JP4249349B2 - Solid-state imaging device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a solid-state imaging device used as an area image sensor, and more particularly to a frame transfer type solid-state imaging device.
[0002]
[Prior art]
Since the mass production technology of CCD (Charge Coupled Device) has been established, video cameras, electronic still cameras, etc. that use CCD type solid-state imaging devices as area image sensors have been rapidly spread. CCD type solid-state imaging devices are classified into several types depending on their structures. One of them is a frame transfer type solid-state imaging device (hereinafter, this solid-state imaging device is abbreviated as “FT-CCD”). There is.
[0003]
In the FT-CCD, a photosensitive portion is set on one surface of a semiconductor substrate, and an output transfer path is formed outside the photosensitive portion. There is a type of FT-CCD in which a CCD storage unit for one frame is provided between the photosensitive unit and the output transfer path, and a type of FT-CCD in which the output transfer path is directly connected to the outside of the photosensitive unit.
[0004]
A predetermined number of charge transfer channels are formed in the semiconductor substrate. Individual charge transfer channels move the photosensitive area in a fixed direction D._VTo the output transfer path. Also, the direction D_VDirection D that intersects_HA plurality of transfer electrodes that cross each of the charge transfer channels in plan view are formed on the semiconductor substrate via an electrical insulating film.
[0005]
Each intersecting portion of the charge transfer channel and the transfer electrode in plan view functions as one charge transfer stage. For this reason, the plurality of transfer electrodes are adjacent to each other.
[0006]
The FT-CCD is a solid-state imaging device in which a part of the charge transfer stage formed in the photosensitive portion is also used as a pixel photoelectric conversion element (photodiode). For this reason, a transfer electrode for forming a charge transfer stage to be used as a photoelectric conversion element for pixels has a desired light transmission property (this transfer electrode is referred to as “transparent transfer electrode for pixels”). Normally, all of the intersections in plan view between the charge transfer channel and the pixel transparent transfer electrode are used as pixel photoelectric conversion elements.
[0007]
In order to accumulate the signal charge generated by the photoelectric conversion in the pixel photoelectric conversion element in the pixel photoelectric conversion element and transfer the signal charge toward the output transfer path, only for the transfer of the signal charge The charge transfer stage used is formed. This charge transfer stage is in the direction D_VAre formed between two adjacent photoelectric conversion elements for pixels.
[0008]
In other words, a transfer electrode for forming a charge transfer stage that is used only for signal charge transfer (this transfer electrode is referred to as a “transfer-dedicated transfer electrode”) has the direction D described above._VA necessary number of pixels are disposed between two transparent transfer electrodes for pixels adjacent to each other. For example, when signal charges are transferred using four-phase drive pulses, there are three charge transfer stages that are used only for transferring signal charges between two adjacent photoelectric conversion elements for pixels. Formed.
[0009]
Light is applied to each pixel photoelectric conversion element in a state where a high-level voltage for potential well formation is applied to each of the pixel transparent transfer electrodes and a low-level voltage for potential barrier formation is applied to the transfer-dedicated transfer electrodes. When incident, signal charges are accumulated in the pixel photoelectric conversion element. These signal charges are transferred toward the output transfer path by applying, for example, a four-phase drive pulse to each of the transfer electrodes (transparent pixel transfer electrode and transfer transfer electrode).
[0010]
The output unit is connected to one end of the output transfer path. The output unit receives the signal charge from the output transfer path, converts this signal charge into a signal voltage, and further amplifies it.
[0011]
The FT-CCD has an advantage that the pixel density can be easily increased as compared to the interline transfer type solid-state imaging device because a part of the charge transfer stage also serves as a photoelectric conversion element for pixels. If the pixel density is increased, the quality of the reproduced image can be easily improved. In addition, since the aperture ratio of each pixel can be easily increased, there is an advantage that the resolution is easily increased. Furthermore, the type of FT-CCD that is not provided with a CCD storage unit has an advantage that the chip size can be easily reduced.
[0012]
[Problems to be solved by the invention]
In general, a microlens array is disposed on the photosensitive portion in order to increase the photoelectric conversion efficiency in the pixel photoelectric conversion element. In the FT-CCD for color imaging, the color filter array is disposed separately from the microlens array.
[0013]
In providing the microlens, first, a planarizing film is formed on the photosensitive portion. This planarization film is also used as a focus adjustment layer. In a FT-CCD for monochrome imaging, a microlens array including a predetermined number of microlenses is disposed on the planarizing film.
[0014]
On the other hand, in an FT-CCD for color imaging, a color filter array is formed on the planarizing film. For this reason, the microlens array is formed on the second planarization film after further providing a second planarization film on the color filter array.
[0015]
In both FT-CCD for monochrome imaging and color imaging, each microlens is formed separately so as to cover one pixel photoelectric conversion element in plan view.
[0016]
However, in the conventional FT-CCD, stray light tends to enter the pixel photoelectric conversion element due to the wraparound of light. When stray light enters a pixel photoelectric conversion element, noise is likely to occur.
[0017]
An object of the present invention is to provide an FT-CCD in which stray light is unlikely to enter a pixel photoelectric conversion element.
[0018]
[Means for Solving the Problems]
  According to one aspect of the present invention, a photosensitive portion set on one surface of a semiconductor substrate, an output transfer path formed in a region outside the photosensitive portion on one surface of the semiconductor substrate, and the photosensitive portion are fixed. Direction D_VA plurality of charge transfer channels that cross the line and reach the output transfer path, and are formed on the photosensitive portion, and the direction D_VDirection D that intersects_HAcross each of the plurality of charge transfer channels alongAnd the direction D _V Are arranged at a constant pitch alongA transparent transfer electrode for a plurality of pixels,AboveThe direction D out of all the intersections in plan view with a plurality of charge transfer channels_VAnd D_HIn each of the intersections selected every other one of the above, a plurality of transparent transfer electrodes for pixels that constitute a photoelectric conversion element for pixels together with the charge transfer channel, and on each of the photoelectric conversion elements for pixel There is provided a solid-state imaging device including a light shielding film having only an opening.
[0020]
Any of the above FT-CCDs includes a light shielding film having an opening only on each of the pixel photoelectric conversion elements. For this reason, even if the microlens array is disposed on the photosensitive portion, stray light is unlikely to enter the pixel photoelectric conversion element due to light wraparound.
[0021]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a plan view schematically showing an FT-CCD 100 according to the first embodiment. However, this figure is a schematic diagram in which the number of pixel photoelectric conversion elements is 24 in order to make the drawing easier to see and to make the explanation easier to understand. In an actual FT-CCD, the number of photoelectric conversion elements for pixels reaches 1000 to 10 million.
[0022]
The illustrated FT-CCD 100 includes a photosensitive unit 10 set on the surface of the semiconductor substrate 1, a CCD storage unit 30 formed outside the photosensitive unit 10, and an output transfer formed outside the CCD storage unit 30. Four pulse supply terminals 60a and 60b for supplying a predetermined drive pulse to each of the path 40, the output section 50 connected to one end of the output transfer path 40, and the transfer electrodes formed in the photosensitive section 10. , 60c, 60d, four pulse supply terminals 70a, 70b, 70c, 70d for supplying a predetermined drive pulse to each of the transfer electrodes constituting the CCD accumulating unit 30, and transfer electrodes constituting the output transfer path 40 Are provided with two pulse supply terminals 80a and 80b for supplying a predetermined drive pulse to each of them.
[0023]
The photosensitive unit 10 is an area in which a predetermined number of pixel photoelectric conversion elements 25 and a charge transfer stage for transferring signal charges stored in the pixel photoelectric conversion elements 25 to the CCD storage unit 30 are disposed. It is.
[0024]
The CCD storage unit 30 is light-shielded by a light shielding film described later. The signal charges accumulated in each of the pixel photoelectric conversion elements 25 are transferred to the CCD accumulating unit 30 at a high speed in order to minimize signal mixing due to light incident during the signal charge transfer period. The CCD accumulation unit 30 sequentially transfers the transferred signal charges to the output transfer path 40.
[0025]
The output transfer path 40 sequentially transfers the received signal charges toward the output unit 50.
[0026]
The output unit 50 converts the received signal charge into a signal voltage and amplifies the signal voltage.
[0027]
Hereinafter, taking as an example the case where the semiconductor substrate 1 made of an n-type silicon substrate having a p-type well is used, the structure of the photosensitive unit 10 and the structures of the CCD storage unit 30, the output transfer path 40, and the output unit 50 will be sequentially described. To do.
[0028]
First, the configuration of the photosensitive portion 10 will be described with reference to FIGS. 1, 2, 3 (a), and 3 (b). FIG. 1 is a plan view schematically showing the FT-CCD 100 of the first embodiment as described above.
[0029]
FIG. 2 is a schematic view of a cross section taken along line AA in FIG.
[0030]
FIG. 3A is a plan view schematically showing the first transparent transfer electrode 12 for pixels shown in FIG. 1, and FIG. 3B is a plan view of the second transparent transfer electrode 13 for pixels shown in FIG. It is a top view which shows an outline.
[0031]
The photosensitive unit 10 is moved in a certain direction D_VA total of seven charge transfer channels 11 are formed on the surface of the semiconductor substrate 1 so as to traverse along the line (indicated by arrows in FIG. 1) and reach the output transfer path 40. Each charge transfer channel 11 has a band shape in plan view. Each of the charge transfer channels 11 is obtained by forming an n-type region 3 at a predetermined location of a p-type well 2 formed on one surface side of the semiconductor substrate 1 as shown in FIG.
[0032]
Three first transparent transfer electrodes for pixels 12 and two second transparent transfer electrodes for pixels 13 are formed on the photosensitive portion 10 (see FIG. 1). These pixel transparent transfer electrodes 12 and 13 are arranged in the direction D described above._H, Each of the charge transfer channels 11 is crossed in plan view. Each of the pixel transparent transfer electrodes 12 and 13 has a predetermined pitch P in the order of the pixel first transparent transfer electrode 12 and the pixel second transparent transfer electrode 13 from the upstream side of the photosensitive portion 10.₁Are formed alternately one by one.
[0033]
The transparent transfer electrodes 12 and 13 for each pixel are made of a transparent conductive material such as polysilicon, tin oxide, ITO, or ruthenium oxide. These transparent transfer electrodes 12 and 13 for pixels are formed on an electric insulating film 4 made of a silicon oxide film formed on the surface of the semiconductor substrate 1 as shown in FIG.
[0034]
In this specification, since the transfer direction of the signal charge is the direction from the photosensitive part to the output transfer path, the side close to the output transfer path in the photosensitive part, CCD storage part, etc. is referred to as “downstream” and output. The side far from the transfer path is called “upstream”. The boundary between “downstream” and “upstream” is not distinguished.
[0035]
One pixel first transparent transfer electrode 12 has a total of four photoelectric conversion element formation regions 12P for pixels (see FIG. 3A). Two adjacent photoelectric conversion element forming regions 12P for pixels are formed at a predetermined pitch P by a connection portion 12C having a strip shape.₂Are connected. Connections of a predetermined length are also provided on the left side of the leftmost pixel photoelectric conversion element formation region 12P in FIG. 3A and on the right side of the rightmost pixel photoelectric conversion element formation region 12P in FIG. A portion 12C is formed. The maximum line width in the pixel photoelectric conversion element formation region 12P is wider than the line width of the connection portion 12C.
[0036]
The direction D between the connection portion 12C and the pixel photoelectric conversion element formation region 12P adjacent to the connection portion 12C._VAssuming the boundary line extending to, each pixel photoelectric conversion element formation region 12P has a substantially octagonal shape in plan view. The line width of each connection portion 12C is substantially the same as the length of the boundary line.
[0037]
The second transparent transfer electrode 13 for pixels has a shape that is substantially point-symmetric with the first transparent transfer electrode 12 for pixels in plan view. The pixel transparent transfer electrode 13 also has a total of four photoelectric conversion element formation regions 13P for pixels. These pixel photoelectric conversion element formation regions 13P are formed at a predetermined pitch P by connecting portions 13C having a strip shape.₂(See FIG. 3B). Connections of a predetermined length are also provided on the left side of the leftmost pixel photoelectric conversion element formation region 13P in FIG. 3B and on the right side of the rightmost pixel photoelectric conversion element formation region 13P in FIG. A portion 13C is formed. The maximum line width in the pixel photoelectric conversion element formation region 13P is wider than the line width of the connection portion 13C.
[0038]
Each of the pixel photoelectric conversion element formation regions 12P and each of the pixel photoelectric conversion element formation regions 13P has the pitch P described above.₂About half of the direction D_H(See FIG. 1). Here, “pitch P” in this specification₂About 1/2 "is P₂In addition to / 2, it also includes values within a range where “pixel shifting arrangement” described later can be performed.
[0039]
A total of three first transfer-dedicated transfer electrodes 15 are formed, one on each plane view, between the first transparent transfer electrode 12 for pixels and the second transparent transfer electrode 13 for pixels adjacent to the downstream side. (See FIGS. 1 and 2).
[0040]
A total of two second transfer-dedicated transfer electrodes 16 are formed one by one between the second transparent transfer electrode 13 for pixels and the first transparent transfer electrode 12 for pixels adjacent to the downstream side thereof. Furthermore, another second transfer-dedicated transfer electrode 16 is provided on the most upstream side of the second transparent transfer electrode 13 for pixels provided on the most downstream side and the first transfer electrodes 31 for storage sections described later. It is formed in plan view with the first transfer electrode 31 for the storage section.
[0041]
One auxiliary transfer electrode 17 is provided on the upstream side of the first transparent transfer electrode 12 for pixels provided on the most upstream side. The auxiliary transfer electrode 17 also has the direction D_HEach charge transfer channel 11 is traversed along a plane view.
[0042]
Each transfer-dedicated transfer electrodes 15 and 16 and auxiliary transfer electrode 17 are made of, for example, polysilicon. These transfer-dedicated transfer electrodes 15 and 16 and the auxiliary transfer electrode 17 are formed on the electrical insulating film 4 (see FIG. 2), similarly to the transparent transfer electrodes 12 and 13 for each pixel.
[0043]
As shown in FIG. 1 or 2, the upstream edge of each transfer-dedicated transfer electrode 15, 16 overlaps the downstream edge of the pixel transparent transfer electrode 12 or 13 adjacent to the upstream side. Yes. The downstream edge of each transfer-dedicated transfer electrode 15, 16 overlaps the upstream edge of the pixel transparent transfer electrode 12 or 13 adjacent to the downstream edge.
[0044]
However, the downstream edge of the second transfer-dedicated transfer electrode 16 provided on the most downstream side overlaps with the upstream edge of the first transfer electrode 31 for the storage portion adjacent to the downstream side. .
[0045]
The downstream edge of the auxiliary transfer electrode 17 overlaps the upstream edge of the pixel first transparent transfer electrode 12 adjacent to the downstream edge.
[0046]
Accordingly, the transparent transfer electrodes 12 and 13 for each pixel, the transfer electrodes 15 and 16 for transfer, and the auxiliary transfer electrode 17 have a so-called overlapping transfer electrode structure. Further, the second transfer dedicated transfer electrode 16 provided on the most downstream side and the first transfer electrode 31 for the storage section provided on the most upstream side also have a so-called overlapping transfer electrode structure. These transfer electrodes are electrically insulated from each other.
[0047]
The two transfer-dedicated transfer electrodes 15 and 16 adjacent to each other via the pixel transparent transfer electrode 12 or 13 are in plan view of the pixel transparent transfer electrode 12 or 13 interposed between these transfer-dedicated transfer electrodes. A gap having a planar shape substantially equivalent to a shape obtained by reducing the shape in the line width direction is defined.
[0048]
The first transfer-dedicated transfer electrode 15 and the auxiliary transfer electrode 17 provided at the most upstream have a predetermined shape in the line width direction of the shape of the first transparent transfer electrode 12 for pixels provided at the most upstream in a plan view. A gap having a planar shape substantially equivalent to the reduced shape is defined.
[0049]
Each of the first transparent transfer electrodes 12 for a pixel intersects with the even-numbered charge transfer channel 11 counted from the left end in FIG. 1 in a plan view in each of the pixel photoelectric conversion element formation regions 12P. 2 intersects with the odd-numbered charge transfer channel 11 in plan view. On the other hand, each pixel second transparent transfer electrode 13 intersects with the odd-numbered charge transfer channel 11 counted from the left end of FIG. 1 in the pixel photoelectric conversion element formation region 13P in plan view, and connected to the connection portion 13C. Each of them intersects with the even-numbered charge transfer channel 11 in plan view. The transfer-dedicated transfer electrodes 15 and 16 and the auxiliary transfer electrode 17 intersect each of the charge transfer channels 11 in a plan view at predetermined positions.
[0050]
Each of the intersections in plan view of each charge transfer channel 11 and each pixel transparent transfer electrode 12, 13 has a function as one charge transfer stage. The same applies to each of the intersections in plan view between each charge transfer channel 11 and each transfer dedicated transfer electrode 15, 16 or auxiliary transfer electrode 17. Therefore, each of the charge transfer channels 11 includes one transparent transfer electrode 12, 13 for each pixel that crosses the charge transfer channel 11 in plan view, each transfer dedicated transfer electrode 15, 16, and the auxiliary transfer electrode 17. A vertical transfer CCD).
[0051]
As shown in FIG. 1 or 2, the light shielding film 20 is formed from the photosensitive portion 10 to the output transfer path 40. An electrical insulating layer 21 made of, for example, a silicon oxide film is interposed between the light shielding film 20 and various transfer electrodes formed on the photosensitive unit 10, the CCD storage unit 30, and the output transfer path 40, respectively.
[0052]
The light shielding film 20 is formed on each intersection of the charge transfer channel 11 and each pixel photoelectric conversion element formation region 12P in plan view, and on each charge transfer channel 11 and each pixel photoelectric conversion element formation region 13P. And an opening 22 having a circular horizontal section on each of the intersecting portions in plan view.
[0053]
Therefore, each of the intersections in plan view of each charge transfer channel 11 and each pixel photoelectric conversion element formation region 12P, and each plan view of each charge transfer channel 11 and each pixel photoelectric conversion element formation region 13P. Each of the upper intersections functions as a pixel photoelectric conversion element (photodiode) 25 (see FIG. 1). Other intersections do not substantially function as photoelectric conversion elements.
[0054]
Therefore, in the FT-CCD 100 shown in FIG. 1, the direction D of the crossing portions of the charge transfer channels 11 and the transparent transfer electrodes 12 and 13 for the pixels in a plan view is described._VAnd D_HEach of the intersections selected every other one functions as a pixel photoelectric conversion element 25. Each of the openings 22 functions as a light receiving portion for one pixel. Hereinafter, the opening 22 may be referred to as “light receiving portion 22”.
[0055]
In the FT-CCD 100, a total of 24 photoelectric conversion elements 25 for pixels are provided over 8 columns and 6 rows. One pixel photoelectric conversion element array is composed of three pixel photoelectric conversion elements 25, and one pixel photoelectric conversion element row is composed of four pixel photoelectric conversion elements 25. Each of the odd-numbered photoelectric conversion element columns counted from the left end of FIG. 1 includes only even-numbered pixel photoelectric conversion elements 25, and each even-numbered photoelectric conversion element array includes only the odd-numbered pixel photoelectric conversion elements 25. including.
[0056]
Each of the pixel photoelectric conversion elements 25 constituting the even-numbered columns has a pitch between the pixel photoelectric conversion elements 25 in each pixel photoelectric conversion element 25 with respect to each of the pixel photoelectric conversion elements 25 constituting the odd-numbered columns. About ½, shifted in the column direction. Each of the pixel photoelectric conversion elements 25 constituting the even-numbered rows has a pitch between the pixel photoelectric conversion elements 25 in each pixel photoelectric conversion element 25 with respect to each of the pixel photoelectric conversion elements 25 constituting the odd-numbered rows. The line is displaced by about 1/2.
[0057]
Here, the “pitch between the pixel photoelectric conversion elements in the pixel photoelectric conversion element row” as used in this specification means the pitch between the intersecting portions in the plan view where the pixel photoelectric conversion elements are formed. Means. In addition, “about 1/2 of the pitch between pixel photoelectric conversion elements” includes 1/2 of the pitch of the pixel photoelectric conversion elements, as well as pixels that occur due to manufacturing errors, design, or mask manufacturing. Although it deviates from 1/2 of the pitch due to factors such as rounding error of the position, it is regarded as substantially equivalent to 1/2 of the pitch in view of the performance of the obtained FT-CCD and the image quality of the image. It also includes possible values.
[0058]
In the present specification, the arrangement of the pixel photoelectric conversion elements as described above is hereinafter referred to as “pixel shifting arrangement”. Of course, in the “pixel shifting arrangement”, each of the odd-numbered pixel photoelectric conversion element columns includes only odd-numbered pixel photoelectric conversion elements, and each of the even-numbered pixel photoelectric conversion element columns is an even-numbered pixel. Only the photoelectric conversion element for use may be included.
[0059]
When light enters from the light receiving unit 22, photoelectric conversion is performed in the pixel photoelectric conversion element 25. By applying a predetermined voltage to the transparent transfer electrodes 12 and 13 for each pixel, the transfer electrodes 15 and 16 for transfer, and the auxiliary transfer electrode 17, signal charges are accumulated in each of the pixel photoelectric conversion elements 25. Can do.
[0060]
Next, the structure of the CCD storage unit 30 will be described with reference to FIGS. The CCD accumulation unit 30 is formed on the surface of the semiconductor substrate 1 between the photosensitive unit 10 and the output transfer path 40.
[0061]
As shown in FIG. 1, each of the charge transfer channels 11 described above moves the CCD accumulator 30 in the direction D._VAnd the output transfer path 40 is reached.
[0062]
Six first storage electrode 31 for storage part and two second transfer electrodes 32 for storage part are formed (see FIG. 1). The storage unit transfer electrodes 31 and 32 are arranged in the direction D described above._H, Each of the charge transfer channels 11 is crossed in plan view. Each of the storage unit transfer electrodes 31 and 32 has a predetermined pitch P in the order of the storage unit first transfer electrode 31 and the storage unit second transfer electrode 32 from the upstream side of the CCD storage unit 30.₁Are formed alternately one by one.
[0063]
Each of the storage unit transfer electrodes 31 and 32 is made of polysilicon, for example. As shown in FIG. 2, the storage unit transfer electrodes 31 and 32 are formed on an electrical insulating film 4 made of a silicon oxide film formed on the surface of the semiconductor substrate 1.
[0064]
Each storage unit first transfer electrode 31 has a strip shape. The shape of each of the second transfer electrodes 32 for the storage unit is the same.
[0065]
The first transfer electrode 31 for each storage unit and the second transfer electrode 32 for each storage unit are the first transparent transfer electrode 12 for pixels, the second transparent transfer electrode 13 for pixels, the transfer electrode 15 for first transfer, and the first transfer electrode 15 for transfer. Similar to the two-transfer dedicated transfer electrode 16, a so-called superimposed transfer electrode structure is formed (see FIGS. 1 and 2). These transfer electrodes are electrically insulated from each other.
[0066]
Each of the crossing portions of the charge transfer channels 11 and the storage unit transfer electrodes 31 and 32 in plan view has a function as one charge transfer stage. Therefore, each of the charge transfer channels 11 constitutes one CCD (vertical transfer CCD) together with the storage unit transfer electrodes 31 and 32 that cross the charge transfer channel 11 in plan view.
[0067]
The most downstream second transfer-dedicated transfer electrode 16 formed in the photosensitive portion 10 and the most upstream first transfer electrode 31 for the storage portion formed in the CCD storage portion form a so-called overlapping transfer electrode structure. Yes. As a result, each of the charge transfer channels 11 forms one CCD (vertical transfer CCD) from the photosensitive unit 10 to the CCD storage unit 30.
[0068]
Each of the crossing portions in plan view of each charge transfer channel 11 and each storage unit transfer electrode 31 and 32 is covered with the above-described light shielding film 20 (see FIGS. 1 and 2). For this reason, each of the intersecting portions in plan view between the charge transfer channels 11 and the storage portion transfer electrodes 31 and 32 does not substantially function as a photoelectric conversion element.
[0069]
The signal charges transferred from the photosensitive unit 10 are sequentially transferred to the output transfer path 40 by the vertical transfer CCDs formed in the CCD storage unit 30. At this time, a predetermined voltage is applied to each of the storage unit transfer electrodes 31 and 32 at a predetermined cycle.
[0070]
In FIG. 1, in order to make each component easy to understand, the charge transfer channel 11 is drawn by a solid line, and the light shielding film 20 is drawn by a two-dot chain line.
[0071]
Next, the structure of the output transfer path 40 will be described. The output transfer path 40 is formed on the surface of the semiconductor substrate 1 so as to be adjacent to the downstream end of the CCD storage unit 30. In FIG. 1, the output transfer path 40 is illustrated in a simplified manner for easy understanding of FIG. 1, and the structure of the output transfer path 40 will be described below with reference to FIG.
[0072]
FIG. 4A is a plan view schematically showing the output transfer path 40. As shown in the figure, a plurality of output transfer paths 40 are provided on the semiconductor substrate 1 via one charge transfer channel 41 formed on the semiconductor substrate 1 and an electrical insulating film (not shown). The transfer electrodes 43, 44, 45, and 46 are formed. An electrical insulating layer (not shown) is formed on the transfer electrodes 43, 44, 45, 46. The light shielding film 20 described above is formed on the electrical insulating layer.
[0073]
The charge transfer channel 41 includes, for example, an n-type region (not shown) and n at a predetermined position of a p-type well formed on one surface side of the semiconductor substrate 1.^-It can be obtained by alternately forming a predetermined number of mold regions (not shown). n^-The n-type impurity concentration in the type region is lower than the n-type impurity concentration in the n-type region. The charge transfer channel 41 has a direction D described above._HIt extends along.
[0074]
Each transfer electrode 43, 44, 45, 46 is in a ratio of one for each charge transfer channel 11, and from the output unit 50 (see FIG. 1) side, the transfer electrode 43, the transfer electrode 44, and the transfer electrode 45. The transfer electrodes 46 are formed in this order. These transfer electrodes 43, 44, 45, 46 are made of, for example, polysilicon.
[0075]
The transfer electrode 43 is formed on one of the n-type regions in plan view. The transfer electrode 43 has a first straight line portion that crosses the charge transfer channel 41 in plan view and a second straight line portion that crosses the charge transfer channel 11 in plan view. The edge portion of the second linear portion on the photosensitive portion 10 side overlaps with the edge portion of the second transfer electrode 32 for the storage portion located downstream and downstream. The shape of the transfer electrode 43 when viewed from above is L-shaped.
[0076]
The transfer electrode 44 has the above n in plan view.^-Formed on one of the mold regions. The transfer electrode 44 includes a first straight portion that crosses the charge transfer channel 41 in plan view and a second straight portion that partially crosses the charge transfer channel 11 in plan view. The shape of the transfer electrode 43 when viewed in plan is also L-shaped.
[0077]
The transfer electrode 45 is formed on one of the n-type regions in plan view. The transfer electrode 45 is a belt-like electrode that crosses the charge transfer channel 41 in plan view.
[0078]
The transfer electrode 46 has the above n in plan view.^-Formed on one of the mold regions. The transfer electrode 46 is also a belt-like electrode that crosses the charge transfer channel 41 in plan view.
[0079]
The edge of the transfer electrode 44 on the transfer electrode 43 side overlaps the transfer electrode 43, and the edge of the transfer electrode 44 on the transfer electrode 45 side overlaps the transfer electrode 45. The edge of the transfer electrode 46 on the transfer electrode 45 side overlaps the transfer electrode 45, and the edge of the transfer electrode 44 other than the leftmost transfer electrode 44 in FIG. 43 is overlaid. That is, each transfer electrode 43, 44, 45, 46 has a so-called overlapping transfer electrode structure.
[0080]
The transfer electrodes 43, 44, 45, and 46 are insulated from each other by an electrical insulating film (not shown). Further, the transfer electrode 43 and the second transfer electrode 32 for the most downstream storage part formed in the CCD storage part 30 are also insulated from each other by an electrical insulating film (not shown).
[0081]
N above^-One of the mold regions and the n^-An intersection in plan view with one transfer electrode 44 or 46 formed on the mold region via an electric insulating film is used as one potential barrier region. An intersection in plan view between one of the n-type regions and one transfer electrode 43 or 45 formed on the n-type region via an electrical insulating film is defined as one potential well region. Used.
[0082]
One potential barrier region and one potential well region formed on the downstream side (meaning the output unit 50 side) are electrically connected to form one charge transfer stage. Therefore, the output transfer path 40 has two charge transfer stages per charge transfer channel 11. That is, the output transfer path 40 is composed of a two-phase drive type CCD.
[0083]
The signal charges transferred from the CCD accumulating unit 30 are sequentially transferred to the output unit 50 by the charge transfer stages formed in the output transfer path 40. At this time, a predetermined voltage is applied to each of the transfer electrodes 43, 44, 45, and 46 at a predetermined cycle.
[0084]
Next, the output unit 50 will be described with reference to FIG.
[0085]
FIG. 4B is a diagram schematically showing the downstream part of the output transfer path 40 and the output part 50 (see FIG. 1) formed in the downstream part.
[0086]
A floating diffusion 51 is formed adjacent to the downstream side of the n-type region formed below the most downstream transfer electrode 43 in the output transfer path 40. An output gate (output gate) electrode 52 is disposed across the floating diffusion 51 and the most downstream transfer electrode 43. Output gate voltage V_ogIs applied to the output gate (output gate) electrode 52 at a predetermined timing.
[0087]
A source follower circuit AMP is connected to the floating diffusion 51.
[0088]
A reset transistor gate region 53 is formed adjacent to the downstream side of the floating diffusion 51, and a reset transistor drain region 54 is formed adjacent to the downstream side thereof.
[0089]
The floating diffusion 51, the reset transistor gate region 53, and the reset transistor drain region 54 constitute a reset transistor together with the reset transistor gate electrode 55 formed above the reset transistor gate region 53.
[0090]
Reset electrode voltage V_reIs applied to the reset transistor gate electrode 55 at a predetermined timing. Reset drain voltage V_rdIs applied to the drain region 54 for the reset transistor
The signal charge transferred to the most downstream charge transfer stage in the output transfer path 40 is transferred to the floating diffusion 51 when a predetermined voltage is applied to the output gate (output gate) electrode 52. This signal charge is supplied to the source follower circuit AMP, where it is amplified and output.
[0091]
The charge after amplification (detection) is applied to the reset transistor gate electrode 55 by the reset electrode voltage V._reIs applied to the reset transistor drain region 54 and is absorbed by the power source.
[0092]
The FT-CCD 100 described above includes the light shielding film 20 having the opening 22 only on each of the pixel photoelectric conversion elements 25. For this reason, even if the microlens array is disposed on the photosensitive portion 10, stray light is unlikely to enter the pixel photoelectric conversion element 25 due to the wraparound of light.
[0093]
Further, a pixel shifting arrangement is performed in which each pixel photoelectric conversion element 25 forms a wide region on every other charge transfer channel 11. For this reason, it is easy to improve a pixel density, suppressing the fall of the area of the light-receiving part 22 in each pixel. As a result, it is easy to increase the resolution. In order to perform four-phase driving, it is only necessary to provide one transfer-dedicated transfer electrode 15 or 16 between two adjacent pixel transparent transfer electrodes 12 and 13.
[0094]
In order to drive the FT-CCD 100 shown in FIG. 1, each pixel transparent transfer electrode 12, 13, each transfer dedicated transfer electrode 15, 16, auxiliary transfer electrode 17, each storage unit transfer electrode 31, 32, output Drive pulse supply means for supplying a predetermined drive pulse to each of the transfer path 40 and the output unit 50 is used.
[0095]
FIG. 5 is a schematic diagram showing an example of the drive pulse supplying means. As shown in the figure, the drive pulse supply means 90 includes, for example, a synchronization signal generator 91, a timing generator 92, a first vertical drive circuit 93a, a second vertical drive circuit 93b, a horizontal drive circuit 94, and the like. Consists of.
[0096]
The synchronization signal generator 91 generates various pulses necessary for signal processing, such as a vertical synchronization pulse and a horizontal synchronization pulse. The timing generator 92 applies four-phase vertical pulse signals to the transparent transfer electrodes 12 and 13 for each pixel, the transfer electrodes 15 and 16 for transfer, and the auxiliary transfer electrode 17 and the transfer electrodes 31 and 32 for the storage unit. A timing signal is generated for the four-phase vertical pulse signal and the two-phase horizontal pulse signal necessary for driving the output transfer path 40.
[0097]
The first vertical drive circuit 93a generates a four-phase vertical pulse signal for the photosensitive portion 10 based on the timing signal. Each of the four-phase vertical pulse signals is transmitted through a pulse supply terminal 60a, 60b, 60c, or 60d to a predetermined pixel transparent transfer electrode 12, 13, a predetermined transfer dedicated transfer electrode 15, 16 or an auxiliary transfer electrode. 17 is applied.
[0098]
The second vertical drive circuit 93b generates a four-phase vertical pulse signal for the storage unit 30 based on the timing signal. Each of the four-phase vertical pulse signals is applied to predetermined storage unit transfer electrodes 31 and 32 via pulse supply terminals 70a, 70b, 70c, and 70d.
[0099]
The horizontal drive circuit 94 generates a two-phase horizontal pulse signal based on the timing signal. Each of the two-phase horizontal pulse signals is applied to predetermined transfer electrodes 43, 44, 45, and 46 via a pulse supply terminal 80a or 80b.
[0100]
After the signal charges are temporarily accumulated in each of the pixel photoelectric conversion elements 25, these signal charges are transferred from the photosensitive unit 10 to the CCD accumulation unit 30. The signal charges are transferred from the photosensitive unit 10 to the CCD storage unit 30 by driving the above-described vertical transfer CCDs formed from the photosensitive unit 10 to the CCD storage unit 30 in four phases.
[0101]
In order to store and transfer signal charges, for example, as shown in FIG. 5, a predetermined pulse supply terminal and a predetermined transfer electrode are electrically connected. In FIG. 5, each second transfer-dedicated transfer electrode 16 and auxiliary transfer electrode 17 are electrically connected to the pulse supply terminal 60a. Each pixel second transparent transfer electrode 13 is electrically connected to the pulse supply terminal 60b. Each first transfer dedicated transfer electrode 15 is electrically connected to a pulse supply terminal 60c. Each pixel first transparent transfer electrode 12 is electrically connected to the pulse supply terminal 60d.
[0102]
In FIG. 5, the fourth, eighth, and twelfth storage unit transfer electrodes counted from the upstream side (both storage unit second transfer electrodes 32) are electrically connected to the pulse supply terminal 70a. The third, seventh, and eleventh storage portion transfer electrodes counted from the upstream side (all of the storage portion first transfer electrodes 31) are electrically connected to the pulse supply terminal 70b. The second, sixth, and tenth storage unit transfer electrodes counted from the upstream side (both storage unit second transfer electrodes 32) are electrically connected to the pulse supply terminal 70c. The first, fifth and ninth storage unit transfer electrodes counted from the upstream side (both storage unit first transfer electrodes 31) are electrically connected to the pulse supply terminal 70d.
[0103]
During a signal charge accumulation period defined by a predetermined pulse generated by the synchronization signal generator 91, a low-level vertical pulse signal (hereinafter referred to as a low-level vertical signal) is supplied from the first vertical drive circuit 93a to the pulse supply terminals 60a and 60c. Change the pulse signal to “V_L". ) Is applied to the pulse supply terminals 60b and 60d, and a high level vertical pulse signal (hereinafter referred to as a high level vertical pulse signal is referred to as "V")._H". ) Is applied.
[0104]
Thereby, a potential well is formed in each of the charge transfer stages formed at the intersections of the transparent transfer electrodes for pixels 12 and 13 and the charge transfer channels 11 in plan view. In addition, each of the charge transfer stages formed at each intersection of the transfer dedicated transfer electrodes 15 and 16 and each charge transfer channel 11 in plan view, and the auxiliary transfer electrode 17 and each charge transfer channel 11 A potential barrier is formed at each of the charge transfer stages formed at each intersection in the plan view.
[0105]
As a result, in each of the charge transfer channels 11, potential barriers and potential wells are alternately formed in this order from the upstream side to the downstream side of the photosensitive portion 10.
[0106]
When light enters each pixel photoelectric conversion element 25 in this state, the signal charge generated by the photoelectric conversion in each pixel photoelectric conversion element 25 is accumulated in the potential well formed in each pixel photoelectric conversion element 25. Is done. That is, signal charges for one frame are accumulated in the photosensitive portion 10.
[0107]
The signal charges for one frame are transferred from the photosensitive unit 10 to the CCD storage unit 30 by using a four-phase drive pulse. This is done by forming sequentially.
[0108]
In the following description of the operations (1) to (4), in order to make the description easy to understand, attention is paid to one arbitrarily selected charge transfer channel 11. Generation of potential barriers and potential wells in one arbitrarily selected charge transfer channel 11 and generation of potential barriers and potential wells in each of the other charge transfer channels 11 in substantially the same pattern and in the same pattern. Disappearance occurs.
[0109]
(1) After accumulating signal charges, a potential barrier is formed in the charge transfer stage formed at the intersection with the auxiliary transfer electrode 17 in plan view, and from the charge transfer stage to the downstream end of the CCD accumulation unit 30. A potential barrier is formed every three charge transfer stages. As a result, one potential well spanning three charge transfer stages is formed between two adjacent potential barriers.
[0110]
Hereinafter, one potential well that spans n (n is an integer of 1 or more) charge transfer stages is referred to as “n series potential wells”, and one potential barrier that spans n charge transfer stages is referred to as “n series potential wells”. It is called “potential barrier”.
[0111]
The signal charge accumulated in each pixel photoelectric conversion element 25 spreads in a triple potential well including the pixel photoelectric conversion element 25.
[0112]
At this time, when all of the three potential wells in the photosensitive portion 10 are viewed in plan, these three potential wells are distributed in a matrix. Accordingly, the signal charges accumulated in the pixel photoelectric conversion elements 25 are also distributed in a matrix, and all the signal charges have substantially the same phase shift.
[0113]
(2) Each of the triple potential wells formed in (1) is contracted by one charge transfer stage from the upstream side to the downstream side to form a double potential well. Each of the series of potential barriers formed in the above (1) is extended downstream by one charge transfer stage to form a series of potential barriers.
[0114]
(3) Each of the two series of potential wells formed in the above (2) is extended downstream by one charge transfer stage to form a triple series of potential wells. Each of the two series of potential barriers formed in the above (2) shrinks the upstream side to the downstream side by one charge transfer stage to form a series of potential barriers.
[0115]
(4) In the following, until all the signal charges for one frame are transferred to the output transfer path 40, the formation of two potential wells and the three potential wells according to the operations (2) to (3) above. The formation is repeated alternately.
[0116]
For example, V is applied to the pulse supply terminal 60a._LIs applied to each second transfer dedicated transfer electrode 16 and auxiliary transfer electrode 17 via the pulse supply terminal 60a._LIs applied. As a result, the second transfer dedicated transfer electrodes 16 and the charge transfer channels 11 are formed at intersections in plan view, and the auxiliary transfer electrodes 17 and the charge transfer channels 11 are formed at intersections in plan view. A potential barrier is formed in each of the charge transfer stages.
[0117]
Conversely, V is applied to the pulse supply terminal 60a._HIs applied to the crossing portion in plan view of each second transfer dedicated transfer electrode 16 and each charge transfer channel 11 and the crossing portion in plan view of each auxiliary transfer electrode 17 and each charge transfer channel 11. A potential well is formed in each of the charge transfer stages.
[0118]
Therefore, in performing the operations (1) to (4), the pulse supply terminals 60a, 60b, 60c, 60d, and 70a are formed so that a desired potential well or potential barrier is formed in a desired charge transfer stage. , 70b, 70c, and 70d, the level of the vertical pulse signal to be applied is appropriately set.
[0119]
It is also possible to store signal charges in the pixel photoelectric conversion elements 25 in the state (1) from the beginning.
[0120]
When all of the signal charges for one frame accumulated in the photosensitive portion 10 have been transferred to the CCD accumulation portion 30, the next signal charge accumulation period can be started.
[0121]
Following the transfer of the signal charge for one frame from the photosensitive unit 10 to the CCD storage unit 30, the transfer of the signal charge for one frame from the CCD storage unit 30 to the output transfer path 40 is performed.
[0122]
The signal charges are transferred from the CCD accumulation unit 30 to the output transfer path 40 in order from the signal charges transferred to the most downstream charge transfer stage in the CCD accumulation unit 30. As shown in FIG. 1, the most downstream charge transfer stage in the CCD accumulation unit 30 is formed at the intersection of the most downstream accumulation unit second transfer electrode 32 and each charge transfer channel 11 in plan view. Each charge transfer stage. Signal charges accumulated in the pixel photoelectric conversion elements 25 constituting two adjacent pixel photoelectric conversion element rows are transferred to these charge transfer stages substantially simultaneously.
[0123]
The transfer of signal charges from the CCD storage unit 30 to the output transfer path 40 is performed, for example, as follows. First, as shown in FIG. 4, the pulse supply terminal 80a and the transfer electrodes 43 and 44 are electrically connected in advance, and the pulse supply terminal 80b and the transfer electrodes 45 and 46 are electrically connected in advance. Keep it.
[0124]
Then, V is applied to the second transfer electrode 32 for the most downstream storage unit._HIs applied to each of the transfer electrodes 43 and 44, a high level horizontal pulse signal (hereinafter referred to as a high level horizontal pulse signal "H")._H". ) And a low level horizontal pulse signal (hereinafter referred to as a low level horizontal pulse signal “H”) to each transfer electrode 45 and 46._L". ) Is applied. Next, the voltage of the second transfer electrode 32 for the most downstream storage unit is set to V_HTo V_LTo change.
[0125]
As a result, each signal charge that has been transferred to each of the charge transfer stages formed at each intersection in the plan view of the second transfer electrode 32 for the storage section on the most downstream side and each charge transfer channel 11 is The charge transfer channel 41 and each transfer electrode 43 are transferred to each potential well formed at each crossing portion in plan view. That is, the signal charge is transferred from the CCD accumulation unit 30 to the output transfer path 40.
[0126]
As described above, the output transfer path 40 is composed of a two-phase drive type CCD. Therefore, the signal charge transferred to the output transfer path 40 is transferred from the horizontal drive circuit 94 (see FIG. 5) to the pulse supply terminal 80a._LAnd H is applied to the pulse supply terminal 80b._HIs transferred once to the output unit 50 side by one charge transfer stage.
[0127]
The transfer of signal charges in the output transfer path 40 is performed by alternately performing the following operations (a) and (b) at a predetermined cycle.
[0128]
(a) H from the horizontal drive circuit 94 to the pulse supply terminal 80a_LAnd H is applied to the pulse supply terminal 80b._HApply.
[0129]
(b) H from the horizontal drive circuit 94 to the pulse supply terminal 80a_HAnd H is applied to the pulse supply terminal 80b._LApply.
[0130]
The second vertical drive circuit 93b applies a predetermined vertical pulse signal to each of the pulse supply terminals 70a, 70b, 70c, and 70d until all of the signal charges for one frame are transferred to the output transfer path 40. Further, the horizontal drive circuit 94 applies a predetermined horizontal pulse signal to each of the pulse supply terminals 80a and 80b until all of the signal charge for one frame is transferred to the output unit 50.
[0131]
The output unit 50 sequentially converts and amplifies the signal charges received from the output transfer path 40 to a signal voltage, and outputs the amplified signal voltage to a predetermined circuit.
[0132]
The FT-CCD 100 according to the first embodiment described above is an FT-CCD having a simple structure as the FT-CCD. In an actual FT-CCD, a microlens is usually disposed in order to increase the photoelectric conversion efficiency in each pixel photoelectric conversion element 25. In addition, a color filter array is provided in an FT-CCD for color imaging.
[0133]
In providing the microlens, first, a planarizing film is formed on the photosensitive portion 10. This planarization film is also used as a focus adjustment layer. In a FT-CCD for monochrome imaging, a microlens array including a predetermined number of microlenses is disposed on the planarizing film. On the other hand, in an FT-CCD for color imaging, a color filter array is formed on the planarizing film. For this reason, the microlens array is formed on the second planarization film after further providing a second planarization film on the color filter array. In both FT-CCD for monochrome imaging and color imaging, each microlens is formed separately so as to cover the light receiving portion of one pixel in plan view.
[0134]
FIG. 6 is a partial cross-sectional view of an FT-CCD 200 according to the second embodiment. This figure shows a part of a cross section from the photosensitive part to the CCD accumulating part in the FT-CCD 200. The FT-CCD 200 is obtained by adding a color filter array and a microlens array to the FT-CCD 100 of the first embodiment. That is, the FT-CCD 200 is an FT-CCD for color imaging.
[0135]
6 that are the same as those shown in FIG. 2 are assigned the same reference numerals as those used in FIG. 2, and descriptions thereof are omitted.
[0136]
In the FT-CCD 200, the first planarization film 110 is formed so as to cover the light shielding film 20 and the light receiving portions 22 of the individual pixels. The first planarizing film 110 is formed from above the photosensitive portion to above the output transfer path (not shown). A color filter array 120 is provided on the first planarization film 110 above the photosensitive portion. The second flattening film 130 is above the color filter array 120 above the photosensitive portion, and above the first flattening film 110 above the CCD storage portion and above the output transfer path (not shown). Is formed. A microlens array including a predetermined number of microlenses 140 is formed on the second planarization film 130 above the photosensitive portion.
[0137]
The first planarization film 110 is formed by applying a transparent resin such as a photoresist to a desired thickness by, for example, a spin coating method.
[0138]
The color filter array 120 is formed by, for example, forming a green filter 121G, a red filter not shown in FIG. 6, and a blue filter not shown in FIG. 6 in a predetermined pattern. The color filter array 120 can be produced, for example, by forming a resin (color resin) layer in which a pigment or dye of a desired color is dispersed at a predetermined position by a method such as photolithography.
[0139]
In the FT-CCD 200 of the second embodiment, the pixel shift arrangement is performed as in the FT-CCD 100 of the first embodiment. Therefore, each of the red filter, the green filter, and the blue filter constituting the color filter array 120 has, for example, one diagonal line in the direction D._VIt presents a diamond that extends (see FIG. 1).
[0140]
The arrangement pattern of each color filter in the color filter array 120 is selected as follows, for example. That is, based on the signal charge accumulated in each pixel photoelectric conversion element constituting a predetermined one or two pixel photoelectric conversion element row in the FT-CCD provided with the color filter array 120, the color signal Alternatively, the color difference signal is selected.
[0141]
FIG. 7 is a plan view partially showing the color filter array 120 described above. In the color filter array 120 shown in the figure, a color filter row composed of only the green filter 121G and a color filter row in which the blue filter 121B and the red filter 121R are alternately arranged are alternately formed. .
[0142]
Each of the color filters 121R, 121G, and 121B is formed so as to cover the light receiving unit 22 of a separate pixel in plan view. The alphabets R, G, and B in each color filter shown in FIG. 7 represent the color of the color filter.
[0143]
The second planarizing film 130 shown in FIG. 6 is formed by applying a transparent resin such as a photoresist to a desired thickness by, for example, a spin coating method.
[0144]
Each of the microlenses 140 shown in FIG. 6 is formed so as to cover the light receiving portion 22 of one pixel in plan view. For example, the microlens 140 is formed by partitioning a layer made of a transparent resin (including a photoresist) having a refractive index of approximately 1.3 to 2.0 into a predetermined shape by a photolithography method or the like, and then performing heat treatment on each partition. The transparent resin layer is melted, and the corners are rounded by surface tension, followed by cooling. The shape of each section in plan view is appropriately selected according to the shape of the target microlens in plan view, such as square (including rhombus), hexagon, and circle.
[0145]
The FT-CCD 200 has the same structure as the FT-CCD 100 of Example 1 except that the FT-CCD 200 includes the first planarization film 110, the color filter array 120, the second planarization film 130, and the microlens array. Have. Therefore, in the FT-CCD 200, for the same reason as in the FT-CCD 100 of the first embodiment, stray light is unlikely to enter the pixel photoelectric conversion element 25 due to light wraparound. Further, it is easy to improve the pixel density while suppressing a decrease in the area of the light receiving unit 22 in each pixel. As a result, it is easy to increase the resolution.
[0146]
Next, the FT-CCD according to the third embodiment will be described with reference to FIG. 8, FIG. 9 (a), FIG. 9 (b), FIG. 10 (a), FIG. 10 (b), FIG. To do.
[0147]
FIG. 8 is a plan view schematically showing the FT-CCD 300 according to the third embodiment. However, this figure is a schematic diagram in which the number of pixel photoelectric conversion elements 225 is 28 in order to make the drawing easy to see and to make the explanation easy to understand. In an actual FT-CCD, as described above, the number of pixel photoelectric conversion elements 225 reaches 1000 to 10 million.
[0148]
The FT-CCD 300 shown in FIG. 8 includes a photosensitive portion 210 set on the surface of the semiconductor substrate 201, an adjustment portion 230 formed outside the photosensitive portion 210, and an output formed outside the adjustment portion 230. Four pulses for supplying a predetermined drive pulse to each of the transfer path 240, the output section 250 connected to one end of the output transfer path 240, and the transfer electrodes formed in the photosensitive section 210 and the adjusting section 230 Supply terminals 260a, 260b, 260c, and 260d, and four pulse supply terminals 280a, 280b, 280c, and 280d for supplying a predetermined drive pulse to each of the transfer electrodes constituting the output transfer path 240 are provided. .
[0149]
The roles of the photosensitive unit 210, the output transfer path 240, and the output unit 250 are substantially the same as the roles of the photosensitive unit 10, the output transfer path 40, or the output unit 50 in the FT-CCD 100 of the first embodiment described above. .
[0150]
However, the output transfer path 240 is composed of a four-phase drive type CCD as will be described later. Accordingly, the four pulse supply terminals 280a, 280b, 280c, and 280d are provided to supply a predetermined drive pulse to each of the transfer electrodes constituting the output transfer path 240.
[0151]
In the FT-CCD 300, an adjustment unit 230 is provided instead of the CCD storage unit 30 in the FT-CCD 100. The adjusting unit 230 is for smoothly transferring the signal charges accumulated in the photosensitive unit 210 to the output transfer path 240 with the same phase shift.
[0152]
Hereinafter, taking the case of using a semiconductor substrate 201 (see FIG. 8) made of an n-type silicon substrate having a p-type well as an example, the configuration of the photosensitive unit 210 and the adjustment unit 230, the output transfer path 240, and the output unit 250, respectively. The structure will be described sequentially.
[0153]
First, the configuration of the photosensitive portion 210 will be described with reference to FIGS. 8, 9A, 9B, 10A, and 10B.
[0154]
FIG. 8 is a plan view schematically showing the FT-CCD 300 of the third embodiment as described above.
[0155]
FIG. 9A is a plan view showing an outline of the charge transfer channel 211a shown in FIG. 8, and FIG. 9B is a plan view showing an outline of the charge transfer channel 211b shown in FIG.
[0156]
FIG. 10A is a plan view showing an outline of the first transparent transfer electrode for pixels 212 shown in FIG. 8, and FIG. 10B shows the second transparent transfer electrode for pixels 213 shown in FIG. It is a top view which shows an outline.
[0157]
As shown in FIG. 8, the photosensitive portion 210 is moved in a certain direction D._VA total of four charge transfer channels 211 a and a total of three charge transfer channels 211 b reaching the output transfer path 240 are formed on the surface of the semiconductor substrate 201. The charge transfer channel 211a and the charge transfer channel 211b are alternately formed in this order from the left or right in the figure. Each of the charge transfer channels 211a and 211b is obtained by forming an n-type region at a predetermined position of a p-type well formed on one surface side of the semiconductor substrate 201.
[0158]
Each of the charge transfer channels 211a has a total of four photoelectric conversion element formation regions 211aP for pixels (see FIG. 9A). Two adjacent photoelectric conversion element formation regions 211aP for one pixel in one charge transfer channel 211a are arranged at a constant pitch P by a connecting portion 211aC having a strip shape.₁₁Are connected. A connecting portion 211aC having a predetermined length is provided on the upstream side of the most upstream pixel photoelectric conversion element formation region 211aP and on the downstream side of the most downstream pixel photoelectric conversion element formation region 211aP in each charge transfer channel 211a. Is formed. The maximum line width in the pixel photoelectric conversion element formation region 211aP is wider than the line width of the connection portion 211aC.
[0159]
The direction D between the connection portion 211aC and the pixel photoelectric conversion element formation region 211aP adjacent to the connection 211aC._HAssuming a boundary line extending to (see FIGS. 1 and 8), each of the pixel photoelectric conversion element formation regions 211aP has an octagonal shape in plan view. The line width of each connection portion 211aC is substantially the same as the length of the boundary line.
[0160]
Each of the charge transfer channels 211b has the same shape as the charge transfer channel 211a described above. However, each of the pixel photoelectric conversion element formation region 211aP constituting the charge transfer channel 211a and each of the pixel photoelectric conversion element formation region 211bP (see FIG. 9B) constituting the charge transfer channel 211b are described above. Pitch P₁₁About half of the direction D_VIt is shifted to.
[0161]
Here, “pitch P” in this specification₁₁About 1/2 "is P₁₁/ 2 and P₁₁Also, a value in the vicinity of / 2 and a value within a range where the target pixel shifting arrangement can be performed is included.
[0162]
As shown in FIG. 8, four first transparent transfer electrodes 212 for pixels and two second transparent transfer electrodes 213 for pixels are formed on the photosensitive portion 10. These pixel transparent transfer electrodes 212 and 213 are arranged in the direction D described above._H, Each of the charge transfer channels 211a and 211b is crossed in plan view. Each of the pixel transparent transfer electrodes 212 and 213 has the pitch P in the order of the first transparent transfer electrode 212 for pixels and the second transparent transfer electrode 213 for pixels from the upstream side of the photosensitive portion 210.₁₁Are formed alternately one by one.
[0163]
Each of the pixel transparent transfer electrodes 212 and 213 is made of polysilicon, for example. These pixel transparent transfer electrodes 212 and 213 are formed on an electrical insulating film formed on the surface of the semiconductor substrate 201.
[0164]
As shown in FIG. 10A, the first transparent transfer electrode 212 for pixels has the same shape as the first transparent transfer electrode 12 for pixels in the FT-CCD 100 of the first embodiment described above. As shown in FIG. 10B, the second transparent transfer electrode for pixels 213 has the same shape as the second transparent transfer electrode for pixels 13 in the FT-CCD 100 of the first embodiment described above.
[0165]
Each of the pixel photoelectric conversion element formation regions 212P (see FIG. 10A) constituting the pixel first transparent transfer electrode 212 and the pixel photoelectric conversion element formation region 213P constituting the pixel second transparent transfer electrode 213. Each of (see FIG. 10B) is a predetermined pitch P₁₂About half of the direction D_H(See FIG. 8).
[0166]
Here, “pitch P” in this specification₁₂About 1/2 "is P₁₂/ 2 and P₁₂Also, a value in the vicinity of / 2 and a value within a range where the target pixel shifting arrangement can be performed is included.
[0167]
A total of four first transfer-dedicated transfer electrodes 215 are formed, one on each plane view, between the first transparent transfer electrode 212 for pixels and the second transparent transfer electrode 213 for pixels adjacent to the downstream side. (See FIG. 8). In addition, a total of three second transfer-dedicated transfer electrodes 216 are formed one by one between the second transparent transfer electrode 213 for pixels and the first transparent transfer electrode 212 for pixels adjacent to the downstream side thereof. .
[0168]
One third transfer-dedicated transfer electrode 217 is formed between the second transparent transfer electrode for pixel 213 provided on the most downstream side and the first transfer electrode for adjustment unit 231 described later. In addition, one auxiliary transfer electrode 218 is disposed close to the upstream side of the first transparent transfer electrode 212 for pixels provided on the most upstream side.
[0169]
Each transfer-dedicated transfer electrode 215, 216, 217 and auxiliary transfer electrode 218 are connected to each charge transfer channel 211a, 211b in the direction D._HAlong the plane view. These transfer-dedicated transfer electrodes 215, 216, 217 and auxiliary transfer electrode 218 are made of polysilicon, for example. Each transfer-dedicated transfer electrode 215, 216, 217 and auxiliary transfer electrode are formed on an electrical insulating film formed on the surface of the semiconductor substrate 201, like the pixel transparent transfer electrodes 212, 213.
[0170]
The pixel transparent transfer electrodes 212 and 213 and the transfer dedicated transfer electrodes 215 and 216 are the same as the pixel transparent transfer electrodes 12 and 13 and the transfer dedicated transfer electrodes 15 and 16 in the FT-CCD 100 of the first embodiment described above. In addition, a so-called superimposed transfer electrode structure is formed (see FIG. 8). The third transfer dedicated transfer electrode 217 and the pixel second transparent transfer electrode 213 provided on the most downstream side, and the auxiliary transfer electrode 218 and the first transparent transfer electrode 212 for the pixel provided on the most upstream side are also so-called overlapping. A combined transfer electrode structure is formed (see FIG. 8). These transfer electrodes are electrically insulated from each other.
[0171]
Each of the pixel photoelectric conversion element formation regions 212P in the pixel first transparent transfer electrode 212 intersects with the pixel photoelectric conversion element formation region 211aP in the charge transfer channel 211a in plan view. Each of the connection portions 212C in the pixel first transparent transfer electrode 212 intersects with the connection portion 211bC in the charge transfer channel 211b in plan view. On the other hand, each of the pixel photoelectric conversion element formation regions 213P in the pixel second transparent transfer electrode 213 intersects with the pixel photoelectric conversion element formation region 211bP in the charge transfer channel 211b in plan view. Each of the connection portions 213C in the pixel second transparent transfer electrode 213 intersects with the connection portion 211aC in the charge transfer channel 211a in plan view. Each transfer-dedicated transfer electrode 215, 216, 217 and auxiliary transfer electrode 218 intersect each of the charge transfer channels 211a or 211b in a plan view at a predetermined location.
[0172]
Each of the intersecting portions of the charge transfer channels 211a and 211b and the transparent transfer electrodes for pixels 212 and 213 in plan view has a function as one charge transfer stage. The same applies to each crossing portion of each charge transfer channel 211a, 211b and each transfer dedicated transfer electrode 215, 216, 217 or auxiliary transfer electrode 218 in plan view. Therefore, each of the charge transfer channels 211a and 211b includes the transparent transfer electrodes 212 and 213 for pixels and the transfer electrodes 215, 216 and 217 for transfer and the auxiliary transfer electrodes 218 that cross the charge transfer channels 211a and 211b in plan view. Together with this, one CCD (vertical transfer CCD) is constructed.
[0173]
As shown in FIG. 8, the light shielding film 220 is formed from the photosensitive portion 210 to the output transfer path 240. Between the light shielding film 220 and various transfer electrodes formed on the photosensitive section 210, the CCD storage section 230, and the output transfer path 240, for example, an electrical insulating layer (not shown) made of a silicon oxide film, for example. Intervenes.
[0174]
The light shielding film 220 is horizontally disposed on the intersection of the pixel photoelectric conversion element formation region 212P in the pixel first transparent transfer electrode 212 and the pixel photoelectric conversion element formation region 211aP in the charge transfer channel 211a in plan view. The opening 222 has a circular cross section. Also, the horizontal cross section is circular on the intersection of the pixel photoelectric conversion element formation region 213P in the pixel second transparent transfer electrode 213 and the pixel photoelectric conversion element formation region 211bP in the charge transfer channel 211b in plan view. The opening 222 is provided.
[0175]
For this reason, each of the intersections in the plan view of each pixel photoelectric conversion element formation region 212P and each pixel photoelectric conversion element formation region 211aP is a pixel photoelectric conversion element (photodiode) 225 (see FIG. 8). Function. The same applies to each of the intersecting portions in plan view of each pixel photoelectric conversion element formation region 213P and each pixel photoelectric conversion element formation region 211bP. Other intersections do not substantially function as photoelectric conversion elements.
[0176]
Therefore, in the FT-CCD 300 shown in FIG. 8, the direction D of the crossing portions of the charge transfer channels 211 a and 211 b and the pixel transparent transfer electrodes 212 and 213 in plan view is described._VAnd direction D_HIn any of the above, every other selected intersection functions as the pixel photoelectric conversion element 225. Each of the openings 222 functions as a light receiving unit for one pixel. Hereinafter, the opening 222 may be referred to as a “light receiving part 222”.
[0177]
In the FT-CCD 300, a total of 28 of these pixel photoelectric conversion elements 225 are arranged so as to be shifted by 7 pixels and 8 rows. One pixel photoelectric conversion element array is composed of four pixel photoelectric conversion elements 225. Each of the odd-numbered pixel photoelectric conversion element rows includes four pixel photoelectric conversion elements 225. Each of the even-numbered pixel photoelectric conversion element rows includes three pixel photoelectric conversion elements 225.
[0178]
Each of the odd-numbered photoelectric conversion element columns counted from the left end of FIG. 8 includes only the odd-numbered pixel photoelectric conversion elements 225, and each of the even-numbered photoelectric conversion element columns includes only the even-numbered pixel photoelectric conversion elements 225. including.
[0179]
When light enters from the light receiving unit 222, photoelectric conversion is performed in the pixel photoelectric conversion element 225. By applying predetermined voltages to the transparent transfer electrodes for pixels 212 and 213, the transfer electrodes for transfer 215, 216, and 217, and the auxiliary transfer electrode 218, signal charges are applied to the photoelectric conversion elements for the pixels 225, respectively. Can be accumulated.
[0180]
Next, the structure of the adjustment unit 230 will be described with reference to FIG. The adjusting unit 230 is formed on the surface of the semiconductor substrate 201 between the photosensitive unit 210 and the output transfer path 240.
[0181]
As shown in FIG. 8, each of the charge transfer channels 211a and 211b described above moves the adjusting unit 230 in the direction D._VAnd the output transfer path 240 is reached.
[0182]
One adjustment part first transfer electrode 231 and one adjustment part second transfer electrode 232 are formed in this order from the upstream side. These transfer electrodes 231 and 232 for the adjustment section respectively connect the charge transfer channels 211a and 211b in the direction D._HAlong the plane view.
[0183]
Each of the adjustment unit transfer electrodes 231 and 232 is made of, for example, polysilicon. These adjustment unit transfer electrodes 231 and 232 are formed on an electrical insulating film formed on the surface of the semiconductor substrate 201.
[0184]
The shapes of the individual transfer electrodes 231 and 232 for the adjustment unit are both band-like. The line width of the adjustment unit first transfer electrode 231 and the line width of the adjustment unit second transfer electrode 232 are substantially the same.
[0185]
The upstream edge of the adjustment unit second transfer electrode 232 overlaps the downstream edge of the adjustment unit first transfer electrode 231. The transfer electrodes 231 and 232 for the adjustment unit have a so-called overlapping transfer electrode structure. The adjustment unit transfer electrodes 231 and 232 are electrically insulated from each other.
[0186]
The edge on the third transfer dedicated transfer electrode 217 side in the first transfer electrode for adjustment unit 231 overlaps with the edge on the first transfer electrode for adjustment unit 231 side in the third transfer dedicated transfer electrode 217. The first transfer electrode 231 for adjustment unit and the third transfer dedicated transfer electrode 217 have a so-called overlapping transfer electrode structure. Note that the first transfer electrode 231 for the adjustment unit and the third transfer dedicated transfer electrode 217 are electrically insulated from each other.
[0187]
Each of the crossing portions of the charge transfer channels 211a and 211b and the adjustment unit transfer electrodes 231 and 232 in plan view has a function as one charge transfer stage. Each of the charge transfer channels 211a and 211b constitutes one CCD (vertical transfer CCD) together with the adjustment unit transfer electrodes 231 and 232 that cross the charge transfer channels 211a and 211b in plan view.
[0188]
As described above, the third transfer dedicated transfer electrode 217 and the adjustment unit first transfer electrode 231 have a so-called superimposed transfer electrode structure. As a result, each of the charge transfer channels 211a and 211b forms one CCD (vertical transfer CCD) from the photosensitive unit 210 to the adjustment unit 230.
[0189]
The light shielding film 220 described above is formed from the photosensitive portion 210 to the output transfer path 240 (see FIG. 8). The light shielding film 220 substantially covers each of the adjustment unit transfer electrodes 231 and 232 in plan view through an electrical insulating layer.
[0190]
The signal charges transferred from the photosensitive unit 210 are sequentially transferred to the output transfer path 240 by the vertical transfer CCDs formed in the adjustment unit 230. At this time, a predetermined voltage is applied to the adjustment unit transfer electrodes 231 and 232 at a predetermined cycle.
[0191]
In FIG. 8, the charge transfer channels 211a and 211b are drawn with solid lines and the light shielding film 220 is drawn with a two-dot chain line for easy understanding of the constituent members.
[0192]
Next, the structure of the output transfer path 240 will be described. The output transfer path 240 is formed on the surface of the semiconductor substrate 201 so as to be adjacent to the downstream end of the adjustment unit 230. In FIG. 8, the output transfer path 240 is simplified for easy understanding of FIG. 8, and the structure of the output transfer path 240 will be described below with reference to FIG. 11.
[0193]
FIG. 11 is a plan view schematically showing the output transfer path 240. As shown in the figure, a plurality of output transfer paths 240 are provided on the semiconductor substrate 201 via one charge transfer channel 241 formed on the semiconductor substrate 201 and an electric insulating film (not shown). The transfer electrodes 243, 244, 245, and 246 are formed. Electrical insulating layers (not shown) are formed on the transfer electrodes 243, 244, 245 and 246. The light shielding film 220 described above is formed on the electrical insulating layer.
[0194]
The charge transfer channel 241 is obtained by forming an n-type region at a predetermined position of a p-type well formed on one surface side of the semiconductor substrate 201. The charge transfer channel 241 has the direction D described above._HIt extends to.
[0195]
Each transfer electrode 243, 244, 245, 246 is in a ratio of one set to one charge transfer channel 211a or 211b, and from the output unit 250 (see FIG. 8) side, the transfer electrode 246, transfer electrode 245, transfer The electrodes 244 and the transfer electrodes 243 are formed in this order. These transfer electrodes 243, 244, 245, 246 are made of, for example, polysilicon.
[0196]
The transfer electrode 243 is a strip electrode that covers the downstream end of one charge transfer channel 211a or 211b in plan view and crosses the charge transfer channel 241 in plan view. The edge of the transfer electrode 243 on the adjustment unit second transfer electrode 232 side overlaps the edge of the adjustment unit second transfer electrode 232 on the output transfer path 240 side. Each of the transfer electrodes 244, 245, and 246 is a belt-like electrode that crosses the charge transfer channel 241 in plan view without overlapping any of the charge transfer channels 211a and 211b in plan view.
[0197]
The edge of the transfer electrode 244 on the transfer electrode 243 side overlaps the transfer electrode 243, and the edge of the transfer electrode 244 on the transfer electrode 245 side overlaps the transfer electrode 245. The edge of the transfer electrode 246 on the transfer electrode 245 side overlaps the transfer electrode 245, and the edge of the transfer electrode 244 other than the rightmost transfer electrode 244 in FIG. 4 is on the transfer electrode 246. overlapping. Each transfer electrode 243, 244, 245, 246 has a so-called superposition transfer electrode structure.
[0198]
The transfer electrodes 243, 244, 245, and 246 are insulated from each other by an electrical insulating film that is not shown. Further, the transfer electrode 243 and the first transfer electrode 231 for the most downstream adjustment unit are also insulated from each other by an electrical insulating film (not shown).
[0199]
Each crossing portion of the charge transfer channel 241 and each transfer electrode 243, 244, 245, 246 in plan view is used as one charge transfer stage. Therefore, the output transfer path 240 has four charge transfer stages per charge transfer channel 211a or 211b. The output transfer path 240 is composed of a four-phase drive type CCD.
[0200]
The signal charges transferred from the adjustment unit 230 are sequentially transferred to the output unit 250 by the charge transfer stages formed in the output transfer path 240. At this time, a predetermined voltage is applied to each of the transfer electrodes 243, 244, 245, and 246 at a predetermined cycle.
[0201]
Next, the output unit 250 will be described. As shown in FIG. 8, the output unit 250 is connected to one end of the output transfer path 240. The output unit 250 converts the signal charge sent from the output transfer path 240 into a signal voltage by a floating capacitor (not shown), and uses the signal voltage using a source follower circuit (not shown) or the like. Amplify. The electric charge after being detected (converted) is absorbed by the power supply via, for example, a reset transistor. A specific configuration is the same as that of the output unit 50 (see FIG. 1) described with reference to FIG.
[0202]
The FT-CCD 300 described above includes the light shielding film 220 having the opening 222 only on each of the pixel photoelectric conversion elements 225. For this reason, even if the microlens array is disposed on the photosensitive portion 210, stray light is unlikely to enter the pixel photoelectric conversion element 225 due to light wraparound.
[0203]
In addition, a pixel shift arrangement in which each pixel photoelectric conversion element 225 forms a wide region on every other charge transfer channel 211a or 211b is performed. For this reason, it is easy to improve the pixel density while suppressing a decrease in the area of the light receiving unit 222 in each pixel. As a result, it is easy to increase the resolution.
[0204]
Furthermore, the maximum line width of the pixel photoelectric conversion element formation region 211aP in the charge transfer channel 211a is wider than the line width of the connection portion 211aC in the charge transfer channel 211a. Similarly, the maximum line width of the pixel photoelectric conversion element formation region 211bP in the charge transfer channel 211b is wider than the line width of the connection portion 211bC in the charge transfer channel 211b. For this reason, it is easier to increase the size of each pixel photoelectric conversion element 225 in plan view than in the above-described FT-CCD 100 of the first embodiment. Accordingly, the size of the light receiving unit 222 of each pixel in a plan view can be easily increased. As a result, it is easier to improve the sensitivity than the FT-CCD 100 of the first embodiment.
[0205]
In order to drive the FT-CCD 300 shown in FIG. 8, each pixel transparent transfer electrode 212, 213, each transfer dedicated transfer electrode 215, 216, 217, auxiliary transfer electrode 218, each adjustment unit transfer electrode 231, 232 are driven. Drive pulse supply means for supplying predetermined drive pulses to the output transfer path 240 and the output unit 50 is used.
[0206]
FIG. 12 is a schematic diagram showing an example of the drive pulse supplying means. As shown in the figure, the drive pulse supply means 290 includes, for example, a synchronization signal generator 291, a timing generator 292, a vertical drive circuit 293, a horizontal drive circuit 294, and the like.
[0207]
The synchronization signal generator 291 generates various pulses necessary for signal processing, such as a vertical synchronization pulse and a horizontal synchronization pulse. The timing generator 292 is a four-phase vertical pulse signal applied to each pixel transparent transfer electrode 212, 213, each transfer dedicated transfer electrode 215, 216, 217, auxiliary transfer electrode 218, and each adjustment unit transfer electrode 231, 232. And a timing signal for a four-phase horizontal pulse signal and the like necessary for driving the output transfer path 240.
[0208]
The vertical drive circuit 293 generates a four-phase vertical pulse signal based on the timing signal. Each of the four-phase vertical pulse signals is transmitted through a pulse supply terminal 260a, 260b, 260c, or 260d, a predetermined pixel transparent transfer electrode 212, 213, a predetermined transfer dedicated transfer electrode 215, 216, 217, an auxiliary device. The voltage is applied to the transfer electrode 218 or the predetermined adjustment unit transfer electrodes 231 and 232.
[0209]
The horizontal drive circuit 294 generates a four-phase horizontal pulse signal based on the timing signal. Each of the four-phase horizontal pulse signals is applied to predetermined transfer electrodes 243, 244, 245, and 246 via pulse supply terminals 280a, 280b, 280c, and 280d.
[0210]
In the FT-CCD 300, signal charges are temporarily accumulated in each of the pixel photoelectric conversion elements 225, and then these signal charges are transferred from the photosensitive unit 210 to the adjustment unit 230. Signal charges are transferred from the photosensitive unit 210 to the adjusting unit 230 by driving the above-described vertical transfer CCDs formed from the photosensitive unit 210 to the adjusting unit 230 in four phases.
[0211]
In order to store and transfer signal charges, for example, as shown in FIG. 12, a predetermined pulse supply terminal and a predetermined transfer electrode are electrically connected. In FIG. 12, each second transfer-dedicated transfer electrode 216, third transfer-dedicated transfer electrode 217, and auxiliary transfer electrode 218 are electrically connected to the pulse supply terminal 260a. Each transparent first bright transfer electrode 212 for pixels and first transfer electrode 231 for adjustment unit are electrically connected to the pulse supply terminal 260b, respectively. Each of the first transfer dedicated transfer electrodes 215 and the second transfer electrode for adjustment unit 232 are electrically connected to the pulse supply terminal 260c, respectively. Each pixel second transparent transfer electrode 213 is electrically connected to a pulse supply terminal 260d.
[0212]
By performing the same operation as that in the FT-CCD 100 of the first embodiment described above during the signal charge accumulation period defined by the predetermined pulse generated by the synchronization signal generator 291, the signal charge for one frame is obtained. Can be accumulated in the photosensitive section 210. At this time, the low level vertical pulse signal V is supplied from the vertical drive circuit 293 to the pulse supply terminals 260a and 260c._LIs applied to the pulse supply terminals 260b and 260d and a high level vertical pulse signal V is applied to the pulse supply terminals 260b and 260d._HIs applied.
[0213]
The signal charge for one frame is adjusted from the photosensitive portion 210 by performing the same operations as the operations (1) to (4) described in the description of the FT-CCD 100 of the first embodiment. Can be transferred to the unit 230.
[0214]
However, all of the signal charges for one frame cannot be stored in the adjustment unit 230. The signal charges transferred to the most downstream charge transfer stages in the adjustment unit 230 are sequentially transferred to the output transfer path 240.
[0215]
As shown in FIG. 8 or FIG. 12, the most downstream charge transfer stage in the adjustment unit 230 is formed at the intersection of the adjustment unit second transfer electrode 232 and the charge transfer channels 211a and 211b in plan view. Each charge transfer stage. In these charge transfer stages, each of the signal charges accumulated in each pixel photoelectric conversion element 225 constituting two adjacent pixel photoelectric conversion element rows is substantially subjected to the same phase shift. It is transferred at the same time.
[0216]
The transfer of signal charges from the adjustment unit 230 to the output transfer path 240 is performed as follows, for example.
[0217]
First, as shown in FIG. 11, the pulse supply terminal 280a and each transfer electrode 243, the pulse supply terminal 280b and each transfer electrode 244, the pulse supply terminal 280c and each transfer electrode 245, and the pulse supply terminal 280d and each transfer electrode 246 are electrically connected to each other in advance.
[0218]
Then, V is applied to the second transfer electrode 232 for the adjustment unit._HIs applied from the horizontal drive circuit 294 to each of the pulse supply terminals 280a and 280b._HAnd a low level horizontal pulse H from the horizontal drive circuit 294 to each of the pulse supply terminals 280c and 280d._LApply. Next, the voltage of the second transfer electrode 232 for the adjustment unit is set to V_HTo V_LTo change.
[0219]
As a result, each signal charge that has been transferred to each of the charge transfer stages formed at the intersection of the second transfer electrode 232 for adjustment unit and each charge transfer channel 211a, 211b in plan view is transferred to the charge transfer unit. The channel 241 and each transfer electrode 243, 244 are transferred to each of the two series of potential wells formed at each intersection in plan view. That is, the signal charge is transferred from the adjustment unit 230 to the output transfer path 240.
[0220]
Each of the signal charges transferred to the output transfer path 240 is sequentially transferred through the output transfer path 240 toward the output unit 250. At this time, the following operations (a) and (b) are alternately performed at a predetermined cycle.
[0221]
(a) Each of the two potential wells is extended to the downstream side (output unit 250 side) by one charge transfer stage to form a triple potential well.
[0222]
(b) Each of the triple potential wells is reduced by one charge transfer stage from the upstream side to the downstream side (output unit 250 side) to form a double potential well.
[0223]
The output unit 250 sequentially converts and amplifies the signal charge received from the output transfer path 240 into a signal voltage, and outputs the amplified signal voltage to a predetermined circuit.
[0224]
Since the FT-CCD 300 of the third embodiment described above does not have a CCD accumulating unit, it is suitable as an area image sensor for a still image capturing device having a mechanical shutter. When the FT-CCD 300 is used as the area image sensor, the mechanical shutter is opened during the signal accumulation period and closed during the other periods.
[0225]
The illustrated FT-CCD 300 is an FT-CCD having a simple structure as an FT-CCD. In an actual FT-CCD, a microlens array is usually provided in order to increase the photoelectric conversion efficiency in each pixel photoelectric conversion element 225. In addition, a color filter array is provided in an FT-CCD for color imaging.
[0226]
The microlens array and the color filter array can be arranged by the method described in the description of the FT-CCD 200 of the second embodiment, for example.
[0227]
Next, an FT-CCD according to a fourth embodiment will be described with reference to FIGS.
[0228]
FIG. 13 is a plan view schematically showing an FT-CCD 400 according to the fourth embodiment. FIG. 14 is a plan view schematically showing the pixel transparent transfer electrode 312 shown in FIG. FIG. 13 is a schematic diagram in which the number of pixel photoelectric conversion elements 325 is 24 in order to make the drawing easier to see and to make the explanation easier to understand. In an actual FT-CCD, as described above, the number of pixel photoelectric conversion elements 325 reaches 1000 to 10 million.
[0229]
The FT-CCD 400 shown in FIG. 13 is formed on the surface of the semiconductor substrate 301, the CCD storage unit 330 formed outside the photosensitive unit 310, and the outside of the CCD storage unit 330. The output transfer path 340, the output unit 350 connected to one end of the output transfer path 340, and four pulse supply terminals for supplying a predetermined drive pulse to each of the transfer electrodes formed in the photosensitive part 310 360 a, 360 b, 360 c, 360 d, four pulse supply terminals 370 a, 370 b, 370 c, 370 d for supplying a predetermined drive pulse to each of the transfer electrodes formed in the CCD accumulator 330, and an output transfer path 340 Four pulse supply terminals 380a, 380b, 380c, 38 for supplying a predetermined drive pulse to each of the transfer electrodes constituting It is and a d.
[0230]
The roles of the photosensitive unit 310, the CCD storage unit 330, the output transfer path 340, and the output unit 350 are as follows: the photosensitive unit 10, the CCD storage unit 30, the output transfer path 40, or the output unit in the FT-CCD 100 of the first embodiment described above. It is substantially the same as 50 roles.
[0231]
Hereinafter, taking the case of using a semiconductor substrate 301 (see FIG. 13) made of an n-type silicon substrate having a p-type well as an example, the configuration of the photosensitive unit 310, the CCD storage unit 330, the output transfer path 340, and the output unit 350, respectively. The structure will be described sequentially.
[0232]
As shown in FIG. 13, the photosensitive portion 310 is moved in a certain direction D._VA total of six charge transfer channels 311 are formed on the surface of the semiconductor substrate 301 to reach the output transfer path 340 across the line (see FIG. 13). Each of the charge transfer channels 311 is obtained by forming an n-type region at a predetermined position of a p-type well formed on one surface side of the semiconductor substrate 301. Each charge transfer channel 311 has a band shape in plan view.
[0233]
Each charge transfer channel 311 is moved in the direction D_HA total of four transparent transfer electrodes 312 for pixels crossing in plan view along the direction D_VA constant pitch P along₂₀It is formed with. Each pixel transparent transfer electrode 312 is made of, for example, polysilicon. These pixel transparent transfer electrodes 312 are formed on an electrical insulating film formed on the surface of the semiconductor substrate 301.
[0234]
Each of the pixel transparent transfer electrodes 312 has the direction D described above._HMeandering along. The transparent transfer electrode 312 for each pixel has the direction D described above._HA total of six photoelectric conversion element formation regions 312P for pixels extending in a strip shape (see FIG. 13) (see FIG. 14).
[0235]
In the first, third, and fifth pixel photoelectric conversion element formation regions 312P counted from the left end in FIG. 14, a connection portion 312C that extends diagonally to the left in the drawing is provided at the left end.₁Is connected to the right end of the connecting portion 312C extending diagonally downward to the right in the figure.₂It is continuing. Therefore, in the second and fourth pixel photoelectric conversion element formation regions 312P counted from the left end in FIG. 14, the left end is a connection portion 312C extending obliquely upward to the left in the drawing.₂Is connected to the right end of the connecting portion 312C extending diagonally to the right in the figure.₁It is continuing.
[0236]
The leftmost connecting portion 312C in FIG.₁At the left end of the direction D_HA connecting portion 312C extending in a strip shape_ThreeIt is continuing. At the right end of the sixth pixel photoelectric conversion element formation region 312P counted from the left end of FIG._HA connecting portion 312C extending in a strip shape_FourIt is continuing.
[0237]
The first transfer-dedicated transfer electrode 315, the second transfer-dedicated transfer electrode 316, and the third transfer-dedicated transfer electrode 317 are formed one by one between two adjacent pixel transparent transfer electrodes 312. Further, the first transfer-dedicated transfer electrode 315, the second transfer-dedicated transfer electrode 316, and the third transfer-dedicated transfer electrode 317 are also formed on the upstream side of the most upstream pixel transparent transfer electrode 312. Yes. These transfer-dedicated transfer electrodes 315, 316, and 317 are formed in this order from the upstream side to the downstream side of the photosensitive portion 310.
[0238]
Each of the transfer-dedicated transfer electrodes 315, 316, and 317 has a shape in plan view that is almost the same as the shape in plan view of the pixel transparent transfer electrode 312 except that the line width is narrower than the pixel transparent transfer electrode 312. The same. Each transfer-dedicated transfer electrode 315, 316, 317 is made of polysilicon, for example. These transfer-dedicated transfer electrodes 315, 316, and 317 are formed on an electrical insulating film formed on the surface of the semiconductor substrate 301, similarly to the pixel transparent transfer electrode 312.
[0239]
The upstream edge of each first transfer dedicated transfer electrode 315 is the downstream edge of the pixel transparent transfer electrode 312 adjacent to the upstream, excluding the first transfer dedicated transfer electrode 315 provided at the most upstream. It overlaps the part. The downstream edge of each first transfer dedicated transfer electrode 315 overlaps with the upstream edge of the second transfer dedicated transfer electrode 316 adjacent to the downstream side.
[0240]
The upstream edge of each third transfer dedicated transfer electrode 317 overlaps the downstream edge of the second transfer dedicated transfer electrode 316 adjacent to the upstream side. The downstream edge of each third transfer dedicated transfer electrode 315 is the upstream edge of the pixel transparent transfer electrode 312 adjacent to the downstream thereof except for the third transfer dedicated transfer electrode 317 provided on the most downstream side. It overlaps the part.
[0241]
Each pixel transparent transfer electrode 312 and each transfer dedicated transfer electrode 315, 316, 317 have a so-called overlapping transfer electrode structure. These transfer electrodes are electrically insulated from each other.
[0242]
A fourth transfer-dedicated transfer electrode 318 is formed downstream of the most downstream first transfer-dedicated transfer electrode 315. A fifth transfer-dedicated transfer electrode 319 having a strip shape is formed downstream of the fourth transfer-dedicated transfer electrode 318. These transfer-dedicated transfer electrodes 318 and 319 respectively connect the charge transfer channels 411 in the direction D in plan view._HCross along.
[0243]
The fourth transfer dedicated transfer electrode 318 and the fifth transfer dedicated transfer electrode 319 are made of, for example, polysilicon. These transfer-dedicated transfer electrodes 318 and 319 are formed on an electrical insulating film formed on the surface of the semiconductor substrate 301, similarly to the pixel transparent transfer electrode 312.
[0244]
The upstream edge portion of the fourth transfer-dedicated transfer electrode 318 enters under the downstream edge portion of the first transfer-dedicated transfer electrode 315 provided on the most downstream side. The downstream edge of the fourth transfer dedicated transfer electrode 318 overlaps the upstream edge of the fifth transfer dedicated transfer electrode 319.
[0245]
The most downstream first transfer electrode 315, fourth transfer electrode 318, and fifth transfer electrode 319 have a so-called overlapping transfer electrode structure. These transfer-dedicated transfer electrodes are electrically insulated from each other.
[0246]
Each of the pixel photoelectric conversion element formation regions 312P in the pixel transparent transfer electrode 312 intersects the charge transfer channel 311 in plan view. Each transfer-dedicated transfer electrode 315, 316, 317, 318, 319 intersects each of the charge transfer channels 311 at a predetermined position in plan view.
[0247]
Each intersecting portion in plan view between each charge transfer channel 311 and each pixel photoelectric conversion element formation region 312P has a function as one charge transfer stage. The same applies to each crossing portion of each charge transfer channel 311 and each transfer dedicated transfer electrode 315, 316, 317, 318, 319 in plan view.
[0248]
Each of the charge transfer channels 311 has one CCD (vertical transfer CCD) together with the pixel transparent transfer electrode 312 and the transfer dedicated transfer electrodes 315, 316, 317, 318, and 319 that cross the charge transfer channel 311 in plan view. Configure.
[0249]
As shown in FIG. 13, the light shielding film 320 is formed from the photosensitive portion 310 to the output transfer path 340. Between the light shielding film 320 and various transfer electrodes formed on the photosensitive section 310, the CCD storage section 330, and the output transfer path 340, an electrical insulating layer (not shown) made of, for example, a silicon oxide film. Intervenes.
[0250]
The light shielding film 320 has an opening 322 having a circular horizontal cross section on each of the intersecting portions of each pixel photoelectric conversion element forming region 312P and each charge transfer channel 311 in plan view.
[0251]
For this reason, each of the intersecting portions of each pixel photoelectric conversion element formation region 312P and each charge transfer channel 311 in plan view functions as a pixel photoelectric conversion element (photodiode) 325 (see FIG. 13). Other intersections do not substantially function as photoelectric conversion elements.
[0252]
Accordingly, in the FT-CCD 400 shown in FIG. 13, all of the crossing portions of the charge transfer channels 311 and the pixel transparent transfer electrodes 312 in plan view function as the pixel photoelectric conversion elements 325. Each of the openings 322 functions as a light receiving portion of one pixel. Hereinafter, the opening 322 may be referred to as a “light receiving part 322”.
[0253]
In the FT-CCD 400, a total of 24 of these pixel photoelectric conversion elements 325 are arranged so as to be shifted in pixels over 6 columns and 8 rows. One pixel photoelectric conversion element array is constituted by four pixel photoelectric conversion elements 325, and one pixel photoelectric conversion element row is also constituted by four pixel photoelectric conversion elements 325. Each odd-numbered photoelectric conversion element column counted from the left end in FIG. 13 includes only odd-numbered pixel photoelectric conversion elements 325, and each even-numbered photoelectric conversion element column includes only even-numbered pixel photoelectric conversion elements 325. including.
[0254]
When light enters from the light receiving unit 322, photoelectric conversion is performed in each of the pixel photoelectric conversion elements 325. By applying a predetermined voltage to each pixel transparent transfer electrode 312 and each transfer dedicated transfer electrode 315, 316, 317, 318, 319, signal charges are accumulated in each pixel photoelectric conversion element 325. Can do.
[0255]
Next, the structure of the CCD storage unit 330 will be described with reference to FIG. The CCD storage unit 330 is formed on the surface of the semiconductor substrate 301 between the photosensitive unit 310 and the output transfer path 340.
[0256]
As shown in FIG. 13, each of the charge transfer channels 311 moves the CCD storage unit 330 in the direction D._VAnd the output transfer path 340 is reached.
[0257]
Each of the charge transfer channels 311 moves in the direction D_HA total of four first transfer electrodes 331 for the storage section and a total of three second transfer electrodes 332 for the storage section that cross in plan view along the direction D._VAre formed alternately one by one. Further, one third storage portion transfer electrode 333 is formed on the downstream side of the most downstream storage portion first transfer electrode 331. The third transfer electrode 333 for the storage portion also has each of the charge transfer channels 311 in the direction D._HAlong the plane view.
[0258]
Each storage unit transfer electrode 331, 332, 333 is made of, for example, polysilicon. The storage unit transfer electrodes 331, 332, and 333 are formed on an electrical insulating film formed on the surface of the semiconductor substrate 301. Each of the storage unit transfer electrodes 331, 332, and 333 has a strip shape.
[0259]
The upstream edge of each storage unit second transfer electrode 332 overlaps with the downstream edge of the storage unit first transfer electrode 331 adjacent to the upstream in plan view. The downstream edge of each storage unit second transfer electrode 332 overlaps the upstream edge of the storage unit first transfer electrode adjacent to the downstream side in plan view. The upstream edge of the storage unit third transfer electrode 333 overlaps with the downstream edge of the storage unit first transfer electrode 331 adjacent to the upstream in plan view. Each storage unit transfer electrode 331, 332, 333 has a so-called superposition transfer electrode structure.
[0260]
Each crossing portion of each charge transfer channel 311 and each storage portion transfer electrode 331, 332, 333 in plan view functions as one charge transfer stage. Each of the charge transfer channels 311 constitutes one CCD (vertical transfer CCD) together with the storage unit transfer electrodes 331, 332, and 333 that cross the charge transfer channel 311 in plan view.
[0261]
The fifth transfer-dedicated transfer electrode 319 formed in the photosensitive section 310 and the most upstream storage section first transfer electrode 331 formed in the CCD storage section 330 form a so-called superimposed transfer electrode structure. As a result, each of the charge transfer channels 311 forms one CCD (vertical transfer CCD) from the photosensitive unit 310 to the CCD storage unit 330.
[0262]
The light shielding film 320 described above is formed from the photosensitive portion 310 to the output transfer path 340 (see FIG. 13). The light shielding film 320 covers the storage unit transfer electrodes 331, 332, and 333 in plan view with an electric insulating layer interposed therebetween.
[0263]
The signal charges accumulated in the photosensitive portion 310 are sequentially transferred toward the output transfer path 340 by the vertical transfer CCDs formed from the photosensitive portion 310 to the CCD accumulation portion 330.
[0264]
In FIG. 13, in order to make each component easy to understand, each charge transfer channel 311 is drawn by a solid line, and the light shielding film 320 is drawn by a two-dot chain line.
[0265]
The output transfer path 340 in the FT-CCD 400 shown in FIG. 13 has basically the same structure as the output transfer path 40 shown in FIG. Further, the output unit 350 in the FT-CCD 400 shown in FIG. 13 has basically the same structure as the output unit 50 shown in FIG. However, the output transfer path 40 shown in FIG. 4 is configured to receive the signal charges transferred through the eight charge transfer channels 11, whereas the output transfer path 340 shown in FIG. The signal charge transferred through the charge transfer channel 311 is received.
[0266]
The FT-CCD 400 described above includes the light shielding film 320 having the opening 322 only on each of the pixel photoelectric conversion elements 325. For this reason, even if the microlens array is disposed on the photosensitive portion 310, stray light is unlikely to enter the pixel photoelectric conversion element 225 due to light wraparound.
[0267]
Further, since the pixel shifting arrangement is performed, it is easy to improve the pixel density while suppressing a decrease in the area of the light receiving portion 322 in each pixel. As a result, it is easy to increase the resolution.
[0268]
Furthermore, said direction D_VSince the three charge transfer stages are formed between the two pixel photoelectric conversion elements 325 adjacent to each other along the line, the following advantages are obtained. That is, smearing and color mixing between the pixel photoelectric conversion elements 325 are further suppressed as compared to the FT-CCD 100 of the first embodiment and the FT-CCD 300 of the second embodiment. be able to.
[0269]
In order to drive the FT-CCD 400 shown in FIG. 13, each pixel transparent transfer electrode 312, each transfer dedicated transfer electrode 315, 316, 317, 318, 319, each storage unit transfer electrode 331, 332, 333, Drive pulse supply means for supplying a predetermined drive pulse to the output transfer path 340 and the output unit 350 is used.
[0270]
FIG. 15 is a schematic diagram showing an example of the drive pulse supplying means and the FT-CCD 400. As shown in the figure, the drive pulse supply means 390 is similar to the drive pulse supply means 90 shown in FIG. 5, for example, with a synchronization signal generator 391, timing generator 392, first vertical drive circuit 393a, 2 includes a vertical drive circuit 393b, a horizontal drive circuit 394, and the like.
[0271]
The synchronization signal generator 391, the timing generator 392, the first vertical drive circuit 393a, the second vertical drive circuit 393b, and the horizontal drive circuit 394 each play a role in the synchronization signal generator 91 in the drive pulse supply means 90 shown in FIG. The role of the timing generator 92, the first vertical driving circuit 93a, the second vertical driving circuit 93b, or the horizontal driving circuit 94 is the same.
[0272]
After the signal charges are temporarily accumulated in each of the pixel photoelectric conversion elements 325, these signal charges are transferred from the photosensitive unit 310 to the CCD accumulation unit 330. The transfer of signal charges from the photosensitive unit 310 to the CCD storage unit 330 is performed by, for example, driving each of the vertical transfer CCDs formed from the photosensitive unit 310 to the CCD storage unit 330, for example, in four phases.
[0273]
In order to store and transfer signal charges, for example, as shown in FIG. 15, a predetermined pulse supply terminal and a predetermined transfer electrode are electrically connected. In FIG. 15, each pixel transparent transfer electrode 312 is electrically connected to a pulse supply terminal 360a. Each first transfer dedicated transfer electrode 315 is electrically connected to the pulse supply terminal 360b. Each of the second transfer-dedicated transfer electrodes 316 and the fourth transfer-dedicated transfer electrode 318 is electrically connected to the pulse supply terminal 360c. Each third transfer-dedicated transfer electrode 317 and the fifth transfer-dedicated transfer electrode 319 are electrically connected to the pulse supply terminal 360d, respectively.
[0274]
In FIG. 15, the first and fifth storage unit transfer electrodes (both storage unit first transfer electrodes 331) counted from the photosensitive unit 310 side are electrically connected to the pulse supply terminal 370 a. . The second and sixth storage unit transfer electrodes (both storage unit second transfer electrodes 332) counted from the photosensitive unit 310 side are electrically connected to the pulse supply terminal 370b, respectively. The third and seventh storage unit transfer electrodes (both storage unit first transfer electrodes 331) counted from the photosensitive unit 310 side are electrically connected to the pulse supply terminal 370c, respectively. Each of the fourth and eighth storage unit transfer electrodes (second storage unit transfer electrode 332 and third storage electrode transfer unit 333) counted from the photosensitive unit 310 side is electrically connected to the pulse supply terminal 370d. ing.
[0275]
During a signal charge accumulation period defined by a predetermined pulse generated by the synchronization signal generator 391, a low-level vertical pulse signal V is sent from the first vertical drive circuit 393a to each pulse supply terminal 360b, 360c, 360d._LIs applied, and a high-level vertical pulse signal V is applied to the pulse supply terminal 360a._HIs applied.
[0276]
As a result, a potential well (a series of potential wells) is formed in each of the pixel photoelectric conversion elements 325, and signal charges for one frame can be accumulated in the photosensitive portion 310. At this time, a triple potential barrier is formed between two adjacent potential wells. Triple potential barriers are also formed on the upstream side of the most upstream potential well and on the downstream side of the most downstream potential well.
[0277]
The transfer of the signal charge in the photosensitive portion 310 can be performed by, for example, repeating the following operations (a) and (b) alternately in this order.
[0278]
(a) Each of a series of potential wells is extended downstream by two charge transfer stages to form a triple series of potential wells. In other words, each of the triple potential barriers is reduced downstream by two charge transfer stages to form a single potential barrier.
[0279]
(b) Each of the triple potential wells formed in (a) is shrunk downstream by one charge transfer stage to form a double potential well. In other words, each of the series of potential barriers formed in (a) is extended downstream by one charge transfer stage to form a series of potential barriers.
[0280]
Each signal charge transferred to each of the charge transfer stages formed at the intersection of the fifth transfer dedicated transfer electrode 319 and each charge transfer channel 311 in plan view is, for example, the following (i) and (ii) ) In this order, the image is transferred from the photosensitive unit 310 to the CCD storage unit 330.
[0281]
(i) A high level horizontal pulse signal H from the horizontal drive circuit 394 to each of the pulse supply terminals 80a and 80b._HAnd a low level horizontal pulse signal H from the second vertical drive circuit 393b to each of the pulse supply terminals 80c and 80d._LApply. At this time, in each of the above-described vertical transfer CCDs in the CCD accumulation unit 330, two potential wells and two potential barriers are alternately formed in this order from upstream to downstream.
[0282]
(ii) The state of (i) is changed from the first vertical drive circuit 393a to the pulse supply terminal 360d._LContinue until is applied.
[0283]
In transferring the signal charge in the CCD accumulating unit 330, the magnitude of the vertical pulse signal applied to the accumulating unit transfer electrodes 331, 332, 333 is appropriately changed, and the charge accumulating region (potential well), potential barrier, Are arranged alternately. The signal charge can be transferred by, for example, repeating the following operations (A) and (B) alternately in this order.
[0284]
(a) Each of the two potential wells (see (i) above) is shrunk downstream by one charge transfer stage to form a single potential well.
[0285]
(b) Each of the series of potential wells is extended downstream by one charge transfer stage to form a series of potential wells.
[0286]
The transfer of the signal charge from the CCD storage unit 330 to the output transfer path 340 can be performed by the same operation as that in the FT-CCD 100 of the first embodiment described above. Further, the transfer of the signal charge in the output transfer path 340 can be performed by the same operation as the operation in the FT-CCD 100 of the first embodiment described above.
[0287]
The output unit 350 sequentially converts and amplifies the signal charge received from the output transfer path 340 into a signal voltage, and outputs the amplified signal voltage to a predetermined circuit.
[0288]
After the transfer of the signal charge for one frame to the output transfer path 340, the signal charge accumulation period described above is defined again by a predetermined pulse generated by the synchronization signal generator 391 as necessary, and the photosensitive portion 310 The signal charge for one frame is accumulated in the frame.
[0289]
The FT-CCD 400 of the fourth embodiment described above is an FT-CCD having a simple structure as the FT-CCD. In an actual FT-CCD, a microlens array is usually provided in order to increase the photoelectric conversion efficiency in the pixel photoelectric conversion element 325. In addition, a color filter array is provided in an FT-CCD for color imaging.
[0290]
The microlens array and the color filter array can be arranged by the method described in the description of the FT-CCD 200 of the second embodiment, for example.
[0291]
While the FT-CCD of the present invention has been described with reference to the embodiments, the present invention is not limited to the above-described embodiments. It will be apparent to those skilled in the art that various modifications, improvements, combinations, and the like can be made.
[0292]
For example, the FT-CCD of each embodiment is such that a pixel photoelectric conversion element (photodiode) or the like is formed on an n-type semiconductor substrate having a p-type well. However, all the conductivity types can be reversed. For example, an FT-CCD can be obtained by forming a pixel photoelectric conversion element (photodiode) or the like on a p-type semiconductor substrate.
[0293]
In this specification, the term “semiconductor substrate” also includes a semiconductor layer for forming a pixel photoelectric conversion element (photodiode) or the like on one surface of a substrate made of a material other than a semiconductor. To do.
[0294]
The charge transfer channel formed from the photosensitive portion to the output transfer path may have a strip shape as exemplified in the first embodiment or the fourth embodiment, or exemplified in the third embodiment. It may have such a pixel photoelectric conversion element formation region. Furthermore, it may have a meandering shape.
[0295]
It is possible to further increase the pixel density by meandering the charge transfer channel in a plan view and disposing the pixels.
[0296]
When the pixel photoelectric conversion element formation region is provided in the charge transfer channel, the shape of the pixel photoelectric conversion element formation region in a plan view can be appropriately selected from, for example, a rectangle, a hexagon, an octagon, and a barrel shape.
[0297]
In this specification, “the shape of the pixel photoelectric conversion element formation region in a plan view” as referred to with respect to the charge transfer channel means a pixel photoelectric conversion element formation region and a connection portion following the pixel photoelectric conversion element formation region. During the direction D_HThis means the shape of the pixel photoelectric conversion element forming region defined in plan view when a boundary line extending in the direction is assumed.
[0298]
The shape of the transparent transfer electrode for pixels in plan view can also be selected as appropriate. When the pixel photoelectric conversion element formation region is provided in the pixel transparent transfer electrode, the shape of the pixel photoelectric conversion element formation region in plan view may be, for example, rectangular, hexagonal, octagonal, barrel shape, or the like. it can.
[0299]
In this specification, “the shape of the pixel photoelectric conversion element formation region in a plan view” as referred to with respect to the pixel transparent transfer electrode means a connection following the pixel photoelectric conversion element formation region and the pixel photoelectric conversion element formation region. Direction D between_VThis means the shape of the pixel photoelectric conversion element forming region defined in plan view when a boundary line extending in the direction is assumed.
[0300]
The shape of the opening provided in the light shielding film in plan view, and hence the shape of the light receiving portion of the pixel in plan view, for example, rectangular (including rhombus), hexagon, octagon, ellipse, circle, etc. Can be selected.
[0301]
Each of the pixel photoelectric conversion elements may be arranged so as to be shifted in pixels, or may be arranged in a normal matrix (lattice).
[0302]
The signal charge transfer method is not limited to the transfer method described in the embodiment, and can be appropriately changed according to the intended use of the FT-CCD. Along with this, the number of pulse supply terminals for supplying a predetermined drive pulse to each transfer electrode and the specification of the connection between the pulse supply terminal and each transfer electrode are also the signals in the target FT-CCD. It can be appropriately changed according to the charge transfer method. The same applies to the output transfer path.
[0303]
The CCD accumulating unit and the adjusting unit are not essential components. The signal charge may be transferred and connected immediately from the photosensitive portion to the output transfer path.
[0304]
The shape of each transfer electrode in the CCD accumulating unit or the adjusting unit is not limited to the belt shape illustrated in each example. For example, a plurality of types of electrodes having different shapes may be arranged with the same specification as the arrangement specification of each transfer electrode in the photosensitive portion.
[0305]
When one microlens is provided above each light receiving portion so as to cover the light receiving portion of each pixel in plan view, the shape of each microlens in plan view is rectangular (including rhombus), A shape with rounded corners of a rectangle, a pentagon or more polygon with all interior angles being obtuse, a shape with rounded corners of the polygon, a circle, an ellipse, etc. can be selected as appropriate. . The shape of the microlens in plan view can be selected as appropriate according to the shape of the light receiving portion in each pixel.
[0306]
The direction D in these microlenses_VThe pitch of the direction D_VThe pitch may be the same as or slightly different from the pitch of the pixel photoelectric conversion elements. Said direction D_VWhen the pitch of the microlens is made different from the pitch of the pixel photoelectric conversion element, each microlens is moved under the following viewpoint, for example.
[0307]
That is, the image forming position by the microlens is more advantageous in obtaining a desired location in the light receiving portion of the pixel, for example, a desired sensitivity or resolution, corresponding to the change in the incident direction of the incident light according to the change in the position in the light receiving portion. It moves so that it may be displaced to the location. In order to increase the sensitivity or resolution of the pixel, it is preferable that the photoelectric conversion region exists over the widest possible range around the image forming position by the microlens.
[0308]
For the same reason, the direction D in each of the above microlenses._HThe pitch of the direction D_HThe pitch may be the same as or slightly different from the pitch of the pixel photoelectric conversion elements.
[0309]
When the relative positional relationship between the pixel photoelectric conversion element and the microlens is substantially the same in all pixels, the position of the image point formed on the pixel photoelectric conversion element by the microlens is The central portion of the pixel photoelectric conversion element row is different from the upper or lower portion in the row direction. In order to prevent the position of the image point formed on the photoelectric conversion element for the pixel by the microlens from deviating from the desired position, for example, the microlens can be shifted as shown in (1) to (3) below. preferable.
[0310]
(1) As schematically shown in FIG. 16A, the positions of the microlenses 510 in the upper and lower column directions of each pixel photoelectric conversion element column 500 are defined as the pixel photoelectric conversion element column 500. As you move away from the center of the column, shift to the center of the column. The arrow in the figure indicates the direction in which the microlens 510 is shifted. Reference numeral 520 indicates a photosensitive portion.
[0311]
(2) As schematically shown in FIG. 16B, the position of the microlens 510 at the end in the row direction in each pixel photoelectric conversion element row 505 increases as the distance from the center of the photosensitive portion 520 increases. Direction D_HAre moved toward the center of the photosensitive portion 520. The arrow in the figure indicates the direction in which the microlens 510 is shifted.
[0312]
(3) As schematically shown in FIG. 16C, as the microlens 510 is moved away from the center of the photosensitive portion 520, the direction D is increased._HAnd said direction D_VAre moved toward the center of the photosensitive portion 520. The arrow in the figure indicates the direction in which the microlens 510 is shifted.
[0313]
Luminance shading can be improved by shifting the microlenses as described in (1) to (3) above.
[0314]
When a color filter array is provided in the FT-CCD, the color filter array only needs to be configured by a color filter that enables color imaging. As such a color filter array, there is a so-called complementary color filter array in addition to the three primary color (red, green, blue) color filter arrays mentioned in the embodiments.
[0315]
Complementary color filter arrays include, for example, (i) green (G), cyan (Cy) and yellow (Ye) color filters, (ii) cyan (Cy), yellow (Ye) and white or colorless (W) Each color filter, (iii) Cyan (Cy), magenta (Mg), yellow (Ye) and green (G) color filters, or (iv) cyan (Cy), yellow (Ye), green (G) and white Alternatively, it can be constituted by colorless (W) color filters or the like.
[0316]
17A is a plan view showing an example of the complementary color type color filter array 600 of (i), and FIG. 17B is a plan view showing an example of the complementary color type color filter array 610 of (ii). FIG. FIG. 17C is a plan view showing an example of the complementary color type color filter array 620 of (iii), and FIG. 17D is a plan view showing an example of the complementary color type color filter array 630 of (iv). FIG.
[0317]
In each of FIGS. 17A to 17D, each hexagon surrounding the alphabets G, Cy, Ye, W, and Mg in the drawing represents one color filter. Alphabets G, Cy, Ye, W, and Mg in the drawing represent colors of individual color filters.
[0318]
The arrangement pattern of the color filters in the three-primary color filter array is not limited to the pattern shown in FIG. Similarly, the arrangement pattern of the color filters in the complementary color filter array is not limited to the patterns shown in FIGS. 17 (a) to 17 (d).
[0319]
FIG. 18A is a plan view showing some examples of arrangement patterns of color filters in a primary color filter array used in an FT-CCD in which pixel photoelectric conversion elements are arranged in a square lattice pattern. is there. FIGS. 18A and 18I show a color filter array 700 having an arrangement pattern called a Bayer type, and FIGS. 18A and 18I show a color filter array 710 having an arrangement pattern called an interline type. . 18 (a) (iii) shows an arrangement pattern color filter array 720 called a G stripe RB checkered pattern, and FIGS. 18 (a) (iv) show an arrangement pattern color filter called a G stripe RB complete checkered pattern. An array 730 is shown. 18 (a) (v) shows a color filter array 740 having an arrangement pattern called a stripe type, and FIGS. 18 (a) (vi) show a color filter array 750 having an arrangement pattern called an oblique stripe type. . 18 (a) (vii) shows a color filter array 760 having an arrangement pattern called a primary color difference type.
[0320]
FIG. 18B is a plan view showing some examples of color filter arrangement patterns in a complementary color filter array used in an FT-CCD in which pixel photoelectric conversion elements are arranged in a square lattice pattern. is there. 18 (b) (i) shows an arrangement pattern color filter array 800 called a field color difference sequential type, and FIG. 18 (ii) shows an arrangement pattern color filter array 810 called a frame color difference sequential type. Show. 18 (b) (iii) shows an arrangement pattern color filter array 820 called a MOS type, and FIGS. 18 (b) (iv) show an arrangement pattern color filter array 830 called an improved MOS type. . 18 (b) (v) shows a color filter array 840 having an arrangement pattern called a frame interleave type, and FIG. 18 (b) (vi) shows a color filter array 850 having an arrangement pattern called a field interleave type. Yes. 18 (b) (vii) shows a color filter array 860 having an arrangement pattern called a stripe type.
[0321]
Also in the FT-CCD in which the pixel shifting arrangement is performed, the color filter arrangement pattern is appropriately changed according to the above-described color filter array for the FT-CCD in which the photoelectric conversion elements for pixels are arranged in a square lattice shape, for example. Can be selected.
[0322]
In the FT-CCD of each embodiment, a pixel photoelectric conversion element (photodiode) is formed on a p-type well formed on an n-type semiconductor substrate. Accordingly, these FT-CCDs can be provided with a vertical overflow drain structure. Along with this, an electronic shutter can be attached.
[0323]
In order to attach the vertical overflow drain structure to the FT-CCD of each embodiment, a structure capable of applying a reverse bias to the p-type well and the lower part of the n-type semiconductor substrate (region below the p-type well) is added. . By providing the vertical overflow drain structure, it becomes easy to suppress blooming.
[0324]
【The invention's effect】
As described above, in the FT-CCD of the present invention, stray light is difficult to enter the pixel photoelectric conversion element.
[0325]
Therefore, according to the present invention, it becomes easy to obtain an FT-CCD having a high quality of reproduced images.
[Brief description of the drawings]
FIG. 1 is a plan view schematically showing an FT-CCD according to a first embodiment.
FIG. 2 is a schematic diagram of a cross section taken along line AA shown in FIG.
3A is a plan view schematically showing one of the first transparent transfer electrodes for pixels shown in FIG. 1, and FIG. 3B is a diagram for the pixels shown in FIG. It is a top view which shows roughly one 2nd transparent transfer electrode.
4A is a plan view schematically showing the output transfer path shown in FIG. 1, and FIG. 4B is a diagram schematically showing the configuration of the output unit shown in FIG. 1; It is.
5 is a diagram showing an example of a driving pulse supply unit used when driving the FT-CCD shown in FIG. 1. FIG.
FIG. 6 is a partial cross-sectional view schematically showing an FT-CCD of a second embodiment.
FIG. 7 is a plan view showing an example of a color filter array.
FIG. 8 is a plan view schematically showing an FT-CCD according to a third embodiment.
9 (a) is a plan view schematically showing one of the charge transfer channels shown in FIG. 8, and FIG. 9 (b) is a diagram of another charge transfer channel shown in FIG. It is a top view which shows one schematically.
10A is a plan view schematically showing one of the first transparent transfer electrodes for pixels shown in FIG. 8, and FIG. 10B is a diagram for the pixels shown in FIG. It is a top view which shows roughly one 2nd transparent transfer electrode.
11 is a plan view schematically showing the output transfer path shown in FIG. 8. FIG.
12 is a diagram showing an example of drive pulse supply means used when driving the FT-CCD shown in FIG.
FIG. 13 is a plan view schematically showing an FT-CCD according to a fourth embodiment.
14 is a plan view schematically showing a transparent transfer electrode for a pixel shown in FIG.
15 is a diagram showing an example of drive pulse supply means used when driving the FT-CCD shown in FIG.
FIGS. 16 (a), 16 (b), and 16 (c) are diagrams for explaining the direction in which the microlens is shifted when the microlens is shifted. .
FIGS. 17A, 17B, 17C, and 17D are plan views showing examples of complementary color type color filter arrays, respectively.
18 (a) and 18 (b) show color filter arrangement patterns in a color filter array used in an FT-CCD in which pixel photoelectric conversion elements are arranged in a square lattice pattern, respectively. It is a top view which shows some examples.
[Explanation of symbols]
1, 201, 301 ... Semiconductor substrate, 2 ... p-type well, 10, 210, 310 ... photosensitive portion, 11, 211a, 211b, 311 ... charge transfer channel, 12, 212, 312 ... first transparent transfer electrode for pixels, 13, 213, 313... Second transparent transfer electrode for pixels, 15, 215, 315... First transfer dedicated transfer electrode, 16, 216, 316... Second transfer dedicated transfer electrode, 17, 218. 220, 320 ... light shielding film, 22, 222, 322 ... opening (pixel light receiving part), 25, 225, 325 ... pixel photoelectric conversion element (photodiode), 30, 330 ... CCD accumulating part, 40, 240 340: Output transfer path 50, 250, 350: Output unit 90, 290, 390 ... Driving pulse supply means 100, 200, 300, 400 ... FT CCD, 120 ... color filter array, 140 ... microlenses 230 ... adjustment section, the transparent transfer electrode 312 ... pixel 317 ... third transfer forwarding electrodes

Claims (7)

A photosensitive portion set on one surface of a semiconductor substrate;
An output transfer path formed in a region outside the photosensitive portion on one surface of the semiconductor substrate;
And a plurality of charge transfer channels reaching the output transfer path across along the exposed portion in a predetermined direction D _V,
Wherein formed on the photosensitive portion, along said direction D _V and direction D _H which intersects the plan view across each of said plurality of charge transfer channels are arranged at a constant pitch along the direction D _V A plurality of transparent transfer electrodes for pixels selected every other one of the directions _DV and _DH from all the intersections in plan view with the charge transfer channels. In each of the intersections, a plurality of pixel transparent transfer electrodes that constitute a pixel photoelectric conversion element together with the charge transfer channel;
A solid-state imaging device comprising: a light shielding film having an opening only on each of the pixel photoelectric conversion elements. Each of the pixel photoelectric conversion elements constituted by an odd-numbered charge transfer channel of the plurality of charge transfer channels and the plurality of pixel transparent transfer electrodes, an even-numbered charge transfer channel, and the plurality and each of the pixels for photoelectric conversion elements constituted by the pixel transparent transfer electrodes of this is, according to claim 1, characterized in that half the deviation of the pitch of the direction D _V to the pixel photoelectric conversion element solid Imaging device. And a plurality of transfer-dedicated transfer electrodes formed on the photosensitive portion and traversing each of the plurality of charge transfer channels in a plan view along the direction _DH. each of the electrodes, the direction D along the _V phase Tonariru one by one between the two pixels for the transparent transfer electrodes, while being in contact on the plan view in both of the phase Tonariru two pixels transparent transfer electrodes The solid-state imaging device according to claim 1 or 2, wherein the solid-state imaging device is formed. The plurality of transparent transfer electrodes for pixels intersect with odd-numbered charge transfer channels of the plurality of charge transfer channels when viewed in a plane and intersect with even-numbered charge transfer channels in a plan view. The width of a portion where the first transparent transfer electrode for pixels wider than the width of the pixel and the even-numbered charge transfer channel intersects the odd-numbered charge transfer channel in plan view. A plurality of second transparent transfer electrodes for a plurality of pixels,
Are formed alternately with each of the second transparent transfer electrode each said pixel to first transparent transfer electrode the pixels along the direction D _V,
Each of the plurality of pixel photoelectric conversion elements includes a cross section of the odd-numbered charge transfer channel and the plurality of first transparent transfer electrodes for pixels in plan view, and the even-numbered charge transfer. 4. The solid-state imaging device according to claim 1, wherein the solid-state imaging device is formed at each intersecting portion of the channel and the plurality of second transparent transfer electrodes for pixels in plan view.   The solid-state imaging device according to claim 1, wherein a microlens that covers the opening in plan view is provided on each of the openings.   The solid-state imaging device according to claim 5, wherein a color filter that covers the opening in plan view is provided between the opening and the microlens.   The solid-state imaging device according to claim 1, further comprising a CCD storage unit between the photosensitive unit and the output transfer path.

Images