JP2011082289A - Drawing device and drawing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accurately form a pattern without causing a pattern displacement even when the mechanism of a drawing device has an accuracy error. <P>SOLUTION: In the drawing device provided with an exposure head, a substrate for measurement is loaded on a drawing table and an exposure operation is performed on the basis of the pattern data of a reference spot pattern. Then, the position of a spot pattern is measured by a measuring instrument outside the device and the position coordinate data are stored in a memory in the drawing device. When drawing a pattern, the pattern data PD of a drawing pattern are divided into respective rectangular areas stipulated by the reference spot pattern, and closed area pattern data PD1 and PD2 of the respective rectangular areas LM1 and LM2 are corrected matched with deformed rectangular areas LN1 and LN2 deformed or displaced based on the measured position coordinates of the spot pattern. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a maskless exposure apparatus that directly draws a pattern using a spatial light modulator such as a DMD (Digital Micro-mirror Device), and more particularly to correction of pattern deformation caused by the mechanism of the exposure apparatus.

  In a maskless exposure apparatus equipped with a spatial light modulation element such as a DMD, an exposure operation is performed by a light modulation array in which spatial light modulation elements (cells) are two-dimensionally arranged, and a pattern is directly formed on a drawing surface such as a substrate or a film. To do. During the exposure, the drawing table on which the substrate is mounted moves relative to the illumination light projection area, and the drawing light is raster-scanned while reciprocating the drawing table to form a drawing pattern.

  The substrate and the film itself are deformed by heat treatment and lamination. Therefore, an alignment adjustment mark is provided on the substrate in advance, and when the mark is measured with the substrate deformed, the drawing data is corrected based on the measurement position. Further, since local deformation occurs in the drawing area, the amount of deformation is detected for each of a plurality of rectangular areas defined by dividing the drawing area, and the drawing data is corrected according to the deformed rectangular area ( For example, see Patent Document 1).

  On the other hand, because there is an accuracy error in the parts of the drawing table transport mechanism such as the guide rail, the table meanders by the yawing motion during the exposure operation, and the pattern deformation along the main scanning direction occurs due to the speed fluctuation. Unless the drawing table is moved at a stable speed along the scanning direction, it is difficult to form an accurate drawing pattern on the substrate.

  In order to eliminate such deformation of the drawing pattern caused by the operation of the drawing apparatus itself, the amount of change in the table moving direction is measured in advance by a measuring instrument such as a reference scale. Then, the substrate position is corrected in accordance with the position variation amount of the table, and a drawing pattern is formed at an appropriate position (see Patent Document 2). Further, the position of the alignment mark is measured by a camera provided in the drawing apparatus, and the drawing data is corrected based on the deviation of the mark position so as to cancel yawing (see Patent Documents 2 and 3).

JP 2008-3504 A JP 2006-309023 A JP 2005-316411 A

  As the pattern becomes finer and more complex in recent years, the size of the projection area of the illumination light is adjusted according to the pattern content (specifically, the magnification of the exposure optical system is changed). When the size of the projection area is changed, the scanning path of the projection area being exposed (projection area passing path) also changes depending on the drawing pattern.

  On the other hand, table position fluctuations caused by accuracy errors of the transport mechanism and the like are caused by various operational disturbances such as yawing and movement speed fluctuations. Therefore, the amount of variation in the position of the substrate mounted on the table cannot be determined uniformly for one substrate, and differs at each point on the substrate.

  Therefore, in order to form the drawing pattern at an accurate position on the substrate, the position variation amount must be acquired for every position on the substrate. On the other hand, since it is necessary to quickly process drawing data with a huge amount of data and improve the throughput of the substrate, the correction processing of the drawing data based on the positional variation amount of the substrate must be performed efficiently.

  The drawing method of the present invention is a drawing method in a drawing apparatus that forms a pattern using a light modulation element array (light modulator) including a spatial light modulation element such as a DMD, and includes a plurality of spatial light modulation elements. An object to be measured is mounted on a drawing table of a drawing apparatus equipped with a light modulation array, and an exposure operation is performed while moving the drawing table based on pattern data of a spot pattern (hereinafter referred to as a reference spot pattern) that defines a grid. Execute.

  The reference spot pattern represents a two-dimensional array pattern arranged at a predetermined interval along two orthogonal directions so as to define a grid. When the exposure operation is executed based on the pattern data, the table yawing or the like occurs due to the accuracy error (variation) of the drawing table mechanism parts, and pattern position deviation occurs. For this reason, the spot pattern formed on the substrate or the like does not match the reference spot pattern (the pattern is not formed at the position where it should be).

  In the present invention, the position of the spot pattern formed on the measurement object is measured by a measuring instrument before pattern drawing. The measuring instrument can measure the position of the spot pattern without moving the drawing table. For example, the position of the spot pattern can be determined without moving the drawing object for measurement, such as a two-dimensional coordinate measuring instrument for verifying the exposure result by the exposure apparatus. A measuring instrument capable of measuring is used.

  A pattern drawing object in which the same coordinate system as the measurement drawing object is defined is mounted on the drawing table, and an exposure operation is performed based on the pattern data of the drawing pattern. In the present invention, the pattern data of the drawing pattern is divided into a plurality of closed area pattern data according to each rectangular area of the grid. The pattern data is represented as vector data, for example, and the drawing pattern is expressed by the position coordinates of feature points such as end points and vertices along the outline of the drawing pattern. Then, closed region pattern data is represented by the position coordinates (vector data) of feature points such as vertices and end points in each rectangular region.

  The grid defined by the actually measured spot pattern is not composed of rectangular regions of the same size as the grid of the reference spot pattern, and there may be a deviation along the main scanning direction or the sub scanning direction. Each rectangular area is deformed by a spot pattern position shift, and even if the size is substantially the same, the position of the entire rectangular area is displaced (varied) by yawing or the like. As a result, partial pattern position errors are scattered throughout the entire substrate.

  In the present invention, the closed region pattern data of each rectangular region is corrected so that there is no pattern deformation based on the displacement / deformation of each rectangular region resulting from the positional deviation of the measurement spot pattern with reference to the reference spot pattern. That is, each closed region pattern data is corrected so as to cancel out pattern position errors caused by a table mechanism or the like. In accordance with the corrected pattern data, the exposure operation is executed while moving the drawing table.

  In this way, by measuring the pattern position fluctuation amount in advance using the spot pattern that defines the grid, it becomes possible to detect the position fluctuation amount at any position of the drawing object, and to change the grid size of the grid, etc. As a result, the position variation amount can be detected without being limited by the position interval of the alignment mark, the size of the projection area (exposure area), or the like.

  Then, the drawing data is divided into a plurality of data according to the grid, and the data in each rectangular area (hereinafter referred to as closed area pattern data) is corrected according to the deformation and displacement of each rectangular area. As a result, a drawing pattern having no pattern displacement as a whole is formed on a drawing object such as a substrate.

  On the other hand, a drawing apparatus of the present invention includes a light modulation array composed of a plurality of spatial light modulation elements, a drawing table that moves relative to an exposure area that is a projection area of illumination light during an exposure operation, Exposure means for executing an exposure operation for forming a pattern is provided. A light modulation element array such as a DMD or an LCD has a plurality of spatial light modulation elements (cells) that guide illumination light from a light source to a drawing object according to a pattern and selectively guide the illumination light to the drawing object or the outside of the drawing object. ).

  When scanning is performed by moving the drawing table in order to irradiate the drawing object with illumination light, for example, a step-and-repeat method in which the exposure area is relatively moved intermittently, or a continuous movement method in which the exposure table is continuously moved is applied. By controlling the spatial light modulator in accordance with the pattern (for example, ON / OFF control) during scanning, a predetermined pattern is formed on the drawing object. For example, vector data input to the apparatus as pattern data is converted into raster data, and exposure data for controlling each light modulation element is generated from the raster data.

  The drawing apparatus of the present invention includes data storage means for storing the position of a spot pattern formed on a measurement drawing object as data in a memory based on pattern data of a reference spot pattern that defines a grid. Further, the drawing apparatus divides the pattern data of the drawing pattern into a plurality of closed area pattern data according to each rectangular area of the grid, and each rectangle resulting from the positional deviation of the measured spot pattern with respect to the reference spot pattern Data correction means is provided for correcting the closed area pattern data of each rectangular area so as not to cause a pattern position shift in accordance with the deformation / displacement of the area. In accordance with the corrected pattern data, an exposure operation is performed on the pattern drawing object.

  When the pattern data is represented by vector data, the closed area pattern data of each rectangular area is represented by the position coordinates of the end points and vertices of the pattern outline. Considering that the pattern data correction is performed according to the displacement and deformation of the rectangular area, the data correction means will divide the feature points (vertices, end points) of the closed area pattern data along the boundary line of each rectangular area. It is preferable to calculate the ratio and correct the position of the feature point so that the internal division ratio (line segment ratio) is maintained even in the rectangular region where the deviation occurs. As a result, the data feature point on the boundary line is subjected to the data correction process while maintaining the distance relationship from both end points (closed area vertices) of the boundary line, and the data correction process can be performed in accordance with the deformation of the rectangular area. .

  For example, if there is a feature point of the closed region pattern on the boundary line of the rectangular region, the data correction means can move the feature point to a position where the internal ratio matches along the boundary line of the deformed and displaced rectangular region. Good. Alternatively, in the case of a feature point inside the rectangular area, data correction is performed according to the internal ratio of each boundary line.

  The scanning path of the exposure area is changed by adjusting the projection magnification of the illumination optical system. In consideration of detecting the position variation amount in any scanning path, the interval along the sub-scanning direction of the reference spot pattern should be set to be approximately equal to or smaller than the size of the exposure area by the light modulation element array. On the other hand, the interval along the main scanning direction of the reference spot pattern is approximately equal to or less than the pitch interval corresponding to the operation cycle of the drive mechanism that drives the drawing table (for example, a distance corresponding to one rotation of the bearing used in the guide mechanism). It is good to decide.

  According to the present invention, even if there is an accuracy error in the mechanism of the drawing apparatus, a pattern can be formed with high accuracy without causing a shift in the pattern formation position.

It is the perspective view which showed typically the drawing apparatus which is this embodiment. It is the figure which showed the scanning path | route at the time of exposure operation | movement. It is a block diagram of a drawing apparatus. It is the figure which showed the spot pattern used for measurement. It is the figure which showed the position of the spot pattern actually detected by a measuring device. It is the figure which showed the position shift of the pattern. It is the flowchart which showed the procedure which correct | amends pattern data. It is the figure which showed correction | amendment of the pattern data shown in FIG. It is the figure which showed correction | amendment of the vector data in a rectangular area. It is the figure which showed the pattern data after correction | amendment. It is a flowchart of the exposure operation process executed by the drawing control unit.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a perspective view schematically showing a drawing apparatus according to the present embodiment. FIG. 2 is a diagram showing a scanning path during the exposure operation.

  A drawing apparatus (exposure apparatus) 10 is a maskless exposure apparatus that forms a pattern by irradiating light onto a substrate SW coated or affixed with a photosensitive material such as a photoresist. 14. A table drive mechanism 19 that supports the drawing table 18 is mounted on the base 14, and the substrate SW is placed on the drawing table 18 by a suction mechanism (not shown).

  The gate-like structure 12 is provided with light sources 16A and 16B, and two exposure heads 20A and 20B for forming a pattern on the surface of the substrate SW are installed at a predetermined interval. Further, a CCD camera 13 for detecting the deformation state of the substrate SW is installed. Each of the exposure heads 20A and 20B has a DMD in which a plurality of micromirrors (for example, 1024 × 768) are arranged in a matrix, an illumination optical system that irradiates parallel light on the DMD, and images the reflected light of the DMD on the substrate SW. An imaging optical system is provided.

  The rectangular substrate SW is a substrate for electronic circuits such as a silicon wafer, a printed substrate, a dry film, a glass substrate, and a copper-clad laminate, and is a drawing table in a blank state that has been subjected to processing such as application of a photoresist. 18 is mounted. Further, as described later, a measurement substrate is prepared separately from the pattern formation substrate SW.

  The substrate SW, that is, the drawing table 18, defines an XYZ coordinate system orthogonal to each other, and the drawing table 18 can be moved along the X and Y directions by the table driving mechanism 19. The table drive mechanism 19 includes two pairs of guide rails facing each other along the X direction and the Y direction, and a linear drive unit (both not shown), and the drawing table 18 extends along the guide rail by the linear drive unit. Moving. Here, the X direction is defined as the main scanning direction, and the Y direction is defined as the sub scanning direction.

  In the drawing apparatus 10, an exposure operation is executed and controlled by a drawing control unit (not shown here). An input device such as a monitor and a keyboard (both not shown) is connected to the drawing control unit, and drawing processing is performed according to the operation of the operator.

  Here, a multiple exposure method using a continuous movement method is applied as the exposure method. That is, the drawing table 18 moves at a constant speed, and each micromirror is ON / OFF controlled at a predetermined exposure pitch interval accordingly. The exposure operation may be performed by a multiple exposure method using a step & repeat method.

  Light emitted from the light sources 16A and 16B is guided to the exposure heads 20A and 20B by an illumination optical system (not shown). Based on the drawing pattern data, the DMDs of the exposure heads 20A and 20B form pattern light. That is, each micromirror of the DMD is selectively switched on / off according to the drawing pattern data, and the light reflected on the DMD is irradiated onto the substrate SW as illumination light.

  As the substrate SW moves along the main scanning direction (X direction), projection areas (hereinafter referred to as exposure areas) EA1 and EA2 by the exposure heads 20A and 20B move relative to the substrate SW ( (See FIG. 2). The size of the exposure areas EA1 and EA2 depends on the optical magnification of the imaging optical system provided in the exposure heads 20A and 20B.

  The substrate SW can be arranged on the drawing table 18 with a slight inclination with respect to the main scanning direction. In this case, when the drawing table 18 moves along the main scanning direction, the exposure area moves relative to the longitudinal direction of the substrate SW, that is, in a state inclined with respect to the main scanning direction.

  While the drawing table 18 moves along the main scanning direction, an exposure operation is performed based on the relative position of the exposure area with respect to the substrate SW. That is, each micromirror of the DMD is independently turned on / off according to the pattern to be formed in the area. When the exposure for the scanning band for one line is completed, the drawing table 18 moves in the sub-scanning direction (Y direction), and the exposure operation is performed along the next scanning band.

  Raster scanning proceeds while relatively moving the exposure area along the main scanning direction, and a pattern is formed on the entire substrate. As shown in FIG. 2, the exposure heads 20A and 20B are spaced apart by a predetermined distance along the sub-scanning direction, and each forms a pattern on a half region of the substrate SW. The exposure areas EA1 and E2 by the exposure heads 20A and 20B proceed along the scanning bands K1 and K2, respectively. When the drawing process is completed, a development process, etching or plating, a resist stripping process, and the like are performed, and a substrate on which a pattern is formed is manufactured.

  The CCD camera 13 disposed on the slider 31 of the drawing apparatus 10 measures the alignment mark formed on the substrate SW. The CCD camera 13 can be moved in the sub-scanning direction by a drive unit (not shown). Alignment correction is performed based on the positions of the alignment marks formed at the four corners of the substrate SW.

  FIG. 3 is a block diagram of the drawing apparatus.

  The drawing control unit 30 is connected to an external workstation (not shown), and the exposure control unit 30 outputs a control signal to each circuit such as the data calculation unit 42 and the light source control unit 44. A program for controlling the drawing process is stored in advance in a ROM (not shown) in the system control circuit 32.

  Pattern data transmitted from a workstation (not shown) is vector data (CAD / CAM data) having drawing pattern position information, and a position based on an XY coordinate system having a reference point on the substrate SW. Expressed as coordinate data. The vector data input to the data input unit 41 is stored in the data buffer 43 and then sent to the data operation unit 42. In the data calculation unit 42, the vector data is converted into raster data which is two-dimensional dot pattern data (ON / OFF data).

  In the DMD control unit 34, the raster data sent from the data calculation unit 42 is output to the DMDs 22A and 22B provided in the exposure heads 20A and 20B, respectively, at a predetermined timing. The raster data to be output controls the micromirrors of each of the DMDs 22A and 22B as exposure data.

  The table position control unit 38 controls the table drive mechanism 19 via a drive circuit (not shown), and thereby the moving speed of the drawing table 18, the substrate feed direction, and the like are controlled. Further, a position detection sensor (not shown) detects the position of the drawing table 18 that is relatively moved, and thereby the relative position of the exposure area of each exposure head is detected.

  The alignment mark detection unit 40 detects the position of the alignment mark formed on the substrate SW based on the image signal from the CCD camera 13. When the substrate SW is deformed by the influence of heat or the like, a displacement of the alignment mark is detected. The drawing data (vector data) stored in the data buffer 43 is corrected based on the misalignment of the alignment mark.

  As will be described later, before the exposure operation for forming the drawing pattern on the substrate SW, the pattern misalignment measurement operation using the dedicated measuring instrument 50 is performed. A measurement substrate SW0 different from the substrate SW on which the drawing pattern is formed is mounted on the drawing table 18, and the measurement pattern is formed on the substrate SW0. The measurement substrate SW0 and the substrate SW are coordinate systems having the same reference position, and the same place has the same position coordinate. When the measurement pattern is formed on the substrate SW0, the position of the spot pattern is measured by a conventionally known two-dimensional coordinate measuring instrument 50.

  The pattern position information measured by the measuring instrument 50 includes information on the pattern position deviation caused by the accuracy error of the table drive mechanism 19, is input to the drawing apparatus 10 via a recording device, a computer, etc., and is stored in the memory 33. The In the data calculation unit 42, drawing data (vector data) is corrected based on the pattern position information.

  FIG. 4 is a diagram showing spot patterns used for measurement. FIG. 5 is a diagram showing the position of the spot pattern actually detected by the measuring instrument.

  A spot pattern (hereinafter referred to as a reference spot pattern) PX0 shown in FIG. 4 is a dedicated pattern for measuring a pattern position deviation, and a plurality of minute circular patterns are spaced along the main scanning direction by an interval PX and a sub scanning direction. Are arranged in a matrix at intervals PY.

  The reference spot pattern PX0 is a regular pattern that defines the grid GL, and the points (lattice points) at which the grid GL intersects coincide with the spot center, and the straight lines connecting the spot centers are parallel or orthogonal to each other. However, when an exposure operation is performed based on the pattern data of the spot pattern PX0, as shown in FIG. 5, a spot pattern PX having a misalignment is formed on the substrate SW0.

  The position of the spot pattern PX actually formed on the substrate SW0 does not coincide with the position of the spot pattern PX0 shown in FIG. Each spot is not formed so as to define the grid GL shown in FIG. This is because the spot pattern is displaced due to the accuracy error of the table driving mechanism 19.

  As described above, the table drive mechanism 19 includes the guide rail and the linear drive unit. However, the drawing table is caused by a mechanical accuracy error due to the sphericity of the ball bearing provided on the guide rail, the straightness of the guide rail, or the like. 18 does not move linearly and does not move at a constant speed. The drawing table 18 moves while varying the speed along the main scanning direction while yawing.

  Therefore, each spot center position of the spot pattern PX that is actually formed deviates from the lattice point of the grid GL that should originally be. The measuring device 50 is a two-dimensional coordinate measuring device that detects the positional deviation of the spot. The grid GL is defined on the surface of the substrate SW0, and the grid point of the grid serving as the reference coordinate and the grid point are actually measured. A deviation from the center position of the formed spot is detected.

FIG. 5 shows the positional deviation of the pattern for one spot pattern SP _31. The spot center is shifted from the lattice point L by ΔP (MPX31, MPY31). Such misregistration also occurs in other spot patterns.

  In the present embodiment, pattern data correction processing is performed so as to eliminate (compensate) errors in pattern formation positions caused by the table drive mechanism 19. While the drawing table 18 is moving, the positional deviation of the pattern occurs in the entire substrate SW0, and the deviation amount varies depending on the pattern forming position. By forming the spot pattern PX that is finely two-dimensionally arranged on the substrate SW0, it is possible to detect the amount of displacement at an arbitrary position of the substrate SW0.

  Pattern position shifts along the main scanning direction due to speed fluctuations appear in the main scanning direction. Therefore, the spot interval PX of the spot pattern PX0 is determined, for example, as a distance for one rotation of the support ball bearing of the guide rail.

  On the other hand, the pattern position shift due to the yawing operation appears as a shift in the movement path (scanning path) of the exposure area. For this reason, the spot interval PY in the sub-scanning direction is determined by the bandwidth BW of the exposure area. Here, as shown in FIG. 4, each spot is formed at the center position along the Y direction of the scanning band SB.

  FIG. 6 is a diagram showing the positional deviation of the spot pattern.

  In FIG. 6, two rectangular areas (lattice areas) LM1 and LM2 to which a part of the pattern PD belongs are taken up. The rectangular areas LM1 and LM2 are grid areas of the grid GL defined by the previous spot pattern PX0. When no positional deviation occurs in the spot pattern, the spot center position of the spot pattern actually measured coincides with the vertices (grid points) a to f of the rectangular areas LM1 and LM2.

  However, since an accuracy error of the table driving mechanism 19 occurs, a deviation occurs between the center position of the spot pattern actually formed on the substrate SW and the apexes a to f. The center position of the spot pattern to be measured is represented by “a1 to f1” here, and a rectangular area defined by connecting the center positions a1 to f1 of the spot pattern (hereinafter referred to as a deformed rectangular area) is “LN1”, This is expressed as “LN2”.

  In the spot pattern position shift shown in FIG. 6, the center positions b1 and e1 of the spot pattern are shifted by the shift amount PX1 along the main scanning direction, and are shifted by the shift amounts PY1 and PY2 respectively along the sub-scanning direction. Yes. Further, the center positions d1 and c1 of the spot pattern are shifted by a shift amount PX2 along the main scanning direction.

  The pattern data is corrected in consideration of such a positional deviation of the spot pattern. Specifically, pattern data represented by vector data is divided for each rectangular area, and the pattern data is corrected according to the position variation and deformation of each rectangular area based on the positional deviation of the spot pattern.

  FIG. 7 is a flowchart showing a procedure for correcting pattern data.

  After the exposure operation is performed based on the pattern data of the reference spot pattern, the actual spot pattern position is measured by the measuring instrument 50 (S101, S102). Then, measurement data is input to the drawing apparatus 10 via a network connection or a portable memory (S103). When drawing pattern data is input to the drawing apparatus, pattern data correction processing described below is performed (S104, S105).

  FIG. 8 is a diagram showing correction of the pattern data shown in FIG.

  The pattern data PD represents a drawing pattern by vector information of a point group that represents a pattern feature such as a vertex or an end point of the drawing pattern. That is, the pattern data PD is represented as frame line information when the drawing pattern is a closed figure.

  In the present embodiment, the drawing pattern is divided for each rectangular area, and the pattern area in each rectangular area is regarded as one closed pattern area. Then, the pattern data PD is divided for each rectangular area, and vector data based on the frame information of the closed figure is defined. Pattern data of each rectangular area (hereinafter referred to as closed area pattern data) is corrected according to the deformed rectangular area of the grid defined by the spot pattern actually formed.

  FIG. 8 shows a part of the pattern data PD shown in FIG. Data portions belonging to the rectangular areas LM1 and LM2 are referred to as closed area pattern data PD1 and PD2, respectively. The closed region pattern data PD1 is represented by a triangle having K, G, and b as vertices. The closed region data PD2 is represented by a hexagon having b, H, I, d, J, and K as vertices.

  The deformed rectangular areas LN1 and LN2 defined by the actually measured spot patterns a1 to f1 are greatly different from the rectangular areas PD1 and PD2 defined on the data. Therefore, the closed region pattern data PD1 and PD2 are corrected according to the deformed rectangular regions LN1 and LN2.

  Specifically, the internal division ratio (line segment ratio) is calculated for the intersection between the boundary line (side) of each rectangular area (hereinafter referred to as the reference rectangular area) on the data without deformation and the closed area pattern data. The closed region pattern data PD1 and PD2 are corrected so as to maintain the internal division ratio. The internal ratio is calculated based on the position coordinates of each vertex of the reference rectangular area, and the closed area pattern data in the deformed rectangular area whose vertex is the point at which the internal ratio is maintained (hereinafter, the deformed closed area pattern data PDW1, PDW2).

The closed region pattern data PD1 will be described as an example. When the deformed rectangular region LN1 is defined by the actually measured spot patterns a1, b1, e1, and f1, the vertex G of the closed region pattern data PD1 maintains the internal division ratio. While moving to GG. The vertex GG is calculated based on the following equation (where ab = PX, a1b1 = PX + PX1).

aG: ab = a1GG: a1b1 (1)

  The position coordinate of the intersection G between the reference rectangular area LM1 and the closed area pattern data PD1 is one vector data representing the pattern shape of the closed area pattern data PD1. Therefore, one vector data GG of the deformation area pattern data PDW1 is obtained by the expression (1).

Similarly, the vertex K of the closed region pattern data PD1 is moved to KK based on the following expression (where eb = PY, e1b1 = PY1-PY2).

eK: eb = e1KK: e1b1 (2)

When the position coordinates of KK and GG are obtained, the position coordinates of b1, KK and GG, that is, modified closed region pattern data PDW1 represented by vector data is generated.

  Also for the closed region pattern data PD2, points JJ, KK, HH, and II on the boundary line of the deformed rectangular region LN2 corresponding to the intersections J, K, H, and I are calculated. Then, the modified closed region pattern data PDW2 is obtained from the position coordinates of the six intersections e1, b1, HH, II, d1, and JJ on the boundary line of the deformed rectangular region LN2.

  In FIG. 8, the vector data of the closed area pattern data is located on the boundary of the reference rectangular area. However, when the vector data representing the closed area pattern data is inside the reference rectangular area, that is, the end points of the drawing pattern are When belonging to the inside of the reference rectangular area, the closed area pattern data is corrected as follows.

  FIG. 9 is a diagram showing correction of vector data in the rectangular area.

  In FIG. 9, the end point t of the closed region pattern data PDS in a certain rectangular region exists inside the reference rectangular region LT having D1 to D4 as vertices. With respect to the vector data of the end point t represented by the position coordinates (tX, tY), internal ratios (tX / PX, tY / PY) along the X direction and the Y direction are obtained. And the position of the end point T in the inside of the deformation | transformation rectangular area | region LTW which makes C0-C3 a vertex is defined so that the internal ratio may be maintained.

  Specifically, the line segment C0-C1 and the line segment C2-C3 are arranged on the boundary lines C2-C3, C0-C1 of the deformed rectangular area LTW corresponding to the boundary lines D2-D3, D0-D1 of the reference rectangular area LT. Points S and U having an internal ratio tX / PX with respect to are respectively determined. Similarly, on the boundary lines C1-C2, C0-C3 of the deformed rectangular area LTW corresponding to the boundary lines D1-D2, D0-D3, there are internal division ratios tY / PY with respect to the line segments C1-C2, C3-C0. Points R and V are determined. The correction position T of the end point t is obtained according to the calculated position coordinates of the points S, R, U, and V.

  FIG. 10 is a diagram showing the corrected pattern data.

  As shown in FIGS. 8 and 9, vector data correction processing is performed on the closed region pattern data of each reference rectangular region, and as a result, closed region pattern data corresponding to the deformation and displacement of each rectangular region is generated separately. The Pattern data PD 'is generated by connecting the corrected closed region pattern data. However, in FIGS. 8 to 10, the deformation of the rectangular area is exaggerated in order to explain the data correction process.

  FIG. 11 is a flowchart of the exposure operation process executed by the drawing control unit 30. After the spot pattern position measurement and the measurement data input to the drawing apparatus are completed, the execution is started.

  In step S201, drawing pattern pattern data is input. In step S202, it is determined whether or not the input pattern data is pattern data used in the previous exposure process. If the pattern data has been used before, the pattern data correction processing is not performed and the process proceeds to step S207.

  On the other hand, if the pattern data is input for the first time, the process proceeds to step S203, and the position of the alignment mark is measured. In step S204, correction processing for the pattern data is performed based on the position coordinates of the measured spot pattern recorded in advance. That is, the pattern data is divided for each rectangular area, and the closed area pattern data of each rectangular area is corrected in accordance with the displacement amount, deformation, etc. of the rectangular area. As a result, the pattern data is corrected so as to cancel the pattern position error at each location on the substrate.

  In step S205, the measured position of the alignment mark is corrected according to the measured positional deviation of the spot pattern. Based on the corrected alignment mark position, the reference position coordinates of the pattern data during the exposure operation are corrected. Further, by detecting the deformation amount of the entire substrate from the position of the alignment mark, conventionally known correction amounts such as the rotation amount, offset amount, and scale ratio are obtained, and the pattern data is corrected. In step S206, the pattern data corrected for each rectangular area is temporarily stored in the data buffer 43.

  In step S207, the corrected pattern data is read from the data buffer 43, and a raster conversion process is executed in the data calculation unit. In step S208, the DMDs 22A and 22B are controlled based on the raster data. As a result, a pattern is formed on the entire substrate.

  As described above, according to the present embodiment, in the drawing apparatus 10 including the exposure heads 20A and 20B, the measurement substrate SW0 is mounted on the drawing table 18, and an exposure operation is performed based on the pattern data of the reference spot pattern PX0. . Then, the position of the spot pattern is measured by the measuring instrument 50 outside the apparatus. The position coordinate data of the measured spot pattern is stored in the memory 33 of the drawing apparatus 10.

  When drawing a pattern, the pattern data PD of the drawing pattern is divided for each rectangular area defined by the reference spot pattern PX0. Then, the closed region pattern data of each rectangular region is corrected in accordance with the rectangular region where displacement or deformation has occurred based on the measured position coordinates of the spot pattern. Thereby, the corrected closed area pattern data is generated for each rectangular area.

  Due to the straightness (straightness) of the guide rail of the table drive mechanism 19 and the accuracy error of the ball bearing, yawing and speed fluctuation occur during the movement of the table 18. However, the characteristics of the table 18 are detected in advance, and the pattern data is corrected for each minute rectangular area so as not to cause pattern deformation. For this reason, a drawing pattern having no pattern displacement as a whole is formed on the substrate SW. Further, since the same correction pattern data is used while the same drawing pattern is formed, it does not hinder throughput improvement.

  Since the intervals PX and PY in the main scanning direction and the sub-scanning direction of the reference spot pattern are determined according to the exposure area width and the rotation distance of the ball bearing, it is possible to detect the pitch fluctuation and yawing of the table 18 in detail. The pattern data suitable for each position is corrected reliably. Even if the scanning band (scanning path) is changed by changing the magnification of the imaging optical system of the exposure head, the pattern data is corrected based on a small rectangular area, so the data correction affects the change in magnification. There is no.

  Note that the intervals PX and PY may be determined to be substantially equal to or smaller than the exposure area width and one rotation distance of the ball bearing in order to detect a more precise spot pattern position shift. Alternatively, when the pattern accuracy is not so required, the intervals PX and PY may be set to a distance that also has a large exposure area width and one rotation distance of the ball bearing. Further, the intervals PX and PY may be determined according to parts that cause an accuracy error of the table drive mechanism 19 other than the guide rail.

  In the present embodiment, pattern data that is vector data is corrected. However, data correction processing may be executed on pattern data other than vector data. Further, the present invention is not limited to pattern data correction based on the internal ratio, and data correction processing may be performed so as to cancel the pattern position deviation in accordance with the displacement and deformation of each rectangular area.

10 Drawing device (exposure device)
18 Drawing table 19 Table drive mechanism 22A, 22B DMD (light modulation element array)
30 Drawing controller 50 Measuring instrument SW board PX0 Reference spot pattern LM1, LM2 Reference rectangular area PD1, PD2 Closed area pattern data

Claims (5)

A measurement object is mounted on a drawing table of a drawing apparatus having a light modulation element array composed of a plurality of spatial light modulation elements,
Based on the pattern data of the reference spot pattern that defines the grid, the exposure operation is executed while moving the drawing table,
Measure the position of the spot pattern formed on the measurement object with a measuring instrument,
Dividing the pattern data of the drawing pattern into a plurality of closed area pattern data according to each rectangular area of the grid,
According to the displacement or deformation of each rectangular area resulting from the positional deviation of the measurement spot pattern based on the reference spot pattern, correct the closed area pattern data of each rectangular area,
A drawing method comprising: performing an exposure operation while moving the drawing table on which a pattern drawing object is mounted according to the corrected pattern data. A light modulation array composed of a plurality of spatial light modulation elements;
A drawing table that moves relative to the exposure area;
Data storage means for storing the position of the spot pattern formed on the drawing object for measurement based on the pattern data of the reference spot pattern defining the grid as data in a memory;
The pattern data of the drawing pattern is divided into a plurality of closed region pattern data according to each rectangular region of the grid, and according to the displacement or deformation of each rectangular region resulting from the positional deviation of the measurement spot pattern based on the reference spot pattern, Data correction means for correcting the closed area pattern data of each rectangular area;
An exposure apparatus comprising: an exposure unit that performs an exposure operation on the pattern drawing object in accordance with the corrected pattern data.   The data correction means calculates the internal ratio of feature points of the closed area pattern data along the boundary line of each rectangular area, and maintains the internal ratio in the rectangular area where displacement or deformation has occurred. The drawing apparatus according to claim 2, wherein the position of the feature point is corrected.   The drawing measure according to claim 2, wherein an interval along the sub-scanning direction of the reference spot pattern is equal to or smaller than an exposure area size by the light modulation element array. 4. The drawing according to claim 2, wherein an interval along the main scanning direction of the reference spot pattern is equal to or less than a pitch interval corresponding to an operation cycle of a driving mechanism that drives the drawing table. apparatus.

Images