JP2010153603A - Solid state imaging apparatus - Google Patents

Abstract

Provided is a solid-state imaging device with improved sensitivity while suppressing smear.
A solid-state imaging device includes a photodiode (PD) 12, a VCCD 13, an element isolation region 15, a light shielding film 41, and an opening 18. A plurality of PDs 12 are formed in a honeycomb arrangement on the surface of the semiconductor substrate. The VCCD 13 is provided in every two rows along the DP 12 row, and transfers the charges accumulated in the left and right two rows of PDs 12. The element isolation region 15 is provided between the adjacent PD12 rows without passing through the VCCD 13 every two rows along the PD12 row, and separates the adjacent PD12 into each of them, thereby transferring charges between the PD12. prevent. The light shielding film 41 prevents light from entering the VCCD 13. The opening 18 is provided in the light shielding film 41 so that each PD 12 is exposed, and the distance from the VCCD 13 is equal to or greater than a predetermined value P, and the shortest distance Q via the element isolation region passes through the VCCD 13. Smaller than the shortest interval R.
[Selection] Figure 3

Description

  The present invention relates to a solid-state imaging device using a CCD in a transfer path of charges accumulated in a photodiode, and more specifically, one VCCD is provided between two adjacent photodiodes, and adjacent to each other with the VCCD interposed therebetween. The present invention relates to a solid-state imaging device in which two rows of photodiodes share a VCCD provided therebetween for transfer of accumulated charges.

  An image sensor (solid-state imaging device) that captures an image of a subject by photoelectrically converting light from the subject with a photodiode formed on a semiconductor substrate has become widespread. As an image sensor, a CCD type image sensor or a CMOS type image sensor is known in accordance with a method for acquiring a signal from charges accumulated in a photodiode.

  The CCD type image sensor directly transfers a signal charge itself from a photodiode by a CCD and acquires an image signal. Such CCD image sensors are further classified into several types depending on the structure and transfer method. For example, there are photodiode columns and charge transfer paths (so-called VCCDs) that read and transfer signal charges from each photodiode. An interline transfer type CCD image sensor arranged alternately is known.

  In this interline transfer type CCD image sensor, a charge transfer path (so-called HCCD) for transferring a signal charge transferred by the VCCD in a direction perpendicular to the VCCD is below the light receiving surface on which the photodiodes are arranged. It is provided in connection with. The HCCD is pulse-driven at a high frequency of about several tens of MHz in order to improve the frame rate. For this reason, in an interline transfer type CCD image sensor, power consumption by driving the HCCD is extremely large.

  When a signal is read out at a constant frame rate, the frequency of the HCCD drive pulse is determined according to the number of VCCDs connected to the HCCD. For this reason, the number of VCCDs connected to the HCCD is reduced by thinning out the VCCDs usually provided for each column of photodiodes so that two rows of photodiodes share one VCCD. Image sensors with reduced power consumption are known (Patent Documents 1 and 2).

In recent years, there has been known an image sensor having a special structure and imaging mode that compensates for the drawbacks of the conventional image sensor. For example, an image sensor is known in which two types of pixels having different sensitivities are provided, and high-sensitivity imaging using high-sensitivity pixels and low-sensitivity imaging using low-sensitivity pixels are performed substantially simultaneously. The high-sensitivity photographed image and the low-sensitivity photographed image acquired here are combined, and an image having a dynamic range expanded beyond the characteristics of the image sensor is acquired. In this image sensor, since high-sensitivity shooting and low-sensitivity shooting must be performed almost simultaneously, it is necessary to drive the HCCD with a pulse of higher frequency, and the power consumption in the HCCD is significantly increased. For this reason, even if photodiodes are included in the same column, signal charges are alternately read out to the two left and right VCCDs, and pixels with different sensitivities are arranged by devising the arrangement of color filters. Even in such a case, an image sensor that can efficiently use these is known (Patent Document 3).
JP 2001-168315 A JP-A-55-160477 JP 2004-55786 A

  In recent years, in order to obtain higher quality images, image sensors are required to have higher pixels. As described above, an image sensor that has two types of pixels and sometimes shoots without using all of the photodiodes at the same time is strongly demanded to increase the number of pixels. In order to meet such demands and to increase the number of pixels of an image sensor of a certain size, naturally, the size of each photodiode is reduced, but a new problem arises due to the reduction in the size of the photodiode.

  In a CCD image sensor, smear occurs when light enters a charge transfer path such as a VCCD. Therefore, a light shielding film is provided so that light does not enter unnecessary portions including the VCCD, and a photodiode is provided on the light shielding film. An opening for exposing at a predetermined size is provided. For this reason, when the image sensor is increased in pixel size, the size of the opening is also reduced in accordance with the miniaturization of the photodiode. However, if the size of the aperture is also miniaturized simultaneously with the miniaturization of the photodiode, there is a problem that the amount of light incident on the photodiode per unit time is reduced and the sensitivity is deteriorated. In particular, when the size of the aperture is about the wavelength of visible light (about several hundred nm to 1 μm), the amount of light incident on the photodiode is remarkably reduced, and the sensitivity is extremely deteriorated.

  Further, in order to keep the sensitivity at a certain level or more while miniaturizing the photodiode, it is necessary to enlarge the size of the opening from the conventional ratio. However, when the size of the opening is enlarged, the distance from the edge of the opening to the VCCD becomes small, so that there is a problem that light easily leaks from the opening to the VCCD and smear is likely to occur.

  The trade-off between sensitivity and smear characteristics as described above is the same for the image sensors described in Patent Documents 1 and 2 as well as the interline transfer type CCD image sensor disclosed in Patent Document 3 and the like. There is a need for improvement.

  The present invention has been made in view of the above-described problems, and an object of the present invention is to provide a solid-state imaging device with improved sensitivity while suppressing occurrence of smear when the number of pixels is increased.

  The solid-state imaging device of the present invention is provided with a plurality of photoelectric conversion elements formed in a predetermined arrangement on the surface of a semiconductor substrate and every two columns along the photoelectric conversion element column, and the photoelectric conversions in two columns on the left and right The charge transfer path for transferring the charge accumulated in the element, and the charge transfer between the adjacent photoelectric conversion element columns without passing through the charge transfer path every two columns along the photoelectric conversion element column An element isolation region provided on the semiconductor substrate and separated to each of the adjacent photoelectric conversion elements to prevent the movement of charges between the photoelectric conversion elements; and to the charge transfer path A light-shielding film that prevents the light from entering, and the light-shielding film so that each of the photoelectric conversion elements is exposed, and the distance between the charge transfer path is a predetermined value or more and the element isolation region The shortest interval through the charge transfer Characterized in that it and a aperture provided smaller than the shortest distance through.

  The opening may expose a plurality of the photoelectric conversion elements adjacent to each other through the element isolation region.

  Further, the opening is provided so that two adjacent photoelectric conversion elements are exposed through the element isolation region.

  Further, the openings are provided so as to expose the two rows of the photoelectric conversion elements between the charge transfer paths, and the light shielding film is provided in a belt shape covering the charge transfer paths.

  Further, a color filter for determining a color of light incident on each of the photoelectric conversion elements is provided in an arrangement in which two adjacent photoelectric conversion elements have the same color.

  In addition, when color filters of the same color corresponding to two adjacent photoelectric conversion elements are used as one unit of color filter, the color filters are arranged in a Bayer array.

  In addition, the color filters are arranged so as to have the same color for two adjacent photoelectric conversion elements via the element isolation region.

  In addition, the color filters are arranged so as to have the same color for two photoelectric conversion elements adjacent to each other through the charge transfer path.

  Further, the photoelectric conversion elements are arranged in such a manner that the photoelectric conversion elements in the same row are arranged at a constant pitch, and the positions of the photoelectric conversion devices are 1 / of the pitch with respect to the photoelectric conversion devices in adjacent rows. The honeycomb arrangement is arranged by being shifted by two.

  In addition, the arrangement of the photoelectric conversion elements is a square lattice arrangement.

  According to the present invention, it is possible to provide a solid-state imaging device with improved sensitivity while suppressing smearing when the number of pixels is increased.

[First Embodiment]
As shown in FIG. 1, an image sensor 11 (solid-state imaging device) is a CCD type image sensor that directly transfers a signal charge accumulated in a photodiode to a bucket relay type to obtain an image signal. (PD) 12, VCCD 13, element isolation region 15, HCCD 16, floating diffusion amplifier (FDA) 17, light shielding film (see FIG. 2), opening 18 and the like.

  The PD 12 is a photoelectric conversion element that converts incident light into signal charges and accumulates them, and is formed on a semiconductor substrate. A plurality of PDs 12 are formed in a row at a constant pitch in the vertical direction (vertical direction in FIG. 1), and a plurality of rows are formed in the horizontal direction (lateral direction in FIG. 1). At this time, the center position of the PD 12 is shifted by ½ of the vertical arrangement pitch with respect to each PD 12 in the adjacent PD row. Therefore, the PD 12 has a honeycomb arrangement in which a square lattice is rotated by 45 degrees. As will be described later, the PD 12 is formed in a region formed between the VCCD 13 and the element isolation region 15 provided in a meandering manner. Further, the incidence of light on the PD 12 is limited to the opening 18 provided in the light shielding film.

  Further, a color filter of any one of red (R), green (G), and blue (B) is laminated on each PD 12. For this reason, each PD 12 selectively photoelectrically converts light of a color transmitted through the color filter according to the color of the stacked color filters, and accumulates signal charges based on the photoelectric conversion. In the image sensor 11, the color filter has a color distribution of R: G: B = 1: 2: 1, and a so-called Bayer array color filter is rotated 45 degrees in accordance with the honeycomb array of the PD 12. It has become.

  The VCCD 13 is a charge transfer path for transferring the signal charge accumulated in the PD 12 in the vertical direction, and is formed on the semiconductor substrate. A plurality of VCCDs 13 are provided along the rows of the PDs 12 in the vertical direction and meandering in the horizontal direction at a constant pitch along the outer periphery of the PDs 12. The VCCD 13 is connected to the PD 12 via the read gate 14, and the signal charge accumulated in the PD 12 is read to the VCCD 13 via the read gate 14. The VCCD 13 is controlled by a four-phase transfer electrode (not shown), and transfers the signal charges read from each PD 12 in the vertical direction. The end of each VCCD 13 is connected to the HCCD 16, and the signal charge transferred in the vertical direction by the VCCD 13 is transferred to the HCCD 16.

  Further, the VCCD 13 is provided every two rows with respect to the row of the PD 12. Therefore, if only the PD12 row and the VCCD 13 are viewed, the VCCD 13 is sandwiched between the two PD12s on the surface of the image sensor 11, and the horizontal direction (the PD12 row, the VCCD13, the PD12 row), (the PD12 The PD 12 and the VCCD 13 are arranged so as to be arranged in the order of column, VCCD 13, PD 12),. For this reason, each VCCD 13 is shared by two right and left PDs 12, and the signal charges accumulated in the corresponding left and right PDs 12 are alternately read to the left and right.

  The element isolation region 15 is a channel stop provided to isolate each PD 12 and prevent a signal charge from moving between adjacent PDs 12 and is formed on a semiconductor substrate. The element isolation region 15 is provided in the vertical direction between the columns of the PD 12 where the VCCD 13 is not provided. For this reason, the element isolation regions 15 are provided every two columns with respect to the columns of the PD 12. In the image sensor 11, the width of the element isolation region 15 is substantially the same as that of the VCCD 13.

  The element isolation region 15 is provided so as to meander in the horizontal direction at a constant pitch along the outer periphery of the PD 12. The meandering pitch of the element isolation region 15 is shifted from the meandering pitch of the VCCD 13 by ½ pitch. Therefore, a plurality of regions surrounded by the element isolation region 15 and the VCCD 13 are formed, and the PD 12 is formed in this region.

  The HCCD 16 is connected to the end of each VCCD 13 and transfers the signal charge transferred from each VCCD 13 to the FDA 17 in the horizontal direction. The FDA 17 converts the signal charge transferred by the HCCD 16 into a voltage signal (imaging signal) and outputs it.

  On the structure constituting the image sensor 11 described above, a light-shielding film (in order to prevent unnecessary light entering the VCCD 13 or the like from entering the VCCD 13 or the like, which causes a smear to cause a smear or the like. 2). The light shielding film is provided with openings 18 having predetermined positions and sizes corresponding to the respective PDs 12. Light passing through the opening 18 is incident on the PD 12.

  A cross section taken along the line XY of the image sensor 11 is shown in FIG. The image sensor 11 is formed from an n-type semiconductor substrate 26. A p-well layer 27 is formed on the surface of the n-type semiconductor substrate 26. Further, an n-type layer 31 constituting the PD 12, an n-type layer 32 constituting the VCCD 13 and a p + layer serving as the element isolation region 15 are formed on the surface layer. 33, p + layer 34 for separating PD 12 and VCCD 13 is formed.

  The PD 12 includes a junction of the p-well layer 27 and the n-type layer 31. When light enters these junctions, electron-hole pairs are generated and electrons are accumulated in the n-type layer 31. Further, the n-type layer 31 constituting the PD 12 is separated from the n-type layer 31 of the adjacent PD 12 by the p + layer 33 and separated from the n-type layer 32 constituting the VCCD 13 by the p + layer 34.

  The VCCD 13 includes an n-type layer 32 and a transfer electrode 37 provided above the n-type layer 32. When the signal charge of the PD 12 accumulated in the n-type layer 31 is transferred to the n-type layer 32, the cross-sectional direction (direction perpendicular to the paper surface) is determined according to the voltage applied to the transfer electrode 37 at a predetermined timing. Transferred. The transfer electrode 37 is made of polysilicon.

  The n-type layer 31 constituting the PD 12 and the n-type layer 32 constituting the VCCD 13 are separated by the p-well layer 27. The read gate 14 includes a p-well layer 27 that separates the n-type layer 31 and the n-type layer 32, and a transfer electrode 38 provided above this portion. Therefore, the potential of the p-well layer 27 between the n-type layer 31 and the n-type layer 32 is controlled by the transfer electrode 38, and the n-type layer 31 and the n-type layer 32 are separated from each other during the signal charge accumulation period. Charges are accumulated in the layer 31, and the signal charges accumulated from the n-type layer 31 to the n-type layer 32 are transferred during a period for reading the signal charge from the PD 12. The transfer electrode 38 is made of polysilicon.

  The element isolation region 15 is a potential barrier including a p + layer 33 provided between adjacent n-type layers 31 and prevents signal charges from moving between the n-type layers 31 of adjacent PDs 12.

  As described above, the transfer electrodes 37 and 38 are formed on the surface of the n-type semiconductor substrate 26 on which the PD 12, the VCCD 13, and the element isolation region 15 are formed on the surface layer, and the light shielding film 41 and the planarization layers 42a and 42b. Color filters 43R and 43G, microlenses 44, and the like are formed.

  The light shielding film 41 is formed so as to entirely cover the surface of the n-type semiconductor 26. Further, the light shielding film 41 is provided with an opening 18 for exposing the n-type layer 31. Therefore, the light shielding film 41 is provided to cover the transfer electrodes 37 and 38 and the p + layer 33. Further, the light shielding film 41 is provided so as to protrude above the n-type layer 31 by a predetermined amount from the portion covering the transfer electrodes 37 and 38, and the end on the p + layer 33 side is connected to the p + layer 33 and n. It is a boundary of the mold layer 31. For this reason, the opening 18 is not provided in the center of the region surrounded by the VCCD 13 and the element isolation region 15, but is extended to the element isolation region 15 side as compared with the case where the opening is provided in the center of this region. It has become a shape.

  The color filters 43R and 43G are formed above each PD 12 via a planarizing layer 42a made of BPSG or the like, and determine the color of light incident on each PD 12. The color filter 43R restricts the color of light incident on the corresponding direct PD 12 to red light, and the color filter 43G limits the color of light incident on the corresponding direct PD 12 to green. In addition to these, the image sensor 11 is also provided with a color filter (not shown) that restricts the color of light incident on the PD 12 to blue in the above-described arrangement. These color filters 43R and 43G and a blue color filter (not shown) are formed in order of color filters of respective colors. At this time, the end of the color filter (color filter 43G in FIG. 2) formed later rides over the end of the color filter (color filter 43R in FIG. 2) formed due to the limitation of the fine processing accuracy. Provided. Such overlapping portions between the color filters prevent light from entering the PD 12 and lead to a decrease in the sensitivity of the PD 12.

  The microlens 44 is provided for each PD 12 and is a lens for efficiently guiding light to the PD 12, and is formed on the uppermost surface of the image sensor 11 through a planarization layer 42 b made of BPSG or the like.

  In the image sensor 11 configured as described above, the opening 18 provided in the light-shielding film 41 for exposing the PD 12 has a position and size that prevents smearing, and a sufficient amount of light is incident on the PD 12. It is the size that is secured.

  As shown in FIG. 3, the shortest interval P between the end of the opening 18 and the VCCD 13 is the minimum interval necessary to prevent light from entering the VCCD 13 and causing smear. For this reason, the end of the opening 18 on the side adjacent to the VCCD 13 is larger than the interval P at any position. Further, the minimum interval Q between the openings 18 adjacent via the element isolation region 15 is smaller than the minimum interval R between the openings 18 adjacent via the VCCD 13. For this reason, the opening 18 is provided at a position biased toward the element isolation region 15 in a region surrounded by the VCCD 13 and the element isolation region 15.

  Further, as shown by a broken line in FIG. 3, the opening of the image sensor 11 is compared with the opening 51 provided at equal intervals from the VCCD 13 and the element isolation region 15 in the center of the region surrounded by the VCCD 13 and the element isolation region 15. The size of 18 is expanded to the element isolation region 15 side.

  As described above, the image sensor 11 is provided with the VCCDs 13 every two rows so that the two rows of PDs 12 share one VCCD 13, and the openings 18 that expose the PD 12 are formed at a predetermined interval P. The openings 18 are provided so that the minimum distance Q between the openings 18 that are separated from the VCCD 13 and that are adjacent via the element isolation region 15 is smaller than the minimum distance R that is adjacent to the opening 18 via the VCCD 13. As a result, the image sensor 11 can increase the amount of incident light on each PD 12 and improve the sensitivity of each PD 12 as compared to the case where the opening 51 is provided while sufficiently preventing the occurrence of smear.

[Second Embodiment]
In the first embodiment described above, the example in which the opening 18 is provided in the light shielding film 41 corresponding to each PD 12 has been described. However, in general, when the size of the opening that exposes the PD 12 is about the wavelength of light (several hundred nm to 1 μm) or less, the light transmission through the opening can be achieved only by providing a light shielding film and the opening. The rate is significantly reduced, and the amount of light incident on the PD 12 may be insufficient. The same applies to the opening 18 that exposes substantially the entire surface of the PD 12 to the extent that smear does not occur as in the first embodiment described above. For this reason, it is preferable that the size of the opening that exposes the PD 12 is sufficiently large with respect to the wavelength of the transmitted light so that more light reaches the PD 12. Hereinafter, as the second embodiment, an example in which an opening having a sufficiently large size with respect to the wavelength of light used for photographing will be described. Note that the description of the same parts as in the first embodiment is omitted.

  For example, as in the image sensor 61 shown in FIG. 4, one opening 62 is provided in common for two adjacent PDs 12 with the element isolation region 15 interposed therebetween. The opening 62 exposes two PDs 12 adjacent to each other via the element isolation region 15 and a part of the element isolation region 15 provided therebetween. Therefore, the substantial size of the PD 12 exposed from the opening 62 is a region surrounded by the element isolation region 15 exposed from the opening 62 and the opening 62 as indicated by a broken line in FIG.

  In the image sensor 11 of the first embodiment, the opening 62 has a shape and a size in which two adjacent openings 18 are connected via the element isolation region 15. Therefore, as in the first embodiment, the shortest interval P between the opening 62 and the VCCD 13 is equal to the interval P in the first embodiment, and a sufficient interval is secured to prevent the occurrence of smear. Also in the image sensor 61, the shortest distance Q between the openings 62 adjacent via the element isolation region 15 is smaller than the shortest distance R between the openings 62 via the VCCD 13.

  As described above, by providing one opening 62 for two PDs 12 adjacent to each other via the element isolation region 15, the size of the opening 62 is sufficiently large with respect to the wavelength of the light transmitted therethrough. . For this reason, the opening 62 does not hinder the incidence of light on the PD 12, a sufficient amount of incident light on the PD 12 is secured, and the sensitivity of the PD 12 is improved.

  In the image sensor 61 described above, the example in which the upper right PD 12 and the lower left PD 12 are connected to form the opening 62 among the PDs 12 adjacent via the element isolation region 15 has been described. The opening 62 may be formed by connecting the openings of the upper left PD 12 and the lower right PD 12. In this case, the same effect as described above can be obtained.

  Further, for example, as in the image sensor 66 shown in FIG. 5, the opening 67 is provided in common to all the PDs 12 adjacent to each other through the element isolation region 15. The openings 67 are provided in a band shape across two rows of PDs 12 that transfer signal charges to different VCCDs 13, and are provided between all of the two rows of PDs 12 sandwiched between the two VCCDs 13 and these PDs 12. A part of the element isolation region 15 is exposed. For this reason, the light shielding film 41 has a belt-like shape meandering in the horizontal direction so as to cover each VCCD 13. Further, the substantial size of the PD 12 exposed from the opening 67 is a region surrounded by the element isolation region 15 exposed from the opening 67 and the opening 67 as shown by a broken line in FIG.

  Further, the opening 67 has a shape and a size in which all the adjacent openings 18 are connected in the vertical direction via the element isolation region 15 in the image sensor 11 of the first embodiment. Therefore, as in the first embodiment, the shortest interval P between the opening 67 and the VCCD 13 is equal to the interval P in the first embodiment, and a sufficient interval is secured to prevent the occurrence of smear. In the image sensor 67, the shortest distance Q between the openings 67 adjacent via the element isolation region 15 is 0, and is smaller than the shortest distance R between the openings 67 via the VCCD 13.

  Thus, by connecting the openings for two rows in the vertical direction to form a band-shaped opening 67, the size of the opening 67 is sufficiently large with respect to the wavelength of light. For this reason, the opening 67 does not hinder the incidence of light on the PD 12, a sufficient amount of incident light on the PD 12 is secured, and the sensitivity of the PD 12 is further improved.

  In the second embodiment described above, the example in which the openings 62 for exposing the two PDs 12 are provided and the example in which the openings 68 for exposing the PDs 12 for two rows are provided have been described. When the size of the opening is such that the size of the opening does not hinder the transmission of light, the number of PDs 12 exposed from one opening may be arbitrarily determined to be 2 or more. For example, an opening may be provided so that three (4,...) PDs 12 are exposed from one opening.

[Third Embodiment]
In the second embodiment described above, in the first embodiment (and the second embodiment), the color filter provided corresponding to each PD 12 is obtained by rotating the Bayer array by 45 degrees according to the array of the PD 12. However, the arrangement of the color filters is not limited to this. In the following, as the third embodiment, an example will be described in which the color filter arrangement is particularly suitable in the image sensors 11, 62, and 67 of the first and second embodiments described above. In addition, description is abbreviate | omitted about the part similar to the image sensor 11 of 1st Embodiment.

  For example, like the image sensor 71 shown in FIG. 6, it is preferable to provide the color filter 72 so that two PDs 12 adjacent to each other through the element isolation region 15 have the same color. The arrangement of the color filters 72 is an arrangement that becomes a Bayer arrangement when the adjacent color filters of the same color are regarded as one unit, as partially shown by a frame and hatching in FIG.

  The color filter 72 is provided in one step for each of R, G, and B colors. For this reason, there is no overlapping portion of the color filters as described with reference to FIG. 2 between the PDs 12 adjacent to each other via the element isolation region 15 and having the same color filter formed therein, thus forming a flat single color filter. Yes. In particular, in the arrangement of the color filters 72 of the image sensor 71, the PDs 12 on which the green color filter indicated by G is formed are arranged in a line in the direction of 45 degrees oblique to the image sensor 71. For this reason, the PDs 12 on which the green color filter is formed are all connected in an oblique direction, and are flat color filters without overlapping portions. Therefore, in FIG. 6, in order to explain the arrangement, the PD 12 on which the green color filter is formed also has a frame drawn with two PDs 12 as a unit, but the green color filter has a short frame. There is no overlap between the color filters along the side direction.

  In this way, by using the same color for the color filters provided in two adjacent PDs 12 via the element isolation region 15, the number of overlapping portions between different color filters that are difficult to avoid in terms of processing accuracy can be reduced. Can do. For this reason, arranging the color filters as in the image sensor 71 suppresses the decrease in the sensitivity of the PD 12 due to the overlap between the different color filters, and further improves the sensitivity of the PD 12 compared to the arrangement of the other color filters. Can do. In particular, when the opening for exposing the PD 12 is expanded as in the image sensors 11, 62, and 67 of the first and second embodiments, the overlapping portion between the different color filters is close to the opening and is easily affected. . For this reason, the sensitivity of the PD 12 is further improved by applying the arrangement of the color filters 72 as in the image sensor 71.

  Further, for example, as in the image sensor 76 shown in FIG. 7, it is preferable to provide a color filter 77 so that two adjacent PDs 12 have the same color via the VCCD 13. The arrangement of the color filters 77 is an array that becomes a Bayer arrangement when the adjacent color filters of the same color are regarded as one unit, as partially shown by a frame and hatching in FIG.

  According to the arrangement of the color filters 77, similarly to the color filter 72 (FIG. 6) of the image sensor 71 described above, a decrease in sensitivity of the PD 12 due to overlapping between different color filters is suppressed, and the sensitivity of the PD 12 is further improved. be able to.

  Furthermore, the image sensor 76 to which the arrangement of the color filters 77 is applied can easily realize high-sensitivity driving that doubles the sensitivity as compared with the case where the signal charges are read out from all the PDs 12 and imaged.

  Further, the image sensor 76 to which the arrangement of the color filters 77 is applied can easily add the signal charges from the PD 12 adjacent to each other via the VCCD 13 and having the same color filter formed therein. For this reason, the image sensor 76 can easily realize high-sensitivity driving that doubles the sensitivity as compared with the case where the signal charges are read independently from all the PDs 12 and imaged. Further, in the case of high-sensitivity driving in this way, the image sensor 76 can add the signal charges from the two PDs 12 by using the reading operation of the signal charges from the respective PDs 12, so that it is necessary to add the signal charges. The time is short. As a result, the image sensor 76 can improve the frame rate when driven with high sensitivity as compared with the case where it is driven with high sensitivity using the HCCD 16 or the line memory. The influence can be suppressed.

  In the first to third embodiments described above, the image sensor in which the PDs 12 are arranged in the honeycomb arrangement has been described as an example. However, the present invention is preferably used also in an image sensor in which the PDs are arranged in a square lattice pattern. Can do.

  As shown in FIG. 8, in the image sensor 91 in which the PDs 12 are arranged in a square lattice pattern and the VCCDs 13 are provided every two rows of the PDs 12, the shortest distance P to the VCCD 13 is such that smear does not occur. A rectangular opening 92 is provided corresponding to each PD 12 while maintaining a predetermined interval or more. At this time, the shortest interval Q between the openings 92 adjacent to each other in the horizontal direction through the element isolation region 15 is smaller than the interval R between the openings 92 adjacent to each other in the horizontal direction via the VCCD 13. Provide. Even when the arrangement of the PDs 12 is a square lattice, by providing the openings 92 in this way, the same effects as those of the first embodiment described above can be obtained.

  Further, by connecting a plurality of adjacent openings 92 with the image sensor 91, the same effect as in the second embodiment described above can be obtained. When the plurality of adjacent openings 92 are thus connected by the image sensor 91, the openings 92 may be connected in the horizontal direction or the openings 92 may be connected in the vertical direction if the VCCD 13 is not exposed. . Further, in correspondence with the image sensor 68 of FIG. 5, an opening may be provided in which the light shielding film covers only the VCCD 13 in a band shape and the openings 92 corresponding to all the PDs 12 between the VCCDs 13 are connected.

  Further, in the image sensor 91 in which the arrangement of the PDs 12 is a square lattice, it is preferable that the arrangement of color filters is the same as that in the third embodiment. For example, as in the third embodiment described above, it is preferable to arrange the color filters so that the color filters provided in the PDs 12 adjacent to each other through the element isolation region 15 have the same color. Further, when the image sensor 91 is driven with high sensitivity, it is preferable that the color filters are arranged so that the color filters provided in the adjacent PDs 12 via the VCCD 13 have the same color.

  In the first to third embodiments described above, the example using the primary color filters of three colors of red (R), green (G), and blue (B) has been described. Other known color filters such as a complementary color filter composed of magenta and yellow, a color filter using white together, and the like may be used.

  In the first to third embodiments described above, the example in which the width of the element isolation region 15 is set to be substantially the same as that of the VCCD 13 has been described. However, the width of the element isolation region 15 increases the area where the PD 12 is exposed. Therefore, it is preferable that the adjacent PDs 12 be as small as possible within the range in which the adjacent PDs 12 can be accurately separated, and at least smaller than the width of the VCCD 13.

It is explanatory drawing which shows the structure of an image sensor roughly. It is sectional drawing which shows the layer structure of an image sensor typically. It is explanatory drawing which shows the aspect of opening. It is explanatory drawing which shows the example which connects opening of adjacent PD through an element isolation region. It is explanatory drawing which shows the example which connects opening of adjacent PD via VCCD. It is explanatory drawing which shows the example which provides the color filter of the same color on PD adjacent via an element isolation area | region. It is explanatory drawing which shows the example which provides the color filter of the same color on PD adjacent via VCCD. It is explanatory drawing which shows the example of the image sensor by which PD was arranged in square lattice shape.

Explanation of symbols

11, 62, 67, 71, 76, 91 Image sensor (solid-state imaging device)
12 Photodiode (photoelectric conversion element)
13 VCCD
14 readout gate 15 element isolation region 16 HCCD
17 FDA
18, 62, 68, 92 opening 26 n-type semiconductor substrate 27 p-well layer 31, 32 n-type layer 33, 34 p + layer 37, 38 transfer electrode 41 light-shielding film 42a, 42b planarization layer 43R, 43G, 72, 77 color Filter 44 Micro lens

Claims (10)

A plurality of photoelectric conversion elements formed in a predetermined arrangement on the surface of the semiconductor substrate;
A charge transfer path that is provided every two rows along the row of photoelectric conversion elements and transfers charges accumulated in the two rows of photoelectric conversion elements;
The photoelectric conversion elements adjacent to each other along the row of the photoelectric conversion elements are provided with a width equal to or smaller than the charge transfer path between the adjacent columns of the photoelectric conversion elements without passing through the charge transfer path. Element separation regions that prevent charge movement between the photoelectric conversion elements,
A light-shielding film provided on the semiconductor substrate and preventing light from entering the charge transfer path;
Provided in the light shielding film so that each of the photoelectric conversion elements is exposed, and the distance from the charge transfer path is not less than a predetermined value, and the shortest distance through the element isolation region is the charge An opening provided smaller than the shortest interval through the transfer path;
A solid-state imaging device comprising:   The solid-state imaging device according to claim 1, wherein the opening exposes a plurality of the photoelectric conversion elements adjacent to each other through the element isolation region.   The solid-state imaging device according to claim 2, wherein the opening is provided so that two adjacent photoelectric conversion elements are exposed through the element isolation region.   The opening is provided so as to expose the two rows of the photoelectric conversion elements between the charge transfer paths, and the light shielding film is provided in a band shape covering the charge transfer paths. Solid-state imaging device.   5. The color filter for determining the color of light incident on each of the photoelectric conversion elements is provided in an array in which two adjacent photoelectric conversion elements have the same color. The solid-state imaging device described.   6. The solid-state imaging device according to claim 5, wherein when the same color filter corresponding to two adjacent photoelectric conversion elements is used as one unit of color filter, the color filter is in a Bayer array. .   7. The solid-state imaging device according to claim 5, wherein the color filters are arranged so as to have the same color for the two photoelectric conversion elements adjacent to each other through the element isolation region.   7. The solid-state imaging device according to claim 5, wherein the color filters are arranged so as to have the same color for two adjacent photoelectric conversion elements via the charge transfer path.   The photoelectric conversion elements are arranged in such a manner that the photoelectric conversion elements in the same row are arranged at a constant pitch, and the positions of the photoelectric conversion devices are ½ of the pitch with respect to the photoelectric conversion devices in adjacent rows. The solid-state imaging device according to any one of claims 1 to 8, wherein the solid-state imaging device has a honeycomb arrangement that is shifted and arranged.   The solid-state imaging device according to claim 1, wherein the array of the photoelectric conversion elements is a square lattice array.

Images