JP2019144367A - Exposure method - Google Patents

Abstract

To improve a production yield of products by suppressing deviation in an exposure light quantity on a substrate depending on a shape of an exposure pattern and by forming a resist pattern having a desired processed shape.SOLUTION: An exposure method includes steps of: (a) preparing an original image data; (b) subjecting the original image data to an image processing filter to obtain a filtered image data; (c) dividing the original image data by the filtered image data to obtain a corrected image data; (d) subjecting the corrected image data to the image processing filter to obtain a new filtered image data; and (e) determining whether the light quantity in a region irradiated with light in the new filtered image data is constant or not. When the light quantity in the region irradiated with light does not show a desired value in the step (e), the corrected image data is replaced by the original data, the new filtered image data is replaced by the filtered image data, and then the steps (c), (d) and (e) are repeated.SELECTED DRAWING: Figure 2

Description

  The present invention relates to an exposure method, and can be suitably used for an exposure method using, for example, a DMD (Digital Micromirror Device).

  For example, Japanese Patent Laid-Open No. 2006-186364 (Patent Document 1) discloses a data operation arranged to perform transformation by adding a point spread function matrix in a pseudo inverse matrix format to a column vector representing a requested pattern. A lithographic apparatus comprising a device is described. The point spread function matrix contains information about the shape and relative position of the point spread function of each light spot exposed on the substrate by one of the radiation sub-beams at a given time.

JP 2006-186364 A

  In the exposure method using DMD, the exposure pattern is projected onto the substrate (the surface of the substrate, the upper surface of the substrate) by turning on or off the micromirrors arranged in a grid (switching the inclination of the micromirror). . The light irradiated from the light source to the DMD through the illumination optical system is reflected by the ON micromirror, and further projected through the reduction projection optical system onto the substrate to be exposed. However, the amount of exposure on the substrate depends on the shape of the exposure pattern, and a resist pattern having a desired processing shape cannot be formed on the substrate. As a result, there is a problem in that the production yield of the product decreases.

  Other problems and novel features will become apparent from the description of the specification and the accompanying drawings.

  In an exposure method according to an embodiment, (a) a step of preparing original image data, (b) an image processing filter is applied to the original image data, and filter image data that approximates an exposure pattern projected on the surface of the substrate is obtained. (C) a process of acquiring corrected image data by dividing the original image data by the filter image data, and (d) applying an image processing filter to the corrected image data to form an exposure pattern projected on the surface of the substrate. A step of acquiring new filter image data to be approximated. Further, (e) includes a step of determining whether or not the amount of light in the region that is irradiated with light in the new filter image data is constant. In step (e), when the amount of light in the region that is exposed to light indicates a desired value, the corrected image data is converted into gradation grayscale drawing data. Further, when the amount of light in the region where the light hits does not indicate a desired value in the step (e), the corrected image data is replaced with the original image data, the new filter image data is replaced with the filter image data, and (e Step (c), Step (d) and Step (e) are repeated until the amount of light in the region where the light hits shows a desired value.

  According to one embodiment, it is possible to form a resist pattern having a desired processed shape on a substrate while suppressing the bias of the exposure amount on the substrate depending on the shape of the exposure pattern. As a result, the production yield of products can be improved.

1 is a schematic block diagram of an exposure apparatus according to an embodiment. It is a flowchart explaining the production method of the correction image data by one embodiment. It is a light quantity distribution figure of the point spread function of 1 pixel by one embodiment. FIG. 6 is a light amount distribution diagram in which a light spreading region of one pixel according to an embodiment is divided into a plurality of portions, and the light amounts of the respective divided portions are shown in a graph. FIG. 4 is a light amount distribution diagram in which a light spreading area of one pixel according to an embodiment is divided into a plurality of portions, and the amount of light at each of the divided portions is indicated by a numerical value. (A) is a top view which shows the original image data by one Embodiment, (b) is a light quantity distribution figure which shows the light quantity in the AA 'line of the same figure (a). (A) is a top view which shows the filter image data which applied the image processing filter to the original image data by one Embodiment, (b) is light quantity distribution which shows the light quantity in the AA 'line of the same figure (a). FIG. It is a figure explaining the correction calculation method of image data by one embodiment. (A) is a top view which shows the correction image data by one Embodiment, (b) is a light quantity distribution figure which shows the light quantity in the AA 'line of the same figure (a). (A) is a top view which shows the new filter image data which applied the image processing filter to the correction image data by one Embodiment, (b) shows the light quantity in the AA 'line of the same figure (a). It is a light quantity distribution diagram. (A) is a top view which shows the correction image data after repeating correction | amendment calculation of the original image data by one Embodiment 4 times, (b) is the light quantity in the AA 'line of the same figure (a). FIG. (A) is a top view which shows the new filter image data which applied the image processing filter to the correction image data after repeating the correction calculation of the original image data by one Embodiment 4 times, (b) is the figure It is a light quantity distribution diagram which shows the light quantity in the AA 'line of (a). (A) is a plan view of a resist pattern formed on a substrate by exposure using original image data, and (b) is an exposure using corrected image data after repeating correction calculation of the original image data four times. It is a top view of the resist pattern formed on the board | substrate by FIG.

  Hereinafter, embodiments will be described in detail with reference to the drawings. Note that components having the same function are denoted by the same or related reference symbols throughout the drawings for describing the embodiments, and the repetitive description thereof is omitted. In addition, when there are a plurality of similar members (parts), a symbol may be added to the generic symbol to indicate an individual or specific part. In the following embodiments, the description of the same or similar parts will not be repeated in principle unless particularly necessary.

  In the drawings used in the embodiments, hatching may be omitted even in a cross-sectional view so as to make the drawings easy to see. Further, even a plan view may be hatched to make the drawing easy to see.

  In the cross-sectional view and the plan view, the size of each part does not correspond to the actual device, and a specific part may be displayed relatively large for easy understanding of the drawing. Even when the cross-sectional view and the plan view correspond to each other, a specific part may be displayed relatively large in order to make the drawing easy to understand.

(task to solve)
In an exposure method using DMD, an exposure pattern is projected onto a substrate by turning on or off micromirrors arranged in a grid. The light emitted from the light source through the illumination optical system to the DVD is reflected by the ON micromirror, and further projected through the reduction projection optical system onto the substrate to be exposed. In the reduction projection optical system, light reflected by, for example, a square micromirror having a side of 10.8 μm constituting the DMD is reduced to a square light spot having a side of 0.5 μm on the substrate.

  However, when a rectangular light spot having a side of 0.5 μm is projected on the substrate through the reduction projection optical system, a square having a side of 0.5 μm due to a distribution called a point spread function (point spread function). The light spreads around. For this reason, when one micromirror is ON, the amount of light reflected by the one micromirror and incident on the substrate changes depending on whether the surrounding micromirror is OFF or ON. .

  For example, when one micromirror is ON and all the surrounding micromirrors are OFF, light enters from the surroundings on the projection surface on which the light reflected by the one micromirror is projected. Not come. However, when one micromirror is ON and some or all of the surrounding micromirrors are ON, the projection surface on the substrate on which the light reflected by the one micromirror is projected is from the periphery. Light enters and the amount of light increases.

  That is, the exposure amount on the substrate is biased depending on the shape of the exposure pattern, and a resist pattern having a desired processed shape cannot be formed on the substrate. Specifically, an exposure pattern with a small number of pixels with micromirrors ON, for example, a thin exposure pattern, or an exposure pattern with a large number of pixels with micromirrors ON, for example, a thin exposure pattern or a corner portion of the exposure pattern. In the pattern, the exposure amount becomes excessive. As a result, there is a problem that the production yield of the product is lowered.

(Embodiment)
《Exposure device》
An exposure apparatus according to the present embodiment will be described with reference to FIG. FIG. 1 shows a schematic block diagram of an exposure apparatus according to the present embodiment.

  As shown in FIG. 1, the exposure apparatus 10 includes an exposure optical system 30 that projects an exposure pattern on the surface of the substrate 20 during exposure, and an autofocus that detects a focused position (a focused position) on the surface of the substrate 20. Optical system 40. Of the autofocus optical system 40, an optical system that projects a focus pattern onto the surface of the substrate 20 has a configuration common to the exposure optical system 30.

  The exposure optical system 30 projects an exposure pattern onto the surface of the substrate 20 placed on the stage 50. The exposure optical system 30 includes an exposure light source 31, a collimator lens 32, mirrors 33a, 33b, and 33c, a DMD 34 that is a spatial light modulator, a relay lens 35, an objective lens 36, and the like.

  Photosensitive light (short wavelength light such as blue light or ultraviolet light) output from the exposure light source 31 is incident on the DMD 34 via the collimator lens 32 and the mirrors 33a and 33b, and the light reflected by the DMD 34 is mirror 33c. The light enters the surface of the substrate 20 through the relay lens 35 and the objective lens 36. Thus, the exposure apparatus 10 according to the present embodiment is a maskless exposure apparatus.

  The exposure light source 31 is composed of, for example, an LED (Light Emitting Diode) having a main wavelength of 405 nm.

  The collimator lens 32 shapes the light output from the exposure light source 31 into parallel light, and the mirrors 33 a and 33 b cause the light transmitted through the collimator lens 32 to enter the DMD 34.

  The DMD 34 is configured as a single panel in which fine micromirrors (also referred to as “mirror elements”) are spread on a semiconductor element in a grid pattern (lattice pattern), and each micromirror is independent of each other. By changing the tilt angle, the projection of the light output from the exposure light source 31 can be turned on or off. A controller (not shown) is electrically connected to the DMD 34, and a micromirror at a predetermined position in the DMD 34 reflects light. Thereby, the DMD 34 forms an exposure pattern having a desired shape projected onto the surface of the substrate 20.

  The mirror 33 c causes the light reflected by the DMD 34 to enter the relay lens 35, and the relay lens 35 corrects the magnification of the exposure pattern formed by the DMD 34. For example, when projecting an exposure pattern formed by the DMD 34, the micromirror of the DMD 34 is a quadrangle having a side of 10.8 μm, and when the projection is reduced to a size of 1/20, for example, by the objective lens 36, the substrate 20. On the surface, a square exposure pattern with a side of 0.54 μm is formed. However, by correcting the magnification with the relay lens 35, a square exposure pattern with a side of 0.5 μm can be reduced and projected onto the surface of the substrate 20.

  In the present embodiment, a square exposure pattern corresponding to one micromirror and projected on the surface of the substrate 20 and having a side of 0.5 μm is defined as one pixel, but it is needless to say that the present invention is not limited to this.

  The objective lens 36 condenses the light transmitted through the relay lens 35 on the surface of the substrate 20.

  The autofocus optical system 40 projects a focus pattern onto the surface of the substrate 20 placed on the stage 50 and receives light reflected from the surface of the substrate 20. The autofocus optical system 40 includes a focus light source 41, a collimator lens 42, a half mirror 43, mirrors 33a, 33b, and 33c, a DMD 34, a relay lens 35, an objective lens 36, a half mirror 44, an imaging lens 45, and a CCD (Charge). Coupled Device) 46 and the like. Among these, the mirrors 33 a, 33 b, 33 c, DMD 34, relay lens 35, and objective lens 36 are configured to be used in common in the exposure optical system 30 and the autofocus optical system 40.

  Light (long wavelength light such as infrared rays) output from the focus light source 41 is incident on the DMD 34 via the collimator lens 42, the half mirror 43 and the mirrors 33a and 33b, and the light reflected by the DMD 34 is the mirror 33c and the relay lens 35. Then, the light enters the surface of the substrate 20 through the objective lens 36. Further, the light reflected by the surface of the substrate 20 forms an image on the CCD 46 via the objective lens 36, the half mirror 44 and the imaging lens 45.

  The focus light source 41 is composed of, for example, a red light emitting diode.

  The collimator lens 42 shapes the light output from the focus light source 41 into parallel light, and the half mirror 43 has mirrors 33a and 33b with the optical axis of the light transmitted through the collimator lens 42 being coaxial with the optical axis of the exposure light source 31. To enter.

  The DMD 34 constituting the focusing optical system 40 forms a focusing pattern, for example, a stripe pattern, of a desired shape projected onto the surface of the substrate 20.

  The objective lens 36 also serves as an imaging optical system that focuses light reflected from the surface of the substrate 20 on the imaging surface of the CCD 46 as focusing light. The light reflected from the surface of the substrate 20 is again the objective lens. 36 passes through and enters the half mirror 44.

  The light incident from the objective lens 36 is reflected by the half mirror 44 and enters the imaging lens 45. The CCD 46 is a photoelectric conversion element that receives the light transmitted through the imaging lens 45 and converts it into an electrical signal, and also functions as an imaging unit.

<Exposure method>
An exposure method according to this embodiment will be described with reference to FIG.

  As shown in FIG. 1, first, the substrate 20 is placed on the stage 50.

  Next, the substrate 20 is moved so that the exposure pattern is projected from the DMD 34 to a predetermined position on the surface of the substrate 20.

  Next, autofocus processing is performed using the focusing optical system 40. First, when focus light (long wavelength light) is output from the focus light source 41, the output light is shaped into parallel light by the collimator lens 42, and is coaxial with the optical axis of the output light of the exposure light source 31 by the half mirror 43. Then, the light is reflected by the mirrors 33a and 33b and enters the DMD 34.

  Subsequently, when the focus pattern is reflected by the micromirror of the DMD 34, the focus pattern is reflected by the mirror 33 c, corrected for magnification by the relay lens 35, and projected from the objective lens 36 onto the surface of the substrate 20.

  Subsequently, when the focus pattern is projected onto the surface of the substrate 20, the reflected light of the substrate 20 is again transmitted through the objective lens 36 and enters the half mirror 44. Subsequently, the reflected light is incident on the imaging lens 45 by the half mirror 44.

Subsequently, the light transmitted through the imaging lens 45 is received by the CCD 46, and the focusing position of the substrate 20 is determined by analyzing the focusing pattern received by the CCD 46. Thereafter, the substrate 20 is moved to the in-focus position by an instruction signal from a control unit (not shown).

  Next, an exposure process is performed using the exposure optical system 30. First, when exposure light (short wavelength light) is output from the exposure light source 31, the output light is shaped into parallel light by the collimator lens 32, passes through the half mirror 43, and then is reflected by the mirrors 33a and 33b. Incident on DMD 34.

  Subsequently, when the exposure pattern is reflected by the micromirror of the DMD 34, the exposure pattern is reflected by the mirror 33c, corrected for magnification by the relay lens 35, and projected from the objective lens 36 onto the surface of the substrate 20. Thereby, an exposure pattern having a desired shape is projected onto the surface of the substrate 20. Thus, in the present embodiment, a desired exposure pattern can be exposed to the resist applied on the surface of the substrate 20 by repeating the autofocus process and the exposure process.

  The exposure method according to the present embodiment is characterized in that the exposure amount on the substrate 20 is adjusted for each pixel in accordance with the shape of the exposure pattern in order to reduce the bias of the exposure amount on the substrate 20. That is, the original image data is binary data that is ON or OFF for each pixel, but the original image data is converted into corrected image data having, for example, an 8-bit luminance gradation. And the brightness | luminance for every pixel is adjusted according to the shape of an exposure pattern. For example, a pixel with few ON pixels in the surrounding area such as a thin exposure pattern is bright, and a pixel having many ON pixels in the surrounding area such as a thick exposure pattern is darkened. Then, by performing exposure according to the brightness of the corrected image data, the bias of the exposure amount on the substrate depending on the shape of the exposure pattern is suppressed.

  Hereinafter, a method of creating corrected image data according to the present embodiment will be described in detail with reference to FIGS.

  FIG. 2 is a flowchart for explaining a method of creating corrected image data. FIG. 3 is a light amount distribution diagram of a point spread function of one pixel obtained from a theoretical formula. FIG. 4 is a light amount distribution diagram in which the light spreading region of one pixel is divided into a plurality of portions, and the amount of light at each of the divided portions is shown in a graph. FIG. 5 is a light amount distribution diagram in which the light spreading region of one pixel is divided into a plurality of portions, and the amount of light at each of the divided portions is indicated by a numerical value.

  FIGS. 6A and 6B are a plan view and a light amount distribution diagram showing the original image data, respectively. FIGS. 7A and 7B are a plan view and a light amount distribution diagram showing filter image data obtained by applying an image processing filter to original image data, respectively. FIG. 8 is a diagram for explaining a correction calculation method for image data. FIGS. 9A and 9B are a plan view and a light amount distribution diagram showing the corrected image data, respectively. FIGS. 10A and 10B are a plan view and a light amount distribution diagram showing new filter image data obtained by applying an image processing filter to the corrected image data, respectively. FIGS. 11A and 11B are a plan view and a light amount distribution diagram showing corrected image data after the correction calculation of the original image data is repeated four times, respectively. 12A and 12B are a plan view and a light amount distribution diagram, respectively, showing new filter image data obtained by applying an image processing filter to corrected image data after the correction calculation of the original image data is repeated four times. .

<Step 1 (Process P1 shown in FIG. 2)>
As shown in FIG. 3, the spread of light of one pixel is obtained from a theoretical formula.

  FIG. 3 shows a light amount distribution of a point spread function when the light amount of one pixel composed of a square having a side of 0.5 μm is 1. Specifically, in a region where the X direction is 5 μm and the Y direction orthogonal to the X direction is 5 μm (hereinafter referred to as a light spreading region), the region is located at the center of the region, and is formed of a square having a side of 0.5 μm. A light amount distribution of a point spread function when normalized with the light amount of the pixel as 1 (normalized with a maximum value of 1 and a minimum value of 0) is shown. In the present embodiment, one pixel corresponds to one micromirror constituting the DMD.

<Step 2 (Process P2 shown in FIG. 2)>
As shown in FIG. 4 and FIG. 5, an image processing filter is created by dividing the light spreading area of one pixel into a plurality of parts and obtaining the amount of light at each part.

  By applying this image processing filter to black-and-white image data (corresponding to original image data or corrected image data described later) (also called “convolution operation”), the amount of light irradiated on the work surface (substrate surface) is simulated. can do.

  For example, a light spreading area consisting of a square with a side of 5 μm is divided into nine in the X direction and divided into nine in the Y direction, and 81 divided parts are formed in a square of 0.5 μm on a side and arranged in a grid. Provide. In the center of the light spreading region, one pixel having a side of 0.5 μm and a light amount of 1 is arranged. Then, the amount of light at each segment is obtained, and an image processing filter is created from the amount of light distribution in the light spreading region.

<Step 3 (Process P3 shown in FIG. 2)>
As shown in FIGS. 6A and 6B, original image data OD is prepared.

  In the present embodiment, as an example, original image data OD having a line and space of 0.5 μm is used. The original image data OD is represented by binary data in which the maximum value of light quantity is 1 and the minimum value of light quantity is 0.

<Step 4 (Process P4 shown in FIG. 2)>
As shown in FIGS. 7A and 7B, the filter image processing is performed on the original image data OD to obtain the filter image data FDn.

  By applying an image processing filter to the original image data OD, filter image data FDn that approximates an image (exposure pattern) projected on the surface of the substrate can be acquired.

  However, in the filter image data FDn obtained by performing the filter image processing on the original image data OD, the amount of light in the area where the light hits is not constant. That is, in the exposure using the original image data OD, the light amount is increased or decreased at the end of the line & space compared to the center of the line & space, and the exposure amount on the surface of the substrate is the line & space pattern. It can be seen that it is biased depending on the shape or pattern position.

  Therefore, the original image data OD is corrected so that the amount of light in the area where the light hits in the filter image data FDn is constant. Specifically, as described in the following <Step 5> to <Step 10>, the original image data OD is corrected, the filter image processing is performed on the corrected original image data OD, and the corrected original image data New filter image data of OD is acquired. Then, the correction of the original image data OD is repeated until the amount of light in the area where the light hits in the new filter image data becomes constant.

<Step 5 (Process P5 shown in FIG. 2)>
As shown in FIGS. 8 and 9A and 9B, the original image data OD is divided by the filter image data FDn to obtain corrected image data CD (= OD / FDn) obtained by correcting the original image data OD. .

  That is, for all the pixels, the corrected image data CD is obtained by calculating using the light quantity of the pixels whose positions in the original image data OD and the positions in the filter image data FDn are the same. The original image data OD is represented by binary data, but the corrected image data CD is represented by a multi-bit luminance gradation, for example, a 16-bit luminance gradation.

<Step 6 (Process P6 shown in FIG. 2)>
As shown in FIGS. 10A and 10B, the filter image processing is performed on the corrected image data CD obtained by correcting the original image data OD to obtain new filter image data FD (n + 1).

  By applying an image processing filter to the corrected image data CD, new filter image data FD (n + 1) that approximates an image (exposure pattern) projected on the surface of the substrate can be acquired.

  In the new filter image data FD (n + 1) obtained by performing the filter image processing on the corrected image data CD, the amount of light in the area where the light hits is the filter image data FDn shown in FIGS. 7A and 7B. Closer to a certain value.

<Step 7 (Step P7 shown in FIG. 2)>
In the new filter image data FD (n + 1) obtained by performing the filter image processing on the corrected image data CD, it is determined whether or not the amount of light in the area where the light hits satisfies a desired constant value.

  In the new filter image data FD (n + 1), when the amount of light in the area where the light hits does not satisfy a desired value, the process proceeds from <Step 8> to <Step 9>, and the new filter image data FD (n + 1) When the amount of light in the area where the light hits satisfies a desired constant value, the process proceeds to <Step 10>.

<Step 8 (Step 8 shown in FIG. 2)>
In the new filter image data FD (n + 1) shown in FIGS. 10A and 10B described above, it is determined that the amount of light in the area where the light hits does not satisfy a desired constant value, and the corrected image data CD Is further corrected.

  Therefore, the corrected image data CD is replaced with the original image data OD, and the new filter image data FD (n + 1) is replaced with the filter image data FDn.

<Step 9 (Steps P5, P6, P7 shown in FIG. 2)
Using the corrected image data CD replaced with the original image data OD and the new filter image data FD (n + 1) replaced with the filter image data FDn, the above-mentioned <Step 5>, the above-mentioned <Step 6>, and the above-mentioned <Step Repeat 7>.

  First, in the same manner as described in <Step 5> above, a new filter image obtained by replacing the original image data OD (corrected image data CD replaced with the original image data OD) with the filter image data FDn (filter image data FDn). Divided by the data FD (n + 1)), the corrected image data CD (= OD / FDn) is obtained again.

  Subsequently, in the same manner as described in <Step 6> above, the filter image processing is performed on the corrected image data CD to obtain new filter image data FD (n + 1).

  Subsequently, in the same manner as described in <Step 7> above, in the new filter image data FD (n + 1), it is determined whether or not the amount of light in the area to which the light hits satisfies a desired constant value.

  Then, in the new filter image data FD (n + 1), the corrected image data CD is further corrected when the amount of light in the area where the light hits does not satisfy a desired constant value. That is, in the same manner as described in <Step 8> above, the corrected image data CD is replaced with the original image data OD, and the new filter image data FD (n + 1) is replaced with the filter image data FDn. 5>, <Step 6> described above, and <Step 7> described above are repeated.

  In this way, the original image data OD until the amount of light in the area that is irradiated with light in the new filter image data FD (n + 1) reaches a desired constant value, in other words, until the luminance of the pixels with the micromirrors ON is sufficiently aligned. Repeat the correction calculation. Then, corrected image data CD that can obtain new filter image data FD (n + 1) in which the amount of light in the region to which the light hits has a desired constant value is obtained by the filter image processing.

  In the new filter image data FD (n + 1), if the amount of light in the area where the light hits satisfies a desired constant value, the process proceeds to <Step 10>.

  In the present embodiment, new filter image data FD5 (FIG. 11) obtained by performing filter image processing on the corrected image data CD4 (FIGS. 11A and 11B) obtained by repeating the correction calculation of the original image data OD four times. 12 (a) and 12 (b)), the amount of light in the region where the light hits could satisfy a desired constant value.

<Step 10 (process P9 shown in FIG. 2)>
In the new filter image data FD (n + 1), when the amount of light in the area that is exposed to light satisfies a desired constant value, the corrected image data CD is converted into desired gradation grayscale drawing data. Then, using this drawing data, a desired exposure pattern is projected onto the surface of the substrate.

  In the present embodiment, as shown in FIGS. 12A and 12B, the amount of light in a region that is exposed to light in the new filter image data FD5 has a substantially constant value. That is, since the light amount is substantially constant at the end and center of the line & space, the exposure using the corrected image data CD4 shown in FIGS. 11A and 11B exposes the surface of the substrate. Since the amount does not depend on the pattern shape or pattern arrangement of the line and space, a line and space resist pattern having a desired processing shape can be formed on the surface of the substrate.

  In the present embodiment, the correction calculation of the original image data OD is repeated four times to obtain the corrected image data CD4 in which the exposure amount on the surface of the substrate 20 does not depend on the shape of the exposure pattern. Needless to say, the number of corrections is not limited to this.

  FIG. 13A is a plan view of a resist pattern formed on a substrate by exposure using original image data, and FIG. 13B shows correction calculation of the original image data according to the present embodiment a plurality of times. It is a top view of the resist pattern formed on the board | substrate by exposure using the correction image data after repeating.

  In the exposure using the original image data shown in FIG. 13A, the line located outside the repeated pattern of lines and spaces is thin and short. On the other hand, in the exposure using the corrected image data according to the present embodiment shown in FIG. 13B, the thickness and length of the line located outside the repetitive pattern of the line and space are in other areas. The thickness and length of the line positioned are almost the same, and a desired pattern of repeated lines and spaces is formed.

  As described above, according to the present embodiment, it is possible to create corrected image data in which the bias of the exposure amount is reduced by repeating the correction for applying the filter image processing to the original image data a plurality of times. Thereby, the bias of the exposure amount on the substrate depending on the shape of the exposure pattern can be suppressed, and a resist pattern having a desired processing shape can be formed on the substrate. As a result, the production yield of products can be improved.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the present invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Needless to say.

DESCRIPTION OF SYMBOLS 10 Exposure apparatus 20 Substrate 30 Exposure optical system 31 Exposure light source 32 Collimator lens 33a, 33b, 33c Mirror 34 DMD
35 relay lens 36 objective lens 40 autofocus optical system 41 focus light source 42 collimator lens 43 half mirror 44 half mirror 45 imaging lens 46 CCD
50 stages

Claims (4)

An exposure method for projecting an exposure pattern on the surface of a substrate by turning on or off each of a plurality of micromirrors arranged in a grid in a DMD,
(A) a step of preparing original image data;
(B) applying an image processing filter to the original image data to obtain filter image data that approximates an exposure pattern projected on the surface of the substrate;
(C) correcting the original image data by dividing the original image data by the filter image data to obtain corrected image data;
(D) applying the image processing filter to the corrected image data to obtain new filter image data that approximates an exposure pattern projected on the surface of the substrate;
(E) a step of determining whether or not the amount of light in a region that is exposed to light in the new filter image data is constant;
Including
In the step (e), when the amount of light in the area where the light hits indicates a desired value, the corrected image data is converted into gradation grayscale drawing data,
In the step (e), when the amount of light in the area where the light hits does not indicate a desired value, the corrected image data is replaced with the original image data, and the new filter image data is replaced with the filter image data. Furthermore, the exposure method which repeats the said (c) process, the said (d) process, and the said (e) process. The exposure method according to claim 1, wherein
An exposure method in which the image processing filter is an amount of light obtained from each divided portion after the light spread area of one pixel is divided into a plurality of areas after the light spread of one pixel is obtained from a theoretical formula. The exposure method according to claim 2, wherein
The one pixel is a square having a side of 0.5 μm,
The light spreading region is a quadrilateral with a side of 5 μm,
The one pixel is positioned at the center of the light spreading region, and the light processing amount of the light spreading region is normalized by setting the light amount of the one pixel to 1, thereby obtaining the image processing filter of the light spreading region. , Exposure method. The exposure method according to claim 2, wherein
The exposure method in which the one pixel corresponds to one micromirror.