JP7065054B2 - Photoelectric conversion element, image reader and image forming device - Google Patents

Description

The present invention relates to a photoelectric conversion element, an image reading device, and an image forming device.

Conventionally, a CCD has been used as a photoelectric conversion element used in a scanner, but due to the recent demand for higher speed, a CMOS linear image sensor (CMOS sensor) is attracting attention. The CMOS sensor is the same as the CCD in that the incident light is photoelectrically converted by a photodiode (PD). However, the CMOS sensor differs from the CCD in that it performs charge-voltage conversion in the vicinity of the pixel and outputs it to the subsequent stage. Further, since a CMOS sensor uses a CMOS process, it is possible to incorporate a circuit such as an ADC (Analog Digital Converter), which is more advantageous than a CCD in terms of high speed.

However, in a CMOS linear image sensor, for example, a plurality of circuits such as a transfer transistor that transfers a PD charge to an FD (Floating Diffusion), a reset transistor that resets the FD, and an amplification transistor (source follower) that outputs a signal to a subsequent stage are used. It needs to be configured in the pixel.

Therefore, in the CMOS linear image sensor, the drive signal of each circuit and a plurality of wirings such as power supply and GND are arranged in multiple layers in the pixel, and the aperture of the pixel (the area where light can be incident) is opened by these multi-layer wiring. It will be limited. In particular, since the pixel aperture of the CMOS linear image sensor affects the incident angle of light, a scanner having a different incident angle of light for each pixel position causes unevenness (shading) in the amount of light, and S in the main scanning direction. There was a problem that unevenness of / N may occur.

In response to the above problem, for example, in Patent Document 1, in the peripheral portion of each imaging area, the positions of the microlens and the aperture are arranged so as to be offset from the corresponding photoelectric conversion region toward the center of each imaging area. The solid-state image sensor is disclosed.

However, in the configuration in which the aperture position is changed for each pixel, the aperture position is fixed according to the incident angle of the light, so that there is a problem that the optical system used for the incident of the light is limited.

The present invention has been made in view of the above, and is a photoelectric conversion element capable of reducing unevenness in the amount of light received by each pixel without limiting the optical system used for incident light, and image reading. It is an object of the present invention to provide an apparatus and an image forming apparatus.

In order to solve the above-mentioned problems and achieve the object, the present invention comprises a plurality of light receiving elements arranged in a straight line for each color of light received by outputting an analog image signal by photoelectric conversion for each pixel. The light receiving element comprises a wiring formed in a wiring layer and enabled as at least one of a signal line, a power supply, and a ground used for the light receiving element or a peripheral circuit of the light receiving element, and the light receiving element has a light receiving surface. The light that is vertically incident on the photoelectric conversion element is incident on the light receiving surface through the opening that is sandwiched between the wirings located on the straight line and opens without changing the incident angle. Is characterized in that it has a first light-shielding region and a second light-shielding region formed by light-shielding by each of the wirings located on the straight line across the opening on the light-receiving surface.

According to the present invention, there is an effect that unevenness in the amount of light received by each pixel can be reduced without limiting the optical system used for incident light.

FIG. 1 is a diagram showing an outline of a reduction optical system of an image reader. FIG. 2 is a diagram illustrating an outline of the pixel configuration of the photoelectric conversion element. FIG. 3 is a schematic cross-sectional view of the photoelectric conversion element showing a decrease in sensitivity due to the incident angle of light in the first main scanning direction. FIG. 4 is a graph showing the effective sensitivity of light that changes depending on the positions of the pixels arranged in the main scanning direction and the angle of incidence of the light on the pixels. FIG. 5 is a schematic cross-sectional view in the main scanning direction of the photoelectric conversion element showing a configuration example of an opening sandwiched between wirings and opened. FIG. 6 is a graph showing the effective sensitivity of light to the positions of pixels in the main scanning direction in the configuration shown in FIG. 5 (b). FIG. 7 is a diagram showing an opening with respect to the PD and a light receiving range of the PD in the configuration shown in FIG. 5 (b). FIG. 8 is a graph showing the positions of pixels arranged in the main scanning direction and the effective sensitivity of light that changes according to the incident angle of light with respect to the pixels according to the type of lens. FIG. 9 is a schematic cross-sectional view of the photoelectric conversion element according to the embodiment in the main scanning direction. FIG. 10 is a graph showing the positions of pixels arranged in the main scanning direction and the effective sensitivity of light that changes according to the incident angle of light with respect to the pixels according to the type of lens. FIG. 11 is a schematic cross-sectional view in the main scanning direction of a modified example of the photoelectric conversion element according to the embodiment. FIG. 12 is a graph showing the effective sensitivity of light corresponding to the positions of the pixels arranged in the main scanning direction and the incident angle of the light with respect to the pixels. FIG. 13 is a diagram showing an outline of the configuration of a photoelectric conversion element that prevents a difference in shading for each color of received light. FIG. 14 is a diagram showing a read wiring from a pixel and a first configuration example for suppressing crosstalk in the read wiring. FIG. 15 is a diagram showing a second configuration example of suppressing crosstalk in the read wiring from the pixel and the read wiring. FIG. 16 is a diagram showing a reading wiring from a pixel and a third configuration example for suppressing crosstalk in the reading wiring. FIG. 17 is a diagram illustrating the overall configuration of the photoelectric conversion element. FIG. 18 is a diagram showing an outline of an image forming apparatus including an image reading apparatus having a photoelectric conversion element.

First, the background leading to the present invention will be described. FIG. 1 is a diagram showing an outline of a reduction optical system of an image reader. The reduction optical system of the image reader reduces the reflected light from the document onto the CMOS linear image sensor (photoelectric conversion element) 11 by the condenser lens 10 to form an image.

In the reduced optical system shown in FIG. 1, it is shown that the reflected light from the central portion in the main scanning direction of the document and from the edge portion of the document are incident on the photoelectric conversion element 11. Here, the solid line indicates the main ray, and the dotted line is the left / right ray.

The reflected light (main light ray) from the central portion in the main scanning direction of the document is incident on the light receiving surface of the photoelectric conversion element 11 substantially perpendicularly to the light receiving surface via the lens 10. The reflected light from the edge of the document is incident on the light receiving surface of the photoelectric conversion element 11 at a predetermined angle θ. The larger the angle of view determined by the focal length of the lens 10 and the like, the larger the incident angle θ. Further, the incident angle θ is small near the central portion in the main scanning direction of the document, but increases as the distance from the central portion increases.

When this is viewed from the photoelectric conversion element 11 side, the incident angle θ of the light is substantially 0 in the pixel to which the reflected light from the vicinity of the central portion in the main scanning direction is incident. The incident angle θ gradually increases as the distance from the center of the photoelectric conversion element 11 increases, and becomes maximum at the pixel at the end of the photoelectric conversion element 11.

As described above, the photoelectric conversion element 11 used in the reduction optical system is incident with light having a predetermined angle. However, in the photoelectric conversion element 11, if the configuration of the reduction optical system is changed, the incident angle θ of light with respect to the same pixel also changes. In FIG. 1, the lens 10 is used to collect the reflected light from the document, but it does not necessarily have to be a lens, and a curved mirror may be used.

FIG. 2 is a diagram illustrating an outline of a pixel configuration of a CMOS linear image sensor (photoelectric conversion element). The CMOS linear image sensor used in the image reader is the same as the CCD in that the incident light is photoelectrically converted by a photodiode (PD), but the charge-voltage conversion is performed near the pixel and output to the subsequent stage. Is different.

The pixel of the CMOS linear image sensor has a light receiving element (photodiode: PD) that photoelectrically converts incident light, and a pixel circuit (PIX_BLK) that converts the charge (analog image signal) accumulated in the PD into a voltage signal. PIX_BLK includes a floating diffusion (FD) that performs charge-voltage conversion, a reset transistor that resets the FD, a transfer transistor that transfers the charge of the PD to the FD, and the like, but is not shown in FIG.

Further, each pixel is arranged in one direction for each color of RGB, and the image signal converted into a voltage by PIX_BLK is read out to a subsequent stage via a read line (R, G, B). A PGA (Programable Gain Amplifier), an ADC (Analog Digital Converter), or the like (not shown) is connected to the subsequent stage of the reading line.

The CMOS linear image sensor differs from the CCD in that it has a pixel circuit for each pixel. As a result, the CMOS linear image sensor has a large number of pixels such as a read wiring for reading an image signal from PIX_BLK, a control signal (reset signal, transfer signal) of PIX_BLK (reset signal, transfer signal), a power supply, and a GND, which are not shown in FIG. Will be placed around.

In particular, when there are many wires around the PD, the sensitivity is lowered because the area of the PD needs to be reduced in order to secure the wiring area. Therefore, in a CMOS linear image sensor, it is common to make the wiring layer a multi-layer structure and arrange the wiring using a plurality of wiring layers.

FIG. 3 is a schematic cross-sectional view of a photoelectric conversion element (CMOS linear image sensor) showing a decrease in sensitivity due to an incident angle of light in a first main scanning direction. As shown in FIG. 3, the photoelectric conversion element has a PD for each pixel, and a pixel separation region (STI: Shallow-Trench-Isolation) 12 for separating each pixel is formed. Further, in the photoelectric conversion element, the wiring used for the read wiring for reading the signal from the pixel, the power supply, the GND (ground), etc. is configured so as to straddle, for example, two wiring layers (M1 and M2). A color filter (CF) is arranged on the top. Between each layer is an insulating film layer.

Also, the wiring is formed, for example, above the STI. Then, the photoelectric conversion element allows light to pass through a region (opening for opening the PD) in which wiring does not exist. That is, the aperture width of the pixel (the width at which the light receiving surface of the PD can be irradiated) is determined by the width of the region where the region where the wiring does not exist and the region where the PD exists overlap with respect to the incident direction of the light. In FIG. 3, the width in which the wiring does not exist and the width of the PD are the same.

As shown in FIG. 1, the light incident on the pixel at the center of the photoelectric conversion element is incident substantially perpendicular to the light receiving surface of the PD, and the aperture width (Wc) covers the entire PD (FIG. 3 (FIG. 3). a)). On the other hand, since the light incident on the pixels at the end of the photoelectric conversion element is incident at a predetermined angle, the aperture width (We) is relative to the aperture width (Wc) at the center. It decreases as Wd (Wc> We: FIG. 3 (b)).

When light having an angle with respect to the light receiving surface of the PD is incident, the light incident by the wiring of M1 is kicked as shown in FIG. 3 (b). That is, the wiring casts a shadow on the PD. The influence of this wiring is more likely to occur as the number of wirings (wiring layers) increases, which is particularly disadvantageous for CMOS linear image sensors.

As shown in FIG. 3, when the width of the PD and the width in which the wiring provided above the PD does not exist are the same, the sensitivity is such that the pixels at the ends are compared with the pixels at the center of the photoelectric conversion element. There is a problem of shading (variation in the amount of received light) that reduces the amount of light received. This shading causes unevenness in S / N and causes deterioration of image quality, and the influence is particularly remarkable in a CMOS linear image sensor having a large number of wirings between pixels.

Since the CCD generally has no (or few) wires between the pixels, there is almost no influence of the sensitivity decrease due to the incident angle of the light described above. Further, in a 1x optical system such as CIS (Contact Image Sensor), there is no problem because the light having an angle from the original is not incident in the first place. That is, the above-mentioned shading is a problem that has a particularly large influence on the CMOS linear image sensor in the reduced optical system. Specifically, as shown in FIG. 2, in the CMOS linear image sensor, the signal output from the PD is output from the read wiring arranged for each pixel from PIX_BLK, whereas in the area sensor, one in one row. This is because it is common that the main read wiring is provided. That is, the CMOS linear image sensor has a larger influence because the number of wires arranged between the PDs is larger than that of the area sensor.

FIG. 4 is a graph showing the effective sensitivity of light that changes depending on the positions of the pixels arranged in the main scanning direction and the angle of incidence of the light on the pixels. FIG. 4A shows the influence of wiring on shading for each position of the pixels arranged in the main scanning direction. As shown in FIG. 3, in the CMOS linear image sensor in the reduced optical system, the light is kicked by the wiring according to the incident angle of the light on the pixel, so that the position away from the pixel in the center of the main scanning direction. The sensitivity decreases as the pixels are arranged in (FIG. 4A). Further, when focusing on the pixel at the end of an arbitrary main scanning direction, the sensitivity decrease becomes large depending on the incident angle of light (FIG. 4 (b)).

Note that FIG. 4 shows the distribution of the effective sensitivity seen from each pixel, and when the light is not kicked by the wiring, the ideal characteristic is flat as shown by the dotted line. Actually, there is attenuation of the amount of incident light due to the cos4 power rule of the lens, etc., but in FIG. 4, for the sake of simplification of the explanation, the attenuation other than the kicking of light by the wiring is excluded.

FIG. 5 is a schematic cross-sectional view in the main scanning direction of a photoelectric conversion element (CMOS linear image sensor) showing a configuration example of an opening sandwiched between wirings and opened. FIG. 5A is the same as FIG. 3A. FIG. 5B shows the state of incident light on the pixels at the end of the photoelectric conversion element as in FIG. 3B, but the wiring of the M1 and M2 layers is connected to the positions of PD and STI. It differs from FIG. 3 (b) in that the positions are shifted by Δ1 and Δ2, respectively.

In FIG. 3 (b), the incident light was kicked by the wiring, but in the configuration shown in FIG. 5 (b), the position of the opening is shifted for each pixel so that the light is not kicked by the wiring. Therefore, shading is reduced. In the example shown in FIG. 5B, the pixels at the end of the photoelectric conversion element are shown, but the same applies to the pixels other than the ends, and the shift amount (Δ1, Δ2) of the wiring is the main scanning direction. It is arranged so as to be slightly different for each pixel according to the pixel position of.

FIG. 6 is a graph showing the effective sensitivity of light to the positions of pixels in the main scanning direction in the configuration shown in FIG. 5 (b). In the configuration shown in FIG. 5B, the influence of shading is reduced because the position of the wiring is shifted for each pixel so as to minimize the kicking of light.

FIG. 7 is a diagram showing an opening with respect to the PD and a light receiving range of the PD in the configuration shown in FIG. 5 (b). In the configuration shown in FIG. 5B, shading is reduced by optimizing the position (that is, the opening) of the wiring for each pixel so as to correspond to the incident angle of the light in advance. As shown in FIG. 7, for example, when the angle of view is narrower than that of the condenser lens used in FIG. 5 (b), that is, when the incident angle of light is small, kicking by wiring occurs and shading deteriorates. Resulting in.

As described above, in the configuration shown in FIG. 5B, the wiring is arranged at the position optimized for the incident angle predetermined for each pixel, but the light is emitted at different incident angles. When irradiated, the conditions are not optimal and light is kicked. Therefore, in the technique of shifting the aperture position for each pixel to reduce shading, the angle of view, that is, the condensing lens that can be used is limited by the CMOS linear image sensor, and the versatility is lost.

Further, when the image sensor combined with the microlens (on-chip lens) has the configuration shown in FIG. 5B, if the incident angle of the light is changed, the influence of the kicking of the light becomes large. It ends up. That is, when the light collected by the microlens is kicked by the wiring, almost all the light that should be incident on the pixel is kicked, so that the effect is not limited to the extent that can be corrected by the shading correction, and the pixel is missing. You end up creating pixels that can't read the image called.

FIG. 8 is a graph showing the positions of pixels arranged in the main scanning direction and the effective sensitivity of light that changes according to the incident angle of light with respect to the pixels according to the type of lens. When a condenser lens having an optimum angle of view for the aperture position changed for each pixel is used for the configuration shown in FIG. 5 (b), the influence of shading is reduced, but it is not optimal. When a condenser lens with an angle of view (light incident angle is small) is used for the configuration shown in FIG. 5 (b), the incident angle should have been small and the effect of shading should have been small. On the contrary, the shading is deteriorated (FIG. 8 (a)).

That is, as can be seen from the characteristics of the sensitivity to the light incident angle at the pixel at an arbitrary end shown in FIG. 8 (b), the shift amount of the wiring is optimized for each pixel at an arbitrary incident angle (θ). When is determined, the decrease in sensitivity is suppressed at the incident angle θ, but at other incident angles (θ'), the shading is deviated from the optimum conditions, and the shading is worsened rather than reduced. That is, only a condenser lens having an incident angle of θ can be used. In this way, with a photoelectric conversion element in which the shift amount of the wiring is changed for each pixel, shading deteriorates when light is incident at an incident angle that is not optimized, so the range of lenses that can be used is limited. Will be done.

The area image sensor is mainly used for cameras, and the distance (conjugated length) between the area image sensor and the condenser lens is generally as narrow as several cm. Therefore, the angle of view determined by the conjugate length has a small difference regardless of which condenser lens is selected, and has a small effect on shading. However, the linear image sensor is mainly used for scanners and needs to be compatible with various optical methods, so that the conjugate length is wide, ranging from tens to several tens of centimeters. Therefore, it can be said that it is a problem peculiar to the linear image sensor that the difference in the angle of view due to the condenser lens is large and the shading is deteriorated by the condenser lens as described above.

FIG. 9 is a schematic cross-sectional view of the photoelectric conversion element (CMOS linear image sensor) according to the embodiment in the main scanning direction. In the configuration of the photoelectric conversion element shown in FIG. 3B described above, the optical path was not in contact with the wiring on the side where the light was not kicked. This is because the end of the wiring and the end of the PD were formed so as to be aligned, but conversely, it means that there is still light that is not received by the PD and is not kicked by the wiring. do. The photoelectric conversion element according to the embodiment shown in FIG. 9 is configured such that a PD region exists under the wiring when the photoelectric conversion element is viewed from above. Further, the PDs are arranged for each color on a straight line in the main scanning direction, for example.

FIG. 9A shows the state of light incident on the pixel at the center of the photoelectric conversion element in the main scanning direction. The configuration shown in FIG. 9 (a) is formed so that the PD region is wider on both outer sides of the straight line extending in the main scanning direction by Δ than the configuration shown in FIG. 3 (a). The difference is that PD exists under each of the wirings of the M1 layer forming the portion. Here, since the opening by the M1 layer and the opening by the M2 layer are the same, the final opening width in the pixel in the central portion of the configuration shown in FIG. 9A is the same Wc as in FIG. 3A. be. Hereinafter, the two light-shielding regions (and the corresponding light-shielding regions) formed on both sides of the PD shown in FIG. 9A and having a width in the main scanning direction of Δ are referred to as a first light-shielding region and a second light-shielding region, respectively. I have something to note.

FIG. 9B shows the state of light incident on the pixel at the end of the photoelectric conversion element in the main scanning direction. In the configuration shown in FIG. 9 (b), first, the light kicking by the wiring occurs in the M2 layer, and the point that the substantial opening width is reduced by Wd is shown in FIG. 3 (b). Same as the configuration. However, unlike the configuration shown in FIG. 3B, the configuration shown in FIG. 9B is different from the configuration shown in FIG. 3B because the PD region is formed under the wiring on the side where the light is not kicked. Due to the angled light incident, the light entering under the wiring is applied to the PD. That is, while there is an aperture width (Wd) that decreases due to the kicking of light, the aperture that increases because the light that was not illuminated on the PD in the configuration shown in FIG. 3 (b) is irradiated on the PD. There will be a width (Wu). That is, the final opening width (We) is expressed by the following equation 1.

We = Wc-Wd + Wu ... (1)
We: Aperture width at the edge pixel
Wc: Aperture width at the central pixel
Wd: Aperture width of the end pixel reduced by kicking in the wiring
Wu: Aperture width of end pixels increased by widening PD width

As described above, the configuration shown in FIG. 9 (b) is improved by the increase due to the spread and formation of the PD, and the decrease in light due to kicking is improved, and is shown in FIG. 3 (b). The amount of light received increases by the amount of Wu more than the aperture width (= Wc—Wd) in the above configuration.

Further, FIG. 9 (c) shows a case where the incident angle of light is smaller than that of FIG. 9 (b). In this case, the aperture width is also determined by the above equation 1 and the light is kicked. The decrease in PD is improved by the increase due to the expansion of PD.

Here, in the configuration shown in FIG. 9B, the amount of light kicked and the amount of increase due to the expansion of the PD change according to the incident angle, and the aperture width increases as the incident angle decreases. It has a monotonous tendency like this. This is because the configuration does not have a fixed opening direction for each pixel as in the configuration shown in FIG. 5 (b). That is, since the shading does not specifically deteriorate depending on the incident angle of light as shown in FIG. 7, the limitation of the condenser lens (angle of view) used in the image reading device is reduced.

In FIG. 10, in the configuration of the photoelectric conversion element shown in FIG. 9, the effective sensitivity of light that changes according to the positions of the pixels arranged in the main scanning direction and the incident angle of light with respect to the pixels is determined according to the type of lens. It is a graph which shows. In the configuration shown in FIG. 9, the influence of shading is reduced regardless of the angle of view (light incident angle) of the condenser lens (FIG. 10A). Since this is not a configuration in which the photoelectric conversion element has a fixed aperture direction as in the configuration shown in FIG. 5 (b), the decrease in sensitivity at a certain incident angle (θ) in an arbitrary end pixel is suppressed. This is because the decrease in sensitivity is further suppressed at an incident angle (θ') smaller than the incident angle (θ) (FIG. 10 (b)). Therefore, in the configuration of the photoelectric conversion element shown in FIG. 9, it is improved that the range of usable lenses as shown in FIG. 8 is limited.

In this way, the photoelectric conversion element is positioned on a straight line across the opening when the light passing through the opening sandwiched between the wirings located on the straight line is incident perpendicular to the light receiving surface. By being formed so as to have two regions (first light-shielding region and second light-shielding region) that are shielded by each of the wirings to be light-shielded, it is possible to compensate for the amount of light kicked by the wiring to the light receiving element, and the condenser lens. It is possible to reduce shading due to the difference in (angle of view). Further, since the configuration shown in FIG. 9B can be uniformly the same for all pixels, the development cost can be suppressed as compared with the configuration shown in FIG.

FIG. 11 is a schematic cross-sectional view in the main scanning direction of a modified example of the photoelectric conversion element (CMOS linear image sensor) according to the embodiment. In the configuration shown in FIG. 9, the effect of reducing shading can be obtained regardless of the incident angle of light, but shading itself is not always sufficient because it occurs. Further, as shown in FIG. 9B, the light kicking that causes shading contributes more to the higher wiring (at a position farther from the PD). For example, in the case shown in FIG. 9B, the wiring of the M1 and M2 layers has the same position in the main scanning direction, but when the incident angle is attached to the light, the wiring of the M2 layer far from the PD makes the light light. Was kicked, and the wiring of the M1 layer did not contribute to the kick at all. This means that the higher the wiring farther from the PD, the greater the influence on kicking, and it is important to reduce the kicking in this higher layer for shading reduction.

Therefore, in the modification of the photoelectric conversion element according to the embodiment, the shading is further reduced by arranging the wiring so that the wiring layer is farther from the PD and the opening (the region through which light can pass) becomes larger. It is configured to be. Specifically, in the modification of the photoelectric conversion element shown in FIG. 11, as compared with the photoelectric conversion element shown in FIG. 9, two wires are arranged in each of the M1 and M2 layers, but three wires are arranged in M1. It is changed so that one is arranged in M2.

In the photoelectric conversion element shown in FIG. 11, the aperture width (Wm2) due to M2 is larger than the aperture width (Wm1) due to M1. FIG. 11A shows the state of light incident on the pixel in the center of the photoelectric conversion element as in the case shown in FIG. 9A, but is perpendicular to the light receiving surface of the PD. When light is incident, the amount of light received does not change.

On the other hand, in the incident of light on the end pixels of the photoelectric conversion element shown in FIG. 11 (b), the kicking of light is reduced as compared with the case shown in FIG. 9 (b). This is because the opening in the M2 layer is enlarged so that the light is not kicked by the wiring in the M2 layer, and the same effect as the shift of the wiring position shown in FIG. 5B can be obtained. It is thought that there is. However, since the photoelectric conversion element shown in FIG. 11 has a configuration in which the opening direction is not fixed according to the direction in which the light is incident and the opening direction is widened, the light as shown in FIG. 7 is used. Even if the incident angle of the light is small, the kicking of light does not worsen.

Further, in the configuration shown in FIG. 11, the light is not kicked by the wiring of the M2 layer, and the light is kicked by the wiring of the M1 layer. At this time, the aperture width (We') when the incident angle of light is attached is expressed by the following equation 2.

We'= Wc'-Wd' + Wu'... (2)
We': Aperture width at the edge pixel
Wc': Aperture width at the central pixel
Wd': Aperture width of the end pixel reduced by kicking in the wiring
Wu': Aperture width of end pixels increased by expanding PD

Further, since the thickness of the wiring (in the height direction in the figure) is small with respect to the width, Wd'≈Wu', and the relationship of the following equation 3 holds.

We'≒ Wc' ・ ・ ・ (3)

That is, in the configuration of the photoelectric conversion element shown in FIG. 11, the decrease and the increase of the received light are canceled out, and the sensitivity of the central pixel and the sensitivity of the edge pixel are substantially the same. As described above, in the photoelectric conversion element according to the embodiment, the farther the wiring is from the PD, the larger the opening is taken, so that the kicking of light by the wiring is offset and shading can be suppressed.

Although the configuration shown in FIG. 11 shows an example in which the number of wirings between pixels is four, the same effect can be obtained even if the number of wirings is not four. Further, in the configuration shown in FIG. 11, the opening width is changed by changing the number of wirings arranged in each wiring layer, but the photoelectric conversion element may have the same number of wirings in each wiring layer, for example. , The same effect may be obtained by changing the width of the wiring and the like. Further, although FIG. 11 shows an example of the pixels arranged in the main scanning direction, the photoelectric conversion element has the same configuration between the pixels in the sub-scanning direction, so that the same effect can be obtained. It may be configured to occur.

In the configuration shown in FIG. 11, while the opening of M2 is large, the opening of M1 is small. Therefore, the sensitivity of the central pixel is reduced, but in general, in a CMOS linear image sensor, the width of the wiring is about 1/20 of the width of the PD, and the effect of the sensitivity reduction in the central pixel is small. It can be ignored.

FIG. 12 is a graph showing the effective sensitivity of light corresponding to the positions of the pixels arranged in the main scanning direction and the incident angle of light with respect to the pixels in the configuration of the photoelectric conversion element shown in FIG. As shown in FIG. 12, in the configuration of the photoelectric conversion element shown in FIG. 11, since the decrease in sensitivity is offset regardless of the incident angle of light, shading does not occur.

Next, the configuration of the photoelectric conversion element that prevents the shading from being different for each color of the received light will be described. The wiring shown in FIG. 9 includes the reading wiring from PIX_BLK shown in FIG. At this time, paying attention to the read wiring arranged between PDs for each color shown in FIG. 2, there is no read wiring in Red, Red read wiring is arranged in Green, and Red / Green in Blue. Read wiring is arranged respectively. That is, the read wiring arranged between the pixels is different for each color. This is caused by reading the signal in any one direction (in FIG. 2, it is an example of reading in the lower rear stage). At this time, the shading for each color will be different due to the difference in the wiring arranged around the PD.

FIG. 13 is a diagram showing an outline of the configuration of a photoelectric conversion element that prevents a difference in shading for each color of received light. As shown in FIG. 13, in the photoelectric conversion element, the dummy read wiring shown by the broken line is also formed between the pixels of Red, so that the read wiring arranged between PDs for each color is the same. Has been made.

For example, in the configuration shown in FIG. 13, the Red read wiring originally requires only the wiring portion below PIX_BLK in order to read the signal to the subsequent stage (not shown) located at the lower part of the figure, but is shown in FIG. The wiring shown by the broken line is extended with respect to the configuration. The same applies to Green pixels. Therefore, the read wiring of R and G is arranged between PD and PD for each color, and there is no difference in the configuration for each color. As described above, in the photoelectric conversion element, the wiring arranged between the pixels for each color is formed in the same manner for each color, so that the shading difference for each color is suppressed. Although wiring other than the read wiring (R, G, B) necessary for explanation is not shown in FIG. 13, the power supply, GND, and various control lines are actually arranged.

Next, a configuration for suppressing crosstalk between signal lines will be described. In the configuration shown in FIG. 11, many wirings are arranged in the lower M1 layer in order to increase the opening in the upper M2 layer. At this time, since a plurality of wirings are arranged in the M1 layer, crosstalk between the wirings may become a problem. In particular, as shown in FIG. 13, a signal reading line from PIX_BLK is arranged between PD and PD, but the distance between wirings is generally set to the minimum width so as not to sacrifice the opening width. It is a target. Therefore, the influence of crosstalk on the output image cannot be ignored.

FIG. 14 is a diagram showing a read wiring from a pixel and a first configuration example for suppressing crosstalk in the read wiring. As shown in FIG. 14A, the image signal output by the PD is buffered with respect to the signal read line and output. That is, the photoelectric conversion element is configured to read the buffered image signal. Specifically, as shown in FIG. 14A, the photoelectric conversion element has a source follower (SF) configured in PIX_BLK, and reads a signal from the output of this SF to a read line (R, G). It is said to be composed.

FIG. 14 (b) shows how the signal changes in the configuration shown in FIG. 14 (a). As shown in FIG. 14 (b), when the signal level of the peripheral wiring (here R) changes with respect to the read line (here G), the change is superimposed on G via the parasitic capacitance between the wirings. Be done (crosstalk). However, the signal is returned to the original level by the SF connected to the read wiring G, and is eventually returned to the original level of G. This is interpreted as a faster (returning) response of the signal on the read line due to the low impedance of the SF. In FIG. 14B, the dotted line shows the state of the signal change when there is no SF, and the signal has not returned to the original level as compared with the case where there is SF. In this case, it causes intercolor crosstalk between RG and appears as mixed colors or false colors on the image.

As described above, since the photoelectric conversion element is configured to output the buffered signal to the read wiring between the pixels, the influence of crosstalk is reduced. Since the photoelectric conversion element, which is a CMOS linear image sensor, reads RGB (all pixels) independently for each color, the read wiring is configured for each RGB (pixel). Therefore, there are a plurality of wirings arranged between the pixels, and there are adjacent wirings, so that crosstalk can be a problem. On the other hand, in the area sensor, column parallel processing (column processing) is common, and since the signal of the pixel is read out for each row, the wiring arranged between the pixels is single. That is, the above-mentioned crosstalk (particularly crosstalk between colors) is a problem peculiar to the linear sensor.

Further, even with the configuration shown in FIG. 14, considering that the responsiveness of SF is finite, in the case of operating PIX_BLK at high speed, simply reducing the impedance of the read line does not necessarily reduce crosstalk. It may not be enough. FIG. 15 is a diagram showing a second configuration example of suppressing crosstalk in the read wiring from the pixel and the read wiring. As shown in FIG. 15, in the second configuration example for suppressing crosstalk of the photoelectric conversion element, the GND is sandwiched between an arbitrary read line and another signal line (FIG. 15 (a)). , (B)).

In the configuration shown in FIGS. 15A and 15B, the GND functions as a shield when viewed from the reading line. At this time, as shown in FIG. 15 (c), when the signal level of the peripheral wiring (here R) changes with respect to the read line (here G), the change to GND via the parasitic capacitance between the wirings. However, since GND is a low impedance line, the change is suppressed, and the change cannot be seen in GND (it is AC grounded). Therefore, even if a parasitic capacitance exists between the read line G and the GND, the change in the read line R is not transmitted to G. In other words, the influence of crosstalk will not be transmitted in the first place. The dotted line in FIG. 15 (c) shows the same state as described in FIG. 14 (b).

Although the configuration shown in FIG. 15 shows an example in which the GND is sandwiched between the signal lines, the same effect can be obtained by sandwiching the power supply between the signal lines. Further, the wiring arranged in the M2 layer of FIG. 15B, which is not shown in FIG. 15A, can be assigned an arbitrary signal.

FIG. 16 is a diagram showing a reading wiring from a pixel and a third configuration example for suppressing crosstalk in the reading wiring. In the configuration shown in FIG. 16, crosstalk is suppressed without providing a dedicated wiring for suppressing crosstalk. The CMOS linear image sensor may have a column configuration in which a plurality of pixels are processed by one processing circuit in order to speed up the operation. In this case, as for the signal of the pixel and the pixel circuit (PIX_BLK), the RGB signal is simultaneously read out to the column circuit, but after that, the signal is processed one pixel at a time in chronological order (FIG. 16A). At this time, a source follower (SF2) is configured in the column circuit (COL), and a signal is output to the subsequent stage by SF2, but a latter stage circuit (see FIG. 17) such as PGA or ADC is arranged on the other side of the pixel. In the case of the circuit layout, the output signal line of SF2 must straddle the pixels.

In the configuration shown in FIGS. 16A and 16B, the signal line of SF2 described above is arranged instead of the GND that suppresses the crosstalk shown in FIG. 15A. Since the output of SF2 is a buffered signal, it becomes a low impedance signal line like the GND shown in FIG. That is, the configuration shown in FIG. 16 can suppress crosstalk between the read wirings RG in the same manner as the configuration shown in FIG.

Here, although the signal line of SF2 has a low impedance, it is a signal line unlike the power supply / GND, so that the signal change in SF2 is superimposed on the read wiring R and G (causing crosstalk). obtain). However, the period during which the read wiring (R, G) is valid and the period during which the SF2 signal line is valid are different without overlapping. Even if the crosstalk of the signal line of SF2 is superimposed on the read wiring R and G sides, the valid period of the pixel signal has expired, so that the valid pixel signal is not affected at all. On the contrary, even if the crosstalk of the read wirings R and G is superimposed on the signal line side of the SF2, the effective period of the output signal of the SF2 has expired, so that it has no effect in the same manner. That is, since the effective timings of the read wiring and the operation in SF2 are different, the influence thereof does not matter even if there is crosstalk in each. As a result, crosstalk is suppressed by the signal line of SF2 as well as the GND shown in FIG. 15 (a).

In this way, for example, in a CMOS linear image sensor with a column configuration, even if a dedicated power supply / GND is not provided, the originally existing low impedance signal lines such as the output of SF2 are arranged between the read lines. , It is possible to suppress crosstalk.

FIG. 17 is a diagram illustrating the overall configuration of the CMOS linear image sensor (photoelectric conversion element) 40 to which the above configuration is applied. Each of PIX (R) 20, PIX (G) 22, and PIX (B) 24 has about 7,000 PDs and is configured for each RGB color. Further, PIX_BLK (R) 21, PIX_BLK (G) 23, and PIX_BLK (B) 25 each have about 7,000 pixel circuits (PIX_BLK) and are configured for each RGB color. That is, each PD is provided with a pixel circuit (PIX_BLK).

Each pixel circuit (PIX_BLK) converts the charge accumulated by the PD into a voltage signal, and outputs the signal to the analog memory through the read line. The PIX_BLK includes a transfer transistor that transfers the charge of the PD to the FD, a reset transistor that resets the FD, and a source follower transistor that buffers the FD voltage and outputs it to the read line. Unlike the area sensor, the linear sensor reads the signal independently from each pixel of RGB, so that the read line exists independently for each pixel.

The AMEM 26 has, for example, about 7,000 analog memories for each RGB color, holds a signal for each pixel, and sequentially outputs an image signal in column units. By holding the signal by the AMEM 26, a global shutter system is realized in which the operation timings of PIX and PIX_BLK, that is, the exposure timings are simultaneous in RGB. The SF2 described above with reference to FIG. 16 is configured on the output side of, for example, AMEM26.

The ADC 27 has the same number of AD converters as the number of columns, and sequentially AD-converts image signals in column units. The ADC 27 has the same number of AD converters as the number of columns and performs parallel processing, thereby realizing high speed as a photoelectric conversion element while suppressing the operating speed of the AD converter.

The signal AD-converted by the ADC 27 is held for each pixel by the parallel-serial converter (P / S) 28, and the held signal is sequentially output to the LVDS 29. The photoelectric conversion element 40 processes parallel data processed in parallel for each pixel in the main scanning direction on the upstream side of the P / S 28, but processes serial data for each RGB color on the downstream side from the P / S 28. .. The signal output by the P / S 28 is converted by the LVDS 29 into a low voltage differential serial signal and output to the subsequent stage. The timing control unit (TG) 30 controls each unit constituting the photoelectric conversion element 40.

Next, an image forming apparatus including an image reading apparatus having a photoelectric conversion element according to the embodiment will be described. FIG. 18 is a diagram showing an outline of an image forming apparatus 50 including an image reading apparatus 60 having, for example, a photoelectric conversion element 40. The image forming apparatus 50 is, for example, a copying machine or an MFP (Multifunction Peripheral) having an image reading device 60 and an image forming unit 70.

The image reader 60 includes, for example, a photoelectric conversion element 40, an LED driver (LED_DRV) 600, and an LED 602. The LED driver 600 drives the LED 602 in synchronization with a line synchronization signal or the like output by the timing control unit (TG) 30. The LED 602 irradiates the original with light. The photoelectric conversion element 40 receives the reflected light from the document in synchronization with the line synchronization signal or the like, and a plurality of PDs (not shown) generate electric charges and start storage. Then, the photoelectric conversion element 40 outputs the image data to the image forming unit 70 after performing AD conversion, parallel serial conversion, and the like.

The image forming unit 70 has a processing unit 80 and a printer engine 82, and the processing unit 80 and the printer engine 82 are connected to each other via an interface (I / F) 84.

The processing unit 80 includes an LVDS800, an image processing unit 802, and a CPU 804. The CPU 804 controls each part constituting the image forming apparatus 50 such as the photoelectric conversion element 40. Further, the CPU 804 (or the timing control unit 30) controls each PD to start generating electric charges according to the amount of light received substantially at the same time.

The photoelectric conversion element 40 outputs, for example, image data of an image read by the image reading device 60, a line synchronization signal, a transmission clock, and the like to the LVDS 800. The LVDS800 converts the received image data, the line synchronization signal, the transmission clock, and the like into parallel 10-bit data. The image processing unit 802 performs image processing using the converted 10-bit data, and outputs the image data or the like to the printer engine 82. The printer engine 82 prints using the received image data.

As described above, the photoelectric conversion element according to the embodiment opens when the light passing through the opening of the light receiving element sandwiched between the wirings located on a straight line is vertically incident on the light receiving surface. Since it is formed so as to have a first light-shielding region and a second light-shielding region that are shielded by wirings located on a straight line across the portion, the optical system used for light incident is not limited. It is possible to reduce unevenness in the amount of light received by the pixels.

10 Lens 11,40 Photoelectric conversion element 20 PIX (R)
21 PIX_BLK (R)
22 PIX (G)
23 PIX_BLK (G)
24 PIX (B)
25 PIX_BLK (B)
26 AMEM
27 ADC
28 P / S
29 LVDS
30 TG
50 Image forming device 60 Image reading device 70 Image forming unit PD Light receiving element (photodiode)
M1, M2 wiring PIX_BLK pixel circuit

Japanese Unexamined Patent Publication No. 2002-170944

Claims (9)

Multiple light receiving elements arranged in a straight line for each color of light received by outputting an analog image signal by photoelectric conversion for each pixel, and
Wiring formed in the wiring layer and enabled to be used as at least one of the signal line, the power supply, and the ground used for the light receiving element or the peripheral circuit of the light receiving element.
Equipped with
The light receiving element is
Has a light receiving surface
The light vertically incident on the photoelectric conversion element is incident on the light receiving surface through the opening sandwiched between the wirings located on the straight line and opened without changing the incident angle.
Photoelectric conversion characterized by having a first light-shielding region and a second light-shielding region formed by the light being shielded by the wirings located on the straight line across the opening on the light-receiving surface. element. The light receiving element is
When the light passing through the opening is obliquely incident on the light receiving surface, one of the light-shielding regions is increased according to the incident angle of the incident light, and the other light-shielding region is decreased. The photoelectric conversion element according to claim 1, wherein the photoelectric conversion element is to be used. The wiring is
The first or second claim is characterized in that the opening is formed in a plurality of wiring layers, the opening is formed for each wiring layer, and the opening is wider as the distance from the light receiving element is longer. The photoelectric conversion element described. The peripheral circuit
It has a pixel circuit that buffers and outputs the signal output by the light receiving element.
At least one of the wires
The photoelectric conversion element according to any one of claims 1 to 3, wherein the signal line transmits a signal output by the pixel circuit. The photoelectric conversion element according to claim 4, wherein the wiring enabled as a power supply or ground is formed adjacent to the signal line for transmitting the signal output by the pixel circuit. The photoelectric conversion element according to claim 4, wherein a plurality of signal lines for transmitting signals output by the plurality of pixel circuits are formed so as to be adjacent to each other. The photoelectric conversion element according to any one of claims 1 to 6, wherein the wiring formed between the pixels uniformly has the same configuration for each pixel. An image reading device comprising the photoelectric conversion element according to any one of claims 1 to 7. The image reader according to claim 8 and
An image forming apparatus comprising an image forming portion for forming an image read by the image reading apparatus.

Images