JPH1168076A - Method for manufacturing color filter array - Google Patents

Abstract

[PROBLEMS] To provide a method of manufacturing a color filter array which can prevent an imaging defect due to a difference in signal output at a low cost or can prevent an imaging defect from the past. A CCD formed on the wafer W to the size L ₁ a plurality of pixels P ₁ of (solid-state imaging device) 10, a color filter 31 of the corresponding first color in one-to-one with the size L ₁ As formed, the reduction lens 24 and the wafer W
Is adjusted, and the resist 41 serving as the color filter 31 is exposed and patterned. Next, in the same way, a reduction lens to fit the size L ₂ of the pixel P ₂ 24
The second color filter 32 is patterned by adjusting the distance H between the wafer and the wafer W, and the third color filter 33 is similarly patterned so as to match the size L ₃ of the pixel P _3. Then, the color filter array 30 is formed directly on the CCD 10.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a method for manufacturing a color filter array formed directly on a solid-state imaging device such as a charge-coupled device (hereinafter, referred to as a CCD).

[0002]

2. Description of the Related Art A color solid-state imaging device (for example, a CCD) is provided with a separation prism and / or a color filter array for reading color signals. Among these, the color filter array transmits only a predetermined color from the color components included in the received light (that is, does not absorb only a specific wavelength among wavelengths in a range from near ultraviolet to near infrared), In addition, a plurality of color filters corresponding to the pixels of the solid-state imaging device are provided in one-to-one correspondence, and the transmitted color has two or three colors as a whole. Thus, for example, the colors transmitted through the color filters are the three primary colors of red, blue, and green, respectively, and when white light is applied to some of the pixel groups, the red color of the pixel group is transmitted. The optical signal is detected from all the pixels corresponding to the red filter, the blue filter transmitting the blue color, and the green filter transmitting the green color. Further, for example, when red light is irradiated to some of the pixel groups, a signal from a pixel corresponding to the red filter in this pixel group is detected, and other blue filters and green filters of this pixel group are detected. Is not detected from the pixel corresponding to. That is, the color solid-state imaging device reads not only image information but also color information depending on which color filter has detected light from a pixel corresponding to the color filter.

FIG. 6 shows a part of a conventional color CCD, which is a color CCD having a so-called on-chip color filter in which a color filter array 8 is formed directly on a monochrome CCD 9. is there. The monochrome CCD 9 has photoelectric conversion regions 1 and 1 'for converting received light into electrons, and gate electrodes 3a, 3b and 3c for transferring the electrons on both sides. In addition,
The outer periphery of the gate electrodes 3a, 3b, 3c is covered with an insulating film 6, and a film 7 (for example, made of an organic resin) for flattening the surface 9a of the monochrome CCD 9 is formed above the insulating film 6. . Note that a plurality of condenser lenses (not shown) are provided above the color filter array 8 as is well known.

At the design stage, that is, ideally, as shown by a dashed line in FIG.
The gate electrodes 3a, 3b, and 3c are formed in between.
When no shielding film is provided between the color filters 8a, 8b, and 8c (providing the shielding film reduces the magnitude of the detected optical signal and prevents this), the color filter is used. So that 8a, 8b, 8c touch
The color filters 8a, 8b, 8c are formed. In addition,
The sizes are all the same, which is the size L 'of the pixel P' of the monochrome 9 (this is, for example, the gate electrodes 3a, 3b, 3c around the photoelectric conversion regions 1, 1 ').
).

[0005]

However, when the center positions of the gate electrodes 3a, 3b, 3c are shifted due to a manufacturing error occurring during manufacturing, for example, as shown by a solid line in FIG. When the gate electrode 3b is shifted to the right, the electrodes 3a, 3a
Since the distance to b increases, the signal output from the photoelectric conversion area 1 decreases. Therefore, as shown in FIG. 6, the color filters 8a, 8b, 8 have the size L 'when the gate electrodes 3a, 3b are ideally formed.
When c is created, the color signal transmitted through the color filter 8b corresponding to the pixel P centered on the photoelectric conversion unit 1 is reduced. Therefore, in general, the size of one color filter 8a, 8b, 8c corresponding to the pixel P at this time is formed according to the size L of the actually formed pixel P, and the signal output difference To prevent imaging defects.
That is, by forming the color filters 8a, 8b, 8c in accordance with the size of the actually formed pixel P having a size different due to a manufacturing error, it is possible to prevent a signal output difference and prevent an imaging defect. I was

Conventionally, as a specific method of matching the size of the pixel P actually formed (ie, a manufacturing error has occurred) with the size of the color filters 8a, 8b, 8c, the color filters 8a, 8b, 8c are conventionally formed. 3 to 4 resized reticles (correction values are usually within ± 0.5 μm (dimensions on a reduced-projected wafer)) in which a correction value is added to a pattern used in the process. )
Was used. Thus, for example, a color filter array was actually created using each reticle, and the completed color CCD was tested. The result was the best (the manufacturing error of each pixel P was different, so for example, ,
A reticle for producing a color CCD when all the color filters 8a, 8b, 8c are considered to have almost the same size as the pixel P, for example, the predetermined color signal is the best for all the colors. Was used to create a subsequent color filter array.

That is, conventionally, a plurality of reticles have been prepared in order to make the size of the actually formed pixel having a manufacturing error coincide with the size of the color filter. It was costly and a lot of time. That is, the cost of the color CCD has been increased in order to prevent image pickup defects due to signal output differences to some extent. Further, in this method, only the reticle in which the best signal is detected is used from among the three or four reticles whose correction amounts are changed, and the size of the actually formed pixel and the size of the color filter are used. It was not always possible to say that they were fully matched. That is, it cannot be said that an imaging defect due to a signal output difference of the color solid-state imaging device having the color filter array can be sufficiently prevented.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems, and has been made to reduce the cost and time required for a reticle. Therefore, it is possible to prevent a certain degree of imaging defect at a lower cost than before, or to make the size of the pixel corresponding to the color filter one-to-one and the size of the color filter more appropriately than before. It is an object of the present invention to provide a method of manufacturing a color filter array which can be formed and can prevent a signal output difference and an imaging defect as compared with the related art.

[0009]

SUMMARY OF THE INVENTION The above objects are attained directly on a solid-state imaging device formed on a wafer.
Only one color of the received light is transmitted, and a plurality of color filters corresponding to the pixels of the solid-state imaging device in a one-to-one manner are projected on a wafer by reducing the pattern of the photomask onto the wafer using one reducing projection lens. The distance between the reduction projection lens and the wafer is set so that the size of all the color filters patterned in the photolithography process where exposure is performed is almost the same as the size of the pixel of the solid-state imaging device after manufacture. In other words, the exposure is performed by adjusting the exposure focal length, and the overall transmitted color is solved by a method of manufacturing a color filter array having two or three colors.

In this way, even if only one reticle serving as a photomask is formed, a color filter array in which the size of most of the color filters matches the size of the actually formed pixels is manufactured. can do. In other words, the cost required to create a plurality of reticles can be reduced, and the time required for that is eliminated. Therefore, the cost of the color solid-state imaging device can be kept low even if imaging defects due to signal output differences are prevented to some extent. In addition, an exposure apparatus that normally performs reduction projection has a leveling stage that is movable to change the distance between the reduction projection lens and the wafer, that is, the exposure focal length, to obtain an image forming surface. And fine-tuning is also possible. Therefore, it is easy to finely adjust the size of the color filter to the size of the actually formed pixels, that is, the size of all the pixels having manufacturing errors.

[0011] The above-described problem is that only one color of light received and formed directly on a solid-state imaging device formed on a wafer is transmitted, and the pixels of the solid-state imaging device are one-to-one. A plurality of color filters corresponding to the size of all the color filters patterned in the photolithography process where exposure is performed using a reduction projection lens,
Using a photomask resized by adding a correction amount to the pattern so as to be almost the same size as the pixel size of the solid-state imaging device after manufacture, and adjusting the distance between the reduction projection lens and the wafer, that is, The exposure focal length is adjusted so that the exposure is performed, and the color transmitted as a whole is solved by a method of manufacturing a color filter array having two or three colors. By doing so, a color filter can be formed with a more optimal correction value than using a reticle that is a plurality of photomasks, so that imaging defects can be sufficiently prevented.

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS In the present invention, only one color of light received and formed directly on a solid-state imaging device formed on a wafer is transmitted, and the solid-state imaging device is formed on a one-to-one basis. Forming a plurality of color filters corresponding to the pixels,
In a method of manufacturing a color filter array having two or three colors transmitted as a whole, in a photolithography process in which exposure is performed by reducing and projecting a pattern of a photomask onto a wafer using one reduction projection lens. The distance between the reduction projection lens and the wafer is adjusted so that the size of all the color filters to be patterned is almost the same as the size of the pixel of the manufactured solid-state imaging device, that is, the exposure focal length. Is a method for manufacturing a color filter array in which two or three colors are transmitted as a whole by adjusting the exposure. That is, when transferring the pattern of the color filter, by adjusting the exposure focal length and changing the size of the pattern transferred on the wafer, the pixel size of the actually manufactured solid-state imaging device and the color Match the size of the filter.

FIG. 5 shows the numerical aperture (N) of the projection lens.
A) = 0.60, illumination condition σ = 0.5, 0.4
The graph shows the relationship between the offset value and the actually obtained line width dimension when a line and space having a width of μm is formed. In FIG. 5, the value of the offset is the amount of deviation from the position of the best focus (ie, the exposure focal position when the photomask pattern is most clearly projected on the wafer (image plane is obtained)). It is. Therefore, in FIG. 5, the fact that the offset value is zero means that the offset value is zero.
This indicates that the camera is at the best focus position. In FIG. 5, a plus value indicates a case where the wafer is moved closer to the reduction projection lens, and a minus value indicates a case where the wafer W is moved away from the reduction projection lens.
As is clear from this figure, the line width, that is, the pattern size can be changed by adjusting the exposure focal position.

As described above, by shifting the exposure focal length at the time of exposure, the size of the color filter formed in one photolithography process can be actually reduced even if only one photomask (reticle) is used. Since it can be adjusted to the size of the pixel of the manufactured solid-state imaging device, it is possible to reduce the cost and time required for producing a photomask as in the related art. That is, the manufacturing cost of the color filter array and the manufacturing cost of the color solid-state imaging device using the color filter array can be reduced. Generally, in an exposure apparatus that performs exposure by reduction projection, the exposure focal length can be set by inputting a numerical value, so that the optimum exposure focal length to be adjusted can be easily and reliably obtained. Can be.

Further, a plurality of light beams formed directly on the solid-state image pickup device formed on the wafer, transmitting only one color of the received light, and corresponding to pixels of the solid-state image pickup device on a one-to-one basis. Color filters, so that the size of all the color filters that are patterned in the photolithography step where exposure is performed using a reduction projection lens is almost the same size as the size of the pixels of the solid-state imaging device after manufacture, Using a photomask resized by adding a correction amount to the pattern, and adjusting the distance between the reduction projection lens and the wafer, that is, adjusting the exposure focal length, forming the pattern so as to perform exposure, and transmitting as a whole The color
This is a method for manufacturing a color filter array having two or three colors. By doing so, a more optimal correction value can be found (that is, the degree of freedom in selecting the correction value is larger) than using only a plurality of reticles that are photomasks. Can be prevented.

Further, the pixel shown here is used for a solid-state imaging device having gate electrodes formed on both sides of a photoelectric conversion region.
If the distance between the centers of adjacent gate electrodes is given,
If the color filters are brought into contact with each other, no light-shielding film or the like is provided between the color filters, so that the obtained optical signal is not reduced.

Furthermore, if this exposure is performed by a step-type projection exposure apparatus, the exposure focal length can be easily changed little by little between shots.
By changing the size of the pattern of the color filter, it is possible to obtain a pattern whose size matches the actually formed pixels of the solid-state imaging device.

In one photolithography process,
In any case, such as when a color filter is created using a method of dyeing a patterned material to be dyed or a photosensitive colored resist, by forming all the color filters that transmit the same color, With the minimum number of processes,
It can be formed without mixing transmitted colors.

[0019]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below with reference to the drawings.
A method for directly manufacturing three-color on-chip color filters is described.

FIG. 1 shows a part of a monochrome CCD 10, which is formed in a plurality (for example, about 60) on a wafer W as shown in FIG. It has a simple structure. That is, as shown in FIG. 1, the silicon substrate 11 is provided with a plurality of photoelectric conversion units 13a, 13b, and 13c that convert received light into electrons. Then, a plurality of gate electrode portions 14 made of polysilicon are formed on the silicon substrate 11 with an oxide film 12 interposed therebetween.
a, 14b, 14c, 14d are the photoelectric conversion units 13a, 1
It is formed between 3b and 13c. The gate electrode portions 14a, 14b, 14c, 14d
The transfer section of the CD 10 and the adjacent gate electrode section 14
a, 14b, 14c, and 14d, by applying a voltage with a phase difference between the photoelectric conversion units 13a, 13b,
The electrons (charges) generated in 13c are transferred to an output unit (not shown). If the CCD 10 is ideally manufactured without manufacturing errors, the photoelectric conversion units 13a, 13b, 1
3c, the gate electrode portions 14a, 14b, 14
The centers of c and 14d are located. However, these deviations are usually due to manufacturing errors, and in FIG.
The center lines of the gate electrode portions 14b and 14c are shown to be shifted rightward and leftward from an ideal state (that is, a design state), respectively (due to a manufacturing error).

The gate electrode portions 14a, 14b, 14
c, the outer periphery of the 14d is covered with the insulating film 15, thereby insulates the electrical connection between the pixel _{P 1, P 2, P 3} . Further, the gate electrode portions 14a, 14b, 14c,
Above 14d, a film 16 made of an organic resin or the like is formed, and its surface 16a is flat.

As shown in FIG. 1A, this CCD
A resist 41 having a first color (for example, a red color) (for example, a red dye is mixed in the resist) is formed on the film 10 by, for example, a known method. Apply. Then, from above, light R is applied to a portion where the first color filter 31 is formed,
The light-irradiated portion of the resist 41 is cured. At this time, a stepper (step-type projection exposure apparatus) 20 as shown in FIG. 4 is used, which has a known structure. That is, light from a light source 21 composed of a KrF excimer laser or the like is passed through a known optical system 22 (here, the variation in light intensity is suppressed to be shaped into a two-dimensional parallel light), and a photomask is used. A certain reticle 23 is irradiated. The reticle 23 is formed on a transparent substrate 23a by a shielding film 23b such as chrome (for example, 5 μm square to 10 μm).
A plurality of squares each having the size of an m-square are formed in a pattern (such as an arranged shape or a stripe shape). Therefore, the light R transmitted through the reticle 23, that is, the light R having the pattern shape formed on the reticle 23
Passes through the reduction projection lens 24, is collected, and is projected onto one CCD 10 of the wafer W. The stepper 20 includes a wafer stage XY for positioning the wafer W in the horizontal direction, and a reduction projection lens 24 (not shown).
And a leveling stage for changing the distance between the wafer and the wafer W.

That is, the CC formed on the wafer W
In D10, the distance H between the reduction projection lens 24 and the wafer W is adjusted, that is, the position of the leveling stage on which the wafer W is mounted is adjusted by, for example, numerical input, and the size of the color filter 31 formed is adjusted. To the actually formed pixel P with a corresponding manufacturing error
₁ (This magnitude, adjacent gate electrode portions 14a, 14b is a distance between the centers of in this embodiment) of the fit to the size L ₁ is exposed. Then, using a known developer, the resist 41 is removed from the uncured portion without being irradiated with light. Then, the first color color filter 31, such as shown in B of FIG. 1 is formed by a size L ₁ pixels P ₁ corresponding thereto.

Next, a color filter 32 of a second color (for example, a blue color) is formed by a method similar to the method of forming the color filter 31 of the first color. That is, a resist 42 having a second color is applied on the first color filter 31 as shown in FIG. 2A by a known method, and the pattern of the first color filter 31 is applied to the stepper 20. Have the same shape (shape or stripe shape in which 5 μm square to 10 μm square are arranged), but set a reticle having a pattern in which the arrangement position is different in place of reticle 23 in FIG. Do. Also at this time, of the CCD 10 formed on the wafer W, the distance H between the reduction projection lens 24 and the wafer W is adjusted so that the size of the color filter 32 corresponds to this (has a manufacturing error). and to fit to the size L ₂ of the pixel P _2, which is actually formed, it is exposed.

Thereafter, the resist 42 is removed from the uncured portion without being irradiated with the light R by using a known developing solution. Then, as shown in FIG. 2B, a first color filter 31 and a second color filter 32 are formed. Next, similarly, the third color filter 33
The attached resist coated the color of a color (e.g. green), and adjust the exposure focal length H, it is formed to fit the size L ₃ of the pixels P ₃ corresponding thereto. As a result, as shown in FIG. 3, the color filters 31 having substantially the same size as the sizes L ₁ , L ₂ , L ₃ of the pixels P ₁ , P ₂ , P ₃ of the CCD 10 corresponding to each other are provided. A color filter array 30 is formed in which 32 and 33 are arranged in a checkered pattern or a stripe pattern as a whole. In this embodiment, when patterning the color filters 31, 32, and 33, the exposure focal length H is adjusted. However, when the exposure focal length H is not the best focus position (the position where the image is clearly formed). However, since the shape of the color filter is a checkerboard shape or a stripe shape of about 5 μm square to 10 μm square, the shape is not lost.

For each color, the distance (exposure focal length) H between the reduction projection lens 24 and the wafer W to be adjusted is determined, for example, as follows.

First, assuming that no manufacturing error occurs in the CCD 10, color filters 31, 32, and 33 are formed on one wafer W in an ideal (designed) shape. It is formed on one CCD 10. That is, the pattern of the color filter 31 of the first color,
The pattern of the second color filter 32 and the pattern of the third color filter 33 are exposed at the best focus position to create a color filter array. next,
The completed color CCD is irradiated with, for example, red light, blue light, green light, and white light, and a test is performed to determine whether each of the obtained color signals has a predetermined value.

[0028] In this case, for example, as shown in FIG. 3, the second color of the color filter 32 to the corresponding pixel P ₂ gate electrode portions 14b to form a displaced center position of 14c, the second color (blue) pixel P ₂ corresponding to the color filter 32, when it is larger than the overall shape of the design time, than the predetermined value is a color signal of the color to be detected by the second color of the color filter 32 (here in blue) Detected as smaller. Therefore, this time, for example, the exposure focal length H when exposing the pattern of the second color filter 32 is adjusted within a range of, for example, ± 1.0 μm, and the size of the second color filter 32 becomes slightly larger. It forms like this.

At the same time, the first color (red color)
If the pixel corresponding to the color filter 31 is formed smaller than the shape at the time of the design as a whole, the color signal for detecting the detected red color is larger than a predetermined value.
When the first color filter 31 is formed, the exposure focal length H is adjusted within a range of, for example, ± 1.0 μm so that the red color filter 32 becomes smaller, and the exposure is performed. Perform patterning. That is, based on the output of the color signal of the completed color CCD, the color filters 3 having different colors to be transmitted respectively.
By adjusting the exposure focal lengths H of 1, 32, and 33, respectively,
The size of each of the color filters 31, 32, and 33 formed in one photolithography process is set so as to substantially match the size of the CCD pixel corresponding to the size of each of the color filters.
The most optimal exposure focal length H is determined.

In this embodiment, one pattern corresponds to one
Even if only one reticle is used, by adjusting the distance between the reduction projection lens 24 and the wafer W, that is, the exposure focal length H, the pixels P ₁ , P ₂ , P of the CCD 10 having a manufacturing error in one photolithography process are adjusted. ₃ size L
The size of the color filter 31 can be almost matched to ₁ , L ₂ and L ₃ . Therefore, the color filter array 30 can be manufactured at lower cost than before, and a color fixed imaging device using the color filter array 30 can be manufactured at lower cost.

Although the embodiment of the present invention has been described above, the present invention is, of course, not limited to this, and various modifications can be made based on the technical concept of the present invention.

For example, in the above embodiment, the resist itself forms a color filter using a so-called photosensitive colored resist that transmits only one color of the received light. After patterning using a color filter, the color filter array may be formed by a method of forming a color filter by repeating a process of dyeing the color into a desired color, performing a dye-proofing process, and forming a color filter. In addition, a color filter array is formed by an interference filter in which a material having a low refractive index and a material having a high refractive index are alternately vacuum-deposited to obtain desired color characteristics, and this material is patterned by etching using a photolithography process. You may make it.

In the above embodiment, a negative photosensitive colored resist is used in which the light-irradiated portion does not dissolve in the developing solution. Of course, a positive-type photosensitive resist in which the light-irradiating portion dissolves in the developing solution is used. You may use an acidic coloring resist.

Further, in the above embodiment, the color filters of the three primary colors of red, blue and green were manufactured. However, the colors of the color filters are not limited to these, and may be three complementary colors. . Further, in the present invention, it is also possible to form a two-color filter array (for example, two colors of red and blue) used in a solid-state imaging device having two color filters.

In the above embodiment, the optimum exposure focal length H for forming each color filter was determined while forming and testing a color CCD. However, the optimum exposure focal length H was determined by another method. It may be determined.
For example, many CCDs such as 60 on one wafer W
Is formed, the first color filter 3
The exposure focal length H for forming 1 is changed every few pieces, and 2
Various combinations of color CCDs in which the exposure focal lengths H forming the color filters 32 and 33 for the third color are changed for every several colors and the exposure focal lengths H of the color filters 31, 32 and 33 are changed. To form In this manner, a plurality of color CCDs formed on one wafer W are inspected. Of these, the exposure focal length H when forming the color filters 31, 32, and 33 of the color CCD in which the color signal is detected at the most desired value is checked, and the exposure focal length H is used to determine the second and subsequent wafers W
A color filter array is formed on the CCD 10. In this case, since the exposure focal length H is variably adjusted between shots, the time required to obtain the best exposure focal length H is very short.

In the above embodiment, the size of all the color filters 31, 32, and 33 patterned by one photolithography process is adjusted by adjusting the exposure focal length H using one reticle. , The size L ₁ of the pixels P ₁ , P ₂ , P ₃ of the solid-state imaging device after manufacture,
The size was almost the same as L ₂ and L ₃ .
In addition to this, a combination of a reticle having an optimal correction amount and an exposure focal length H is selected by creating a plurality of reticles in which various correction amounts are added to the pattern and adjusting the exposure focal length H. Thus, a color filter may be formed by using this.

In the above embodiment, all of the color filters 31, 32, and 33 transmitting the same color are formed by using one photolithography process. It is not necessary to limit to forming all the color filters transmitting the same color.

In the above embodiment, the size L ₁ of the pixels P ₁ , P ₂ , P ₃ of the CCD (solid-state imaging device) 10,
As L _2, L _3, the solid-state imaging device of the photoelectric conversion area gate electrode portion 14a formed on both sides of, 14b, 14c, 1
The center-to-center distance was 4d. That is, the color filters 31, 32, and 33 were provided so as to contact each other. However, a shielding film is provided between the color filters 31, 32, and 33, and a size slightly smaller than the distance between the centers of the gate electrodes is set as the size of the fixed imaging device. The exposure focal length may be determined so that the sizes of the color filters 31, 32, and 33 formed by the above are almost the same.

In the above embodiment, the CCD is exemplified as the solid-state imaging device. However, it goes without saying that another solid-state imaging device such as a bucket relay device (BBD) may be used.

[0040]

As described above, according to the method for manufacturing a color filter array of the present invention, the cost and time required for the reticle can be reduced, and the cost can be reduced to a one-to-one correspondence with the color filter. It can be formed substantially in accordance with the size of the corresponding pixel. Alternatively, it is possible to form the color filter and the size of the pixel corresponding thereto one-to-one more accurately than in the past, and to further prevent a signal output difference and reduce an imaging defect. Can be prevented.

[Brief description of the drawings]

FIG. 1 is a front cross-sectional view of an incomplete color filter array showing a manufacturing process for forming a first color filter in an embodiment of the present invention, where A is a state in which a resist serving as a first color filter is applied; B shows a state in which a resist serving as a first color filter is patterned.

FIG. 2 is a front cross-sectional view of an unfinished color filter array showing a manufacturing process for forming a second color filter in an embodiment of the present invention, where A is a state where a resist serving as a second color filter is applied; B shows a state where a resist serving as a second color filter is patterned.

FIG. 3 is a front sectional view of a completed color filter array according to the embodiment of the present invention.

FIG. 4 is a schematic view of an exposure apparatus used in an embodiment of the present invention.

FIG. 5 is a diagram illustrating a relationship between a change in an exposure focal length and a size of an actually formed line width.

FIG. 6 is a front sectional view of a completed color filter array in a conventional example.

[Explanation of symbols]

Reference numeral 10: CDD, 13a, 13b, 13c: photoelectric conversion unit, 14a, 14b, 14c, 14d: gate electrode unit, 20: stepper, 23: reticle, 24: reduction projection lens, 30: color Filter array, 31 ...
.., The first color filter, 32, the second color filter, 33, the third color filter, 41, 42
...... resist, H ...... exposure focal _{length, L 1, L 2, L} 3
…… Pixel size, P ₁ , P ₂ , P ₃ … Pixel, R…
Light, W ... Wafer.

Claims (6)

[Claims] 1. A photolithography process in which a pattern of a photomask is reduced and projected onto a wafer by using a reduction projection lens and exposure is performed, thereby having a gate electrode formed on both sides of a photoelectric conversion region. Directly on the solid-state imaging device formed on the wafer,
A plurality of color filters that are formed, transmit only one color of the received light, and correspond to pixels of the solid-state imaging device in a one-to-one manner;
In the method of manufacturing a color filter array having two or three colors, wherein the color transmitted as a whole has two or three colors, out of the plurality of color filters, all of the color filters patterned in one photolithography step The distance is adjusted between the reduction projection lens and the wafer so that the size is substantially the same as the size of the pixel of the solid-state imaging device after manufacturing, and the exposure is performed. A method for manufacturing a characteristic color filter array. 2. A photolithography process in which a pattern of a photomask is reduced and projected onto a wafer by using a reduction projection lens to perform exposure, thereby having gate electrodes formed on both sides of the photoelectric conversion region. Directly on the solid-state imaging device formed on the wafer,
A plurality of color filters that are formed, transmit only one color of the received light, and correspond to pixels of the solid-state imaging device in a one-to-one manner;
In the method of manufacturing a color filter array having two or three colors that are transmitted as a whole, among the plurality of color filters, all of the color filters patterned in one photolithography step are: The size of the pixel of the solid-state imaging device after manufacture is substantially the same as the size, using a photomask resized by adding a correction amount to the pattern, and the distance between the reduction projection lens and the wafer. The method of manufacturing a color filter array, wherein the exposure is performed by adjusting the color filter array. 3. The pixel according to claim 1, wherein the size of the pixel is a distance between centers of the adjacent gate electrodes.
Alternatively, the method for manufacturing a color filter array according to claim 2. 4. The method according to claim 3, wherein the exposure is performed by a step-type projection exposure apparatus. 5. The method according to claim 4, wherein all the color filters transmitting the same color are formed in the one photolithography step. 6. The method according to claim 5, wherein the solid-state imaging device is a charge-coupled device.