer optical filter 10 according to the present invention is provided on the light incident surface side of the solid-state imaging device 8.
That is, the solid-state imaging device 8 includes a multilayer optical filter 10 and a photoelectric conversion unit (not shown) that photoelectrically converts light transmitted through the multilayer optical filter 10.
The multilayer optical filter 10 is configured by alternately laminating first dielectric layers 12 and second dielectric layers 14 having different refractive indexes, and a multilayer film that selectively transmits light of a specific wavelength. It is configured to include.
All of the first dielectric layer 12 and the second dielectric layer 14 are configured so that the optical film thicknesses of the first dielectric 12 and the second dielectric 14 adjacent in the direction of lamination are selectively transmitted. It is formed to be 1/4 of the wavelength λ.
That is, the optical film thickness of the first dielectric 12 is a value n1 · d1 represented by the product of the refractive index n1 of the first dielectric layer 10 and the physical film thickness d1 of the first dielectric layer 10. .
The optical film thickness of the second dielectric 14 is a value n2 · d2 indicated by the product of the refractive index n2 of the second dielectric layer 12 and the physical film thickness d2 of the second dielectric layer 12. .
Therefore, the optical film thickness of each of the first dielectric 12 and the second dielectric 14 adjacent in the stacking direction and the wavelength λ of light to be selectively transmitted must satisfy the condition of the expression (1). .
n1 · d1 = n2 · d2 = λ / 4 (1)

At least a part of the thickness direction of the multilayer film, which is the direction in which the first and second dielectric layers 12 and 14 are laminated, is formed by laminating the first and second dielectric layers 12 and 14 respectively. It is formed as a spherical portion 20 formed along a spherical surface that is convex in the same direction.
In the present embodiment, all of the multilayer film in the thickness direction is formed as the spherical portion 20.
In the present embodiment, in the spherical portion 20, the spherical surfaces along which the first and second dielectric layers 12, 14 are aligned have the same curvature.
The spherical portion 20 functions as a lens (convex lens) having a power (positive power) for converging light transmitted through the spherical portion 20.

A spacer layer 16 made of a dielectric is provided in place of the first dielectric layer 12 or the second dielectric layer 14 in the middle of the multilayer film in the thickness direction.
The dielectric constituting the spacer layer 16 is made of the same dielectric material as the first dielectric layer 12 or the second dielectric layer 14, or any one of the first dielectric layer 12 and the second dielectric layer 14 is used. Both are made of a dielectric material having a different refractive index.
The multilayer optical filter 20 is configured as a band-pass filter that transmits a desired electromagnetic wave and does not transmit other wavelengths by changing the optical film thickness of the spacer layer 16.
That is, the spacer layer 16 is formed with an optical film thickness determined by a specific wavelength of light that each multilayer film selectively transmits.
In the direction in which the first and second dielectric layers 12 and 14 are laminated, an upper reflector is formed by a multilayer film positioned above the spacer layer 16, and a lower film is formed by the multilayer film positioned below the spacer layer 16. A reflector is configured.
Therefore, the spacer layer 16 is interposed between the upper reflector and the lower reflector.
The spacer layer 16 can be omitted.

The materials of the first and second dielectric layers 12 and 14 will be described.
Of the first dielectric layer 12 and the second dielectric layer 14, one dielectric layer is composed of a low refractive index medium, and the other dielectric layer is composed of a high refractive index medium.
As the medium having a low refractive index in the visible wavelength band, a composite material mainly composed of silicon oxide (SiO ₂ ) and SiO ₂ is suitable.
In addition, magnesium fluoride (MgF ₂ ) or a hollow structure (Air Gap) can be used as a low refractive index medium.
Next, high refractive index media include silicon nitride (Si ₃ N ₄ ), titanium oxide (TiO ₂ ), tantalum oxide (Ta ₂ O ₅ ), zirconium oxide (ZrO ₂ ), niobium oxide (Nb ₂ O _5). ), Oxides and nitrides such as hafnium oxide (HfO ₂ ) are preferable.
When the multilayer optical filter 10 of the present invention is used as an optical element that transmits the near-infrared wavelength region, a semiconductor material such as Si or Ge is used as the material for the first and second dielectric layers 12 and 14. Can do.
The multilayer optical filter of the present invention can be easily realized by alternately laminating these low refractive index materials and high refractive index materials while maintaining the basic structure.

Next, the function and effect of the multilayer optical filter 10 of the present embodiment will be described.
As shown in FIG. 3B, light A ′, B ′, C ′ converged by an optical system (not shown) is incident on the multilayer optical filter 10.
The light A ′ passes near the center of the multilayer film corresponding to the pixel central portion of the solid-state image sensor 8, and the lights B ′ and C ′ pass through the periphery of the multilayer film corresponding to the pixel peripheral portion of the solid-state image sensor 8. To do.
In the multilayer optical filter 10 of the present embodiment, at least a part of the multilayer film has a spherical portion 20, and thus the multilayer film has a curvature.
Therefore, both the light A ′ passing near the center of the multilayer film and the lights B ′ and C ′ passing around the multilayer film pass through the multilayer film in a direction substantially parallel to the direction in which the multilayer films are laminated. Become.
That is, the optical path difference when the light A ′, B ′, C ′ is transmitted through the multilayer film hardly occurs, and the interval between the multilayer films is almost optical with respect to any of the light A ′, B ′, C ′. It will coincide with the film thickness.
As a result, the light A ′, B ′, and C ′ are efficiently interfered with each other in the multilayer film, which is advantageous in improving the transmission efficiency while ensuring the light collection effect of the multilayer film optical filter 10. A multilayer optical filter having transmission characteristics can be realized.
In other words, providing the multilayer film with a curvature reduces the optical path difference due to the incident position and angle of the incident electromagnetic wave, which is advantageous in improving the characteristics of the band pass filter.

Also, as shown in FIG. 3B, since at least a part of the multilayer film forms a spherical portion 20, the multilayer optical filter 10 itself has a convex lens shape and functions as a condensing element. Will also have.
Therefore, in the multilayer optical filter 10 of the present embodiment, the on-chip convex lens 6 in the comparative example of FIG. 3A can be omitted, and the on-chip convex lens 6 in the comparative example can be downsized by the thickness D. This is advantageous in reducing the height.

In order to secure the transmission efficiency of the multilayer optical filter 10, the variation in the curvature of the multilayer film is preferably in the range described below, for example.
That is, in FIG. 3B, the multilayer film on the electromagnetic wave incident side (upper side in the figure) is the first layer, and the multilayer film on the solid-state imaging device 8 side is the Nth layer.
When focusing on the nth multilayer film, if the radius of curvature when the nth multilayer film is approximated to a spherical surface is R, the (n + 1) th or (n-1) th multiphase film (first, first) The radius of curvature of the two dielectric layers 12, 14) is preferably in the range of 0.9R to 1.1R. Note that n is an arbitrary integer between 1 and N.

In other words, according to the multilayer optical filter 10 of the present embodiment, the following effects are achieved.
A spherical portion 20 that functions as a lens having power that converges on the multilayer film is provided.
Therefore, the layer interval in the oblique incident light (the interval between the first dielectric layer 12 and the second dielectric layer 14) and the layer interval in the normal incidence are substantially constant.
Therefore, the difference in the optical path difference between the multilayer films between the electromagnetic wave passing through the central portion of the pixel and the electromagnetic wave passing through the outer peripheral portion is reduced.
Therefore, the optical film thickness of the multilayer film is accurately λ / 4 even when the incident wavefront is inclined by the on-chip condensing element.
That is, the coherence of incident electromagnetic waves is increased, and a color filter with good characteristics can be realized. Alternatively, characteristics equivalent to those of a planar multilayer optical filter can be obtained with a small number of films.
Since the multilayer optical filter is realized by inorganic materials such as SiO2, TiO2, and Si3N4, it is possible to realize a color filter having strong resistance even under severe environments such as ultraviolet rays, high temperature and high humidity.
In addition, the multilayer optical filter is provided with a curvature, and the light collecting function similar to that of the on-chip convex lens is mounted on the multilayer optical filter.
Therefore, when compared with a solid-state imaging device that includes a color filter and an on-chip convex lens individually, the height of the imaging device can be reduced to the extent that the on-chip convex lens is unnecessary.

Next, simulation results of the light transmittance of the multilayer optical filter 10 of the present invention shown in FIG. 3B and the multilayer optical filter of the comparative example shown in FIG. 3A will be described.
In this case, the multilayer film is composed of eight first dielectric layers 12, eight second dielectric layers 14, and a spacer layer 16, and the first dielectric layer 12 is composed of SiO _2. The second dielectric layer 14 is made of Si ₃ N ₄ .
FIG. 4 is a diagram showing transmission characteristics when light having a visible wavelength of 400 nm to 700 nm is vertically incident on the multilayer optical filter as a TE mode.
FIG. 5 is a diagram showing transmission characteristics when light having a visible wavelength of 400 nm to 700 nm is vertically incident on the multilayer optical filter as TM mode.
4 and 5, the horizontal axis represents the electromagnetic wave wavelength.
The vertical axis shows the transmission characteristics normalized by setting the transmittance at a wavelength λ = 550 nm to 1.0 in the multilayer optical filter in the comparative example.
4 and 5, the characteristic of the multilayer optical filter 10 of the present invention shown in FIG. 3B is indicated by the symbol ◆ (corresponding to Convex in the figure), and the multilayer optical of the comparative example shown in FIG. The filter is indicated by symbol □ (corresponding to Flat in the figure).
As shown in FIGS. 4 and 5, the multilayer optical filter 10 of the present invention has an improvement in transmission characteristics of about 1.5 to 2.0 times compared to the comparative example near the transmission wavelength of 550 [nm]. It turned out that it is advantageous in improving the transmission efficiency of the membrane optical filter 10.

(Second Embodiment)
Next, a second embodiment will be described.
6A is a cross-sectional view showing the structure of the multilayer optical filter 10 of the first embodiment, and FIG. 6B is a cross-sectional view showing the structure of the multilayer optical filter 10 of the second embodiment. is there. In the following embodiments, the same or corresponding parts and members as those in the first embodiment will be described with the same reference numerals.
As shown in FIG. 6A, in the first embodiment, all of the multilayer film in the thickness direction is formed as the spherical portion 20, but as shown in FIG. In this embodiment, a part of the multilayer film in the thickness direction is formed as the spherical portion 20.
That is, as shown in FIG. 6B, the remaining portion of the multilayer film excluding the spherical portion 20 is a plane on which the first dielectric layer 12 and the second dielectric layer 14 are laminated in parallel. It is formed as a shape portion 22.
A spacer layer 16 is provided between the spherical portion 20 and the planar portion 22.
Therefore, the upper reflector composed of the multilayer film positioned on the spacer layer 16 constitutes the spherical portion 20.

Also in the second embodiment, the action of the spherical portion 20 makes it difficult for an optical path difference to occur when light passes through the multilayer film, and the distance between the multilayer films as viewed from the light transmitted through the multilayer film is almost equal to the optical film. It will match the thickness.
Therefore, as in the first embodiment, it is advantageous in improving the transmission efficiency while ensuring the light collecting effect of the multilayer optical filter 10.
In other words, giving a curvature to a part of the multilayer film is advantageous in reducing the optical path difference due to the incident position and incident angle of the incident electromagnetic wave and improving the characteristics as a band pass filter.
Also, the omission of the on-chip convex lens 6 is advantageous in reducing the size and height by the thickness D of the on-chip convex lens 6 in the comparative example, as in the first embodiment.

(Third embodiment)
Next, a third embodiment will be described.
In the multilayer optical filter of the third embodiment, the wavelengths of light transmitted through adjacent multilayer films are made different from each other.
FIG. 7 is a cross-sectional view showing the configuration of the multilayer optical filter 10 according to the third embodiment.
FIG. 7 is a schematic diagram in which a part of the multilayer optical filter 10 is extracted, and a photoelectric conversion element, a signal wiring, and the like are omitted.

As shown in FIG. 7, the multilayer films 501, 502, and 503 are arranged in a direction that intersects (orthogonally intersects) the direction in which the first dielectric layer 12 and the second dielectric layer 14 are laminated. Are provided.
Further, the number of the first dielectric layers 12 and the number of the second dielectric layers 14 constituting the multilayer films 501, 502, and 503 are the same.
In other words, multilayer films 501, 502, and 503 having the spherical portion 20 are provided corresponding to the three pixels P1, P2, and P3.
That is, a multilayer film having a convex lens shape is mounted on each pixel P1, P2, P3 of the solid-state imaging device 8.
Spacer layers 504, 505, and 506 made of the same dielectric material are provided in place of the first dielectric layer 12 or the second dielectric layer 14 in the intermediate portions in the thickness direction of the multilayer films 501, 502, and 503, respectively. It has been.
That is, the spacer layer 16 is formed with an optical film thickness that is determined by a specific wavelength of light that each multilayer film 501, 502, 503 selectively transmits.
Accordingly, the thicknesses (physical film thicknesses) of the spacer layers 504, 505, and 506 provided in the multilayer films 501, 502, and 503 are different from each other.

The medium of the spacer layer may be the same as the medium constituting the low refractive index portion or the high refractive index portion of the multilayer film. By changing the thickness of the spacer layer, it is possible to adjust the wavelength of the electromagnetic wave to be transmitted and realize various color filters.
Here, as an example, when the selectively transmitted light is green (G), the thickness (physical film thickness) of the spacer layer is λ / 2n (n is the refractive index of the medium), and the spacer layer is When composed of SiO ₂ , about 189.7 [nm] is preferable.
In FIG. 7, the spacer layer 506 of the multilayer film 503 corresponding to the pixel P3 corresponds to a green filter.
When the selectively transmitted light is red (R), it is preferable that the thickness of the spacer layer is about +30 to 50 [nm] larger than the thickness of the spacer layer 506 as a green filter.
The spacer layer 505 of the multilayer film 502 corresponding to the pixel P2 corresponds to a red filter.
Therefore, when configuring the spacer layer 505 of SiO _2, together with the upper and lower SiO ₂ layer, 220 - 240 [nm] degree is preferred.
When the selectively transmitted light is blue (B), it is preferable that the thickness of the spacer layer is about +100 to 200 [nm] larger than the thickness of the spacer layer 506 as a green filter.
The spacer layer 504 of the multilayer film 501 corresponding to the pixel P1 corresponds to a blue filter.
Therefore, when configuring the spacer layer 504 of SiO _2, together with the upper and lower SiO ₂ layer, 290-390 [nm] degree is preferred.
The thicknesses of the spacer layers 504, 505, and 506 described above are only examples, and members having an arbitrary interval or an arbitrary refractive index can be used as long as desired transmission characteristics can be obtained at a desired electromagnetic wave wavelength even at other intervals. Can be used.

As shown in FIG. 7, the first dielectric layers 12 and the second dielectric layers 14 of adjacent multilayer films having spacer layers having different thicknesses are different in height position.
In other words, in the upper reflector, an offset occurs in the height of the multilayer film between the pixels of each color due to the difference in the thickness of the spacer layer.
Therefore, a gap 22 is provided between adjacent multilayer films having spacer layers having different optical film thicknesses.
The adjacent first dielectric layers 12 and the second dielectric layers 14 of the multilayer film are connected in a straight line or in a curved line in the gap 22.
In other words, no significant step portion is formed between the dielectric layers between the pixels.
Thereby, it is possible to suppress the light reflected by the stepped portion of the adjacent first dielectric layer 12 and the stepped portion of the adjacent second dielectric layer 14 from entering the adjacent pixel. Therefore, it is advantageous in suppressing the adverse effect that the light reflected by the step portion enters the adjacent pixel and causes color mixing.

Further, as shown in FIG. 7, the first dielectric layer 12 positioned at the tip of the spherical projecting direction is formed on the remaining multilayer film excluding the multilayer film having the largest height in the spherical projecting direction of the spherical portion 20. Alternatively, the height adjusting layer 24 provided on the surface of the second dielectric layer 14 and having different heights is provided.
The height adjustment layer 24 is made of a material that transmits visible wavelengths.
The surface of the multilayer film having the largest height in the protruding direction of the spherical surface and the surface of the height adjustment layer of the multilayer film are provided at the same height.
That is, the entire surface of the solid-state imaging device is formed by laminating a height adjusting layer 24 formed of a material that transmits visible wavelengths such as SiO ₂ or Si ₃ N _{4 on} the upper layer of the multilayer optical filter and changing the thickness thereof. It is possible to prevent a significant difference in the height of individual pixels.
Thus, by matching the height position of the surface of the multilayer film, it is advantageous in suppressing variations in incident light incident on the multilayer optical filter 10 or the solid-state imaging device and ensuring light incident characteristics.
Each multilayer film is provided with a height adjusting layer 24 provided on the surface of the first dielectric layer 12 or the second dielectric layer 14 located at the tip in the protruding direction of the spherical surface and having a different height from each other. The surface of the height adjustment layer 24 of the multilayer film may be provided at the same height.
Further, in adjacent multilayer films having spacer layers having different thicknesses, the first and second dielectric layers 12 and 14 constituting the lower multilayer film are arranged so that the height positions of the surfaces of the multilayer films coincide with each other. Some may be omitted.

Of course, according to the third embodiment, the same effects as those of the first embodiment can be obtained.
Further, since the wavelength of light selectively transmitted is made different by changing the optical film thickness of each spacer layer 504, 505, 506, the material of the spacer layer can be a single material, and the material cost can be reduced. This is advantageous for reduction.

(Fourth embodiment)
Next, a fourth embodiment will be described.
In the third embodiment, the case where the spacer layer is formed with an optical film thickness determined by a specific wavelength of light that each multilayer film selectively transmits is described. In contrast, in the fourth embodiment, a case will be described in which the spacer layer has a refractive index determined by a specific wavelength of light that each multilayer film selectively transmits.
FIG. 8 is a cross-sectional view showing the configuration of the multilayer optical filter 10 according to the fourth embodiment.
As shown in FIG. 8, multilayer films 601, 602, and 603 each having a spherical portion 20 are provided corresponding to the three pixels P1, P2, and P3.
In other words, a multilayer film having a convex lens shape is mounted on the top of each pixel P1, P2, P3 of the solid-state imaging device 8.
Spacer layers 604, 606, and 606 made of dielectric are provided in the middle portions of the multilayer films 601, 602, and 603 in the thickness direction in place of the first dielectric layer 12 or the second dielectric layer 14, respectively. Yes.
The spacer layers 604, 606, and 606 have a refractive index that is determined by a specific wavelength of light that each multilayer film 601, 602, and 603 selectively transmits.
In the adjacent multilayer films 601, 602, and 603, the spacer layers 604, 606, and 606 are formed so as to have the same physical film thickness.

Here, as an example, when the light to be selectively transmitted is green (G), and the thickness of the spacer layer (physical thickness) is lambda / 2n (n is the refractive index of the medium), SiO spacer layer _{In the case of 2} (refractive index n = 1.45), the physical thickness of the spacer layer is about 189.7 [nm].
In FIG. 8, the spacer layer 606 of the multilayer film 603 corresponding to the pixel P3 corresponds to a green filter.
When light to be selectively transmitted is red (R), the thickness of the spacer layer in terms of the film thickness of the SiO _2, in than the thickness + 30 to 50 [nm] greater degree value of the spacer layer 606 as a green filter Preferably there is.
The spacer layer 605 of the multilayer film 602 corresponding to the pixel P2 corresponds to a red filter.
That is, when the thickness of the spacer layer 606 as a red filter 189.7 [nm] equivalent to the spacer layer 606 as a green filter, setting the refractive index of the spacer layer 606 as a red filter at about 1.6 to 1.9 Is preferred.
The medium refractive index of about 1.6 to 1.9, in addition to silicon or an organic resin nitride, be implemented by a SiO ₂ and Si ₃ N ₄ and the sub-wavelength structures periodically emphasis in the following scale wavelength it can. The subwavelength structure will be described in detail in paragraphs 0058 and after.
When light to be selectively transmitted is blue (B), is the thickness of the thick + 100 to 200 [nm] a greater extent than the spacer layer 606 as a green filter when converted to the film thickness of the SiO ₂ spacer layer It is preferable.
The spacer layer 604 of the multilayer film 601 corresponding to the pixel P1 corresponds to a blue filter.
That is, when the thickness of the spacer layer 604 as a blue filter 189.7 [nm] equivalent to the spacer layer 606 as a green filter, setting the refractive index of the spacer layer 604 as a blue filter to approximately 2.2 to 3.0 Is preferred.
An example of the medium having a refractive index of about 2.2 to 3.0 is titanium oxide (TiO2) refractive index = 2.5.
Thickness of the spacer layer, the refractive index of the filling medium is an example, even in a medium having another spacer layer spacing and refractive index, as long as the resulting desired transmission characteristic at the desired wavelength, may be arbitrary.

Of course, according to the fourth embodiment, the same effects as those of the first embodiment can be obtained.
In addition, since the spacer layer has a refractive index determined by a specific wavelength of light that each multilayer film selectively transmits, if the physical thickness of the spacer layer is formed to be the same, The height positions of the surfaces of adjacent multilayer films can be matched.
Therefore, as in the third embodiment, the incident characteristic of light incident on the multilayer optical filter 10 or the solid-state imaging device is ensured without stacking the height adjustment layer 24 on the multilayer optical filter. This is advantageous in reducing the manufacturing cost.

FIG. 9 is a cross-sectional view showing the configuration of the multilayer optical filter 10 in a modification of the fourth embodiment.
As shown in FIG. 9, a gap 22 is provided between adjacent multilayer films having spacer layers having different refractive indexes.
Then, a reflection portion 24 made of a single medium (for example, SiO ₂ , Si ₃ N _4, etc.) is formed in the gap 22. Alternatively, the reflective portion 24 having an air gap structure is formed by etching the gap 22.
When such a reflection portion 24 is formed, light incident in an oblique direction with respect to the multilayer film is reflected (total reflection) at the interface between the reflection portion 24 and the multilayer film, and is prevented from entering the adjacent pixels. . For this reason, it is advantageous in suppressing the adverse effect that light incident on the multilayer film in an oblique direction enters the adjacent pixel and causes color mixing.

(Fifth embodiment)
Next, a fifth embodiment will be described.
The fifth embodiment is different from the first to fourth embodiments in the shape of the spherical portion 20 of the multilayer film.
FIG. 10 is a cross-sectional view showing the configuration of the multilayer optical filter 10 of the fifth embodiment.
FIG. 10 is also a schematic diagram in which a part of the multilayer optical filter 10 is extracted as in FIG. 7, and the photoelectric conversion elements and signal wirings are omitted.

As shown in FIG. 10, multilayer films 1401, 1402, and 1403 having a spherical portion 20 are provided corresponding to the three pixels P1, P2, and P3.
In other words, a multilayer film having a convex lens shape is mounted on the top of each pixel P1, P2, P3 of the solid-state imaging device 8.
Spacer layers 1404, 14014, and 1406 made of the same dielectric material are provided in place of the first dielectric layer 12 or the second dielectric layer 14 in the intermediate portions in the thickness direction of the multilayer films 1401, 1402, and 1403, respectively. It has been.
The spacer layers 1404, 1405, and 1406 are formed with a refractive index determined by a specific wavelength of light that each multilayer film 1401, 1402, and 1403 selectively transmits.
Also in the fourth embodiment, the spacer layers 1404, 14014, and 1406 are formed with an optical film thickness determined by a specific wavelength of light that each multilayer film 1401, 1402, and 1403 selectively transmits. Also good.

At least a part of the thickness direction of the multilayer film, which is the direction in which the first and second dielectric layers 12 and 14 are laminated, is formed by laminating the first and second dielectric layers 12 and 14 respectively. It is formed as a spherical portion 20 formed along a spherical surface that is convex in the same direction.
In the present embodiment, all of the multilayer film in the thickness direction is formed as the spherical portion 20.
The spherical portion 20 functions as a lens having a power for converging light transmitted through the spherical portion 20.
In the present embodiment, the spherical portion 20 is different from the first to fourth embodiments in that the center of the spherical surface along which the first dielectric layer 12 and the second dielectric layer 14 are aligned is the same. is doing.
In other words, the multilayer film corresponding to each pixel does not have a similar shape, but has a spherical shell shape in which the curvature decreases toward the upper layer.
In order to secure the transmission efficiency of the multilayer optical filter 10, the variation in the curvature of the multilayer film is preferably in the range described below, for example.
That is, in FIG. 10, the multilayer film on the electromagnetic wave incident side (upper side in the figure) is the first layer, and the multilayer film on the solid-state imaging device 8 side is the Nth layer.
When focusing on the nth multilayer film, if the radius of curvature when the nth multilayer film is approximated to a spherical surface is R, the (n + 1) th or (n-1) th multiphase film (first, first) The radius of curvature of the two dielectric layers 12, 14) is preferably in the range of 0.9R to 1.1R. Note that n is an arbitrary integer between 1 and N.

In the fifth embodiment, it is needless to say that the same effect as in the first embodiment can be obtained.
FIG. 11 is a cross-sectional view showing the configuration of the multilayer optical filter 10 in a modification of the fifth embodiment.
As shown in FIG. 11, multilayer films 1501, 1502, and 1503 having a spherical portion 20 are provided corresponding to the three pixels P1, P2, and P3.
Spacer layers 1504, 15015, and 1506 made of the same dielectric material are provided in place of the first dielectric layer 12 or the second dielectric layer 14 at intermediate portions in the thickness direction of the multilayer films 1501, 1502, and 1503, respectively. It has been.
The spacer layers 1504, 1505, and 1406 are formed with a refractive index determined by a specific wavelength of light that each multilayer film 1401, 1402, and 1403 selectively transmits.
This modification is different from FIG. 10 only in that the curvature of the spherical surface portion 20 is larger than that of the spherical surface portion 20 of FIG.

(Sixth embodiment)
Next, a sixth embodiment will be described.
In the sixth embodiment, the multilayer optical filter 10 shown in FIG. 8A of the first embodiment is mounted on a CMOS solid-state imaging device. Note that the solid-state imaging device on which the multilayer optical filter 10 is mounted is not limited to the COMS solid-state imaging device, and can be applied to various known solid-state imaging devices such as a CCD solid-state imaging device. Of course.
FIG. 12 is a cross-sectional view of a CMOS type solid-state imaging device on which the multilayer optical filter 10 according to the sixth embodiment is mounted. A spacer layer (not shown) is provided in the middle of the multilayer film of the multilayer optical filter 10.
A multilayer optical filter 10 according to the present invention is provided in a pixel 701 of the solid-state imaging device 8.
The solid-state image sensor 8 has a pixel 701.
Each pixel 701 is provided with a photoelectric conversion element 703.
The photoelectric conversion element 703 is electrically separated from adjacent pixels by a pixel separation layer 705.
A signal wiring layer 704 is provided on the pixel separation layer 705.
A portion surrounded by the photoelectric conversion element 703 and the signal wiring layer 704 is a waveguide structure filled with a medium having a higher refractive index than the surroundings (for example, the surrounding is SiO ₂ and the waveguide portion is Si ₃ N ₄ ). Part 706.
The waveguide structure portion 706 realizes a clad core structure by being filled with a medium having a high refractive index with respect to the surroundings. It is possible to guide.

  According to such a CMOS type solid-state imaging device, the transmission efficiency is improved while ensuring the light condensing effect of the multilayer optical filter 10, which is advantageous in ensuring the quality of the captured image signal. Become.

(Seventh embodiment)
Next, a seventh embodiment will be described.
The seventh embodiment is a modification of the sixth embodiment, in which the multilayer optical filter 10 shown in FIG. 8B of the first embodiment is mounted on a CMOS solid-state imaging device. It is.
FIG. 13 is a cross-sectional view of a CMOS type solid-state imaging device on which the multilayer optical filter 10 according to the seventh embodiment is mounted.
As shown in FIG. 13, a part of the multilayer film in the thickness direction is formed as a spherical portion 20.
That is, the remaining part of the multilayer film excluding the spherical portion 20 is formed as a planar portion 22 in which the first dielectric layer 12 and the second dielectric layer 14 are laminated in parallel.
A spacer layer (not shown) is provided between the spherical portion 20 and the planar portion 22.
The solid-state image sensor 8 has a pixel 801.
Each pixel 801 is provided with a photoelectric conversion element 803.
The photoelectric conversion element 803 is electrically separated from adjacent pixels by a pixel separation layer 805.
A signal wiring layer 804 is provided on the pixel separation layer 805.
A portion surrounded by the photoelectric conversion element 803 and the signal wiring layer 804 is a waveguide structure filled with a medium having a higher refractive index than the surroundings (for example, the surrounding is SiO ₂ and the waveguide portion is Si ₃ N ₄ ). Part 806.
The waveguide structure portion 806 realizes a clad core structure by being filled with a medium having a high refractive index with respect to the surroundings. It is possible to guide.

  In such a CMOS type solid-state imaging device, since the transmission efficiency is improved while ensuring the light condensing effect of the multilayer optical filter 10, it is advantageous in ensuring the quality of the captured image signal. .

(Eighth embodiment)
Next, an eighth embodiment will be described.
FIG. 14 is a cross-sectional view of a CMOS solid-state imaging device on which the multilayer optical filter 10 according to the eighth embodiment is mounted.
As shown in FIG. 14, a part of the multilayer film in the thickness direction is formed as a spherical portion 20.
That is, the remaining part of the multilayer film excluding the spherical portion 20 is formed as a planar portion 22 in which the first dielectric layer 12 and the second dielectric layer 14 are laminated in parallel.
A spacer layer (not shown) is provided between the spherical portion 20 and the planar portion 22.
The solid-state image sensor 8 has a pixel 901.
Each pixel 901 is provided with a photoelectric conversion element 903.
The photoelectric conversion element 903 is electrically separated from adjacent pixels by a pixel separation layer 905.
A signal wiring layer 904 is provided above the pixel isolation layer 905.
Here, as the signal wiring layer 904, a Cu wiring layer composed of a plurality of layers is formed in the SiO ₂ substrate 912 by a damascene technique, and silicon nitride is used as a stopper layer (diffusion prevention layer) 914 formed on the upper part of the Cu wiring layer. (Si ₃ N ₄ ) is used.

Therefore, the SiO ₂ substrate 912 functions as the first dielectric layer 12 because it is a low refractive index medium, and the stopper layer is formed because silicon nitride (Si ₃ N ₄ ) is a high refractive index medium. The (diffusion prevention layer) 914 functions as the second dielectric layer 14.
That is, the layered structure of SiO ₂ / Si ₃ N ₄ resulting from these CMOS wiring structures can be used as the lower reflector of the multilayer optical filter.
Therefore, according to the eighth embodiment, it is possible to reduce the number of layers of the multilayer film, as a matter of course, that the same effect as in the seventh embodiment can be obtained. This is advantageous in reducing the overall height of the image sensor.

(Ninth embodiment)
Next, a ninth embodiment will be described.
FIG. 15 is a cross-sectional view of the multilayer optical filter 10 according to the ninth embodiment.
As shown in FIG. 15, the multilayer optical filter 10 is provided with a smoothing layer 26 on the surface of the first dielectric layer 12 or the second dielectric layer 14 located at the tip of the spherical portion 20 in the protruding direction of the spherical surface. It has been.
A lens mounting surface 28 is formed on the surface of the smoothing layer 26 so as to extend on a single plane perpendicular to the protruding direction of the spherical surface of the spherical portion 20.
An on-chip condensing element 30 that functions as a lens having a positive power (convex lens) is attached to the lens mounting surface 28.
Here, a plurality of multilayer films are arranged in a direction intersecting the direction in which the first dielectric layer 12 and the second dielectric layer 14 are laminated, and the height positions of the surfaces of the multilayer films are mutually different. Is different.
Specifically, the thicknesses of the spacer layers 16 provided in each multilayer film are different from each other, or the number of first dielectric layers 12 or the number of second dielectric layers 14 constituting each multilayer film is mutually different. Is different.
In the present embodiment, the heights of the smoothing layers 26 are made different so that the surface of the on-chip condensing element 30 is provided at the same height.

  According to the ninth embodiment, the heights of the pixels of the respective color components can be made uniform over the entire surface of the solid-state imaging device, whereby variation in shading characteristics for each color can be reduced.

(Tenth embodiment)
Next, a tenth embodiment will be described.
The tenth embodiment is a modification of the ninth embodiment, and is different from the ninth embodiment in that the direction in which the spherical portion 20 protrudes is the opposite direction.
FIG. 16 is a cross-sectional view of the multilayer optical filter 10 according to the tenth embodiment.
As shown in FIG. 16, all of the thickness direction of the multilayer film, which is the direction in which the first and second dielectric layers 12 and 14 are laminated, is formed as the spherical portion 20.
In the spherical portion 20, each of the first and second dielectric layers 12 and 14 is formed along a spherical surface having a convex shape in the same direction in the direction in which they are laminated, and the spherical portion 20 protrudes. The direction to do is the same as the light transmission direction.
That is, the spherical portion 20 functions as a lens (concave lens) having a power (negative power) for diverging light transmitted through the spherical portion 20.
In the present embodiment, in the spherical portion 20, the spherical surfaces along which the first and second dielectric layers 12, 14 are aligned have the same curvature.

As shown in FIG. 16, the multilayer optical filter 10 has a smoothing layer on the surface of the first dielectric layer 12 or the second dielectric layer 14 located at the tip of the spherical portion 20 in the direction opposite to the protruding direction of the spherical surface. 26 is provided.
A lens mounting surface 28 is formed on the surface of the smoothing layer 26 so as to extend on a single plane perpendicular to the protruding direction of the spherical surface of the spherical portion 20.
An on-chip condensing element 1102 that functions as a lens having a positive power (convex lens) is attached to the lens mounting surface 28.
Here, a plurality of multilayer films are arranged in a direction intersecting the direction in which the first dielectric layer 12 and the second dielectric layer 14 are laminated, and the height positions of the surfaces of the multilayer films are mutually different. Is different.
Specifically, the thicknesses of the spacer layers 16 provided in each multilayer film are different from each other, or the number of first dielectric layers 12 or the number of second dielectric layers 14 constituting each multilayer film is mutually different. Is different.
Then, as in the ninth embodiment, the heights of the smoothing layers 26 are made different so that the surface of the on-chip condensing element 30 is provided at the same height.

According to the tenth embodiment, the same effects as in the ninth embodiment are achieved.
Further, since the spherical portion 20 functions as a lens having a power to diverge the light transmitted through the spherical portion 20, the spherical portion 20 is used for diffusing or collimating the light collected by the on-chip condensing element 1102. It will be advantageous.
Therefore, it is suitable when diverging light or parallel light is incident on the light receiving surface of the solid-state imaging device.

(Eleventh embodiment)
Next, an eleventh embodiment will be described.
In the eleventh embodiment, a hybrid multilayer optical filter 32 is constructed by combining an inorganic thin film filter (or organic filter) with a multilayer film.
FIG. 17B is a cross-sectional view of a CMOS solid-state imaging device on which the multilayer optical filter 10 of the eleventh embodiment is mounted.
As shown in FIG. 17B, the portion close to the solid-state image sensor in the multilayer film in FIG. 17A is replaced with an inorganic thin film filter 1205 (or organic filter 1205) and mounted.
In other words, in the multilayer optical filter 32 of the eleventh embodiment, the multilayer film in which the first and second dielectric layers 12 and 14 are laminated and the inorganic thin film filter 1205 (or the organic filter 1205) are combined. Configured.

FIG. 17A is a cross-sectional view of a CMOS solid-state imaging device on which the multilayer optical filter 10 is mounted.
That is, FIG. 17A shows a cross-sectional view of a CMOS type solid-state imaging device configured similarly to the sixth embodiment (FIG. 12).
In FIG. 17A, a spacer layer is provided in the middle of the multilayer film of the multilayer optical filter 10, an upper reflector is provided above the spacer layer, and a lower reflector is provided below the spacer layer. However, they are not shown in order to simplify the drawing.
FIG. 17C is a cross-sectional view of a CMOS type solid-state imaging device on which an optical filter 34 composed of an inorganic thin film filter 1205 (or an organic filter 1205) is mounted.

According to the eleventh embodiment, since the multilayer optical filter 32 is configured by combining the multilayer film and the inorganic thin film filter 1205 (or the organic filter 1205), the degree of freedom in designing the multilayer optical filter 32 is ensured. This is advantageous. Therefore, it is advantageous to easily and surely realize a bandpass filter having a desired transmission characteristic.
Moreover, two-dimensional solid-state imaging is obtained by appropriately combining the multilayer optical filter 10 shown in FIG. 17A, the hybrid multilayer optical filter 32 shown in FIG. 17B, and the optical filter 34 shown in FIG. Of course, the element can be configured. In that case, it is advantageous in securing the degree of freedom in designing the solid-state imaging device.

(Twelfth embodiment)
Next, a twelfth embodiment will be described.
As described in the fourth embodiment (FIGS. 8 and 9), each spacer layer 604, 605, 606 has a refractive index determined by a specific wavelength of light that each multilayer film selectively transmits. In this case, it is necessary to adjust the refractive index of the spacer layer.
In this case, the refractive index of the spacer layer can be easily and reliably adjusted by configuring each spacer layer using the sub-wavelength structure described below.

18 is a cross-sectional view showing the configuration of the multilayer optical filter 10 of the twelfth embodiment, and FIG. 19 is a perspective view showing the structure of each sub-wavelength structure. In order to simplify the drawing, each of the sub-wavelength structures 1604 to 1612 is shown as extending in a planar shape. However, in actuality, each sub-wavelength structure has a spherical shape like a multilayer film. Presents.
As shown in FIGS. 18 and 19, the spacer layers 604, 605, and 606 are composed of sub-wavelength structures 1604 to 1612.
In other words, the spacer layers 604, 605, and 606 have sub-wavelength structures in which two types of media are arranged one-dimensionally or two-dimensionally with different refractive indexes and dimensions smaller than the transmitted electromagnetic wave wavelength. Consists of the body.
18 and 19, three types of sub-wavelength structures are shown as examples. The first type is sub-wavelength structures 1604, 1605, and 1606, and the second type is sub-wavelength structures. 1607, 1608, and 1609, and the third type is subwavelength structures 1610, 1611, and 1612.

First, a case where the first type sub-wavelength structures 1604, 1605, and 1606 are used as the spacer layers 604, 605, and 606 will be described.
The thickness directions of the sub-wavelength structures 1604, 1605, and 1606 are defined as the Z axis, and the two axes orthogonal to each other on a plane orthogonal to the thickness direction are defined as the X axis and the Y axis.
In the present embodiment, the sub-wavelength structures 1604, 1605, and 1606 are composed of two types of media having different refractive indexes in two directions of the X axis and the Y axis on a scale (size) that is significantly smaller than the electromagnetic wave wavelength. It is a structure arranged periodically.

In FIG. 18, a portion indicated by white and a portion indicated by fine dots indicate two types of media having different refractive indexes. Hereinafter, for convenience of explanation, the medium shown in white will be described as the first medium 40A, and the medium shown with fine points will be described as the second medium 40B.
Here, the scale that is significantly smaller than the electromagnetic wave wavelength typically has a size that is less than half of the electromagnetic wave wavelength in the medium, and ideally two types of media up to about 1/10 of the electromagnetic wave wavelength. It is still preferable that the structure can be made fine.
The sub-wavelength structures 1604, 1605, and 1606 configured in this manner change in refractive index on a scale that is significantly smaller than the spread of electromagnetic waves.

Therefore, the effective refractive index felt by the electromagnetic waves transmitted through the sub-wavelength structures 1604, 1605 and 1606 is equal to the average refractive index determined by the ratio of the volume density of the two types of media.
Therefore, as shown in the sub-wavelength structures 1604, 1605, and 1606, the refractive index of the sub-wavelength structure can be adjusted by changing the ratio of the two types of media 40A and 40B constituting the sub-wavelength structure. .
More specifically, in the case of the sub-wavelength structures 1604, 1605, and 1606, the two types of media 40A and 40B have a cubic shape that is divided along the X-axis direction and the Y-axis direction.
The sub-wavelength structures 1604, 1605, and 1606 have their refractive indexes changed by the biaxial periodic structure of the X-axis direction and the Y-axis direction.
That is, the volume ratios of the media 40A and 40B in the first type sub-wavelength structures 1604, 1605, and 1606 are 1: 0, 1: 1, and 0: 1, and the refractive index also depends on these ratios. Will be determined.

In the case of the second type sub-wavelength structures 1607, 1608, and 1609, the first is that the two types of media 40A and 40B have a cubic shape divided along the X-axis direction and the Y-axis direction. Similar to the types of subwavelength structures. However, it differs from the first type sub-wavelength structure in that the refractive index is changed by the periodic structure in either the X-axis direction or the Y-axis direction.
As shown in the figure, the volume ratios of the media 40A and 40B in the second type sub-wavelength structures 1604, 1605, and 1606 are 3: 1, 1: 1, and 1: 3, and the refractive indexes are also those. It will be decided according to the ratio.

In the case of the third type sub-wavelength structures 1610, 1611 and 1612, the two types of media 40A and 40B are alternately arranged in a concentric manner. The rate is changing.
As shown in the figure, the refractive index is determined in accordance with the ratio by changing the ratio of the two types of media 40A and 40B.

(Thirteenth embodiment)
Next, a thirteenth embodiment will be described.
In the thirteenth embodiment, a case where both the multilayer optical filter 10 of the present invention and the polarizing filter 50 are provided in a two-dimensional solid-state imaging device will be described.
FIG. 20 is a cross-sectional view of a two-dimensional solid-state imaging device provided with both the multilayer filter 10 and the polarizing filter 50 in the thirteenth embodiment.
As shown in FIG. 20, the solid-state imaging device 8 has a plurality of pixels P1, P2,... Arranged on a two-dimensional plane.
Each pixel is provided with either the multilayer optical filter 10 or the polarizing filter 50. Therefore, the multilayer optical filter 10 and the polarizing filter 50 are arranged on a two-dimensional plane.

The arrangement pattern of the multilayer optical filter 10 and the polarization filter 50 is not limited, for example, the multilayer optical filter 10 and the polarization filter 50 are arranged in a staggered manner for each pixel.
The polarizing filter 50 is composed of a multilayered polarizing element.
Each polarizing element is a structure having a saw-like groove structure with a period smaller than the visible wavelength.
The light incident from above enters the convex portion 52 corresponding to one blade of the saw, but when viewed from the plane constituting the convex portion, the incident light is irradiated obliquely.
Since TE waves and TM waves of obliquely incident light are efficiently reflected, polarized components corresponding to TM wave components are selectively transmitted.
In this way, color information is acquired by detecting light transmitted through the multilayer optical filter 10 arranged on a two-dimensional plane by the pixels of the solid-state imaging device 8. At the same time, the light transmitted through the polarizing filter 50 arranged on the two-dimensional plane is detected by the pixels of the solid-state imaging device 8, whereby the information on the intensity of polarization and the direction of polarization is acquired.

Therefore, according to the thirteenth embodiment, it is possible to obtain color information and information on the intensity of polarization and the direction of polarization, and by using the color information and information on the intensity of polarization and the direction of polarization. This is advantageous in performing various processes on the captured image information.
Examples of the process include a process for removing polarized light, a process for enhancing polarization, and a process for extracting polarized light for each pixel from an image after imaging.

(Fourteenth embodiment)

Next, a fourteenth embodiment will be described.
In the fourteenth embodiment, a specific configuration of the filter array in the two-dimensional solid-state imaging device including the multilayer optical filter 10 of the present invention will be described.
FIG. 21A is an explanatory view showing an arrangement structure of a general RGB filter that separates transmitted light into three wavelength bands of R, G, and B by the multilayer optical filter 10, and FIG. 21B shows the wavelength in this RGB filter. It is a diagram which shows the relationship between a transmittance | permeability.
FIG. 21A shows a portion formed of 4 pixels in the vertical direction and 4 pixels in the horizontal direction from the two-dimensional solid-state imaging device including the multilayer optical filter 10.
In this multilayer optical filter 10, three types of multilayer films that respectively transmit light of wavelengths R (wavelength 650 nm), G (wavelength 550 nm), and B (wavelength 450 nm) are arranged in a staggered pattern.
Here, one unit of the color arrangement is 2 pixels in the vertical direction and 2 pixels in the horizontal direction, and for each unit (4 pixels), a multilayer film that transmits G is provided for two pixels.

FIG. 22A is an explanatory diagram showing an arrangement structure of a filter that separates transmitted light into seven wavelength bands by the multilayer optical filter 10, and FIG. 22B is a diagram showing the relationship between wavelength and transmittance in this filter. is there.
In other words, this multilayer optical filter 10 is an example in which wavelengths from 400 [nm] to 700 [nm] are classified into 7 wavelengths in 50 [nm] steps, and a total of 8 pixels of 2 horizontal pixels and 4 vertical pixels. One unit is composed of pixels.
One unit consists of 2 pixels of multilayer film that transmits G (wavelength 550 nm) and 6 pixels of multilayer film that transmits other 6 types of wavelengths. can do.
Note that the arrangement of these color arrangements is merely an example, and the arrangement structure of the multilayer film, in other words, the color arrangement of the filters is arbitrary.

That is, according to the multilayer optical filter of the present invention, by providing the multilayer portion with the spherical surface portion 20, it is possible to accurately align the optical film thickness of the multilayer film to λ / 4 for both normal incident light and oblique incident light. It becomes. As a result, a bandpass filter having sharp transmission characteristics can be realized.
Although the human eye can only divide the visible wavelength to only three colors of RGB, it can realize a bandpass filter with sharp transmission characteristics as described above, so a narrow band color of 50 [nm] to 100 [nm] A filter can be easily realized.
Therefore, it is advantageous in obtaining a spectral profile with more accurate color expression and a two-dimensional image sensor.

(Fifteenth embodiment)
Next, a fifteenth embodiment is described.
In the fifteenth embodiment, a manufacturing method of the multilayer optical filter 10 will be described.
The multilayer optical filter 10 of the present invention stacks a self-similar multilayer structure by combining sputtering deposition and bias sputtering.
Thereby, a multilayer film is realized with high accuracy and functions as an optical filter exhibiting high transmission characteristics.
In order to stack a plurality of thin films without distortion in a self-similar manner, it is necessary to appropriately adjust the bias condition. For this purpose, the self-cloning technique is known as an excellent technique (for example, Japanese Patent No. 3325825).
Hereinafter, a method for producing the present optical filter by the self-cloning technique will be described.
However, the production method is not limited to the self-cloning method as long as a multilayer film structure having a curvature can be realized with high accuracy, and does not limit the realization of the present optical filter by other manufacturing methods.

First, a structure serving as a base of a multilayer film is formed on a substrate by techniques such as electron beam lithography, photolithography, interference exposure, and etching.
Etching is preferably anisotropic dry etching, and a gas used for etching is preferably a tetrafluoromethane (CF ₄ ) -based etching gas.
In addition, sulfur hexafluoride, trifluoromethane, xenon difluoride, and the like are also suitable. In addition, a nano stamper having a basic structure may be manufactured by electron beam lithography, and the structure may be transferred by a nano imprint technique.

Next, in order to realize a photonic crystal (first and second dielectric layers), a laminated structure of a low refractive index medium and a high refractive index medium is alternately laminated by combining bias sputtering and sputter deposition.
Here, as a medium having a low refractive index in the visible wavelength band, a composite material mainly composed of silicon oxide (SiO ₂ ) and SiO ₂ is suitable.
In addition, magnesium fluoride (MgF ₂ ) can also be used as a low refractive index medium. High refractive index media include silicon nitride (Si ₃ N ₄ ), titanium oxide (TiO ₂ ), tantalum oxide (Ta ₂ O ₅ ), zirconium oxide (ZrO ₂ ), niobium oxide (Nb ₂ O ₅ ), oxidation Oxides and nitrides such as hafnium (HfO ₂ ) are preferable.
In addition, when this filter is used as an optical element that transmits the near-infrared wavelength region, Si, Ge, or the like can be used.
These low-refractive index materials and high-refractive index materials are alternately stacked with an optical film thickness that is ¼ of the wavelength λ of light that is transmitted while maintaining the basic structure.

A method for implementing a self-similar structure by bias sputtering is shown.
Here, bias sputtering refers to a process in which film formation by sputtering and sputter etching proceed simultaneously.
FIG. 23 is an explanatory diagram showing an outline of a bias sputtering apparatus.
The substrate 1805 is connected to the high frequency power supply 1801 through the impedance matching unit 1802. Similarly, the target 1806 is connected to the high frequency power source 1803 via the impedance matching unit 1804.
Here, the high-frequency power source is generally often 13.56 MHz. The vacuum dewar 1807 is filled with a mixed gas mainly composed of an inert gas (for example, Ar gas), and a gas pressure of typically about 0.1 to 10 mTorr is suitable.
Plasma is generated by supplying high-frequency power to the target 1806 and the substrate 1805.
By applying an AC voltage to the target 1806, a DC bias (self-bias effect) due to the non-linearity of the probe characteristics is applied, and the target 1806 becomes a negative potential on a time average.
Accordingly, the positively charged gas ions acquire kinetic energy due to the potential difference and collide with the target 1806.
By this reaction, atoms / molecules on the surface of the target material are scattered, the substance particles adhere to the substrate 1805, and a thin film is laminated on the substrate.

On the other hand, a high frequency power source 1801 and an impedance matching unit 1802 are also connected to the substrate 1805.
Therefore, by adjusting the power supplied to the substrate 1805, the type of gas, and the pressure, the type of ions that collide, the magnitude of kinetic energy, and the effect of sputtering can be controlled.
In the case of performing independent sputter etching without the film formation effect by sputtering, a high frequency power source may be supplied only to the substrate 1805.
A film shaped by these processes can be formed, and a desired shape can be maintained and realized.
Note that although a manufacturing method by bias sputtering in which sputtering and etching proceed simultaneously is shown here, naturally, film formation by sputtering may be performed, and then a procedure for shaping by dry etching or plasma etching may be followed.

The reason why self-similar shapes can be shaped three-dimensionally with high accuracy by the self-cloning method is as follows: (1) Deposition by neutral incidence of neutral particles from the target, (2) Sputter etching by perpendicular incidence of Ar ions, 3) This can be explained by the three actions of reattachment of deposited particles.
Consider the case of forming a multilayer film having a fine structure.
Since the incident particles used for film formation collide with the substrate with a finite extent, when the substrate has a steep concavo-convex structure, the concave portion is partially shielded by the convex portion.
As a result, the bottom of the recess is bent (kinked).
On the other hand, for film formation, a component in which particles are directly incident from the space (primary effect), and a component that has adhered to or collided with the film is ejected by sputter etching or recoil, and the film particles or etching gas particles are incident again. It is thought that there is a (secondary effect).
Since the influence of these secondary effects is more conspicuous in the concave portion than in the convex portion, it becomes a factor that cancels the above-described shielding effect. In this way, by optimizing the balance of the three actions, a multi-layer structure can be realized with high accuracy while maintaining a steady structure.

Here, as an example of the conditions for producing the multilayer film, the following parameters are suitable, for example.
In the film formation of the Si ₃ N ₄ layer, the gas pressure is 0.3 [Pa], and the high frequency power applied to the target 1806 is 300 to 400 [W].
For forming the SiO2 layer, the gas pressure is 0.8 [Pa], and the high frequency power applied to the target 1806 is 300 to 400 [W].
Sputter etching is performed before and after the SiO ₂ layer is formed. The gas pressure is 0.3 [Pa], and the high frequency power applied to the substrate 1805 is about 50 to 100 [W].
The above conditions are merely examples, and adjustment is necessary depending on the medium constituting the multilayer film, gas pressure, applied power, gas flow rate, and the like.

Note that the multilayer film constituting the spherical portion 20 of the multilayer optical filter 10 of the present invention has a structure with little steep unevenness, so that it is not limited to the self-cloning method, and film formation by normal sputtering and etching are performed. It can also be created by alternately repeating the processing by.
A more detailed photonic crystal production method using self-cloning technology is disclosed in Japanese Patent No. 3325825.

Next, a method for producing a spacer layer between multilayer films will be outlined.
Here, the case where the multilayer optical filter constitutes a bandpass filter of three colors, that is, the case where the transmission wavelength band of the multilayer optical filter is RGB is shown.
First, a film thickness corresponding to the spacer layer (1) having the maximum thickness among the spacer layers of the bandpass filter corresponding to the three color components is formed.
Next, in order to make the deposited spacer layer have a thickness (2) corresponding to the second color component, a first resist is applied to the spacer layer, and the thickness of the corresponding pixel portion is adjusted by etching.
Thereafter, after removing the first resist, the second resist is applied, and etching is performed to obtain a film thickness (3) corresponding to the third color component.
Thereafter, the second resist is removed. For the etching of the spacer layer, anisotropic dry etching is suitable.

When the spacer layer is realized by a medium having a different refractive index, for example, a plurality of different types of spacer layers such as SiO2, Si3N4, and TiO2 can be formed by sputtering and the film thickness can be adjusted by etching.
Furthermore, when the spacer layer is realized by a sub-wavelength scale periodic structure of a low refractive index medium and a high refractive index medium, the spacer layer of the low refractive index portion is formed by sputtering and corresponds to the periodic structure of the sub wavelength. A mask pattern resist to be applied is applied and etched.
Then, the resist is removed, and then a high refractive index medium is formed by sputtering. Thereafter, the spacer layer having a sub-wavelength structure of a high refractive index medium and a low refractive index medium is realized by adjusting the layer shape by etching.
After the spacer layer is realized, a multilayer optical filter can be realized by forming a multilayer film by bias sputtering as an upper layer.

(Sixteenth embodiment)
Next, a sixteenth embodiment will be described.
In the sixteenth embodiment, a case where the solid-state imaging device 8 provided with the multilayer optical filter 10 of the present invention is applied to an imaging apparatus such as a digital still camera will be described with reference to FIG.
As shown in FIG. 24, the imaging apparatus 400 includes an imaging optical system 300, an imaging unit 310, a system control unit 320, a drive control unit 330, a memory medium 340, a scanning panel unit 350, a display 360, and the like.

The imaging optical system 300 includes a zoom lens 301, a diaphragm mechanism 302, and the like disposed in a lens barrel, and forms a subject image on a light receiving unit of an image sensor.
The imaging optical system 300 is mechanically driven by the control of the drive control unit 330 based on an instruction from the system control unit 320 to perform control such as autofocus.

The imaging unit 310 performs imaging of a subject using the solid-state imaging device 8 including the multilayer optical filter 10 of the present invention, and outputs an imaging signal (image signal) to the system control unit 320 mounted on the main board. To do.
That is, the imaging unit 310 performs processing such as AGC (automatic gain control), OB (optical black) clamping, CDS (correlated double sampling), and A / D conversion on the output signal of the image sensor described above, and performs digital imaging. Generate and output a signal.

The system control unit 320 is provided with a CPU 321, a ROM 322, a RAM 323, a DSP 324, an external interface 325, and the like.
The CPU 321 uses the ROM 322 and the RAM 323 to send an instruction to each unit of the imaging apparatus and controls the entire system.
The DSP 324 performs various kinds of signal processing on the imaging signal from the imaging unit 310, thereby generating a still image or moving image video signal (for example, a YUV signal) in a predetermined format.
The external interface 325 is provided with various encoders and D / A converters, and with external elements (in this example, the display 360, the memory medium 340, and the operation panel unit 350) connected to the system control unit 320. Various control signals and data are exchanged.

The memory medium 340 can appropriately store images taken on, for example, various memory cards, and can replace the memory medium with the memory medium controller 341, for example. As the memory medium 340, in addition to various memory cards, a disk medium using magnetism or light can be used.
The operation panel unit 350 is provided with input keys for a user to give various instructions when performing a shooting operation with the imaging apparatus. The CPU 321 monitors an input signal from the operation panel unit 350, Various operation controls are executed based on the input contents.
The display 360 is a small display such as a liquid crystal panel incorporated in the imaging apparatus, and displays a captured image.

(Seventeenth embodiment)
Next, a seventeenth embodiment will be described.
In the seventeenth embodiment, a case where the multilayer optical filter 10 of the present invention is applied to a display device will be described with reference to FIG.
FIG. 25 is a block diagram illustrating a configuration of an electronic device 70 such as a television device.
The electronic device 70 includes a display device 60, a first image processing unit 71, a buffer memory 72, a second image processing unit 73, an image compression unit 74, a storage unit 75, a microprocessor 77, and a backlight control. A unit 78 and a timing control unit 79 are included.
The first image processing unit 71 receives an input signal and performs predetermined signal processing.
The buffer memory 72 takes in the image signal supplied from the image processing unit 71.
The second image processing unit 73 receives the image signal supplied from the buffer memory 72 and performs predetermined signal processing.
The image compression unit 74 compresses the image signal supplied from the buffer memory 72 by a predetermined compression method.
The storage unit 75 records the image signal compressed by the image compression unit 74 on the recording medium 76.
The backlight control unit 78 controls the luminance of the backlight 61 of the display device 60.
The timing control unit 79 modulates each pixel of the liquid crystal panel 62 of the display device 60 based on the image signal.
The microprocessor 77 receives a control signal supplied via the user interface control unit 80 and controls each unit 72, 73, 74, 75, 78, 79.

The display device 60 includes a backlight 61, a liquid crystal panel 62, and a filter array 63.
The backlight 61 irradiates white light to the back surface of the liquid crystal panel 62.
The liquid crystal panel 62 modulates the light irradiated by the backlight 61 for each pixel.
That is, a luminance image generating unit is configured to generate a luminance image in which the luminance is modulated for each pixel by the backlight 61 and the liquid crystal panel 62.
The filter array 63 converts the light of each pixel modulated by the liquid crystal panel 62 into light of each color of R, G, B, and is constituted by the multilayer optical filter 10 of the present invention.
That is, the filter array 63 selectively transmits light forming the luminance image for each wavelength set in advance for each pixel.
Therefore, a luminance image obtained by modulating the white light emitted from the backlight 61 for each pixel by the liquid crystal panel 62 is converted into light of each color of R, G, and B by the filter array 63, whereby a color image is displayed on the display device 60. Is displayed.
Note that the luminance image generating means is not limited to the one configured by the backlight 61 and the liquid crystal panel 62, and various conventionally known configurations such as a light emitting LED provided for each pixel can be adopted. It is.

(Eighteenth embodiment)
Next, an eighteenth embodiment will be described.
In the eighteenth embodiment, a case where the multilayer optical filter 10 of the present invention is applied to a communication device will be described with reference to FIG.
FIG. 26 is a block diagram illustrating a configuration of the communication device 90.
The communication device 90 is a device that performs communication by inputting and outputting optical signals.
The communication device 90 includes an optical signal detection unit 91, an image processing unit 92, an image compression unit 93, a storage unit 94, a microprocessor 95, a buffer memory 96, a user interface control unit 97, and an LED light source control unit. 98 and an optical signal transmission unit 99.
The optical signal detection unit 91 receives an optical signal and converts it into an image signal as an electrical signal, and includes a filter that selectively transmits light of a specific wavelength among the light included in the optical signal. As this filter, the multilayer optical filter 10 of the present invention is used.
The image processing unit 92 receives the image signal supplied from the optical signal detection unit 91 and performs predetermined signal processing.
The image compression unit 93 compresses the image signal supplied from the image processing unit 92 by a predetermined compression method.
The storage unit 94 records the image signal compressed by the image compression unit 93 on the recording medium 100.
The buffer memory 96 takes in the image signal supplied from the image processing unit 92.
The microprocessor 95 receives a control signal supplied via the user interface control unit 97 and controls each unit 91, 92, 93, 94, 95, 96, 98.
The LED light source controller 98 supplies a control signal to the optical signal transmitter 99 according to the control of the microprocessor 95.
The optical signal transmission unit 99 drives the LED by supplying an electric signal to the LED as the light source based on the control signal supplied from the LED light source control unit 98, and transmits the optical signal.
The optical signal transmission unit 99 has a filter that selectively transmits light of a specific wavelength among optical signals to be transmitted, and the multilayer optical filter 10 of the present invention is used as this filter.

  8 ... Solid-state imaging device, 10 ... Multilayer optical filter, 12 ... First dielectric layer, 14 ... Second dielectric layer, 20 ... Spherical portion, 60 ... Display device, 90 ... Communication Device, 400...

Claims (19)

It is configured by alternately laminating first dielectric layers and second dielectric layers having different refractive indexes, and includes a multilayer film that selectively transmits light of a specific wavelength,
Each optical film thickness of the first dielectric and the second dielectric adjacent in the direction of lamination is formed to be 1/4 of the wavelength of the selectively transmitted light,
At least part of the thickness direction of the multilayer film, which is the direction in which the first dielectric layer and the second dielectric layer are laminated, is that each of the first dielectric layer and the second dielectric layer is , Formed as a spherical portion formed along a spherical surface having a convex shape in the same direction in the direction in which they are stacked.
Multilayer optical filter. In the spherical portion, the spherical surfaces along which the first dielectric layer and the second dielectric layer are aligned have the same curvature.
The multilayer optical filter according to claim 1. In the spherical portion, the center of the spherical surface along which the first dielectric layer and the second dielectric layer are aligned is the same.
The multilayer optical filter according to claim 1. A spacer layer made of a dielectric is provided in place of the first dielectric layer or the second dielectric layer at an intermediate portion in the thickness direction of the multilayer film.
The multilayer optical filter according to claim 1. The remaining part of the multilayer film excluding the spherical portion is formed as a planar portion in which the first dielectric layer and the second dielectric layer are laminated in parallel.
The multilayer optical filter according to claim 1. The remaining part of the multilayer film excluding the spherical part is formed as a planar part in which the first dielectric layer and the second dielectric layer are laminated in parallel.
A spacer layer made of a dielectric is provided between the spherical portion and the upper surface of the plane.
The multilayer optical filter according to claim 1. The spherical portion functions as a lens having a power to converge light transmitted through the spherical portion,
The multilayer optical filter according to claim 1. The spherical portion functions as a lens having a power to diverge light transmitted through the spherical portion,
The multilayer optical filter according to claim 1. A plurality of the multilayer films are provided side by side in a direction intersecting with a direction in which the first dielectric layer and the second dielectric layer are laminated,
A spacer layer made of a dielectric material is provided in place of the first dielectric layer or the second dielectric layer at an intermediate portion in the thickness direction of the plurality of multilayer films,
The spacer layer is formed with an optical film thickness determined by a specific wavelength of light that each multilayer film selectively transmits.
The multilayer optical filter according to claim 1. The number of the first dielectric layers and the number of the second dielectric layers constituting each multilayer film are the same,
A gap is provided between adjacent multilayer films,
The thickness of the spacer layer provided in each multilayer film is different from each other,
The first dielectric layers and the second dielectric layers of the adjacent multilayer films are connected in a straight line within the gap, or connected in a curved line.
The multilayer optical filter according to claim 9. The thickness of the spacer layer provided in each multilayer film is different from each other,
Each of the multilayer films includes a height adjustment layer provided on the surface of the first dielectric layer or the second dielectric layer located at the tip in the protruding direction of the spherical surface and having a height different from each other.
The surface of the height adjusting layer of each multilayer film is provided at the same height,
The multilayer optical filter according to claim 9. The thickness of the spacer layer provided in each multilayer film is different from each other,
The remaining multilayer film excluding the multilayer film having the largest height in the protruding direction of the spherical surface is provided on the surface of the first dielectric layer or the second dielectric layer located at the tip in the protruding direction of the spherical surface. Height adjustment layers with different heights are provided,
The surface of the multilayer film having the largest height in the protruding direction of the spherical surface and the surface of the height adjustment layer of the multilayer film are provided at the same height.
The multilayer optical filter according to claim 9. A spacer layer made of a dielectric is provided in place of the first dielectric layer or the second dielectric layer in the intermediate portion in the thickness direction of the multilayer film,
The spacer layer has a refractive index determined by a specific wavelength of light that each multilayer film selectively transmits.
The multilayer optical filter according to claim 1. A plurality of the multilayer films are provided side by side in a direction intersecting with a direction in which the first dielectric layer and the second dielectric layer are laminated,
A spacer layer made of a dielectric material is provided in place of the first dielectric layer or the second dielectric layer at the intermediate portion in the thickness direction of each multilayer film,
The spacer layer has a refractive index determined by a specific wavelength of light that each multilayer film selectively transmits,
In the adjacent multilayer film, each of the spacer layers is formed so that the physical thicknesses of the spacer layers are the same.
The multilayer optical filter according to claim 13. The spacer layer has a refractive index different from each other, and is composed of a sub-wavelength structure in which two types of media are arranged one-dimensionally or two-dimensionally with a size smaller than the wavelength of an electromagnetic wave to be transmitted.
The multilayer optical filter according to claim 4. A multilayer optical filter;
A photoelectric conversion unit that photoelectrically converts light transmitted through the multilayer optical filter,
The multilayer optical filter is:
It is configured by alternately laminating first dielectric layers and second dielectric layers having different refractive indexes, and includes a multilayer film that selectively transmits light of a specific wavelength,
Each optical film thickness of the first dielectric and the second dielectric adjacent in the direction of lamination is formed to be 1/4 of the wavelength of the selectively transmitted light,
At least part of the thickness direction of the multilayer film, which is the direction in which the first dielectric layer and the second dielectric layer are laminated, is that each of the first dielectric layer and the second dielectric layer is , Formed as a spherical portion formed along a spherical surface having a convex shape in the same direction in the direction in which they are stacked.
Solid-state image sensor. An imaging unit having a solid-state imaging device, a control unit for controlling the imaging unit, and an operation unit for operating the imaging unit,
The solid-state imaging device is
A multilayer optical filter;
A photoelectric conversion unit that photoelectrically converts light transmitted through the multilayer optical filter,
The multilayer optical filter is:
It is configured by alternately laminating first dielectric layers and second dielectric layers having different refractive indexes, and includes a multilayer film that selectively transmits light of a specific wavelength,
Each optical film thickness of the first dielectric and the second dielectric adjacent in the direction of lamination is formed to be 1/4 of the wavelength of the selectively transmitted light,
At least part of the thickness direction of the multilayer film, which is the direction in which the first dielectric layer and the second dielectric layer are laminated, is that each of the first dielectric layer and the second dielectric layer is , Formed as a spherical portion formed along a spherical surface having a convex shape in the same direction in the direction in which they are stacked.
Imaging device. Luminance image generation means for generating a luminance image in which the luminance is modulated for each pixel;
A filter array that selectively transmits the light forming the luminance image for each wavelength set in advance for each pixel;
The filter array is constituted by a multilayer optical filter,
The multilayer optical filter is:
It is configured by alternately laminating first dielectric layers and second dielectric layers having different refractive indexes, and includes a multilayer film that selectively transmits light of a specific wavelength,
Each optical film thickness of the first dielectric and the second dielectric adjacent in the direction of lamination is formed to be 1/4 of the wavelength of the selectively transmitted light,
At least part of the thickness direction of the multilayer film, which is the direction in which the first dielectric layer and the second dielectric layer are laminated, is that each of the first dielectric layer and the second dielectric layer is , Formed as a spherical portion formed along a spherical surface having a convex shape in the same direction in the direction in which they are stacked.
Display device. Comprising at least one of an optical signal transmitter for transmitting an optical signal and an optical signal detector for receiving an optical signal;
The at least one of the optical signal transmission unit or the optical signal detection unit has a filter that selectively transmits light of a specific wavelength among light included in the optical signal,
The filter is composed of a multilayer optical filter,
The multilayer optical filter is:
It is configured by alternately laminating first dielectric layers and second dielectric layers having different refractive indexes, and includes a multilayer film that selectively transmits light of a specific wavelength,
Each optical film thickness of the first dielectric and the second dielectric adjacent in the direction of lamination is formed to be 1/4 of the wavelength of the selectively transmitted light,
At least part of the thickness direction of the multilayer film, which is the direction in which the first dielectric layer and the second dielectric layer are laminated, is that each of the first dielectric layer and the second dielectric layer is , Formed as a spherical portion formed along a spherical surface having a convex shape in the same direction in the direction in which they are stacked.
Communication device.

Images