JP4502170B2 - Linear image sensor and manufacturing method thereof - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a linear image sensor and a manufacturing method thereof, and more particularly to a linear image sensor capable of realizing a plurality of modes having different resolutions and a manufacturing method thereof.
[0002]
[Prior art]
2. Description of the Related Art Charge transfer devices such as CCD (Charge Coupled Device) type linear image sensors are used as document reading devices such as scanners, copiers, and facsimile machines. This linear image sensor includes a light receiving unit in which photodiodes are arranged in one direction, a charge reading unit that reads signal charges photoelectrically converted and accumulated by the light receiving unit, and a charge transfer unit that transfers signal charges. In addition, the charge transfer unit is usually composed of a two-layer drive CCD shift register, and has a structure in which a pulse line for supplying two-layer drive pulses (φ1, φ2) is arranged close to the charge transfer unit. ing.
[0003]
Then, the signal charge generated in the light receiving unit is accumulated in the period of LOW level when the pulse (φTG) applied to the charge reading unit is read out to the charge transfer unit in the period of HIGH level, and the driving pulse (φ1, φ2) The signal charges sequentially transferred by the signal are output to the outside by an output circuit including an analog circuit such as a signal charge detection unit that converts the signal charges into a signal voltage, a source follower, and an inverter. Two-dimensional information is acquired by mechanically scanning the linear image sensor in the vertical direction (sub-scanning direction) with respect to the arrangement direction (main scanning direction) of the light receiving units.
[0004]
Here, since the charge transfer device sequentially transfers the signal charges accumulated in the photodiode, if the number of pixels constituting the light receiving unit is increased in order to obtain a high-definition image, the transfer time becomes longer accordingly. End up. Therefore, a method has been proposed in which the linear image sensor is driven in two modes of high resolution and low resolution, and in the case of low resolution, the transfer time is shortened by thinning out pixels. For example, Japanese Patent Laid-Open No. 2001-244448 (hereinafter referred to as “first publication”) discloses a light receiving unit in which a large number of pixels are arranged in one direction, and a first signal charge readout provided on one side of the light receiving unit. High-resolution mode means having a first signal charge transfer unit using a CCD shift register and a second signal charge transfer unit provided on the other side of the light receiving unit and a second signal charge transfer using a CCD shift register A linear image sensor comprising a low resolution mode means having a portion is disclosed.
[0005]
In the first publication, the first and second signal charge transfer units are provided on both sides of a single light receiving unit, and the signal charge transfer unit is selectively used according to the mode to be selected. Is read out to the signal charge transfer unit for each normal pixel, and in the low resolution mode, two adjacent pixels are combined and read out to one charge transfer unit so that the two modes can be supported.
[0006]
However, the method described in the first publication has a problem in that a high-resolution signal and a low-resolution signal cannot be processed at the same time because a single light receiving unit is used in combination between different resolution modes. Further, in the linear image sensor, a staggered structure in which two pixel columns are shifted by half a pixel and arranged in a staggered pattern may be used. However, in the method described in the first publication, the linear image sensor is arranged in a single pixel column. Therefore, there is a problem that it cannot be applied to a staggered structure.
[0007]
In view of this, a multi-mode CCD has been proposed in which a plurality of light receiving portions having different pixel pitches are provided in order to simultaneously process a plurality of resolution signals. For example, Japanese Patent Application Laid-Open No. 2001-169050 (hereinafter referred to as second publication) discloses at least one row of first photosensors each having a first active region and at least one second active region. A photo array sensor including a second photo sensor in a row is disclosed, and two types of modes can be realized simultaneously by changing the size and wavelength sensitivity characteristics of the first active region and the second active region. Yes.
[0008]
[Patent Document 1]
JP 2001-244448 A (page 4-7, FIG. 1)
[Patent Document 2]
JP 2001-169050 A (page 3-5, FIG. 1)
[0009]
[Problems to be solved by the invention]
Here, the structure of a conventional multimode type linear image sensor will be described with reference to FIGS. FIG. 8 is a plan view showing a schematic configuration of a conventional linear image sensor, and FIGS. 9 and 10 are enlarged views of portions A and B of FIG. 8, respectively. As shown in FIGS. 8 to 10, the conventional linear image sensor has a pixel row (in the figure, the upper side is a pixel) composed of a plurality of pixels (photodiode N well 1) having different resolutions (pixel pitches) on the same chip 14. A pixel array for low resolution having a pitch 2a and a pixel array for high resolution having a pixel pitch a on the lower side are formed side by side. Each photodiode N well 1 is P ⁺ Separated by a channel stopper 5, there is a charge readout portion adjacent to the pixel column, and a charge transfer portion 3 is provided at the tip. Further, pulse lines φ1 (φ1 wiring 9) and φ2 (φ2 wiring 10) for driving the charge transfer unit 3 and a pulse line φTG (φTG wiring 2) for controlling the read gate 2 are provided.
[0010]
Here, the potential of the photodiode N well 1 is determined by the amount of photodiode N well implantation and the pattern shape of this portion, and particularly when the width of the photodiode N well implantation region is small, the potential becomes shallow due to the narrow channel effect. Therefore, in the conventional linear image sensor shown in FIGS. 8 to 10, since the pixel pitch is changed in each pixel column in order to change the resolution between the high resolution and the low resolution, the width of the photodiode N well implantation region is set. As a result, the potential becomes shallow in the narrow pixel row (high resolution side pixel row) due to the narrow channel effect.
[0011]
This problem will be described with reference to the figure. In the linear image sensor, when the potential relationship between the photodiode N well 1 and the readout gate 2 is appropriate, the signal charge accumulated in the photodiode N well 1 is caused by φTG applied to the readout gate 2. If the potential relationship is not appropriate, as shown in FIG. 11 (c), a part of the signal charge is not read out and remains in the photodiode N well 1 as shown in FIG. As shown in b), a defect occurs in which the potential of the photodiode N well 1 is too shallow to sufficiently accumulate charges in the photodiode.
[0012]
For this reason, the injection condition of the photodiode N well 1 must be controlled so that the potential becomes an appropriate value. However, a multimode linear image sensor in which pixel columns having different resolutions (pixel pitches) are arranged on one chip. In this case, the photodiode N well 1 must be injected under different conditions for each resolution, that is, for each pixel column. As a result, the number of processes increases.
[0013]
In order to prevent the potential mismatch due to the narrow channel effect, it is necessary to arrange the injection for each resolution and to align the potential of each pixel column. For this purpose, as shown in FIG. PR, implantation, and PR peeling (step S105 to 104) of the photodiode N well (for example, high resolution side) and PR, implantation, and PR peeling (step S105 to step S105) of the second photodiode N well (for example, the low resolution side). 107) must be performed separately, and the process for at least one PR, injection, and PR peeling is increased. This increase in the number of processes is a great demerit in the present situation where the cost reduction of devices is required.
[0014]
The present invention has been made in view of the above problems, and its main purpose is a multi-mode linear image capable of realizing two or more modes having different resolutions without increasing the number of steps. It is to provide a sensor and a manufacturing method thereof.
[0015]
[Means for solving problems]
In order to achieve the above object, a linear image sensor of the present invention includes a light receiving unit in which pixels are arranged in one direction, a read gate that reads signal charges from the pixels, and a charge transfer unit that sequentially transfers the signal charges. Multiple detection units with different resolutions In the plurality of detection units, the pixel pitches of the light receiving units are set to be equal, and depending on the resolution of the detection unit, one or a plurality of adjacent pixels are separated from the pixels. The signal charge movement path to the charge transfer section is formed separately.
[0016]
The linear image sensor according to the present invention includes a light receiving unit in which pixels including a well region are arranged in one direction and a channel stopper is formed at least between the pixels, a read gate for reading signal charges from the pixels, and the signal A linear image sensor having a plurality of detection units each including a charge transfer unit configured to sequentially transfer charges, wherein the pixel pitches of the light receiving units in the plurality of detection units are set to be equal; The signal charge movement path from the pixel to the charge transfer unit is formed separately for each pixel by the channel stopper, and the other detection unit is configured to transfer the charge from a set of a plurality of adjacent pixels. The movement path of the signal charge reaching the part is formed separately for each set of pixels by the channel stopper.
[0017]
The linear image sensor according to the present invention includes a light receiving portion in which pixels including well regions formed in a semiconductor substrate are arranged in one direction and a channel stopper is formed at least between the pixels, and a well region formed in the semiconductor substrate. In a linear image sensor having two detection units of high resolution and low resolution, each including a charge transfer unit including a read gate formed on the semiconductor substrate between the light receiving unit and the charge transfer unit. The pixel pitches of the light receiving units in the two detection units are set to be equal, and the detection unit on the high resolution side has the signal charge moving path from each pixel to the charge transfer unit by the channel stopper. The detection unit on the low resolution side is formed separately for each pixel, and the detection unit on the low resolution side is n (n is an integer of 2 or more) adjacent pixels. Movement path of the signal charge from the set leading to the charge transfer portion, in which are formed separately for each set of said pixels by said channel stopper.
[0018]
In the present invention, a plurality of gate electrodes perpendicular to the arrangement direction of the pixels are provided on the charge transfer portion, and each of the gate electrodes is disposed corresponding to the movement path of the signal charge. Preferably, the gate electrode may be composed of first-phase and second-phase gate electrode pairs that are alternately and repeatedly arranged in the arrangement direction of the pixels.
[0019]
In the present invention, at least one of the plurality of detection units includes the readout gate and the charge transfer unit on both sides of the light receiving unit, and the charge transfer units on both sides of the light receiving unit have output terminals thereof. The signal charges of each of the pixels constituting the light receiving unit are alternately moved to the charge transfer unit on each side, and the signal charges of the adjacent pixels are connected on the output end side. It can be set as the structure by which the said path | route is formed so that it may synthesize | combine.
[0020]
The method for manufacturing a linear image sensor according to the present invention includes a light receiving unit in which pixels including well regions are arranged in one direction and a channel stopper is formed at least between the pixels, and a readout gate for reading signal charges from the pixels. And a charge transfer section for sequentially transferring the signal charges. Multiple detection units with different resolutions In the method for manufacturing a linear image sensor, the signal charges are transferred from the pixels to the charge transfer unit by simultaneously injecting the well formation regions of the plurality of detection units formed at an equal pixel pitch under the same injection condition. The channel stopper is formed so that the path is separated for each pixel or for each set of a plurality of adjacent pixels according to the resolution of each of the detection units.
[0021]
In the method for manufacturing a linear image sensor according to the present invention, a light receiving unit in which pixels including well regions formed in a semiconductor substrate are arranged in one direction and a channel stopper is formed at least between the pixels, and the semiconductor substrate is formed. A linear image sensor having two detection units of high resolution and low resolution including a charge transfer unit including a well region and a read gate formed on the semiconductor substrate between the light receiving unit and the charge transfer unit. In the manufacturing method, a plurality of well formation regions of the detection unit formed at an equal pixel pitch are simultaneously injected under the same injection condition, and the detection unit on the high resolution side receives the charge transfer unit from each of the pixels. The signal charge moving path leading to is separated for each of the pixels, and the low-resolution detection unit includes n (n is an integer of 2 or more) adjacent pixels. As the movement path of the signal charge from the set leading to the charge transfer portion is separated for each set of said pixels, and forms the channel stopper.
[0022]
As described above, in the present invention, instead of changing the pixel shape and the pixel pitch in the plurality of light receiving portions (pixel columns), the pixel column structure is the same, and the readout structure (that is, the separation structure by the channel stopper or the charge transfer) By devising the structure of the part), a plurality of modes with different resolutions are realized. As a result, it is possible to share the photodiode N-well injection conditions for each resolution, to suppress an increase in the number of processes, and to make the potentials of a plurality of pixel columns the same and store them in readout characteristics and photodiodes. By bringing the characteristics of the charge amount close to each other, it is easy to estimate the sensitivity of each resolution, and the effect of facilitating the design of the amplifier and photodiode openings can be obtained.
[0023]
DETAILED DESCRIPTION OF THE INVENTION
As described in the prior art, a structure in which a plurality of pixel columns are provided is effective for simultaneously acquiring information of two modes of high resolution and low resolution, but the resolution is provided on the same chip 14 as shown in FIG. In a linear image sensor having two types of pixel columns with different (pixel pitches), when injection into the photodiode N well 1 is performed under the same conditions, as shown in FIG. There arises a problem that the potential of the well 1 becomes relatively shallow, or a part of the signal charge is not read and a reading failure remaining in the photodiode N well 1 occurs.
[0024]
As a result, the readout characteristics of the photodiode and the amount of signal charge that can be accumulated in the photodiode vary greatly from pixel column to pixel column. Therefore, if the injection condition of the photodiode N well 1 is not set for each pixel column, high resolution / low resolution However, if the implantation conditions are changed for each pixel column, at least one PR, implantation, and PR peeling must be added as shown in FIG. Will lead to price increases.
[0025]
Therefore, the present invention is characterized in that the charge readout method is improved instead of changing the injection region for each pixel column so that it can cope with a plurality of modes without increasing the number of steps. That is, the photodiode N well injection region on the low resolution side is made the same as the photodiode N well injection region on the high resolution side, and the potentials of the photodiode portions of both pixel columns are made the same. In addition, a structure in which signal charges are read out from a plurality of photodiode N wells to one charge transfer unit in the charge readout unit on the low resolution side (for example, a path from the photodiode N well to the charge transfer unit on one pixel on the high resolution side) Every P ⁺ The resolution (number of pixels) is adjusted by separating with a channel stopper and adopting a structure in which the path is separated for each of a plurality of pixels on the low resolution side.
[0026]
As a result, it is possible to reduce the difference in photodiode readout characteristics between resolutions and the amount of signal charge that can be accumulated in the photodiode, and it is possible to manufacture under a single photodiode N-well injection condition. Can be suppressed.
[0027]
【Example】
In order to describe the above-described embodiment of the present invention in more detail, examples of the present invention will be described with reference to the drawings.
[0028]
[Example 1]
First, a linear image sensor and a manufacturing method thereof according to a first embodiment of the present invention will be described with reference to FIGS. 1 to 3, 11, and 12. FIG. 1 is a plan view showing a schematic configuration of a linear image sensor according to a first embodiment of the present invention, and FIGS. 2 and 3 are enlarged views of a portion A and a portion B in FIG. 1, respectively. FIG. 11 is a timing chart showing the timing of each clock (φTG, φ1, φ2) and output signal, and FIG. 12 is a flowchart for explaining the process reduction effect of the present invention.
[0029]
As shown in FIGS. 1 to 3, the linear image sensor of this embodiment is formed by arranging pixel rows composed of a plurality of photodiode N wells 1 arranged at the same pixel pitch on the same chip 14. The photodiode N well 1 is P ⁺ Separated by a channel stopper 5, there is a charge reading portion adjacent to the photodiode N well 1, and a charge transfer portion 3 is formed at the tip. A pulse line φ1 (φ1 wiring 9) for driving the charge transfer unit 3, a pulse line φ2 (φ2 wiring 10), and a pulse line φTG for controlling the reading gate 2 are provided. The signal charge is transferred to the charge transfer N-well 6 by the first-layer polysilicon gate 7 and the second-layer polysilicon gate 8 that supply the read gate 2 and the two-layer drive pulses (φ1, φ2). Thus, in this embodiment, the charge readout portion P ⁺ The pattern of the photodiode N well 1 portion is P except that the channel stopper 5 is arranged to read out from the two photodiode N wells 1 to the single charge transfer unit 3. ⁺ It is the same on the high / low resolution side / low resolution side including the channel stopper 5.
[0030]
Here, in the linear image sensor, the photoelectrically converted electrons are read to the charge transfer unit 3 by the pulse φTG applied to the electrode of the charge reading unit, sequentially transferred by the φ1 and φ2 pulses, and signaled by the signal charge detection unit (not shown). And then output to the outside through the output circuit 4. FIG. 11A shows a timing chart of each clock and output signal. When the voltage applied to the readout gate 2 becomes HIGH level, the photoelectrically converted electrons are read out to the charge transfer unit 3. If the potential of the readout gate 2 at this time is shallower than the potential of the photodiode N well 1, FIG. As shown in c), some signal charges remain without being read out, resulting in read failure. If the potential of the photodiode N well 1 is too shallow, the amount of signal charge that can be accumulated in the photodiode N well 1 is insufficient as shown in FIG.
[0031]
Such a problem occurs when the potential of the photodiode N-well portion fluctuates due to the narrow channel effect when the pixel pitch is different for each pixel column. In this embodiment, the photo on the high resolution side and the low resolution side are used. The diode N well injection region is made the same, and P on the low resolution side ⁺ A channel stopper 5 is disposed up to the charge transfer section 3 for each set of adjacent pixels (every two pixels in this case), and the first and second layer polysilicon gates 7 and 8 are provided corresponding to the set of pixels. Since the charges of the plurality of photodiode N wells 1 are synthesized by forming, the potentials of the photodiodes of the pixel columns corresponding to both resolutions can be made the same under a single ion implantation condition, As a result, an increase in the number of processes can be suppressed. This process reduction effect will be described with reference to the flowchart of FIG.
[0032]
Conventionally, as shown in FIG. 12A, after performing P-well implantation (step S101) on the N-type semiconductor substrate 13, first, either one (for example, high-resolution side) photodiode N-well implantation is performed. (Steps S102 to 104) are performed, and then the other (in this case, low resolution side) photodiode N well implantation (Steps S105 to 107) is performed. As shown in (b), since the photodiode N well implantation of both pixel columns can be performed simultaneously in steps S202 to 204, the number of implantation steps of the photodiode N well 1 can be halved. It is not necessary to create a mask for each column.
[0033]
In the present invention, the signals of the photodiodes divided into two parts are not combined on the circuit, but are combined into one at the time of reading from the photodiodes, so that it is not necessary to adjust the timing of pixel addition. Furthermore, since the pixel column patterns corresponding to the respective resolutions are aligned, it is possible to easily estimate the sensitivity of each resolution, and it is possible to easily design the amplifier and the photodiode aperture.
[0034]
[Example 2]
Next, a linear image sensor and a method for manufacturing the same according to a second embodiment of the present invention will be described with reference to FIGS. FIG. 4 is a plan view schematically showing a schematic configuration of the linear image sensor according to the second embodiment of the present invention, and FIG. 5 is an enlarged view of a portion A (low resolution side) of FIG. .
[0035]
In the first embodiment, the case where the pixel pitch on the high resolution side is a, the pixel pitch on the low resolution side is 2a, and the photodiode length is b for both resolutions has been described. However, the present invention is not limited to the above embodiment. The present invention is not limited and can be applied to the case where the pixel pitch on the low resolution side is 3a, 4a... Na (n is a positive number), that is, the ratio of both resolutions is an integral multiple. is there.
[0036]
In that case, P is adjusted to adjust the number of pixels. ⁺ A layout in which the channel stopper 5 is combined into one at the time of reading from the n-divided photodiode N well 1 (for example, P for each set of three adjacent photodiode N wells 1 ⁺ Channel stoppers 5 are arranged up to the charge transfer unit 3 and the signal charges of the three photodiode N wells 1 are transferred to the charge transfer N well 6 under one polysilicon gate (here, the first layer polysilicon gate 7). To be transferred). As an example, the configuration in the case of n = 3 is shown in FIGS.
[0037]
Even with such a configuration, the distance between the photodiode N wells 1 does not change between the pixel columns, so that the implantation can be performed under the same conditions for the pixel columns on the high resolution side and the low resolution side, thereby reducing the number of implantation steps. can do. In addition, since signal charges are combined into one when reading from the photodiode N well 1, it is not necessary to adjust the timing of pixel addition.
[0038]
In the above description, the case where two types of photodiode arrays having different resolutions are arranged on one chip has been described. However, three or more types of photodiode arrays are provided on one chip, and the resolution is the same. The present invention can also be applied to cases where all are integral multiples. For example, a linear image sensor in which photodiode rows having pixel pitches a, 2a, and 3a are present in one chip has a structure in which FIGS. 2, 5, and 10 are combined.
[0039]
[Example 3]
Next, a linear image sensor and a method for manufacturing the same according to a third embodiment of the present invention will be described with reference to FIGS. FIG. 6 is a plan view showing a schematic configuration of a linear image sensor according to the third embodiment of the present invention, and FIG. 7 is an enlarged view of a portion A (low resolution side) of FIG.
[0040]
In the first and second embodiments described above, the readout on the low resolution side is made the same direction for all the pixels. However, the charge transfer units 3 are provided on both sides of the photodiode row, and the readout is alternately performed on both sides every other pixel. For example, the same effect can be obtained even if two pixels are combined at the end of the charge transfer unit 3 by adjusting the path length of the charge transfer units 3 on both sides and the timing of the pulses φ1, φ2, and φTG. 6 and 7 show examples of the configuration.
[0041]
In each of the above embodiments, the case where the photodiode length b is the same for both resolutions has been described. However, if the photodiode length b is sufficiently larger than the pixel pitch a, the photodiode length b may differ between the resolutions. The potential modulation is small, and the present invention is applicable. In each of the above embodiments, the photodiode N wells 1 in different pixel columns are arranged at the same position, but the present invention can also be applied to a staggered structure in which they are shifted by half a pixel and arranged in a staggered manner.
[0042]
【The invention's effect】
As described above, the linear image sensor and the manufacturing method thereof according to the present invention have the following effects.
[0043]
The first effect of the present invention is that even if a plurality of pixel columns having different resolutions are arranged on one chip, it is not necessary to individually set the photodiode N well implantation conditions for each resolution, and the increase in the number of processes is suppressed. Is that you can.
[0044]
The reason for this is that, conventionally, when there are two types of pixel columns having different resolutions in one chip, two types of injection conditions for the photodiode N well are also required. This is because it is possible to perform injection under a single condition.
[0045]
In addition, the second effect of the present invention is that it is easy to estimate the sensitivity of each resolution, and it is easy to design an amplifier or a photodiode aperture.
[0046]
The reason is that in the linear image sensor having such a structure, the sensitivity standard is generally set for each resolution. However, in the present invention, since the pattern of the photodiode portion is made uniform at each resolution, the resolution of each resolution is set. This is because the sensitivity can be estimated accurately.
[Brief description of the drawings]
FIG. 1 is a plan view showing a schematic configuration of a linear image sensor according to a first embodiment of the present invention.
FIG. 2 is a plan view showing a configuration of a low resolution side of the linear image sensor according to the first embodiment of the present invention.
FIG. 3 is a plan view showing a configuration on a high resolution side of the linear image sensor according to the first embodiment of the present invention.
FIG. 4 is a plan view showing a schematic configuration of a linear image sensor according to a second embodiment of the present invention.
FIG. 5 is a plan view showing a configuration of a low resolution side of a linear image sensor according to a second embodiment of the present invention.
FIG. 6 is a plan view showing a schematic configuration of a linear image sensor according to a third embodiment of the present invention.
FIG. 7 is a plan view showing a configuration on a low resolution side of a linear image sensor according to a third embodiment of the present invention.
FIG. 8 is a plan view showing a schematic configuration of a conventional linear image sensor.
FIG. 9 is a plan view showing a configuration on a low resolution side of a conventional linear image sensor.
FIG. 10 is a plan view showing a configuration on a high resolution side of a conventional linear image sensor.
FIG. 11 is a timing chart showing timings of clocks (φTG, φ1, φ2) and output signals.
FIG. 12 is a flowchart for comparing a conventional linear image sensor manufacturing method and a linear image sensor manufacturing method of the present invention.
[Explanation of symbols]
1 Photodiode N-well
2 Read gate (φTG wiring)
3 Charge transfer section
4 Output circuit
5P ⁺ Channel stopper
6 Charge transfer N well
7 First layer polysilicon gate
8 Second layer polysilicon gate
9 φ1 wiring
10 φ2 wiring
11 Contact
12 Pwell
13 N-type semiconductor substrate
14 chips

Claims (10)

A linear image sensor having a plurality of detection units having different resolutions, each including a light receiving unit in which pixels are arranged in one direction, a readout gate that reads signal charges from the pixels, and a charge transfer unit that sequentially transfers the signal charges. And
In the plurality of detection units,
The pixel pitch of the light receiving part is set equal,
A linear movement path of the signal charge from the pixel to the charge transfer unit is formed separately for each one or a plurality of adjacent pixels according to the resolution of the detection unit. Image sensor. A light receiving unit in which pixels including a well region are arranged in one direction and a channel stopper is formed at least between the pixels; a read gate that reads signal charges from the pixels; and a charge transfer unit that sequentially transfers the signal charges. A linear image sensor having a plurality of detection units comprising:
The pixel pitches of the light receiving units in the plurality of detection units are set equal,
In one of the detection units, the signal charge movement path from each pixel to the charge transfer unit is formed separately for each pixel by the channel stopper,
In the other detection unit, a movement path of the signal charge from a set of a plurality of adjacent pixels to the charge transfer unit is formed separately for each set of pixels by the channel stopper. Characteristic linear image sensor. A light receiving unit in which pixels each including a well region formed in a semiconductor substrate are arranged in one direction and a channel stopper is formed at least between the pixels, a charge transfer unit including a well region formed in the semiconductor substrate, and the light receiving unit In a linear image sensor having two high-resolution and low-resolution detection units, each of which includes a readout gate formed on the semiconductor substrate between the first transfer unit and the charge transfer unit.
The pixel pitches of the light receiving parts in the two detection parts are set equal,
In the detection unit on the high resolution side, the signal charge moving path from each pixel to the charge transfer unit is formed separately for each pixel by the channel stopper,
The detection unit on the low resolution side is configured such that a movement path of the signal charge from a set of n (n is an integer of 2 or more) adjacent pixels to the charge transfer unit is connected to the pixel by the channel stopper. A linear image sensor characterized by being formed separately for each set.   A plurality of gate electrodes orthogonal to the arrangement direction of the pixels are provided on the charge transfer section, and each of the gate electrodes is disposed corresponding to the movement path of the signal charge. Item 4. The linear image sensor according to any one of Items 1 to 3.   5. The linear image sensor according to claim 4, wherein the gate electrode includes first-phase and second-phase gate electrode pairs alternately and repeatedly arranged in the arrangement direction of the pixels. At least one of the plurality of detection units includes the readout gate and the charge transfer unit on both sides of the light receiving unit,
The charge transfer units on both sides of the light receiving unit are coupled on the output end side thereof,
The signal charges of each of the pixels constituting the light receiving section are alternately moved to the charge transfer section on each side, and the signal charges of adjacent pixels are combined on the output end side. The linear image sensor according to claim 1, wherein the path is formed. A light receiving unit in which pixels including a well region are arranged in one direction and a channel stopper is formed at least between the pixels; a read gate that reads signal charges from the pixels; and a charge transfer unit that sequentially transfers the signal charges. In a method for manufacturing a linear image sensor having a plurality of detection units having different resolutions ,
Implanting simultaneously in the same implantation conditions to the well formation regions of the plurality of detection units formed with equal pixel pitch,
The channel so that a movement path of the signal charge from the pixel to the charge transfer unit is separated for each pixel or for each set of a plurality of adjacent pixels according to the resolution of each detection unit. A method of manufacturing a linear image sensor, comprising forming a stopper. A light receiving unit in which pixels each including a well region formed in a semiconductor substrate are arranged in one direction and a channel stopper is formed at least between the pixels, a charge transfer unit including a well region formed in the semiconductor substrate, and the light receiving unit In a method for manufacturing a linear image sensor having two detection units of high resolution and low resolution, each of which includes a readout gate formed on the semiconductor substrate between the charge transfer unit and the charge transfer unit.
Implanting simultaneously in the same implantation conditions to the well formation regions of the plurality of detection units formed with equal pixel pitch,
The detection unit on the high resolution side separates the signal charge movement path from each pixel to the charge transfer unit for each pixel, and the low resolution detection unit has n (n is an integer of 2 or more). In the linear image sensor, the channel stopper is formed so that a movement path of the signal charge from the set of adjacent pixels to the charge transfer unit is separated for each set of pixels. Production method.   9. The linear device according to claim 7, wherein a plurality of gate electrodes provided so as to be orthogonal to an arrangement direction of the pixels on the charge transfer portion are formed corresponding to the movement path of the signal charges. A method for manufacturing an image sensor.   The gate electrode includes first-phase and second-phase gate electrode pairs that are alternately and repeatedly arranged in the pixel arrangement direction, and each of the gate electrode pairs corresponds to the movement path of the signal charge. 9. The method for manufacturing a linear image sensor according to claim 7, wherein the linear image sensor is formed.

Images