JP4579503B2 - Solid-state image sensor - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention Solid-state image sensor It is about.
[0002]
[Prior art]
The first conventional technique will be described.
[0003]
FIG. 61 is a schematic cross-sectional configuration diagram of a light-receiving portion of a color solid-state imaging device that is generally used in the past.
[0004]
Reference numeral 41 denotes a color filter, and reference numeral 42 denotes a microlens that is mounted to increase the aperture ratio of the solid-state imaging device. Reference numeral 44 denotes a light shielding portion, 45 denotes a pixel area of one pixel, and 46 denotes a photoelectric conversion element that converts incident light into electric charges.
[0005]
The solid-state imaging device has a plurality of pixel regions 45 shown in FIG. 4 arranged two-dimensionally, performs photoelectric conversion of incident light in a photoelectric conversion device 46 provided for each pixel 45, accumulates the charge, and stores the charge in an electric circuit. The image information is obtained based on the amount of charge periodically sent to a transfer unit (not shown) constituted by
[0006]
FIG. 61 shows a solid-state image pickup device for color, but a monochrome or single-color solid-state image pickup device does not have 41 color filters and does not have a microlens attached to increase the aperture ratio. is there.
[0007]
Such a solid-state imaging device is manufactured by praying a transfer unit (not shown) and portions constituting the solid-state imaging device such as the photoelectric conversion element 46 sequentially formed through a semiconductor process.
[0008]
In addition, the color filter of the color imaging element is also formed by a photolithography method or a printing method, and the microlens mounted for increasing the aperture ratio is also manufactured through a semiconductor process such as a photolithography method or a dry etching process.
[0009]
Currently, there are line sensors, area sensors, etc. as solid-state imaging devices used in such imaging systems, and each one is used in various fields, and the performance of the solid-state imaging device itself also improves year by year. is doing.
[0010]
Next, general characteristics of the solid-state imaging device will be described.
[0011]
FIG. 62 schematically shows the output characteristics of charges with respect to incident light in the pixels of the solid-state imaging device. When the illuminance of incident light is lower than a certain value as shown in FIG. Although it increases, it has a characteristic that the output saturates at more than that.
[0012]
As described above, the solid-state imaging device has a problem that the sensitivity width of photoelectric conversion, that is, the latitude is narrow.
[0013]
In addition, the photoelectrically converted charge is temporarily stored in each pixel, but when the amount of incident light becomes excessive, the amount of charge that can be stored in each pixel exceeds its limit, and surplus charges flow into other pixels. Causes the image to deteriorate significantly.
[0014]
When a solid-state image sensor is used for a camera or the like because of such a problem, an aperture or an ND filter for limiting the amount of light to the solid-state image sensor is provided.
[0015]
However, such a method is not suitable for photographing a subject with a large difference in luminance because it only controls the amount of light applied to the entire solid-state imaging device.
[0016]
This will be specifically described below.
[0017]
FIG. 63 is an example of a subject with a large difference in brightness, and a frame 52 indicates a range captured by a solid-state imaging device.
[0018]
In FIG. 63, 53 is a relatively bright mountain located in the background, 54 is a sun with a large amount of light, and 55 is a person with the sun 54 as the back.
[0019]
In such a case, if an aperture or ND filter is set so that the background mountain 53 is properly exposed, the amount of light of the person 55 becomes insufficient and the person becomes dark.
[0020]
Further, when a diaphragm or ND filter is set so that the person 55 is properly exposed, the relatively bright mountain 53 is overexposed and flies white and does not remain in the image.
[0021]
Further, in the vicinity of the sun 54, the output of the photoelectric conversion element saturates and flows into other pixels, or a charge is accumulated in the charge transfer unit, resulting in image degradation called blooming or smear.
[0022]
In addition to these problems, each pixel is caused by defects in the silicon wafer during manufacturing of the solid-state image sensor, differences in film thickness and film quality of the silicon oxide film, dirt on the cover glass surface of the solid-state image sensor, and scratches. There was a problem of causing sensitivity variations.
[0023]
As a method for solving such a problem, a method of controlling the amount of incident light for each photoelectric conversion element has been devised.
[0024]
According to this method, since the amount of incident light can be controlled for each pixel, the amount of light incident on the pixels corresponding to the periphery of the sun 54 described above can be reduced to prevent image degradation, and the sensitivity variation of each pixel can be controlled by controlling the amount of incident light. In addition, it is possible to compensate for sensitivity reduction (shading) in surrounding pixels, which is a problem in a camera using a solid-state image sensor.
[0025]
In this proposal, as disclosed in Japanese Patent Laid-Open No. 63-35081, several photoelectric conversion elements are combined into one group, and a liquid crystal shutter is provided on the entrance surface for each group. As disclosed in Japanese Patent No. 294059, there is a method in which liquid crystals are individually provided on the entire surface of each photoelectric conversion element, and the amount of light incident on each photoelectric conversion element is controlled independently.
[0026]
FIG. 64 shows a light amount control method using liquid crystal, and a light modulation element 56 is provided between the microlens 42 and the photoelectric conversion element 46 in the solid-state imaging device shown in FIG.
[0027]
The light modulation element 56 is made of a physical element such as an electronic element or a liquid crystal element, and its light transmittance can be changed by applying a voltage or the like.
[0028]
With such a configuration, the subject is observed with a solid-state imaging device prior to shooting, a photoelectric conversion element whose output is likely to be saturated is specified, and the light modulation element 56 corresponding to the photoelectric conversion element whose output is likely to be saturated during actual shooting. Can be driven to reduce the amount of incident light.
[0029]
Therefore, it is possible to perform imaging without image deterioration or overexposure even for a subject with a large difference in brightness.
[0030]
In addition, since a light modulation element is provided for each photoelectric conversion element, the sensitivity can be improved by driving the light modulation element so as to compensate for the sensitivity variation of each photoelectric conversion element during manufacture of a solid-state imaging element that has been measured in advance. Can be made uniform.
[0031]
For example, the transmittance of the light modulation element corresponding to the photoelectric conversion element having the lowest sensitivity among the solid-state imaging elements is maximized, and the transmittance of the light modulation element corresponding to each photoelectric conversion element having higher sensitivity is sequentially changed. This makes it possible to align the sensitivity of all photoelectric conversion elements.
[0032]
Furthermore, shading can also be corrected by setting the transmittance of the light modulation element corresponding to the solid-state image sensor in the central part lower than that in the peripheral photoelectric conversion element in the solid-state image sensor.
[0033]
Next, the second conventional technique will be described.
[0034]
In an imaging apparatus using a solid-state image sensor, a pseudo signal called moire is generated when a component having a frequency higher than half the sampling frequency (Nyquist frequency) determined by the pixel arrangement pitch is included in the image. There is a problem that the image gets worse.
[0035]
Therefore, in order to eliminate this problem, an optical element for removing high-frequency components of an image, that is, an optical low-pass filter, is used in the photographing optical system or in front of the imaging element. A typical example of such an optical low-pass filter is a birefringent plate using birefringence such as quartz. However, since this optical element is expensive, the following optical low-pass filter has been proposed.
[0036]
For example, in Japanese Patent Application Laid-Open No. 3-296013, a photographing light beam is shifted by periodically vibrating a photographing optical system or an image sensor in a direction orthogonal to the optical axis to obtain a low pass effect.
[0037]
Japanese Patent Laid-Open No. 8-43878 discloses a low-pass effect by driving a liquid prism arranged inside a photographing optical system at high speed, periodically oscillating its apex angle, and minutely changing the direction of a photographing light beam. It has gained.
[0038]
In Japanese Patent Application Laid-Open No. 11-215514, a color filter array and a micro prism array on which a dichroic mirror is formed are arranged immediately in front of an image pickup device, and a low pass effect is obtained by utilizing light reflection by the micro prism.
[0039]
Further, in Japanese Patent Registration No. 2539424, optical elements that are parallel to each other and have at least two boundary surfaces are arranged at an angle of about 45 degrees with respect to the optical axis, and the light beam is reflected by the boundary surface multiple times. The low pass effect is obtained.
[0040]
The second prior art will be described.
[0041]
In an imaging apparatus using a solid-state image sensor, a pseudo signal called moire is generated when a component having a frequency higher than half the sampling frequency (Nyquist frequency) determined by the pixel arrangement pitch is included in the image. There is a problem that the image gets worse.
[0042]
Therefore, in order to eliminate this problem, an optical element for removing high-frequency components of an image, that is, an optical low-pass filter is used in the photographing optical system or in front of the imaging element. A typical example of such an optical low-pass filter is a birefringent plate using birefringence such as quartz. However, since this optical element is expensive, the following optical low-pass filter has been proposed.
[0043]
For example, in Japanese Patent Application Laid-Open No. 3-296013, a photographing light beam is shifted by periodically vibrating a photographing optical system or an image sensor in a direction orthogonal to the optical axis to obtain a low pass effect.
[0044]
Japanese Patent Laid-Open No. 8-43878 discloses a low-pass effect by driving a liquid prism arranged inside a photographing optical system at high speed, periodically oscillating its apex angle, and minutely changing the direction of a photographing light beam. It has gained.
[0045]
In Japanese Patent Application Laid-Open No. 11-215514, a color filter array and a micro prism array on which a dichroic mirror is formed are arranged immediately in front of an image pickup device, and a low pass effect is obtained by utilizing light reflection by the micro prism.
[0046]
Further, in Japanese Patent Registration No. 2539424, optical elements that are parallel to each other and have at least two boundary surfaces are arranged at an angle of about 45 degrees with respect to the optical axis, and the light beam is reflected by the boundary surface multiple times. The low pass effect is obtained.
[0047]
[Problems to be solved by the invention]
The first conventional technique has the following problems.
[0048]
The light modulation element 56 described above is formed of a physical element such as an electronic element or a liquid crystal element as described above, but such a physical element is difficult to produce a complete transmission state or a complete light shielding state. There are drawbacks.
[0049]
Therefore, when the light modulation element 56 is arranged on the entire surface of the photoelectric conversion element 46, the amount of light incident on the photoelectric conversion element 45 is reduced even when the transmittance is maximum compared to the state where the light modulation element 56 is not arranged. End up.
[0050]
That is, there is a drawback that the absolute sensitivity of the photoelectric conversion element 46 is lowered.
[0051]
Therefore, when the subject is dark, good shooting cannot be performed.
[0052]
In addition, when the subject is extremely bright, even when the accumulation of the photoelectric conversion element 45 is completed and the charge is transferred, the photoelectric conversion element has a slight incidence of light, which deteriorates the image.
[0053]
This is because light from the outside leaks into the photoelectric conversion element 56 because the light modulation element 56 cannot be completely shielded.
[0054]
Furthermore, the light modulation element 56 has a problem of low response.
[0055]
This is because the operation of changing the transmission state of the light modulation element 56, for example, changing from the maximum transmittance to the minimum is slow, and there is a phenomenon that the operation becomes extremely slow particularly at low temperatures.
[0056]
Such a phenomenon is not a problem because the response is not so much considered in the case of limiting the amount of light incident on each photoelectric conversion element in advance before photographing.
[0057]
However, for the purpose of preventing the incidence of light rays on the photoelectric conversion element 46 at the stage of transferring the charge accumulated in the photoelectric conversion element 46, the responsiveness cannot be allowed. For example, when the charge is transferred, the photoelectric conversion element 46 is further charged. Will accumulate and the charge will also be transferred, degrading the image.
[0058]
When the light modulation element 56 is arranged in correspondence with each photoelectric conversion element 46 in this way, the amount of incident light can be controlled for each pixel, so that a wide dynamic range can be taken even for a subject with a large difference in brightness. A solid-state image sensor can be realized, and the sensitivity variation of each pixel can be supplemented by controlling the amount of incident light. Furthermore, sensitivity reduction (shading) can be corrected at surrounding pixels, which is a problem with cameras using solid-state image sensors. However, there remain problems such as sensitivity reduction and image degradation (light leakage due to low response of the light modulation element 56), which are the most important conditions for the camera.
[0059]
Therefore, in the present invention, taking advantage of the above-described solid-state imaging device using the light modulator 56 (dynamic range expansion, dispersion compensation, shading correction), and problems (decrease in absolute sensitivity, light leakage due to low response) It is an object of the present invention to realize a solid-state imaging device having a high image quality and a wide imaging area (good imaging can be performed even with a subject having a large difference in brightness).
[0060]
Furthermore, it aims to establish a method for stably manufacturing such a solid-state imaging device, and to construct a photographing system for adapting such a solid-state imaging device to a digital camera. .
[0061]
Further, the second conventional technique has the following problems.
[0062]
In the prior art disclosed in JP-A-3-296603 and JP-A-8-43878, a photographing optical system, an image sensor or a liquid prism as a driving target is generally difficult to be driven at high speed because of its large mass. A sufficient low-pass effect cannot be obtained in a shooting scene with a short exposure time. In addition, there is a risk of noise and vibration accompanying the drive.
[0063]
In the prior art disclosed in Japanese Patent Laid-Open No. 11-215514, it is necessary to accurately align each unit prism of the microprism array on a one-to-one basis with each pixel of the image sensor. In general, since the dichroic mirror has a large wavelength dependency with respect to the incident angle, there is also a drawback that the spectral sensitivity characteristics of the image sensor differ between the center and the periphery of the image sensor.
[0064]
In the prior art disclosed in Japanese Patent Registration No. 2539424, the optical element has to be disposed at a predetermined angle with respect to the photographing optical axis, which requires a large space in the direction of the optical axis, which increases the size of the device. End up.
[0065]
Therefore, the present invention provides an optical low-pass filter that is small and has a variable low-pass effect by relatively driving two micro-prism arrays with a predetermined waveform and finely displacing the path of the light beam in the array. It is intended to provide.
[0066]
The problem of the second prior art will be described.
[0067]
However, the above prior art has the following drawbacks.
[0068]
In the prior art disclosed in JP-A-3-296603 and JP-A-8-43878, a photographing optical system, an image sensor or a liquid prism as a driving target is generally difficult to be driven at high speed because of its large mass. A sufficient low-pass effect cannot be obtained in a shooting scene with a short exposure time. In addition, there is a risk of noise and vibration accompanying the drive.
[0069]
In the prior art disclosed in Japanese Patent Laid-Open No. 11-215514, it is necessary to accurately align each unit prism of the microprism array on a one-to-one basis with each pixel of the image sensor. In general, since the dichroic mirror has a large wavelength dependency with respect to the incident angle, there is also a drawback that the spectral sensitivity characteristics of the image sensor differ between the center and the periphery of the image sensor.
[0070]
In the prior art disclosed in Japanese Patent Registration No. 2539424, the optical element has to be disposed at a predetermined angle with respect to the photographing optical axis, which requires a large space in the direction of the optical axis, which increases the size of the device. End up.
[0071]
Therefore, the present invention provides an optical low-pass filter that is small and has a variable low-pass effect by relatively driving two micro-prism arrays with a predetermined waveform and finely displacing the path of the light beam in the array. It is intended to provide.
[0072]
[Means for Solving the Problems]
To solve the above problem, In a solid-state imaging device comprising a light receiving region comprising a plurality of photoelectric conversion elements arranged on the main surface of a semiconductor substrate and a light control region comprising a light control element for controlling the incident state of each photoelectric conversion element, the light control element Is opposed to the movable microprism provided for each photoelectric conversion element, the fixed microprism facing the movable microprism and having an incident surface substantially the same shape as the prism surface of the movable microprism, and the other end of the movable microprism. Drive means for changing the optical waveguide, the first opposing interval of the movable microprism and the fixed microprism, and the second opposing interval of the movable microprism and the optical waveguide, and the first opposing interval and / or the second opposing interval And a light shielding means for shielding stray light due to the emission of light rays on a surface along the main optical axis, and the first facing interval is changed by the driving means. A solid-state image pickup device for controlling the transmission state and the total reflection state of the movable micro prism Rukoto I will provide a.
[0082]
DETAILED DESCRIPTION OF THE INVENTION
(First embodiment)
FIG. 1 is a cross-sectional view of a solid-state image sensor 10 according to a first embodiment of the present invention.
[0083]
In the drawing, the alphabet at the end of the part number of each constituent member is provided for identification for each pixel, and when there is no alphabet at the end of the part number in the sentence, the general name thereof is shown.
[0084]
Reference numeral 11aa denotes a microlens, which is provided corresponding to each of the photoelectric conversion elements 111 described later and has a refractive index of 2.0 or more. ² The transparent bright layer is cut into a lens shape by etching.
[0085]
Reference numeral 12aa denotes a color filter. For example, in the case of a Bayer array primary color filter, red, blue, and green filters are alternately adjacent to each other and arranged as shown in FIG.
[0086]
In FIG. 2, 12aa, 12ac, 12ca and 12cc are red filters, 12ab, 12ac, 12ba, 12bc, 12cb, 12cd, 12da and 12dc are green filters, and 12bb, 12bd, 12db and 12dd are blue filters.
[0087]
Each filter can be formed by forming a resin (color resin) layer containing a pigment or dye of a desired color at a predetermined location by a method such as photolithography.
[0088]
13aa is an optical waveguide, which is similar to a microprism as TIO. ² The transparent bright layer has a cylindrical shape formed by etching by an etching method.
[0089]
The planarization layer 15 is SIO so as to wrap the outer periphery of the optical waveguide 13aa. ² The optical waveguides are coupled to each other.
[0090]
Material SIO of the planarization layer 15 ² The refractive index of the optical waveguide 13 is about 1.5 and the material TIO of the optical waveguide 13 ² Since the refractive index of the optical waveguide 13 is different from the refractive index of about 2.3, the light beam to be emitted from the side surface 14 of the light beam incident on the optical waveguide 13 is totally reflected at the interface with the planarizing layer 15 and guided to the photoelectric conversion element 111. .
[0091]
Therefore, the amount of light guided to the photoelectric conversion element 111 is increased, and the sensitivity of the solid-state imaging element 10 is improved.
[0092]
The planarizing layer 15 is provided with a later-described silicon leaf spring 16aa, and a movable microprism 17aa is formed at the other end of the leaf spring 16aa.
[0093]
The movable microprism 17aa is placed on the optical waveguide 13aa ² However, in the manufacturing process, an extremely thin sacrificial layer is provided between the optical waveguide 13aa and the movable microprism 17aa by removing the sacrificial layer by etching after the movable microprism is formed.
[0094]
The tip of the movable microprism 17aa has a corner cube shape, and the transparent electrode layer 18aa made of ITO and the SIO are on the tip. ² An insulating layer 19aa made of is provided.
[0095]
As described above, the second substrate is formed by the microlens 11, the color filter 12, the optical waveguide 13, the planarizing layer 15, and the leaf spring 16.
[0096]
Similar to the planarization layer 15, SIO ² A photoelectric conversion element 111aa is provided at the outermost end portion of the base layer 110 formed in (1).
[0097]
The photoelectric conversion element 111 has, for example, an n-type impurity region on a matrix on a p-type impurity region (which is omitted as a base layer over the entire lower surface of the photoelectric conversion device 111 but is omitted for the sake of simplicity). And p on each of them ⁺ This is a buried photodiode formed by providing a type impurity region, and the n type impurity region functions as a charge storage region.
[0098]
A read gate 112aa is provided adjacent to the photoelectric conversion element 111aa.
[0099]
The read gate 112aa is a region of the vertical transfer electrode 113aa and is formed of a polysilicon layer.
[0100]
A vertical transfer charge transfer path 114aa is provided to face the vertical transfer electrode 117aa.
[0101]
When a read pulse (for example, 15 V) is applied to the vertical transfer electrode 113aa, the charge accumulated in the photoelectric conversion element (here, the photoelectric conversion element 111aa) corresponding to the vertical transfer electrode 113aa corresponds via the read gate 112aa. Data is read out to the vertical charge transfer path 114aa.
[0102]
Reference numeral 115aa denotes a channel stop region, which is provided so as to surround the periphery of the photoelectric conversion element 111aa other than the readout gate 112aa.
[0103]
The channel stop region 115aa is provided to electrically isolate other adjacent discharge conversion elements and transfer regions.
[0104]
Charge reading from the photoelectric conversion element 111 to the vertical charge transfer path 114 is performed in units of the photoelectric conversion element 111, and there are various reading modes such as interlace scanning, progressive scanning, and all pixel reading.
[0105]
The base layer 110 is provided with a fixed microprism 116aa.
[0106]
Fixed microprism 116aa is TIO ² The surface facing the movable microprism 17aa has a shape obtained by transferring the tip shape.
[0107]
At the tip thereof, a transparent electrode layer 117aa made of ITO and SIO ² An insulating layer 118aa made of is provided.
[0108]
Again, the material SIO of the base layer 15 ² Has a refractive index of about 1.5 and the material TIO of the fixed microprism 116. ² Since the refractive index is different from about 2.3, the light beam entering the fixed microprism 116 is totally reflected at the interface with the base layer 110 and guided to the photoelectric conversion element 111 direction.
[0109]
119aa is an aluminum light-shielding layer, and stray light that enters the planarizing layer 15 and the base layer 110 through the gap (second opposing interval) between the optical waveguide 14 and the movable microprism 17, or the movable microprism 17 and the fixed microprism. The stray light incident on the base layer 110 through a gap (first facing interval) 116 is provided as a light shielding unit for reaching the photoelectric conversion element 111.
[0110]
These base layer 110, photoelectric conversion element 111, readout gate 112, vertical transfer electrode 113, vertical transfer path 114, channel stop region 115, fixed microprism 116, and light shielding layer 119 form a first base.
[0111]
The state of the light beam incident from the microlens 11 in the solid-state imaging device 10 having such a structure will be described below.
[0112]
In the case of the microlens 11aa, the distance between the optical waveguide 13aa and the movable microprism 17aa (second opposing distance) is extremely small, for example, about 0.5 μm, and conversely, the distance between the movable microprism 17aa and the fixed microprism 116aa ( The first facing interval) has a gap of about 1.0 μm to 2.0 μm.
[0113]
In such a case, the light beam 120aa is incident from the optical waveguide 13aa to the movable microprism 17aa through the second facing interval, but the light beam 120aa is folded back at the tip of the movable microprism 17aa as shown in the figure. Reflected to the side.
[0114]
This is because the tip of the movable microprism 17aa and the tip of the fixed microprism 116aa are separated, so that the movable microprism 17aa is movable due to the difference between the refractive index of the movable microprism 17aa (about 2.3) and the refractive index of the air layer (1.0). This is because the light beam 120aa is totally reflected by the corner cube surface at the tip of the microprism 17aa.
Therefore, the light beam that has passed through the microlens 11aa does not reach the photoelectric conversion element 111aa.
[0115]
That is, when the movable microprism 17aa is in such a state, the shutter is closed with respect to the photoelectric conversion element 111aa.
[0116]
On the other hand, in the case of the microlens 17ac, the interval between the optical waveguide 13ac and the movable microprism 17ac (second opposing interval) is, for example, about 1.0 μm, and the interval between the movable microprism 17ac and the fixed microprism 116ac (first ) Is extremely small, about 0.5 μm.
[0117]
Even in such a case, the light beam 120ac is incident from the optical waveguide 13ac to the movable microprism 17ac through the second facing interval therebetween.
[0118]
This is because, although the second counter gap is open, the tip of the optical waveguide 13aa is at a right angle to the incident light beam 120ac, so that total reflection due to the difference in refractive index does not occur.
[0119]
Therefore, the light beam 120ac exits the optical waveguide 13ac and enters the movable microprism 17ac.
[0120]
The light beam 120ac incident on the movable microprism 17ac is not reflected and reflected at the tip portion like the light beam 120aa, but enters the fixed microprism 116ac as it is.
[0121]
This is because the first facing interval is narrow, and the refractive index of the movable microprism (about 2.3) and the refractive index of the fixed microprism (about 2.3) are aligned. This is because total reflection of light does not occur at the tip.
[0122]
The light beam 120ac incident on the fixed microprism 13ac is directly guided to the photoelectric conversion element 111ac.
[0123]
That is, when the movable microprism 17ac is in this state, the shutter is open with respect to the photoelectric conversion element 111ac.
[0124]
The state of the movable microprism 17 and the leaf spring 16 that supports the movable microprism 17 is formed in a state where the second facing interval is opened like the movable microprism 17ac at the time of manufacture.
[0125]
When the movable microprism 17 is moved in the direction of the optical waveguide 13 against the leaf spring 16, the movable microprism 17aa is obtained.
[0126]
Here, a driving method for setting the movable microprism 17 to a position like 17aa will be described.
[0127]
At this time, the transparent electrode layer 18aa of the movable microprism 17aa is positively charged, and similarly, the transparent electrode layer 117aa of the fixed microprism 116aa is also positively charged.
[0128]
Then, the movable microprism 17aa and the fixed microprism 116aa repel each other, and the movable microprism 17aa is pushed in the direction of the optical waveguide 13aa against the spring force of the leaf spring 16aa, so that the first facing interval is opened and the second facing interval is opened. Becomes narrower.
[0129]
As described above, when the movable microprism 17 is not driven, the shutter is opened. When the movable microprism 17 is driven, a normally open shutter is formed.
[0130]
Conversely, when the charging of the two transparent electrodes 18aa and 117aa is released, the movable microprism 17aa returns to the state of the movable microprism 17ac by the restoring force of the leaf spring 16aa.
[0131]
Of course, instead of uncharging the two transparent electrodes 18aa and 117aa, the transparent electrode 117aa may be negatively charged, and the movable microprism 17aa may be attracted to the fixed microprism 116aa side. The microprism 17aa may be returned to the state of the movable microprism 17ac, and then the state may be maintained by the spring force of the leaf spring 16aa.
[0132]
Now, the mass of the movable microprism 17ac is obtained from the pixel pitch in the general solid-state imaging device 10, and the deflection of the movable microprism 17ac is the second opposing interval even when a force (100G) 100 times the mass is applied. Consider the case where the spring constant of the leaf spring 16ac is set to about 1/5 of the above.
[0133]
This is because the support force is set to prevent the movable microprism 17ac from colliding with the optical waveguide 13ac and the fixed microprism 117aa when the solid-state image pickup device 10 is dropped and being destroyed. become.
[0134]
When such a spring constant is set, in order to move the movable microprism 17ac to the position of the movable microprism 17aa against the spring force of the leaf spring 16ac, the two transparent electrode layers 18aa and 117aa are both charged to about 10V. You can let it go.
[0135]
In particular, since the two transparent electrode layers 18aa and 117aa are close to each other at the beginning of driving (the first facing interval is narrow), the driving force sufficient to start moving the movable microprism can be given by applying this voltage. .
[0136]
Generally, when moving a mass point, a large force is required at the beginning of movement due to the influence of friction, etc., but if the gap that generates an electrostatic force at the beginning of movement as described above is narrow, the generated force at that time becomes extremely large. Stable driving against friction is possible.
[0137]
Of course, as described above, there is a driving method using suction instead of a driving method using repulsion.
[0138]
The relationship between the leaf spring 16 and the movable microprism 17 is such that when the second substrate is formed in FIG. 1, the second facing interval is narrow and the first facing interval is widened (position of the movable microprism 17aa). Should be set.
[0139]
In this case, the normal position is the shutter closed like the movable microprism 17aa, and the state of the movable microprism 17ac is created by charging the two transparent electrodes 18aa and 117aa positively and negatively and attracting them to each other.
[0140]
Thus, when the movable microprism 17 is not driven, the shutter is closed, and when it is driven, a normally closed shutter is opened.
[0141]
The spring force of the leaf spring 16 is about 1/5 of the second facing interval even when a force (100 G) that is 100 times the mass of the movable microprism 17 is applied as described above. In the case of setting, the voltage for suction may be a potential difference of about 5 V between the two transparent electrode layers 18aa and 117aa.
[0142]
The difference between the voltage in the case of suction and the voltage in the case of repulsion is that, in the case of suction, the distance between the transparent electrodes 18aa and 117aa becomes narrower and the leaf spring becomes closer to a predetermined distance. This is because the movable microprism 17 is attracted to the fixed microprism 116 regardless of the 16 linear spring force.
[0143]
As shown in FIG. 3, the pair of transparent electrode layers is provided with a transparent electrode layer 121aa and an insulating layer 122aa on the surface facing the movable microprism 17aa of the optical waveguide 13aa, and transparent on the surface facing the optical waveguide 13aa of the movable microprism 17aa. An electrode layer 123aa and an insulating layer 124aa may be provided.
[0144]
In this case, if the transparent electrode layers 121aa and 123aa are charged to the same polarity, the second facing interval is opened due to repulsion and the movable microprism 17ac is opened, and the shutter is opened. 1 is opened, the movable micro prism 17aa is brought into a closed state, and the shutter is closed.
[0145]
If the movable microprism is formed in a state of 17ac at the time of manufacture, a potential difference of about 5V is applied to the two transparent electrode layers 121aa and 123aa, and the shutter can be closed by sucking each other.
[0146]
As a driving method of the movable microprism 17 by suction and repulsion using static electricity, a transparent electrode layer is provided on the movable microprism 17 and the fixed microprism 116 in FIGS.
[0147]
However, the position of the transparent electrode layer is not limited thereto.
[0148]
FIG. 4 shows an example in which a transparent electrode layer is provided on the leaf spring 16.
[0149]
Here, FIG. 4 shows an enlarged view of the central pixel (movable microprism 17ab) in FIGS.
[0150]
In FIG. 4, a transparent electrode layer 125ab and an insulating layer 126ab covering the transparent electrode layer 125ab are provided on the side of the leaf spring 16ab facing the base layer 110.
[0151]
A transparent electrode layer 127ab and an insulating layer 128ab are also provided on the surface of the base layer 110 facing the leaf spring 16ab.
[0152]
FIG. 5 is a plan view of the movable microprism 17 and the leaf spring 16 as viewed from the base layer 15 side. (FIG. 1 and FIG. 3 are AA cross sections of FIG. 5)
As can be seen from FIG. 5, the direction of the leaf spring 16 is changed between adjacent pixels with respect to each movable microprism 17.
[0153]
This is because a spring as long as possible is arranged by effectively using the interval between the movable microprisms 17.
[0154]
The leaf spring 16aa is attached to the planarizing layer 15 by a base portion 16aa1, and the opposite end face 16aa2 is attached to the movable microprism 17aa.
[0155]
Therefore, the movable microprism 17aa is supported so as to be movable with respect to the planarizing layer 15 in the direction perpendicular to the paper surface against the spring force of the leaf spring 16aa.
[0156]
Similarly, the leaf spring 16ab is attached to the flattening layer 15 by the root portion 16ab1, the opposite end face 16ab2 is attached to the movable microprism 17ab, and the leaf spring 16ac is attached to the flattening layer 15 by the root portion 16ac1. 16ac2 is attached to the movable microprism 17aa.
[0157]
Therefore, each microprism 17 can move independently in the direction perpendicular to the paper surface.
[0158]
A transparent electrode layer 125aa (not visible in a plan view) and an insulating layer 126aa are provided in the vicinity of the attachment portion (end face 16aa2) of the movable microprism 17aa of the leaf spring 16aa. Similarly, the movable microprism of the leaf spring 16ab is provided. A transparent electrode layer 125ab (not visible in the plan view) and its insulating layer 126ab are provided near the mounting portion (end surface 16ab2) of 17ab, and transparent near the mounting portion (end surface 16ac2) of the movable microprism 17ac of the leaf spring 16ac. An electrode layer 125ac (not visible in a plan view) and its insulating layer 126ac are provided.
[0159]
The movable microprism 17 is driven in association with the transparent electrode layer 127 provided on the base layer 110 so as to face the transparent electrode layer 125.
[0160]
For example, when the movable microprism 17 and the leaf spring 16 are formed in the state of the movable microprism 17ac shown in FIG. 1a, the shutter is normally opened, and the transparent electrode layers 125 and 127 are given the same polarity charges to each other. By repelling, the movable microprism 17 and the fixed microprism can be separated to close the shutter.
[0161]
Alternatively, when the movable microprism 17 and the leaf spring 16 are formed in the state of the movable microprism 17aa in FIG. 1, the shutter is normally closed, and the transparent electrode layers 125 and 127 are given opposite charges to each other. By suctioning, the distance between the movable microprism 17 and the fixed microprism can be reduced to open the shutter.
[0162]
The voltages required for these driving operations are the same as those shown in FIGS.
[0163]
Thus, when the transparent electrode layer is not provided on the optical path (on the movable microprism 17, the fixed microprism 116, and the optical waveguide 13), the sensitivity of the photoelectric conversion element 111 can be stably manufactured without variation.
[0164]
All of the above driving methods use static electricity, but the driving method is not limited thereto.
[0165]
FIG. 6 shows a case where a piezoelectric member is used as a driving method.
[0166]
In FIG. 6, a piezoelectric member 129ab is provided on the surface of the leaf spring 16ab facing the base layer 110.
[0167]
FIG. 7 is a plan view of the movable microprism 17 and the leaf spring 16 as viewed from the base layer 15 side.
[0168]
In FIG. 7, a piezoelectric member 129aa is provided on the leaf spring 16aa.
[0169]
The piezoelectric member 129aa has piezoelectric characteristics and is formed on the leaf spring 16aa by vapor deposition or the like.
[0170]
By applying a voltage, the piezoelectric member 129aa generates a bending force due to a change in strain with respect to the polarization direction, and the leaf spring 16aa is bent by the force.
[0171]
The bent leaf spring 16aa displaces the movable microprism 17aa fixed to the tip thereof.
[0172]
For example, as described above, the mass of the movable microprism 17 is obtained, and the deflection of the movable microprism 17 is about 1/5 of the second facing interval even when a force (100 G) that is 100 times the mass is applied. When the spring constant of the leaf spring 16 is set, the movable microprism 17 is displaced by about 2 μm in the paper surface direction of FIG. 7 by applying a voltage of about 0.3 V to the piezoelectric member.
[0173]
Therefore, the shutter closing and shutter opening of the solid-state imaging device 10 can be controlled by this displacement.
[0174]
Similarly, a piezoelectric member 129ad is provided on the leaf spring 16ad, and a piezoelectric member 129ac is provided on the leaf spring 16ac.
[0175]
Therefore, by individually applying a voltage to each piezoelectric member, each photoelectric conversion element can be independently controlled to be in a shutter closed or open state.
[0176]
In the conventional example, a variable ND of the liquid crystal is provided for each pixel, and the saturation pixel is controlled by the amount of transmitted light. However, the shutter is closed at an early point, and the saturation control is not performed by the light transmission amount but by the accumulation time.
[0177]
FIG. 8 shows a pixel that does not reach saturation at an assumed appropriate exposure time, and the timing at which the shutter is closed in the curve 130A is the timing at which the charge accumulation is completed and the process proceeds to charge transfer.
[0178]
However, even in the same photographic subject, the pixel in FIG. 9 receives extremely bright subject light, and charge saturation is expected at an appropriate exposure time, so the shutter is closed earlier than the shutter closing timing in FIG. (Curve 130B)
The timing difference 131 is set as an accumulation control time for preventing saturation.
[0179]
In the above driving method, particularly when the movable microprism 17 is driven electrostatically, a large voltage is required.
[0180]
A method for avoiding this will be described below.
[0181]
When the spring constant is set so that the movable microprism 17 is displaced by 1/5 (0.4 μm) of the second facing interval when a force 100 times the mass of the movable microprism 17 is applied as in the above calculation. The resonance frequency determined by the movable microprism 17 and the leaf spring 16 at that time is about 7 Hz.
[0182]
Then, the leaf spring 16 or the like is made of a silicon compound, and friction and viscous resistance are not applied to the movable microprism 17. In the case where air damping is taken into consideration, Q at the resonance point can be expected to be 500 or more.
[0183]
That is, when driving at the resonance frequency, a driving force that is 1/500 of the normal driving is sufficient. For example, in FIG. 1a, the two transparent electrode layers 18aa and 117aa are both charged to about 10V so that the movable microprism 17ac is moved to the position of the movable microprism 17aa against the spring force of the leaf spring 16ac. The voltage applied to the electrodes 18aa and 117aa can be 20 mv.
[0184]
In this way, the amount of the driving force is reduced, for example, to reinforce the spring, and the spring is increased 50 times (thickness is 4 times) with the driving force of 1 V, so that a durable solid-state imaging device 10 can be obtained.
[0185]
In the case of the normally closed shutter type, the period during which resonance driving is performed is during the time when the shutter is closed, and the movable microprism 17 is continuously driven during resonance.
[0186]
Of course, since the drive is performed alternately, the shutter closed state and the shutter open state occur alternately, and although there is no complete light shielding state, there is a possibility that saturation occurs in a specific pixel depending on the state of the subject. Sometimes, during storage, the shutter of the pixel can be driven to resonate to reduce the light incident on the photoelectric conversion element 111 of the pixel.
[0187]
In the conventional example, a variable ND of the liquid crystal is provided for each pixel, and the saturation pixel is controlled by the amount of transmitted light. However, in the present invention, the pixel is saturated by intermittently closing the shutter instead of ND. Prevention is in progress.
[0188]
FIG. 10 is a pixel that does not reach saturation with an appropriate exposure time assumed in the same manner as in FIG. 8, and the timing at which the shutter is closed in the curve 130A is the timing at which the charge accumulation is completed and the process proceeds to charge transfer.
[0189]
However, even in the same photographic subject, the pixel in FIG. 9 receives extremely bright subject light, and charge saturation is expected at an appropriate exposure time, so that the amount of incident light is reduced by opening and closing the shutter during the accumulation time. . (Curve 130C)
Then, the shutter is closed at the same timing (charge transfer timing) as the shutter closing timing in FIG.
[0190]
The alternate shutter opening / closing period 131 is set as an accumulation control time for preventing saturation.
[0191]
Further, a description will be given of performing the complete light-shielding of the solid-state imaging device 10 at the time of charge transfer instead of the saturation control with less power.
[0192]
For example, in the configuration as shown in FIG. 3, when a reverse voltage is applied to the transparent electrode layers 121ac and 123ac, they are attracted to each other and the distance between the movable microprism 17ac and the optical waveguide 13ac (second facing distance) becomes narrow.
[0193]
At this time, when a voltage is alternately applied to the transparent electrode layers 121ac and 123ac, the second facing interval also varies accordingly.
[0194]
However, the electrostatic force becomes extremely large at the timing when the second facing interval becomes narrow in the resonance state.
[0195]
This is because the electrostatic attraction force is proportional to the square of the interval between them.
[0196]
Therefore, when the second facing interval is narrowed, the movable microprism 17ac can be attracted to the optical waveguide 13ac with a small amount of power.
[0197]
As a driving method, first, a voltage is alternately applied to the transparent electrode layers 121ac and 123ac, and after the resonance driving of the movable microprism 17 is stabilized, the transparent electrode layer 121ac is moved at the timing when the movable microprism 17 is closest to the optical waveguide 13ac. , 123ac is applied with a DC voltage of about 2V, for example.
[0198]
At that time, the movable microprism 17ac is attracted to the optical waveguide 13ac.
[0199]
Since the second facing interval is further narrowed after the adsorption, the voltage applied to the transparent electrode layers 121ac and 123ac may be further reduced (for example, 0.1 v).
[0200]
FIG. 12 shows the time change of the voltage applied to the transparent electrode layers 121 and 123. The voltage change is divided into five stages as shown in the figure.
・ Shutter opening period
Period when exposure is started and charge corresponding to the subject image is accumulated in the solid-state imaging device 10 / initial driving period
Period when exposure time is almost over and shutter starts to close to prevent smearing during transfer
During this period, an alternating voltage is applied to the transparent electrode layers 121 and 123 to resonate the movable microprism 17, and the second facing interval with the optical waveguide 13 is made closer or wider.
[0201]
Since the amplitude of resonance increases as the vibration becomes stable, the second facing interval changes greatly accordingly (the timing at which the interval between the movable microprism 17 and the optical waveguide 13 approaches considerably).
Incidentally, in the curve 130D in the figure, only alternating driving for two wavelengths is performed in the initial driving period, but this is to make the figure easier to see, and in actuality, driving with more cycles is performed.
・ Adsorption period
Period of closing the shutter for charge transfer
In the initial drive period, the drive amplitude is stabilized (the amplitude is increased), and in this state, the DC voltage is applied to the transparent electrode layers 121 and 123 at the timing when the second facing interval is the narrowest.
[0202]
At this time, since the second facing interval is narrow, the movable microprism 17 is attracted to the optical waveguide 13 even when a small voltage is applied. (Shutter closed)
・ Stable period
When the movable microprism 17 is attracted to the optical waveguide 13, the second facing interval therebetween becomes extremely narrow, so that the attracted state can be maintained even with a smaller voltage.
[0203]
Therefore, the applied voltage is lowered to save power. (The applied voltage during this stable period may be the same as the adsorption period)
・ Shutter opening period
A time period in which the subject image is guided to the solid-state image sensor 10 even after the photographing is completed so that the photographer can continuously observe the subject image.
[0204]
The voltage application to the transparent electrode layers 121 and 123 is stopped, and the second opposing interval is widened by the spring force of the leaf spring 16. (The first facing interval between the movable microprism 17 and the fixed microprism 116 is narrowed so that the subject light beam can enter the fixed microprism 116 from the movable microprism 17).
In this period, the voltage application to the transparent electrode layers 121 and 123 is not stopped, but the shutters are repelled by applying voltages of the same polarity to each other, and the applied voltage is removed after the shutters are opened. good.
[0205]
Here, how the solid-state imaging device 10 is used will be considered.
[0206]
In FIG. 13, a light beam incident from a lens 133 passes through a diaphragm 135 and is limited in light amount, and then forms an image on the solid-state imaging device 10.
[0207]
The imaging means is composed of a semiconductor imaging device such as a MOS or CCD.
[0208]
The lens 133 moves on the optical axis 132 by the AF drive motor 134 for focus adjustment to adjust the focus.
[0209]
The AF drive motor 134 is driven by the focus drive unit 138, and the aperture 135 of the aperture 135 is determined by the aperture drive unit 137 and controls the light rays incident on the solid-state image sensor 10.
[0210]
The focus driving means 138 and the aperture driving means 137 are controlled by the imaging control means 141, respectively.
[0211]
The photographing control unit 141 determines the diameter of the aperture 135 and the shutter speed of the solid-state image sensor 10 by performing photometry with an image captured by the signal processing unit 142 described later. We are looking for a point of focus.
[0212]
The photographing control unit 141 drives the aperture driving unit 137 to optimize the diameter of the aperture 135, and controls the focus driving unit 138 to drive the AF motor 134 to focus on the subject.
[0213]
Further, when photographing, the subject information is accumulated in each photoelectric conversion element 111 of the solid-state image sensor 10, and when the accumulated charge is transferred or when the accumulated charge may be saturated before the transfer, as described above, it is transparent. The electrode layers 18aa, 117aa, 121aa, 123aa are charged to create a shutter closed state.
[0214]
The video signal output from the solid-state imaging device 10 is converted into a digital signal by the A / D conversion unit 140 and input to the signal processing unit 142.
[0215]
The signal processing circuit 142 performs signal processing such as forming a luminance signal and a color signal to form a color video signal.
[0216]
The formed image vibration is displayed on the display unit 143 and recorded on the recording unit 144.
[0217]
In addition, when the subject is dark, the flash unit 139 controlled by the photographing control device 141 irradiates the subject as necessary at the time of photographing to compensate for the insufficient exposure of the subject.
[0218]
FIG. 1n is a flowchart summarizing the operation of the camera. This flow starts when the main switch of the camera is turned on.
[0219]
In step # 1001, the solid-state imaging device 10 captures the subject image and measures its brightness to obtain the control amount of the diaphragm 135.
[0220]
In Step # 1002, the diaphragm driving unit 137 adjusts the appropriate amount of light to be incident on the solid-state imaging device 10 in accordance with the obtained control amount, and waits until the adjustment is completed.
[0221]
In step # 1003, the subject image formed on the solid-state image sensor 10 is displayed on the display means 143.
[0222]
In step # 1004, the process circulates and waits until the photographer presses the release button halfway sw1 and performs the shooting preparation unit operation. When sw1 is turned on, the process proceeds to step # 1005.
[0223]
In step # 1005, the solid-state image pickup device 10 picks up an image of the subject, the signal processing unit 142 detects the contrast of the image, and the AF control unit 141 and the focus adjustment unit 138 drive the AF motor 134 over the entire focus detection range. 133 is fed out and the feeding position with the highest contrast is obtained.
[0224]
Further, the subject is re-photometered under the current aperture state to obtain an exposure time necessary for photographing.
[0225]
It circulates through step # 1005 and waits until the above operation is completed.
[0226]
In step # 1006, the lens 133 is driven again toward the obtained feeding position, and the lens 133 is stopped at that position.
[0227]
It circulates through step # 1006 and waits until the operation is completed.
[0228]
In step # 1007, an element is selected from which the charge is saturated by the amount of incident subject light in each of the photoelectric conversion elements 111 of the solid-state image sensor 10 or is predicted to be saturated at the time of photographing.
[0229]
Further, an optimum exposure time is obtained for each pixel for which saturation is expected.
[0230]
In step # 1008, the release button is pressed down and the process waits until a shooting instruction is given.
[0231]
Note that this flow returns to the start when the release button half-press sw1 is released during this standby step.
[0232]
In step # 1009, shooting is started.
[0233]
At this time, first, a predetermined reset pulse is given to each photoelectric conversion element 111 to sweep away the accumulated charges.
[0234]
Next, each photoelectric conversion element 111 starts accumulation of subject light.
[0235]
The shutter is closed when the exposure time obtained in step # 1005 has elapsed. At this time, the specific photoelectric conversion element obtained in step # 1007 is not the exposure time obtained in step # 1005 but in step # 1007. The shutter is closed after the obtained exposure time has elapsed.
[0236]
That is, the shutter of the photoelectric conversion element expected to be saturated is closed earlier than the other photoelectric conversion elements.
[0237]
In step # 1010, the shutter is closed for the photoelectric conversion element that is not expected to be saturated.
[0238]
Note that closing the shutter in steps # 1009 and # 1010 means fixing the movable microprism 17aa by applying a voltage to or removing the voltage from the transparent electrode layers 18aa, 117aa, 121aa, and 123aa in FIGS. 1a and 1c. The first facing interval between the microprisms 116aa is increased.
[0239]
In step # 1011, the accumulated charges are transferred through the vertical transfer path 114 or a horizontal transfer path (not shown), amplified and stabilized, and then output.
[0240]
In step # 1012, the output image is displayed on the liquid crystal display means 143 disposed on the back of the camera.
[0241]
In step # 1013, the output image is subjected to a known compression process or the like, and after the amount of information is reduced, it is recorded on a recording medium (recording means 144) that can be attached to and detached from the camera, for example, with a semiconductor memory.
[0242]
In step # 1014, the shutter is opened.
[0243]
As described above, the period during which the shutter is closed in the solid-state imaging device 10 is between step # 1010 and step # 1014, and this time is all very short time that is unrelated to the operation of the photographer. (Eg 0.5 seconds)
On the other hand, from step # 1001 to step # 1009, it is necessary to capture the subject image, and the time depends on the operation of the subject (operation of the release button).
[0244]
Therefore, it is necessary to keep the shutter open for a long time.
[0245]
Therefore, when selecting whether the shutter of the solid-state imaging device 10 is normally open or normally closed, normally open, which does not require driving power when the shutter is open, saves power. preferable.
[0246]
Therefore, the movable microprism 17aa is formed in advance so that the interval between the movable microprism 17aa and the fixed microprism 116aa is narrowed. In the case of FIG. 1, the transparent electrode layers 18aa and 117aa are charged to the same polarity and repel each other to repel the shutter. In the case of FIG. 3, it is desirable to close the shutter by charging the transparent electrode layers 121aa and 123aa to opposite polarities and attracting them to each other.
[0247]
Next, a manufacturing method of the solid-state imaging device 10 having the structure as shown in FIG. 1 will be described.
[0248]
1. First manufacturing process (process for forming the first base and the second base individually)
・ First base
The base layer 110 and the photoelectric conversion element 111 provided in the base layer 110, the read gate 112, the vertical transfer electrode 113, the vertical transfer path 114, the channel stop region 115, and the light shielding layer 119 are manufactured in the same manner as the conventional method.
[0249]
However, the top surface of the photoelectric conversion element 111 is SIO. ² An oxide film 145 such as is provided.
[0250]
Then, resists 146 and 147 are provided on the upper and lower surfaces of the base layer 110 completed as shown in FIG. 15, and the whole substrate is etched to dig a hole.
[0251]
At this time, etching does not proceed in the portions covered with the resists 146 and 147, and since the oxide film 145 is provided on the photoelectric conversion element 111, the etching is stopped in that portion and the photoelectric conversion element 111 is etched. Not.
[0252]
The etching can dig a substantially vertical hole by using a process such as DEEP RIE. (Fig. 16)
Then, the oxide film 145 is removed, and TIO is sputtered into the dug hole. ² Form a layer.
[0253]
Then, resists 148 and 149 are provided again on the upper and lower surfaces of the base layer 110.
[0254]
Resist 148 is TIO ² The layer is provided in a portion that is not eroded when it is applied to the etching solution.
[0255]
For example, in the case of a positive resist, the base layer 110 is first provided by spin coating or the like, and a mask is brought into close contact with the resist area to be left, and the resist composition is changed by applying light from the upper surface to perform etching. Apply the solution to remove the remaining resist.
[0256]
Here, if the mask of the resist 148 has a gradient pattern in which the transmittance at the center is high and the transmittance is lowered toward the periphery as shown in FIG. 17, the remaining resist is as shown in FIG.
[0257]
When etching is performed on the resist 148, the shape of the resist is changed to TIO by etch back. ² A fixed microprism 116 having a conical depression is transferred to the layer. (Fig. 19)
Then, ITO that becomes the transparent electrode layer 117 is formed by sputtering on the formed depression, and the oxide film SIO that becomes the edge layer 118 is further formed thereon. ² To complete the first base.
(Fig. 20)
・ Second base
SIO ² A resist is provided on the upper and lower surfaces of the planarizing layer 15, and the entire substrate is etched to dig a hole.
[0258]
Etching can dig a substantially vertical hole by using a process such as DEEP RIE. (Fig. 21)
Thereafter, etching is performed again from the upper surface to form a tapered surface as shown in FIG.
[0259]
TIO is sputtered into the hole dug in the tapered surface. ² The optical waveguide 13 is formed of layers.
[0260]
The color filter 12 is provided on the upper surface of the planarizing layer 15 and the optical waveguide 13 so as to correspond to each optical waveguide 13. (Fig. 23)
Further, TIO is sputtered on the upper surface of the color filter. ² Layer 11 is formed.
[0261]
Then, resists 150 and 151 are provided again on the upper and lower surfaces of the planarizing layer 15.
[0262]
Resist 150 is TIO ² The layer 11 is provided in a portion that is not eroded when it is applied to the etching solution.
[0263]
For example, in the case of a positive resist, initially TIO ² It is provided on the entire surface of the layer 11 by spin coating or the like, a mask is brought into close contact with the resist area to be left, and the resist composition is changed by applying light from the upper surface, and then the remaining resist is removed by applying to an etching solution. .
[0264]
Here, if the mask of the resist 150 is a gradient pattern in which the transmittance at the center is low and the transmittance increases toward the periphery as shown in FIG. 24, the remaining resist is as shown in FIG.
[0265]
When etching is performed on the resist 150, the shape of the resist is changed to TIO by etch back. ² A microlens 11 having a conical depression is transferred to the layer.
[0266]
A resist layer for protection is provided on the upper surface of the microlens 11.
[0267]
Next, the process proceeds to the formation of the back surface side of the planarizing layer 15.
[0268]
First, a sacrificial layer 151 is formed on the entire surface of the planarizing layer 15 and the back surface of the optical waveguide 13.
[0269]
The sacrificial layer 151 at the base portions 16aa2, 16ab2,... Of the leaf spring 16 is removed in advance by etching or the like. (See the plan view in Fig. 5)
FIG. 26 shows a BB cross section of FIG. 5 in that state.
[0270]
Then, a polysilicon layer 152 is formed on the sacrificial layer by CVD, a resist is provided from the upper surface, and pattern etching is performed to create a spring shape.
[0271]
Thereafter, the upper surface of the leaf spring 16 is again protected with the resist 152, and the resist is removed only on the surface on which the movable microprism 17 is formed. (Fig. 27)
Thereafter, TIO is sputtered onto the sacrificial layer 151 including the end of the leaf spring 16. ² Layer 17 is formed.
[0272]
TIO formed ² A resist layer 153 is further provided on the layer.
[0273]
The resist layer 153 has a corner cube shape.
[0274]
For example, in the case of a positive resist, if the mask of the resist 153 has a gradient pattern in which the transmittance at the center is low and the transmittance increases toward the periphery as shown in FIG. 28, the remaining resist is as shown in FIG.
[0275]
When etching is performed on the resist 153, the shape of the resist is changed to TIO by etch back. ² A movable microprism 17 having a conical protrusion is transferred to the layer.
[0276]
Thereafter, portions other than the movable microprism 17 are protected with a resist, and a transparent electrode layer 18aa and an insulating layer 19aa are provided on the upper surface of the movable microprism 17.
[0277]
The movable microprism 17 is joined to the end (16aa2 etc.) of the leaf spring 16.
[0278]
Thereafter, when the sacrificial layer is removed, as shown in FIG. 30, the leaf spring 16 has one end (base 16aa1, etc.) fixed to the planarizing layer 15 and the other end (end face 16aa2, etc.) fixed to the movable microprism 17, The surface becomes a free surface.
[0279]
Since the sacrificial layer 151 is also provided between the movable microprism 17 and the optical waveguide 13, the movable microprism 17 can move freely with respect to the optical waveguide 13 by removing the sacrificial layer 151.
[0280]
In this plan view, as shown in FIG. 31, the base 16ab1 of the leaf spring 16ab is also fixed to the movable microprisms 17aa and 17ac. (16aa3, 16ab3, 16ac3)
This is to keep the movable microprism 17 stable until the end of the formation of the first substrate. (First formation step)
Therefore, in the final step, the resist is further patterned to remove the joints 16aa3, 16ab3, and 16ac3 while protecting the main parts of the movable microprism 17 and the leaf spring 16, thereby completing the second substrate. (End of second forming step)
2. Second manufacturing process (joining the first base and the second base)
The completed first base and second base are aligned by an aligner, and are bonded to each other by a resist layer provided on the base layer 110, for example.
[0281]
In order to adjust the position of the two bases, the positions may be adjusted by providing an uneven portion on the bases of each other.
[0282]
As described above, a light receiving region including a plurality of pixels in which a plurality of photoelectric conversion elements 111 are arranged on a main surface of a semiconductor substrate (first substrate, base layer 110);
A shutter is configured with a light control region composed of a moving member (movable microprism 17) that controls the incident state of each photoelectric conversion element 111 by being driven in the optical axis direction.
The moving member (movable microprism 17) is disposed on the incident surface side of the photoelectric conversion element 111, and the total reflection state where the incident light is totally reflected and is not guided to the photoelectric conversion element 111, and the incident light is transmitted and the photoelectric conversion element 111 is transmitted. By using optical means (using total reflection of the corner cube) that can select the transmission state leading to the light, improvement in transmittance and improvement in response are realized.
[0283]
The optical means includes a movable microprism 17 provided for each photoelectric conversion element 111, a fixed microprism 116 facing the movable microprism 17 and having an incident surface substantially the same shape as the prism surface of the movable microprism 17, and a movable microprism. The driving means (transparent electrode layers 18, 117, 121, 124, 125, 127, piezoelectric member 129) that changes the facing distance (first facing distance) between the microprism 17 and the fixed microprism 116, and the first Since the configuration is such that the total reflection state and the transmission state are selected by controlling the facing distance, a compact optical means is realized.
[0284]
The drive means selects the total reflection state by widening the first facing interval, and selects the transmission state by narrowing the first facing interval, thereby realizing a compact optical means. ing.
[0285]
In addition, each movable microprism 17 corresponding to the photoelectric conversion element is connected to and supported by the base of the light control region by an elastic means (plate spring 16). Misalignment with each fixed microprism 116 can also be absorbed.
[0286]
The driving means is composed of a first electrode (transparent electrode layer 18) provided on the opposing spacing surface of the movable microprism 17 and a second electrode (transparent electrode layer 117) provided on the opposing spacing surface of the fixed microprism 116. Since the first opposing distance is controlled electrostatically by controlling the charges of the first electrode (transparent electrode layer 18) and the second electrode (transparent electrode layer 117), the driving means is small. It is made.
[0287]
As another driving means, it is composed of a first electrode provided on the elastic means and a second electrode provided on the surface facing the elastic means, and controls the electrostatic charge by controlling the charges of the first electrode and the second electrode. Since the first opposing distance is controlled, the driving means is small.
[0288]
Further, as another driving means, the driving means is configured to control the first facing distance by the piezoelectric member 129 provided on the elastic means (leaf spring 16), so that it can be a small driving means.
[0289]
Further, the driving means independently drives the movable microprism 17 corresponding to each of the plurality of photoelectric conversion elements 111 to individually control the amount of light incident on each photoelectric conversion element 111. The dynamic range can be expanded.
[0290]
In the above light control method and its driving method in general
The driving means is configured to control the amount of light incident on the photoelectric conversion element by alternately driving the moving member (movable microprism 17). In particular, the driving means sets the moving member (movable microprism 17) to its resonance frequency. Since the amount of light incident on the photoelectric conversion element 111 is controlled by driving at (the natural frequency determined by the mass of the movable microprism 17 and the spring constant of the leaf spring 16), a large driving stroke is achieved with a small driving force. It has been realized.
[0291]
The amount of light incident on the photoelectric conversion element 111 is changed by providing a period (accumulation control engine 131) in which the driving unit alternately drives the moving unit (movable microprism 17) and a period in which the driving is not performed. In this way, the incident light to the photoelectric conversion element 111 is controlled to realize the enlargement of photographing dynamics.
[0292]
Alternatively, the driving means controls the amount of light incident on the photoelectric conversion element by driving the moving member alternately (initial driving period in FIG. 12) and then driving in a DC manner (adsorption period in FIG. 12, stable engine). Because of this, a large driving stroke can be reliably maintained with a small driving force.
[0293]
Then, after the driving means alternately drives the moving means (initial driving period in FIG. 12), a period in which the driving means is driven in a direct current (adsorption period, stable engine in FIG. 12) and a period in which the driving of the moving means is not driven. By providing (the shutter opening period of FIG. 12), the incident light to the photoelectric conversion element is controlled so as to change the amount of light incident on the photoelectric conversion element, thereby realizing an increase in photographing dynamics.
[0294]
The shapes of the movable microprism 17 and the fixed microprism 116 in the solid-state imaging device 10 described so far are circular as shown in the movable microprism 17 of FIG. 5 (the three-dimensional shape is a conical shape because the cross section of FIG. 1a is a wedge shape). However, it is not limited to such a top surface shape.
[0295]
FIG. 32 shows a case where the upper surface shape of the movable microprism 17 is a triangle (a three-dimensional shape is a triangular pyramid) and is attached to the base of the leaf spring 16. (Of course, the center of the circular portion of the leaf spring 16 is hollowed out so as not to block incident light.)
Although not shown, the corresponding fixed microprism 116 also has a triangular pyramid shape.
[0296]
In addition, various shapes such as a quadrangular pyramid shape can be taken.
[0297]
In the configuration of the solid-state imaging device 10 described above
Since power is supplied to the moving member (movable micro prism 17) to attenuate the amount of light incident on the photoelectric conversion element 111, power is not required in normal times and power is supplied only at the end of photographing. So you can save power.
[0298]
A first substrate having a light receiving region in which a plurality of photoelectric conversion elements 111 are arranged on a main surface (base layer 110) of the semiconductor substrate;
A solid-state imaging device 10 composed of a second base having a light control region in which light control elements (movable microprisms 17) for controlling the respective incident states corresponding to each of the plurality of photoelectric conversion elements 111 in the light receiving region are arranged. The manufacturing method according to claim 1, wherein the first base and the second base are manufactured separately, and the first base and the second base are aligned and combined to align the first base and the second base. Since the image pickup device 10 is manufactured, both the first and second substrates can be stably manufactured by a normal process.
[0299]
Further, it is possible to prevent the yield of each manufacturing from becoming a product and deteriorating the yield.
[0300]
Also, a plurality of photoelectric conversion elements 111 provided on the main surface (base layer 110) of the semiconductor substrate, and a plurality of light control elements (movable micros) that control the respective incident states corresponding to each of the plurality of photoelectric conversion elements 111. A method of manufacturing a solid-state imaging device 10 having a prism 17), the first forming step of forming each light control element (movable microprism 17) supported by a support member (leaf spring 16), and The solid-state imaging device 10 is manufactured by the second forming step of reducing the supporting force of each supporting member (leaf spring 16) of the light control element (movable microprism 17), and the first forming is described in detail. In the step, a plurality of support members (plate springs 16) that support each of the light control elements (movable microprisms 17) are formed, and in the second formation step, a support portion It has become like a light control element (movable microprisms 17) can be stably produced in the first forming step so that in the manner to reduce the support number of (leaf spring 16).
[0301]
In addition, a first substrate having a light receiving region in which a plurality of photoelectric conversion elements 111 are arranged on the main surface (base layer 110) of the semiconductor substrate, and a respective incident state corresponding to each of the plurality of photoelectric conversion elements 111 in the light receiving region. In the solid-state imaging device 10 having a second base having a light control region in which light control elements (movable microprisms 17) for controlling the light are arranged, each light control element (movable microprism 17) is elastically supported ( By elastically supporting the second base by the leaf spring 16), it is possible to absorb the positional deviation between each light control element and each corresponding photoelectric conversion element.
[0302]
In addition, a light receiving region composed of a plurality of photoelectric conversion elements 111 arranged on the main surface (base layer 110) of the semiconductor substrate, and a plurality of light control elements (movable micros) that control the incident state corresponding to each photoelectric conversion element 111. By driving the light control area composed of the prism 17) and each light control element (movable microprism 17), the incident light to each photoelectric conversion element 111 is shielded, and each light control element (movable microprism 17) is driven. When not, it responds to the drive means (transparent electrode layer 18, 117, 121, 124, 125, 127, piezoelectric member 129) which permits the incident light to each photoelectric conversion element 111, and the accumulation time end of each photoelectric conversion element 111 Then, the solid-state imaging device 10 composed of control means for driving the drive means (transparent electrode layers 18, 117, 121, 124, 125, 127, piezoelectric member 129) is Small number of times of driving by using La, it was possible to perform imaging to the power saving in the driving time.
[0303]
In addition, a light receiving region composed of a plurality of photoelectric conversion elements 111 arranged on the main surface (base layer 110) of the semiconductor substrate, and a plurality of light control elements (movable micros) that control incident light corresponding to each photoelectric conversion element 111. A light control region composed of a prism 17) and driving means (transparent electrode layers 18, 117, 121, 124, etc.) for shielding incident light to each photoelectric conversion element by driving each light control element (movable microprism 17). 125, 127, piezoelectric member 129) and a solid-state imaging device 10 composed of a control means (shooting control means 141) for driving the driving means at an independent timing for each photoelectric conversion element 111 is used for exposure. Shooting with a wide dynamic range is possible using time as a parameter. Specifically, the control means (shooting control means 141) is stored in each photoelectric conversion element 111. Whether to perform shooting according to each predetermined timing from the start of the image, that is, according to the exposure time (step # 1007 in FIG. 1n) of each photoelectric conversion element set according to the imaging state before the start of accumulation of each photoelectric conversion element 111 Alternatively, it is possible to perform shooting with a wide dynamic range by independently controlling the exposure time of each photoelectric conversion element 111 in accordance with the amount of charge accumulated in each photoelectric conversion element 111 during exposure.
[0304]
The light receiving area is composed of a plurality of photoelectric conversion elements 111 arranged on the main surface (base layer 110) of the semiconductor substrate, and the light control area is composed of a light control element that controls the incident state of each photoelectric conversion element 111. In the solid-state imaging device 10, the light control element is a fixed microscopic prism 17 provided for each photoelectric conversion element, and a stationary surface facing the movable microprism 17 and having an incident surface substantially the same shape as the prism surface of the movable microprism 17. The microprism 116, the optical waveguide 13 facing the other end of the movable microprism 17, the first opposing distance between the movable microprism 17 and the fixed microprism 116, and the second opposing distance between the movable microprism 17 and the optical waveguide 13. Driving means (transparent electrode layers 18, 117, 121, 124, 125, 127, piezoelectric member 129 And a light-shielding means (light-shielding layer 119) that shields stray light due to the emission of light rays on the surface along the main optical axis at the first facing distance and / or the second facing distance, and driving means (transparent electrode) The total reflection state and transmission state of the movable microprism 17 are controlled by changing the first facing interval by the layers 18, 117, 121, 124, 125, 127, and the piezoelectric member 129). Since the influence of light can be blocked by the light blocking means (the light blocking layer 119), a good image can be obtained.
[0305]
(Second Embodiment)
In the first embodiment, the operation time of the movable microprism 17 is controlled independently for each photoelectric conversion element 111 to shorten the exposure time for the photoelectric conversion element 111 with strong incident light from the subject. The accumulated charge was prevented from being saturated.
[0306]
Therefore, it has become possible to perform shooting with a wide dynamic range.
[0307]
In the second embodiment, for the purpose of further simplifying the drive and control configuration, the dynamic range is not expanded, and the shutter of the solid-state imaging device 10 is replaced by a physical property element such as an electronic element or a liquid crystal element. This solves only the problem of light beam leakage caused by the decrease in absolute sensitivity and low response, which are the problems when forming the lens.
[0308]
For this purpose, each movable microprism 17 cannot be driven individually, and the distance between the first base and the second base can be changed.
[0309]
FIG. 33 is a cross-sectional view of the second embodiment, in which the transparent electrode 18 on the movable microprism 17, the insulating layer 19, the transparent electrode 117 on the fixed microprism 116, and the insulating layer 118 are omitted.
[0310]
The first base of the solid-state imaging device 10 is fixed to the stator 21 (the coupling portion is omitted in the figure), and the second base is fixed to the slider 25. (Similarly, the joint is omitted in the figure)
A permanent magnet 23 and a yoke 24 are provided above the stator 21, and a coil 22 is provided above the permanent magnet 23 so as to surround the yoke 24.
[0311]
The slider 25 is biased against the coil 22 by a coil spring 26 in a repulsive direction.
[0312]
Of the members constituting the slider 25, the portion inserted into the coil 22 is formed of a magnetic material.
[0313]
As shown in FIG. 33, when the slider 25 is close to the permanent magnet 23, the yoke 24 and the slider 25 are magnetically coupled against the urging force of the coil spring 26, so the first base and the second base are close. It becomes a state.
[0314]
Therefore, the incident light 120 to each photoelectric conversion element 111 enters the fixed microprism 116 from the movable microprism 17 like the incident light 120ac, and charges are accumulated in the photoelectric conversion element 111.
[0315]
Here, when a current in a predetermined direction is passed through the coil 22, the magnetic force of the permanent magnet 23 is weakened by the magnetic force generated in the coil.
[0316]
Therefore, the slider 25 is separated from the yoke 24 by the repulsive force of the coil spring 26 and is in the state shown in FIG.
[0317]
Then, the gap (first opposing interval) between the movable microprism 17 and the fixed microprism 116 becomes wide, and the incident light 120 to the photoelectric conversion element 111 causes total reflection in the movable microprism 17 like 120aa, and the fixed microprism. No incident on 116.
[0318]
When it is desired to bring the first base plate and the second base plate closer again (when it is desired to reduce the first facing interval), a current in the direction opposite to the predetermined direction is passed through the coil 22.
[0319]
Then, since the resultant force of the magnetic force generated by the coil 22 and the magnetic force generated by the permanent magnet 23 exceeds the repulsive force of the coil spring 26, the slider 25 is attracted to the yoke 24.
[0320]
When the slider 25 is brought into contact with the yoke 25, the attracting force is increased, so that the state is maintained even after the current to the coil 22 is cut.
[0321]
In this way, controlling the distance between the first base and the second base with the plunger configuration has the following advantages.
・ Low power consumption
Compared with the case where an electrostatic force is generated by applying a large voltage as in the first embodiment, the size of the actuator (coil 22) can be sufficiently increased, so that a large driving force can be obtained with a small amount of power.
[0322]
Further, only when the distance between the first base and the second base is changed, a current is passed through the coil 22, and the current to the coil 22 is not required to maintain the distance, so that power consumption is small.
・ Stable shutter closed
In the first embodiment, the first facing interval cannot be set large because of the relationship of electrostatic driving. (When the interval is large, the electrostatic force is reduced.) However, in the method of the second embodiment, since the first facing interval can be widened, the incident light leaking from the movable microprism 17 to the fixed microprism 116 is complete even when the shutter is closed. It will disappear.
[0323]
In the second embodiment, since the entire second base moves, the leaf spring 16 that supports the movable microprism 17 is unnecessary, and the movable microprism 17 and the optical waveguide 13 may be formed integrally.
[0324]
In the embodiment of FIGS. 33 and 34, the movable microprism 17 is supported by the leaf spring 16 and separated from the optical waveguide 13 for the following reason.
[0325]
As described in the first embodiment, the first base and the second base are formed by different processes. Therefore, both the first base and the second base are easy to manufacture, and the yields do not affect each other.
[0326]
However, the shape of the first opposing spacing surface of the fixed microprism 116 on the first base, the variation in the pitch of the fixed microprisms 116 with each other, the shape of the first opposing spacing surface of the movable microprism 17 on the second base, Variations in the pitch of the movable microprisms 17 relative to each other and a shift when the first substrate and the second substrate are combined become problems.
[0327]
Therefore, the movable microprism 17 is flexibly supported by the leaf spring 16, and when the first facing interval is narrowed, the movable microprism 17 follows the taper (corner cube surface) with the fixed microprism 116, and the above-described variation and I try to absorb the deviation.
[0328]
As a method of changing the distance between the first base and the second base, not only the plunger method described above but also a voice coil or bimorph may be used. As shown in FIG. The change may be controlled.
In FIG. 35, the stator 21 is provided with a bearing 27, and the shaft of the slider 25 is fitted to the bearing 27.
[0329]
The slider 25 is urged by a coil spring 28 in the stator pressing direction.
[0330]
The cam plate 29 can be rotated by a motor (not shown) or the like, and in the state shown in FIG.
[0331]
Therefore, the first facing interval is widened to create a shutter closed state.
[0332]
When the cam plate 29 is rotated in the direction of the arrow 210 in this state, the slider 25 approaches the stator 21 by the urging force of the coil spring 28 as the rotation is made, and the first facing interval is narrowed to create a shutter open state.
[0333]
When the cam plate 29 is used in this way, the change in the first facing interval can be increased, and since the number of changes in the first facing interval can be set during one rotation of the cam plate 29, You can freely set according to the shooting sequence.
[0334]
As described above, a light receiving region including a plurality of pixels in which a plurality of photoelectric conversion elements 111 are arranged on a main surface of a semiconductor substrate (first substrate, base layer 110);
A shutter is configured with a light control region composed of a moving member (movable microprism 17) that controls the incident state of each photoelectric conversion element 111 by being driven in the optical axis direction.
The moving member (movable microprism 17) is disposed on the incident surface side of the photoelectric conversion element 111, and the total reflection state where the incident light is totally reflected and is not guided to the photoelectric conversion element 111, and the incident light is transmitted and the photoelectric conversion element 111 is transmitted. By using optical means (using total reflection of the corner cube) that can select the transmission state leading to the light, improvement in transmittance and improvement in response are realized.
[0335]
The optical means includes a movable microprism 17 provided for each photoelectric conversion element 111, a fixed microprism 116 facing the movable microprism 17 and having an incident surface substantially the same shape as the prism surface of the movable microprism 17, and a movable microprism. It comprises driving means (coil 22, permanent magnet 23, coil spring 26) that changes the facing distance (first facing distance) between the microprism 17 and the fixed microprism 116, and is totally reflected by controlling the first facing distance. Since it is configured to select a state and a transmission state, a small optical means is realized.
[0336]
Driving means (coil 22, permanent magnet 23, coil spring 26, cam plate 29) drives the entire movable microprism 17 (second base) corresponding to each of the plurality of photoelectric conversion elements 11 and enters the photoelectric conversion element 111. Since the amount of light to be controlled is controlled, reliable light control is possible with a certain driving force.
[0337]
In addition, a first substrate having a light receiving region in which a plurality of photoelectric conversion elements 111 are arranged on the main surface (base layer 110) of the semiconductor substrate, and a respective incident state corresponding to each of the plurality of photoelectric conversion elements 111 in the light receiving region. In the solid-state imaging device 10 having a second base having a light control region in which light control elements (movable microprisms 17) for controlling the light are arranged, each light control element (movable microprism 17) is elastically supported ( The plate spring 16) is elastically supported with respect to the second base to absorb the positional deviation between each light control element (movable microprism 17) and each corresponding photoelectric conversion element 111 (or fixed microprism 116). It can be done.
[0338]
In addition, a first substrate having a light receiving region in which a plurality of photoelectric conversion elements 111 are arranged on the main surface (base layer 110) of the semiconductor substrate, and a light control element corresponding to each of the plurality of photoelectric conversion elements 111 in the light receiving region ( A second base having a light control region in which the movable microprisms 17) are arranged, and a driving means for controlling the incident state to each photoelectric conversion element by relatively driving the first base and the second base By configuring the solid-state imaging device 10 with the coil 22, the permanent magnet 23, the coil spring 26, and the cam plate 29, it is possible to perform reliable light control with a simple structure.
[0339]
(Third embodiment)
FIG. 36 is a cross-sectional view of the solid-state imaging device 30 according to the third embodiment of the present invention.
[0340]
In the drawing, the alphabet at the end of the part number of each constituent member is provided for identification for each pixel, and when there is no alphabet at the end of the part number in the sentence, the general name thereof is shown.
[0341]
Reference numeral 11aa denotes a microlens, which is provided corresponding to each of the photoelectric conversion elements 111 described later and has a refractive index of 2.0 or more. ² The transparent bright layer is cut into a lens shape by etching.
[0342]
Reference numeral 12aa denotes a color filter. For example, in the case of a Bayer array primary color filter, red, blue, and green filters are alternately adjacent to each other and arranged as shown in FIG.
[0343]
In FIG. 37, 12aa, 12ac, 12ca and 12cc are red filters, 12ab, 12ac, 12ba, 12bc, 12cb, 12cd, 12da and 12dc are green filters, and 12bb, 12bd, 12db and 12dd are blue filters.
[0344]
Each filter can be formed by forming a resin (color resin) layer containing a pigment or dye of a desired color at a predetermined location by a method such as photolithography.
[0345]
The partition wall layer 32aa is formed of silicon so as to surround the outer periphery of the space region 31aa serving as an optical waveguide, and divides each space region 31.
[0346]
The partition wall 32aa is provided with a later-described polysilicon flap 37aa.
[0347]
A reflective layer 38aa made of aluminum, nickel, gold or the like is deposited on the surface of the flap 37aa.
[0348]
Further, the transparent electrode layer 33aa made of ITO and the SIO are formed on the surface of the partition wall 32aa. ² An insulating layer 34aa made of is provided.
[0349]
Similarly, the transparent electrode layer 36aa made of ITO and the SIO are also formed on the back surface of the flap 37aa. ² An insulating layer 35aa made of is provided.
[0350]
Similar to the planarization layer 15, SIO ² A photoelectric conversion element 111aa is provided at the outermost end portion of the base layer 110 formed in (1).
[0351]
The photoelectric conversion element 111 has, for example, an n-type impurity region on a matrix on a p-type impurity region (which is omitted as a base layer over the entire lower surface of the photoelectric conversion device 111 but is omitted for the sake of simplicity). And p on each of them ⁺ This is a buried photodiode formed by providing a type impurity region, and the n type impurity region functions as a charge storage region.
[0352]
A read gate 112aa is provided adjacent to the photoelectric conversion element 111aa.
[0353]
The read gate 112aa is a region of the vertical transfer electrode 113aa and is formed of a polysilicon layer.
[0354]
A vertical transfer charge transfer path 114aa is provided to face the vertical transfer electrode 117aa.
[0355]
When a read pulse (for example, 15 V) is applied to the vertical transfer electrode 113aa, the charge accumulated in the photoelectric conversion element (here, the photoelectric conversion element 111aa) corresponding to the vertical transfer electrode 113aa corresponds via the read gate 112aa. Data is read out to the vertical charge transfer path 114aa.
[0356]
Reference numeral 115aa denotes a channel stop region, which is provided so as to surround the periphery of the photoelectric conversion element 111aa other than the readout gate 112aa.
[0357]
The channel stop region 115aa is provided to electrically isolate other adjacent discharge conversion elements and transfer regions.
[0358]
Charge reading from the photoelectric conversion element 111 to the vertical charge transfer path 114 is performed in units of the photoelectric conversion element 111, and there are various reading modes such as interlace scanning, progressive scanning, and all pixel reading.
[0359]
Next, a method for manufacturing the solid-state imaging device 30 having the structure as shown in FIG. 36 will be described.
[0360]
The base layer 110 and the photoelectric conversion element 111 provided in the base layer 110, the read gate 112, the vertical transfer electrode 113, the vertical transfer path 114, the channel stop region 115, and the light shielding layer 119 are manufactured in the same manner as the conventional method.
[0361]
However, the top surface of the photoelectric conversion element 111 is SIO. ² An oxide film 39 such as is provided.
[0362]
A silicon layer 310 that becomes the partition wall layer 32 is further formed thereon.
[0363]
Then, resists 311 and 312 are provided on the upper surface and the lower surface of the completed substrate as shown in FIG. 38, and the entire substrate is etched to dig a hole.
For example, in the case of a positive resist, the silicon layer 310 is first provided by spin coating or the like, a mask is brought into close contact with the resist area to be left, and the resist composition is changed by applying light from the upper surface. And remove the remaining resist.
[0364]
Here, if the mask of the resist 311 has a gradient pattern in which the transmittance at the center is high and the transmittance is lowered toward the periphery as shown in FIG. 39, the remaining resist is as shown in FIG.
[0365]
When etching is performed from the top of the resist 311, the shape of the resist 311 is transferred to the silicon layer 310 by etch back to have a conical depression, and the remaining portion becomes the partition layer 32.
[0366]
Of course, the portion covered with the resist 312 is not etched, and the oxide film 39 is provided on the photoelectric conversion element 111, so that the etching is stopped at that portion and the photoelectric conversion element 111 is not etched.
[0367]
Next, a transparent electrode layer 33 is provided, and a slope shape is formed by etch back as in the case of the partition wall 32, and the same process is repeated to form the insulating layer 34.
[0368]
Next, a sacrificial layer 313 is provided on the base layer 110.
[0369]
The resist surface is processed with the same mask as in FIG.
At this time, if the etching progress time is different from that of the resist 310, the remaining sacrificial layer 313 becomes as shown in FIG.
[0370]
The SIO which becomes the insulating layer 35 on the surface of the sacrificial layer 313 ² A layer is provided, the same procedure as the formation of the partition wall layer 32 is repeated to create a slope shape, and a transparent electrode layer 36 made of ITO is formed on the upper surface by the same procedure.
[0371]
The transparent electrode layer 36 is also formed with a slope by an etch back process, a polysilicon layer 314 serving as a basis for the flap 37 is formed thereon by CVD, and a resist 315 is further provided from the upper surface.
[0372]
The resist 315 is also shaved with the same gradation pattern as described above to form the shape shown in FIG.
[0373]
When etching is performed from above the resist 315, the shape of the resist 315 is transferred to the polysilicon layer 314 by etch back, and the flap 37 is formed.
[0374]
A reflective layer 38 is deposited on the surface of the formed flap as a metal reflective surface such as aluminum.
[0375]
Thereafter, the oxide film 39 and the remaining resist are removed.
[0376]
The color filter 12 is provided on the upper surface of the partition layer 32 so as to correspond to each photoelectric conversion element 111. (Fig. 43)
Further, TIO is sputtered on the upper surface of the color filter. ² Layer 11 is formed.
[0377]
Then, resists 316 and 317 are provided again on the upper and lower surfaces of the planarizing layer base layer 110.
[0378]
Resist 316 is TIO ² The layer is provided in a portion that is not eroded when it is applied to the etching solution.
[0379]
For example, in the case of a positive resist, the first TIO ² A layer is provided on the entire surface of the layer 11 by spin coating or the like, a mask is brought into close contact with the resist area to be left, light is applied from the upper surface, the composition of the resist is changed, and then the remaining resist is removed by applying an etching solution.
[0380]
Here, if the mask of the resist 316 has a gradient pattern in which the transmittance at the center is low and the transmittance increases toward the periphery as shown in FIG. 44, the remaining resist is as shown in FIG.
[0381]
When etching is performed from above the resist 316, the shape of the resist is changed to TIO by etch back. ² A microlens 11 having a conical depression transferred to the layer 11 is formed.
[0382]
The operation of the solid-state image sensor 30 thus completed will be described.
[0383]
In the photoelectric conversion element 111, no voltage is applied to the transparent electrode layers 33aa and 36aa.
[0384]
Therefore, the flap 37aa is along the partition wall 32aa, and the incident light from the subject is guided to the photoelectric conversion element 111aa.
[0385]
Further, since the reflective layer 38aa is reflected by oblique incidence or stray light (light ray 317) and guided to the photoelectric conversion element 111aa, the apparent sensitivity of the photoelectric conversion element 111a is improved, and the solid-state imaging excellent in shading characteristics is also provided. Element 30 is formed.
[0386]
In the photoelectric conversion element 111ab, a voltage of 10 v, for example, is applied to the transparent electrode layers 33ab and 36ab.
[0387]
At this time, since the transparent electrode layers 33ab and 36ab repel each other, the flap 37ab blocks the optical path by closing two flaps as shown in the figure.
[0388]
Therefore, there is no incident light from the subject in the photoelectric conversion element 111ab.
[0389]
Further, by controlling the voltage applied to the transparent electrode layers 33ab and 36ab, it is possible to obtain an effect of narrowing the incident light like a diaphragm instead of completely closing the flap 37ab.
[0390]
The independent light quantity for each photoelectric conversion element 111 is controlled according to the flow of FIG. 14 by the photographing control means 141 of FIG. 13 as in the first embodiment.
[0390]
Due to such a shutter configuration, the following features are obtained.
・ Highly transparent shutter
Since no liquid crystal or electrochromic element is arranged on the front surface of the photoelectric conversion element 111 as in the conventional example, the photoelectric conversion element 111 with high transmittance and high absolute sensitivity can be obtained.
・ Highly responsive shutter
Compared with liquid crystal or the like, the response is high, and the shutter can be controlled with high response even at low temperatures, so that stable imaging performance can be obtained.
・ Expand the dynamic range
In the photoelectric conversion element 111 in which saturation is predicted because the shutter can be opened and closed independently for each photoelectric conversion element, the saturation can be prevented by closing the shutter at an early stage of exposure, and a wide dynamic range is achieved. I can shoot.
[0392]
Alternatively, since the size of the diaphragm can be changed independently for each photoelectric conversion element, in the photoelectric conversion element 111 that is predicted to be saturated, the diaphragm is throttled (the flap 37 is semi-closed) to prevent the saturation. Yes, you can shoot a wide dynamic range.
·Low power consumption
When no power is applied, the shutter is open (the state of the flap 37 in FIG. 36), and the power for closing the shutter is used only for a short time of charge transfer.
・ Further high sensitivity
Since the reflection layer 38 is provided on the surface of the flap 37, light that does not directly enter the photoelectric conversion element 111 can be guided to the photoelectric conversion element 111, and the influence of shading is reduced with high sensitivity.
[0393]
In FIG. 36, the electrostatic force is used to drive the flap 37, but not limited thereto, an electrostrictive member such as PZT, bimorph, or shape memory alloy may be used.
[0394]
FIG. 46 shows an example using an electrostrictive element. The transparent electrodes 33 and 36 and the insulating layer 335 in FIG. 36 are omitted, and an electrostrictive element 317 is deposited instead.
[0395]
The electrostrictive element 317 is distorted by application of voltage, but when the electrostrictive element 317 is integrated with the flap 37, the flap 37 bends due to a difference in distortion caused by voltage application.
[0396]
In FIG. 46, a flap 37 is a case where a voltage is applied to the electrostrictive element 317 to bend the flap 37, and in this state, incident light of the photoelectric conversion element 111ab can be reduced or shielded.
[0397]
For example, when a shape memory alloy is used, since one shape memory alloy is extremely small, heat conduction is fast and a high response can be expected.
[0398]
Furthermore, when there is intense incident light that saturates the photoelectric conversion element 111, the heat generated by the incident causes the shape memory alloy to squeeze or block the incident light of the discharge conversion element 111. Saturation of the element 111 can be prevented.
[0399]
The shapes of the flap 37 and the partition wall 32 are not limited to those shown in FIG. 36, but may be configured as shown in FIG.
[0400]
In FIG. 47, the silicon partition wall 32 has a substantially vertical shape, which is produced by the DEEP RIE manufacturing method described in the manufacturing method of the first embodiment.
By repeating the same process, the transparent electrode layer 33 made of ITO, SIO ² Insulating layer 3 made of polysilicon, flap 37 of polysilicon, transparent electrode layer 36 made of ITO, SIO ² An insulating layer 35 made of
[0401]
As described above, a light receiving region composed of a plurality of pixels in which a plurality of photoelectric conversion elements 111 are arranged on the main surface (base layer 110) of the semiconductor substrate,
The shutter is configured with a light control region composed of a moving member (flap 37) that controls the incident state of each photoelectric conversion element by being driven in the optical axis direction,
The moving member (flap 37) is arranged on the incident surface side of the photoelectric conversion element 111, and selects a light shielding state in which incident light is shielded and not guided to the photoelectric conversion element 111 and a non-light shielding state in which incident light is guided to the photoelectric conversion element 111. It consists of possible moving members (flaps 37),
The moving member (flap 37) includes a light shielding film (flap 37) disposed along the side wall (partition wall 32) of the incident optical path of each photoelectric conversion element 111, and driving means (transparent electrode layers 33, 36, The electrostrictive element 317), and the driving means (transparent electrode layers 33 and 36, electrostrictive element 317) drives the light-shielding film (flap 37) in the incident optical path to select the light-shielding state. In addition, light control can be performed with a more compact configuration.
[0402]
In addition, a reflection layer 38 is provided that reflects light collected on a portion other than the photoelectric conversion element 111 in a state where the light shielding film (flap 37) is along the side surface of the incident optical path and collects the reflected light on the photoelectric conversion element. Therefore, the light beam that has not been used until now is also incident on the photoelectric conversion element 111, so that the sensitivity can be improved.
[0403]
As a driving method of such a light shielding film, the driving means includes a first electrode (transparent electrode layer 36) provided on the surface of the light shielding film (flap 37) facing the side wall (partition 32) of the incident optical path, and the side wall of the incident optical path (partition wall 32). ) Of the second electrode (transparent electrode 33) provided on the surface facing the light-shielding film (flap 37) and driving means for driving the light-shielding film (flap 37), and the first electrode (transparent electrode layer 36) By controlling the charge of the second electrode (transparent electrode 33), the light shielding film (flap 37) is electrostatically moved into the incident optical path, so that the driving means can be realized with a small configuration.
[0404]
As another driving method, the driving means is configured to move the light shielding film (flap 37) into the incident optical path by an electrostrictive member (electrostrictive element 317) provided on the light shielding film (flap 37). Thus, the driving means can be realized.
[0405]
In addition, a light receiving region composed of a plurality of photoelectric conversion elements 111 arranged on the main surface (base layer 110) of the semiconductor substrate, and a plurality of light control elements (flaps 37) for controlling incident light corresponding to each photoelectric conversion element 111. ) And a driving means (transparent electrode layers 33, 36, electrostrictive element 317) for narrowing incident light to each photoelectric conversion element 111 by driving each light control element (flap 37), By using a solid-state imaging device 30 configured with a control unit that drives a driving unit with an independent aperture amount for each photoelectric conversion element 111 in a camera, it is possible to perform imaging with a wide dynamic range using an exposure time as a parameter. Specifically, the control means is set according to each predetermined aperture amount after each photoelectric conversion element 111 starts accumulation, that is, according to the imaging state before each photoelectric conversion element 111 starts accumulation. Photographing is performed according to the aperture amount of each photoelectric conversion element 111, or the aperture amount of each photoelectric conversion element is controlled independently according to the amount of charge accumulated in each photoelectric conversion element 111 during exposure. Shooting with a wide dynamic range is possible.
[0406]
(Fourth embodiment)
48 to 54 are views according to the fourth embodiment of the present invention.
[0407]
First, the configuration of the photographing apparatus using the low-pass filter of the present invention will be described with reference to FIG.
[0408]
In this embodiment, a description will be given by taking as an example a digital still camera in which a photographing apparatus photoelectrically converts a still image into an electrical signal by an imaging unit and records this as digital data. However, the present invention is not limited to a digital still camera, and the same effect can be obtained even when applied to other photographing apparatuses having photoelectric conversion means such as a video camera, an industrial tool camera, and a surveillance camera.
[0409]
An imaging optical system 400 includes a plurality of lens groups, and includes a front lens group 401, a variator lens group 402, and a focusing lens group 403. Reference numeral 405 denotes a known iris diaphragm mechanism, which includes a plurality of diaphragm blades and driving means for the blades. The aperture diameter formed by the blades is adjusted to adjust the amount of light beam passing through the photographing optical system 400. A known shutter mechanism 406 includes a plurality of shutter blades and driving means for the blades. The exposure time is adjusted by adjusting the travel timing of the blades.
[0410]
An imaging unit 150 is disposed at the focal position (planned imaging plane) of the photographing optical system 400. This is a photoelectric conversion means such as a two-dimensional CCD comprising a plurality of photoelectric conversion units for converting irradiated light energy into charges, a charge storage unit for storing the charges, and a charge transfer unit for transferring the charges and sending them to the outside. Is used.
[0411]
The image of the subject formed on the imaging unit 150 is converted into an electric signal as a charge amount for each pixel corresponding to the intensity of the brightness, amplified by the amplifier circuit 441, and then converted by the camera signal processing circuit 442 to A / D conversion is performed, and processing such as pixel interpolation, γ correction, and JPEG compression is performed. The video signal created in this way is recorded in the memory 443. As the memory 443, various types such as a semiconductor memory such as a flash ROM, an optical memory such as a magneto-optical disk, and a magnetic memory such as a magnetic tape can be used.
[0412]
Between the photographing optical system 400 and the image pickup means 150, the optical low-pass filters 100H and 100V of the present invention are arranged. 100H is a low-pass filter for horizontally shifting the imaging light beam in the horizontal direction, and 100V is a low-pass filter for horizontally shifting the imaging light beam in the vertical direction, the configuration of which will be described later.
[0413]
Reference numeral 421 denotes a display such as a liquid crystal display, which displays the subject image acquired by the imaging unit 150 and the operating status of the optical device. A group of operation switches 422 includes a zoom switch, a shooting preparation switch, a shooting start switch, and shooting condition switches for setting an exposure control mode, an AF mode, and the like. A zoom actuator 423 drives the variator lens group 402 to change the focal length of the photographing optical system 400. Reference numeral 424 denotes a focus actuator that drives the focusing lens group 403 to adjust the focus state of the photographing optical system 400.
[0414]
A CPU 431 controls the operation of the entire photographing apparatus. Reference numeral 432 denotes an aperture driving circuit that drives the aperture mechanism 405 described above and adjusts the aperture diameter of the iris aperture to adjust the amount of light beam reaching the image pickup means 150. Reference numeral 433 denotes a shutter drive circuit that drives the shutter mechanism 406 to control the exposure time to the image pickup means 150. An optical low-pass filter driving circuit 434 drives the low-pass filters 100H and 100V of the present invention to obtain a predetermined optical effect.
[0415]
FIG. 48 is a cross-sectional view of the optical low-pass filter of the present invention and the image pickup means disposed behind the optical low-pass filter. In the figure, reference numeral 100H denotes a low-pass filter for horizontally shifting the imaging light flux in the horizontal direction of the imaging screen, and 150 denotes an imaging means such as a CCD. C is an optical axis which coincides with the optical axis of the photographing optical system 400 of FIG. First, the imaging unit 150 will be described.
[0416]
151 is a package of imaging means, 152 is a light receiving unit that photoelectrically converts incident light, 153 is a color filter array arranged in a mosaic pattern for color separation, and 154 is a microlens for condensing incident light on the light receiving unit 152 It is an array. Since the imaging unit 150 uses a known element, detailed description thereof is omitted.
[0417]
Reference numeral 100H denotes an entire optical low-pass filter for laterally shifting an incident photographing light beam in a predetermined direction. Reference numeral 101 denotes a first microprism array. Each prism has an apex angle of approximately 90 degrees and a substantially symmetric lattice prism with respect to the optical axis. These optical elements are arranged at equal pitches in one direction. It has become. In each microprism, a slope group whose normal extends to the upper right is 101a, and a slope group whose normal extends to the upper left is 101b. Reference numeral 102 denotes a second microprism array, which is a lattice-like optical element having the same shape as the first microprism array 101. In each microprism, a slope group whose normal extends to the lower left is denoted as 102a, and a slope group whose normal extends to the lower right is denoted as 102b. The prism portions of the microprism arrays 101 and 102 are arranged to face each other.
[0418]
Reference numeral 103 denotes a prism fixing means for fixing the first microprism array 101 by the fixing portion 103b, and the protrusion 103a is in sliding contact with the lower surface of the outer edge portion of the second microprism array 102. Reference numeral 104 denotes a prism pressing plate, and the protrusion 104 a is in sliding contact with the upper surface of the outer edge of the second microprism array 102.
[0419]
Reference numeral 105 denotes a laminated piezoelectric element that expands and contracts in response to an applied voltage. One end is fixed to the inner wall of the prism fixing means 103, and the other end is fixed to the outer edge side surface of the second microprism array 102. In the figure, no voltage is applied to the piezoelectric element 105 and its free length is A. At this time, in the two microprism arrays, one of the slope groups 101a and 102a is in close contact, and the other slope groups 101b and 102b are provided with a gap Gap1 by a minute amount δH. Then, among the light beams incident from above the optical low-pass filter 100H, the light beam La incident on the slope group 102a of the second microprism array 102 passes straight through the slope group 101a facing closely and is imaged. The light enters the predetermined pixel Pix4 of the means 150. On the other hand, the light beam Lb incident on the slope group 102b of the second microprism array 102 is totally reflected by the slope group 102b and proceeds to the left, passes through the slope groups 102a and 101a, and then totally reflected by the slope group 101b. Then, it proceeds downward and enters a predetermined pixel Pix5 of the imaging means 150.
[0420]
Therefore, about 50% of the light beam incident from above the optical low-pass filter 100H travels straight to reach the image pickup means, and the remaining 50% shifts to the left by one pitch of the micro prism to the image pickup means. Therefore, the average of the total luminous flux shifts to the left by 0.5 pitch of the microprism.
[0421]
FIG. 49 is a diagram showing a state when a predetermined voltage is applied to the piezoelectric element 105 included in the low-pass filter 100H and its length is expanded by δH. At this time, in the two micro prism arrays, the slope groups 101a and 102a in close contact with each other in FIG. 48 are provided with the gap Gap2 by a minute amount δH. In FIG. 48, the slope groups 101b and 102b with a gap are In FIG. 28, they are in close contact. Then, out of the light flux incident from above the optical low-pass filter 100H, the light beam La incident on the slope group 102a of the second microprism array 102 is totally reflected by the slope group 102a and proceeds to the right. After passing through 102b and 101b, the light is totally reflected by the slope group 101a and proceeds downward, and enters a predetermined pixel Pix5 of the image pickup means 150. On the other hand, the light beam Lb incident on the slope group 102b of the second microprism array 102 passes straight through the slope group 101b facing closely and enters the predetermined pixel Pix6 of the imaging means 150.
[0422]
Therefore, about 50% of the light beam incident from above the optical low-pass filter 100H travels straight to reach the image pickup means, and the remaining 50% shifts to the right by one pitch of the micro prism to the image pickup means. Therefore, the average of the entire luminous flux is shifted to the right by 0.5 pitch of the micro prism.
[0423]
Therefore, about 50% of the light beam incident from above the optical low-pass filter 100H travels straight to reach the image pickup means, and the remaining 50% shifts to the right by one pitch of the micro prism to the image pickup means. Therefore, the average of the entire luminous flux is shifted to the right by 0.5 pitch of the micro prism.
[0424]
FIG. 50 is a diagram showing a state when the voltage applied to the piezoelectric element 105 is reduced to about half of the state of FIG. 49 and its length is extended by half of δH from the initial state. At this time, the two microprism arrays are separated. Then, the light beam incident from above the optical low-pass filter 100H is totally reflected by the slope group 102a and the slope group 102b of the second microprism array 102 and returns to the original direction. That is, the second microprism array 102 is a reflective optical member having an action similar to that of a corner cube, and the low-pass filter also has a function as an optical switch that blocks an incident light beam.
[0425]
In the above-described configuration, the optical low-pass filter 100H is changed from the state shown in FIG. 48 to the state shown in FIG. 49 during one exposure time of the image pickup unit 150, whereby the incident light beam is laterally shifted by one pitch of the microprism. An effect can be obtained.
[0426]
In addition, when the exposure of the imaging unit 150 is completed, the effect of blocking the incident light beam can be obtained by changing the optical low-pass filter 100H from the state of FIG. 48 or 49 to the state of FIG.
[0427]
FIG. 51 is a perspective view when the optical low-pass filter of the present invention is applied to an imaging apparatus, showing only the color filter array 153 of the imaging means 150 and the micro-prism array of the low-pass filter, and other members are omitted. . The imaging means is a two-dimensional image sensor, and the color filter array used therefor is also a two-dimensional mosaic filter array. Therefore, the optical low-pass filter also needs to shift the light flux in the two-dimensional direction, that is, both the horizontal direction (indicated by the arrow H) and the vertical direction (indicated by the arrow V) of the photographing screen. Accordingly, the optical low-pass filter is also provided with two sets of a laterally shifting filter unit 100H and a longitudinally shifting filter unit 100V, each of which includes a driving piezoelectric element.
[0428]
FIGS. 52A and 52B are diagrams for explaining a method of driving the two sets of optical low-pass filters shown in FIG. 51. FIG. 52A is a timing chart of the piezoelectric element 105, and FIG. It is.
[0429]
First, referring to FIG. 2A, a method for driving two piezoelectric elements will be described. In FIG. 6A, the horizontal axis represents time t, the vertical axis represents the displacement of the piezoelectric element, the lower stage represents the displacement of the piezoelectric element for the low-pass filter 100H, and the upper stage represents the displacement of the piezoelectric element for the low-pass filter 100V.
[0430]
At time t0, the displacement of both elements is zero, and at time t1, a predetermined voltage is applied to the piezoelectric element 105H to expand it, and the displacement is set to δH. At time t2, a predetermined voltage is applied to the piezoelectric element 105V to expand it, and the displacement is set to δH. Thereafter, at time t3, the voltage applied to the piezoelectric element 105H is released to return the displacement to zero. Finally, at time t4, the applied voltages of the piezoelectric elements 105H and 105V are set to about one half of the predetermined value, and the displacement is set to δH / 2. T from time t0 to t4 corresponds to the charge accumulation time, that is, the total exposure time, and during each exposure time from T0 to T3 divided into four, the position of the emitted light beam is relatively shifted and It has become.
[0431]
When the above control is performed, the shift locus of the light beam incident on the imaging means is shown in FIG. By controlling the voltage to the piezoelectric element from time t0 to time t4, the position of the emitted light beam during each exposure time follows the position indicated by T0 to T3, and an optical low-pass effect can be obtained. Finally, when reaching the center T4, the low-pass filter acts as an optical shutter, so that the imaging light flux to the imaging means 150 is blocked. Therefore, the shutter mechanism 406 of FIG. 53 can be eliminated.
[0432]
At each timing from time t1 to time t3, the microprism arrays 101 and 102 are momentarily separated from each other and the optical shutter function works. However, in this embodiment, a multilayer piezoelectric element that can be deformed at high speed is used, which is actually harmful. Will not come out.
[0433]
FIG. 54 is a control flowchart of the CPU 431 included in the photographing apparatus shown in FIG. Hereinafter, the control flow of the photographing apparatus will be described using 54.
[0434]
Through step S101, in step S102, it is determined whether or not the main switch is turned on by the photographer. If the main switch is not turned on, the process stays in step S102. If it is determined in step S102 that the main switch is turned on, the CPU 431 exits the sleep state and executes step S111 and subsequent steps.
[0435]
In step S111, the photographing apparatus is initialized. Specifically, the photographic optical system in the retracted state is driven out to the shootable state and accepts the setting of the photographic conditions by the photographer.
[0436]
In step S112, the imaging unit 150 and the camera signal processing circuit 442 are driven to obtain a preview image, and the preview image is displayed on the display unit 421 in step S113.
[0437]
In step S114, it is determined whether or not the photographing preparation switch (denoted as SW1 in the flow chart) linked to the first stroke of the release button has been turned on by the photographer. When the on operation is not performed, the process returns to step S112, and the preview image display is repeatedly executed. If it is determined in step S114 that the shooting preparation switch has been turned on, step S114 is skipped and step S121 and subsequent steps are executed.
[0438]
In steps S121 to S122, the focus of the photographing optical system 400 is adjusted. This is so-called hill-climbing servo AF, which is focus adjustment control for stopping the lens by searching for a focus position where the high-frequency component of the image signal is the maximum value. In step S123, it is determined whether or not the subject is in focus. If the subject is not in focus, step S121 and step S122 are repeated. When the in-focus state is achieved, the driving of the focusing lens is stopped, and the process proceeds to step S131.
[0439]
In step S131, photometric calculation is performed. Specifically, the amount of the subject luminous flux received by the image pickup unit 411 is determined, and an aperture value and a shutter time for making the light amount appropriate for photographing are calculated. At this time, when the aperture priority AE mode is selected as the exposure control mode of the photographing apparatus, the aperture value set by the photographer is adopted. On the other hand, when the shutter priority AE mode or the program AE mode is selected, the aperture value is calculated according to a predetermined program built in the photographing apparatus.
[0440]
In step S132, the aperture drive amount is calculated based on the aperture value calculated in step S131. This corresponds to the difference between the current aperture value and the aperture value calculated in step S131. In step S133, the aperture drive is performed based on the aperture drive amount calculated in step S132.
[0441]
In step S134, it is determined whether or not the shooting trigger switch (indicated by SW2 in the flowchart) linked to the second stroke of the release button is turned on. If not, the steps S131 to S133 are performed. Repeatedly. On the other hand, if it is turned on, the process jumps from step S134 to step S141 to execute the photographing operation.
[0442]
In step S141, accumulation of image signals in the imaging unit 150 is started. In step S142, the low-pass filters 100H and 100V are driven in accordance with the timing chart described with reference to FIG. 52, and the emitted light beam is shifted and finally functions as an optical shutter to block the incident light beam to the imaging unit 150. In step S143, the shutter is closed. In step S144, the electric charge accumulated in the imaging unit 150 is read out via the charge transfer line and input to the camera signal processing circuit 442 via the amplifying unit 441. In step S145, the camera signal processing circuit 442 performs A / D conversion on the input analog image signal, performs image processing such as AGC control, white balance, γ correction, and edge enhancement, and further stores it in the CPU 431 as necessary. JPEG compression or the like is performed with the image compression program. In step S146, the image signal obtained in step S145 is recorded in the memory 443, and in step S147, the photographing operation ends.
[0443]
As described above, by driving the low-pass filter of this embodiment during charge accumulation in the photographing means during photographing, the photographing light flux is shifted by a predetermined amount in the horizontal and vertical directions, and an optical low-pass effect is obtained. The harmful moire that occurs in the photographed image can be avoided. Furthermore, the low-pass filter can be made to function as an optical shutter, and in combination with the mechanical shutter means, the shutter speed can be increased and the light shielding performance can be improved. Alternatively, the mechanical shutter can be eliminated.
[0444]
(Fifth embodiment)
The microprism in the fourth embodiment is an aggregate of lattice-like optical elements having an apex angle of about 90 degrees, and one set of microprism units shifts a photographing light beam only in one direction. The microprism of the second embodiment of the present invention described below is an aggregate of quadrangular pyramid optical elements having an apex angle of approximately 90 degrees, and by driving one microprism unit in two directions, horizontal and vertical, It has a function of shifting the photographing light flux in two directions.
[0445]
FIG. 55 is a perspective view of an optical low-pass filter according to the fifth embodiment, and corresponds to FIG. 50 of the fourth embodiment. Also in this figure, only the color filter array 253 of the image pickup means and the micro prism array of the low-pass filter are shown, and other members are omitted.
[0446]
Reference numeral 200 denotes an entire optical low-pass filter for laterally shifting an incident photographing light beam in two predetermined directions. 201 is a first microprism array, and 202 is a second microprism array. In the drawing, the second micro prism array is shown upside down so that the shape of the second micro prism array can be easily understood.
[0447]
Here, each prism constituting the first microprism array 201 is a regular quadrangular pyramid having a square bottom surface and an apex angle of approximately 90 degrees, and these are arranged on a two-dimensional plane at an equal pitch. On the other hand, the individual prisms constituting the second microprism array 201 have a shape in which a regular quadrangular pyramid having an apex angle of approximately 90 degrees is turned over, that is, when they are brought into close contact with the first microprism, the clearance therebetween is spread over the entire surface. The shape is substantially zero, and these are arranged on a two-dimensional plane at an equal pitch. The prism portions of the microprism arrays 201 and 202 are opposed to each other with a minute dimension in the optical axis direction, and the cross-sectional shape thereof is the same as that shown in FIG. 48 of the fourth embodiment.
[0448]
The second microprism array 202 is configured to be driven independently in two directions within a plane orthogonal to the optical axis by a mechanism similar to the mechanism of FIG. When photographing, the second microprism array 202 is driven as shown in the driving locus Tr2 of FIG. However, the length of each side of the drive locus at this time is the same as that described in FIG. 52, that is, δH. Then, in the first pyramid-shaped first microprism, two of the four inclined surfaces are always in close contact with the second microprism, and the inclining inclined surfaces are sequentially moved as the second microprism array is driven. Therefore, the light beam incident on the prism array is shifted in the horizontal and vertical directions on the photographing screen. In other words, in the fourth embodiment, one set of prism units composed of two microprism arrays is used, and one prism array is driven in a two-dimensional direction, so that a low pass equivalent to that in the fourth embodiment is obtained. An effect can be obtained.
[0449]
FIG. 56 is a cross-sectional view of an imaging apparatus incorporating the optical low-pass filter of the fifth embodiment. The present embodiment is different from the first embodiment only in that one set of low-pass filters 200 is arranged and one filter array is driven in two directions by a low-pass filter driving circuit 434. Other configurations are substantially the same as those shown in FIG. The control flow is also the same as that shown in FIG.
[0450]
As described above, in the present embodiment, the optical low-pass effect in the horizontal and vertical directions can be obtained with one set of low-pass filters, and further downsizing and cost reduction of the photographing apparatus can be achieved.
[0451]
(Sixth embodiment)
The microprism in the fourth embodiment is an aggregate of lattice-like optical elements having an apex angle of approximately 90 degrees, and shifts the photographing light flux by moving one microprism array in a direction perpendicular to the optical axis. It was a thing. The microprism of the third embodiment of the present invention shown below is an aggregate of lattice-like optical elements having an apex angle of about 45 degrees, and one of the microprism arrays is driven in the optical axis direction, so that the photographing light beam It has a function of shifting in a direction perpendicular to the axis.
[0452]
FIG. 57 is a cross-sectional view of the optical low-pass filter in the sixth embodiment, and corresponds to FIG. 48 in the fourth embodiment. In the figure, reference numeral 300H denotes a low-pass filter for horizontally shifting the imaging light flux in the horizontal direction of the imaging screen, and 350 denotes imaging means such as a CCD. C is an optical axis, which corresponds to the optical axis of the photographing optical system. Since the imaging unit 350 has the same structure as the imaging unit 150 of the fourth embodiment, description thereof is omitted.
[0453]
Reference numeral 300H denotes an entire optical low-pass filter for laterally shifting an incident photographing light beam in a predetermined direction. Reference numeral 301 denotes a first microprism array. The cross-sectional shape of each prism is a lattice-like optical element in which triangles having apex angles of approximately 45 degrees are arranged at an equal pitch and extend in the same direction in the direction perpendicular to the paper surface. It has become. In each microprism, a slope group whose normal is inclined by 45 degrees with respect to the optical axis is 301a, and a slope group whose normal is orthogonal to the optical axis is 301b. Reference numeral 302 denotes a second microprism array, which is a lattice-like optical element having the same shape as the first microprism array 301. In each microprism, a slope group whose normal is inclined by 45 degrees with respect to the optical axis is 302a, and a slope group whose normal is orthogonal to the optical axis is 302b. The prism portions of the microprism arrays 301 and 302 are arranged to face each other, and the slope groups 301a and 302a and the slope groups 301b and 302b are in close contact with each other.
[0454]
Reference numeral 303 denotes prism fixing means, which fixes the first microprism array 301 by the fixing portion 303b, and the fitting portion 303a is in sliding contact with the end surface of the outer edge of the second microprism array 302.
[0455]
Reference numeral 305 denotes a laminated piezoelectric element that expands and contracts in response to an applied voltage. One end is fixed to the bottom of the inner wall of the prism fixing means 303 and the other end is fixed to the lower surface of the outer edge of the second microprism array 302. Two piezoelectric elements 305 having the same shape are provided. In the figure, no voltage is applied to the piezoelectric element 305 and its free length is A. At this time, in the two microprism arrays, the slope groups 301a and 302a are in close contact, and 301b and 302b are also in close contact. Then, the light ray L incident from above the optical low-pass filter 100H enters the slope group 302a of the second microprism array 302, passes through the slope group 301a that is in close contact with each other, travels straight, and is imaged. Is incident on a predetermined pixel Pix6.
[0456]
Therefore, since all the light beams incident from above the optical low-pass filter 100H travel straight and reach the imaging unit, the light beam shift effect does not occur.
[0457]
FIG. 48 is a diagram showing a state when a predetermined voltage is applied to the piezoelectric element 305 provided in the low-pass filter 300H and its length is expanded by δA. At this time, in the two microprism arrays, the slope groups 301a and 302a that are in close contact with each other in FIG. Then, the light beam L incident from above the optical low-pass filter 300H is totally reflected by the slope group 302a and travels in the left direction. After passing through the slope groups 302b and 301b, the light beam L is totally reflected by the slope group 301a and travels downward. Then, the light enters the predetermined pixel Pix5 of the imaging unit 350.
[0458]
Therefore, all the light beams incident from above the optical low-pass filter 300H are shifted leftward by one pitch of the microprism and reach the image pickup means. Therefore, the optical low-pass effect can be obtained by changing the filter 300H from the state of FIG. 58 to the state of FIG. 59 during the exposure time.
[0459]
FIG. 59 is a perspective view when the optical low-pass filter of the present invention is applied to a photographing apparatus, showing only the color filter array 353 of the imaging means 350 and the micro-prism array of the low-pass filter, and other members are omitted. . The imaging means is a two-dimensional image sensor, and the color filter array used therefor is also a two-dimensional mosaic filter array. Therefore, the optical low-pass filter also needs to shift the light flux in the two-dimensional direction, that is, both the horizontal direction (indicated by the arrow H) and the vertical direction (indicated by the arrow V) of the photographing screen. Accordingly, the optical low-pass filter is also provided with two sets of a laterally shifting filter unit 300H and a longitudinally shifting filter unit 300V, each of which includes a driving piezoelectric element.
[0460]
FIGS. 60A and 60B are diagrams for explaining a method of driving the two sets of optical low-pass filters shown in FIG. 59. FIG. 60A is a timing chart of the piezoelectric element 305, and FIG. It is.
[0461]
First, referring to FIG. 2A, a method for driving two piezoelectric elements will be described. In FIG. 6A, the horizontal axis represents time t, the vertical axis represents the displacement of the piezoelectric element, the lower stage represents the displacement of the piezoelectric element for the low-pass filter 300H, and the upper stage represents the displacement of the piezoelectric element for the low-pass filter 300V.
[0462]
At time t0, the displacement of both elements is zero, and at time t1, a predetermined voltage is applied to the piezoelectric element 305H to expand it, and the displacement is set to δH. At time t2, a predetermined voltage is applied to the piezoelectric element 305V to expand it, and the displacement is set to δH. Thereafter, at time t3, the voltage applied to the piezoelectric element 305H is released to return the displacement to zero. Finally, at time t4, the voltage applied to the piezoelectric element 305V is released to return the displacement to zero. T from time t0 to t4 corresponds to the charge accumulation time, that is, the total exposure time, and during each exposure time from T0 to T3 divided into four, the position of the emitted light beam is relatively shifted and It has become.
[0463]
When the above control is performed, the shift locus of the light beam incident on the imaging means is shown in FIG. By controlling the voltage to the piezoelectric element from time t0 to time t4, the position of the emitted light beam during each exposure time follows the position indicated by T0 to T3, and an optical low-pass effect can be obtained.
[0464]
The configuration and control flow when the optical low-pass filter of the sixth embodiment described above is incorporated in the photographing apparatus are the same as those in FIGS. 52 and 53 of the fourth embodiment, and thus description thereof is omitted. .
[0465]
【The invention's effect】
As described above, according to the present invention, a light beam can be shifted in a short time by driving a small and light microprism array, and noise and vibration can be prevented. A low-pass effect can be obtained. Further, an optical shutter function can be added, and the photographing apparatus can be provided in a small size and at a low cost.
[0466]
In the present invention, attention is paid to the fact that micromechanics can be used to create a small and highly accurate mechanical shutter, and in particular, a mechanical shutter corresponding to each photoelectric conversion element forming the solid-state image sensor is solid-state imaged. Created in the element manufacturing process, such a shutter is individually driven and controlled for each photoelectric conversion element to expand the shooting dynamic range, compensate for variations in photoelectric conversion element sensitivity, correct shading, and further increase the incident transmittance Therefore, a decrease in absolute sensitivity can be prevented and a drive response can be increased, so that light leakage due to low response can be solved.
[0467]
Further, according to the present invention, the light beam can be shifted in a short time by driving a small and light microprism array in a minute amount, so that noise and vibration can be prevented and a sufficient low-pass effect can be obtained even when the exposure time is short. be able to. Further, an optical shutter function can be added, and the photographing apparatus can be provided in a small size and at a low cost.
[Brief description of the drawings]
FIG. 1 is a cross-sectional view of a first embodiment of the present invention.
FIG. 2 is an explanatory diagram of a color filter according to a first embodiment of this invention.
FIG. 3 is a cross-sectional view of a modified example of the first embodiment of the present invention.
FIG. 4 is a cross-sectional view of a modified example of the first embodiment of the present invention.
FIG. 5 is a plan view of a modified example of the first embodiment of the present invention.
FIG. 6 is a sectional view of a modified example of the first embodiment of the present invention.
FIG. 7 is a drive waveform diagram according to the first embodiment of the present invention.
FIG. 8 is a drive waveform diagram according to the first embodiment of the present invention.
FIG. 9 is a drive waveform diagram according to the first embodiment of the present invention.
FIG. 10 is a drive waveform diagram according to the first embodiment of the present invention.
FIG. 11 is a drive waveform diagram according to the first embodiment of the present invention.
FIG. 12 is a drive waveform diagram according to the first embodiment of the present invention.
FIG. 13 is a block diagram of a first embodiment of the present invention.
FIG. 14 is a flowchart of the first embodiment of the present invention.
FIG. 15 shows a manufacturing method according to the first embodiment of the present invention.
FIG. 16 shows a manufacturing method according to the first embodiment of the present invention.
FIG. 17 shows a manufacturing method according to the first embodiment of the present invention.
FIG. 18 shows a manufacturing method according to the first embodiment of the present invention.
FIG. 19 shows a manufacturing method according to the first embodiment of the present invention.
FIG. 20 shows a manufacturing method according to the first embodiment of the present invention.
FIG. 21 is a manufacturing method according to the first embodiment of the present invention.
FIG. 22 shows a manufacturing method according to the first embodiment of the present invention.
FIG. 23 shows a manufacturing method according to the first embodiment of the present invention.
FIG. 24 is a manufacturing method according to the first embodiment of the present invention.
FIG. 25 is a manufacturing method according to the first embodiment of the present invention;
FIG. 26 shows a manufacturing method according to the first embodiment of the present invention.
FIG. 27 shows a manufacturing method according to the first embodiment of the present invention.
FIG. 28 is a manufacturing method according to the first embodiment of the present invention.
FIG. 29 is a manufacturing method according to the first embodiment of the present invention.
FIG. 30 shows a manufacturing method according to the first embodiment of the present invention.
FIG. 31 is a manufacturing method according to the first embodiment of the present invention.
FIG. 32 is a plan view of a modified example of the present invention.
FIG. 33 is a cross-sectional view of a second embodiment of the present invention.
FIG. 34 is a sectional view of a modification of the second embodiment of the present invention.
FIG. 35 is a sectional view of a modification of the second embodiment of the present invention.
FIG. 36 is a cross-sectional view of a third embodiment of the present invention.
FIG. 37 is an explanatory diagram of a color filter according to a third embodiment of the present invention.
FIG. 38 shows a manufacturing method according to the third embodiment of the present invention.
FIG. 39 shows a manufacturing method according to the third embodiment of the present invention.
FIG. 40 shows a manufacturing method according to the third embodiment of the present invention.
FIG. 41 is a manufacturing method according to the third embodiment of the present invention.
FIG. 42 is a manufacturing method according to the third embodiment of the present invention.
FIG. 43 is a manufacturing method according to the third embodiment of the present invention.
FIG. 44 shows a manufacturing method according to the third embodiment of the present invention.
FIG. 45 is a manufacturing method according to the third embodiment of the present invention.
FIG. 46 is a sectional view of a modification of the third embodiment of the present invention.
FIG. 47 is a sectional view of a modification of the third embodiment of the present invention.
FIG. 48 is a cross-sectional view of an optical low-pass filter according to a fourth embodiment of the present invention.
FIG. 49 is an explanatory diagram of a low-pass effect of the optical low-pass filter according to the fourth embodiment of the present invention.
FIG. 50 is an explanatory diagram of an optical shutter effect of the optical low-pass filter according to the fourth embodiment of the present invention.
FIG. 51 is a perspective view of an optical low-pass filter according to a fourth embodiment of the present invention.
FIG. 52 is an explanatory diagram of an optical low-pass filter driving method according to the fourth embodiment of the present invention.
FIG. 53 is a configuration diagram of a photographing apparatus according to a fourth embodiment of the present invention.
FIG. 54 is a control flow diagram for a photographing apparatus according to a fourth embodiment of the present invention.
FIG. 55 is a perspective view of an optical low-pass filter according to a fifth embodiment of the present invention.
FIG. 56 is a configuration diagram of a photographing apparatus according to a fifth embodiment of the present invention.
FIG. 57 is a cross-sectional view of an optical low-pass filter according to a sixth embodiment of the present invention.
FIG. 58 is an explanatory diagram of a low-pass effect of an optical low-pass filter according to a sixth embodiment of the present invention.
FIG. 59 is a perspective view of an optical low-pass filter according to a sixth embodiment of the present invention.
FIG. 60 is an explanatory diagram of an optical low-pass filter driving method according to the sixth embodiment of the present invention.
FIG. 61 is a cross-sectional view of a conventional solid-state imaging device.
FIG. 62 is a characteristic explanatory diagram of a solid-state imaging element according to a conventional technique.
FIG. 63 is an explanatory diagram of problems of a conventional solid-state imaging device.
FIG. 64 is an improved cross-sectional view of a conventional solid-state imaging device.
[Explanation of symbols]
10 Solid-state image sensor
11 Microlens
12 Color filters
13 Optical waveguide
14 Interface
15 Planarization layer
16 leaf spring
17 Movable microprism
18 Transparent electrode layer
19 Insulating layer
21 Stator
22 coils
23 Permanent magnet
24 York
25 slider
26 Coil spring
27 Bearing
28 Coil spring
29 Cam plate
30 Solid-state image sensor
31 Spatial domain
32 Bulkhead layer
33 Transparent electrode layer
34 Insulating layer
35 Insulation layer
36 Transparent electrode layer
37 flaps
38 Reflective layer
110 Base layer
111 photoelectric conversion element
112 Read gate
113 Vertical transfer electrode
114 Vertical transfer charge transfer path
115 channel stop region
116 Fixed microprism
117 Transparent electrode layer
118 Insulating layer
119 Shading layer
120 Incident light
121 Transparent electrode layer
122 Insulating layer
123 Transparent electrode layer
124 Insulating layer
125 Transparent electrode layer
126 Insulating layer
127 Transparent electrode layer
128 Insulating layer
129 Piezoelectric member
130 Shutter aperture curve
131 Accumulation control time
132 Optical axis
133 lens
135 aperture
134 AF drive motor
137 Aperture driving means
138 Focus drive means
139 Flashing means
140 A / D conversion means
141 Shooting control means
142 Signal processing means
143 Display means
144 Recording means

Claims (1)

A light receiving region comprising a plurality of photoelectric conversion elements arranged on the main surface of the semiconductor substrate;
In a solid-state imaging device composed of a light control region composed of a light control element that controls the incident state of each photoelectric conversion element,
The light control element is a movable microprism provided for each photoelectric conversion element,
A stationary microprism facing the movable microprism and having an incident surface substantially the same shape as the prismatic surface of the movable microprism;
An optical waveguide facing the other end of the movable microprism;
Driving means for changing the first facing distance between the movable microprism and the fixed microprism and the second facing distance between the movable microprism and the optical waveguide;
A light shielding means for shielding stray light due to the emission of light rays on a surface along the main optical axis in the first facing distance and / or the second facing distance;
A solid-state imaging device that controls a total reflection state and a transmission state of a movable microprism by changing a first facing interval by a driving unit.

Images