JP6116456B2 - Drawing method and drawing apparatus - Google Patents

Description

  The present invention relates to a drawing method and a drawing apparatus for performing drawing by irradiating light to a plurality of drawing areas provided on a drawing object, and more particularly to a technique for adjusting a drawing position on a drawing object.

  For example, as a method for forming a protective layer or a wiring pattern on a substrate such as a semiconductor wafer, there is a technique of performing drawing by light irradiation. In this technique, a substrate on which a photosensitive layer is formed is used as an object to be drawn, and the photosensitive layer is exposed by irradiating the object to be drawn with light modulated based on drawing data. At this time, it is necessary to adjust the drawing position for drawing at an appropriate position of the drawing object, and techniques for that purpose have been proposed.

  For example, the technique described in Patent Document 1 is to form a new drawing pattern by superimposing a drawing pattern on a substrate pattern previously formed on a substrate. Eliminating misalignment. That is, in this technique, the design data in vector format is rasterized in advance to create run length data, and the data is subjected to drawing while correcting the data for each drawing unit based on the position detection result of the background pattern. The pattern to be drawn and the base pattern are aligned without reducing the processing time.

  Incidentally, a substrate as such an object to be drawn is generally a monolithic structure in which a plurality of chip regions are formed on a single wafer. For example, as described in Patent Document 2, In some cases, a pseudo-wafer formed by arranging a plurality of chips formed separately on a substrate later becomes a drawing target.

JP 2012-0774615 A Japanese Patent No. 4724988

  In a monolithic wafer, since a plurality of chip regions are originally formed as a single unit, there is almost no relative displacement between the plurality of chip regions, and the technique described in Patent Document 1 described above is also available. This is the premise. On the other hand, in the pseudo wafer described in Patent Document 2, a relatively large positional variation on a chip basis can occur on the wafer. For this reason, there is a case where it is not possible to cope with the simple adjustment of the wafer position at the time of drawing or the correction process described in Patent Document 1. In this case, the position is detected for each chip and drawing data matching the position is created. There is a need. Therefore, it takes a relatively long time for the drawing to be completed after the wafer is used for drawing, but in principle, any chip arrangement can be dealt with.

  As described above, the magnitude of distortion in the drawing object used for drawing varies, and there are various correction methods suitable for it. However, in the conventional techniques, only a correction technique determined in advance according to the assumed magnitude of distortion is applied, and an optimal correction process has not been applied for each drawing target. It was. As a result, there has been a problem in that sufficient alignment cannot be performed and a drawing position shift occurs, and the tact time becomes long due to unnecessary processing being executed.

  The present invention has been made in view of the above problems, and in a drawing method and a drawing apparatus for performing drawing by irradiating light to a plurality of drawing regions provided in a drawing object, an optimum method corresponding to the distortion of the drawing object is provided. It is an object of the present invention to provide a technique capable of performing drawing in which a drawing position is adjusted with high accuracy by performing a registration process while eliminating waste of processing time.

  One aspect of the present invention is a drawing method in which light is drawn from a drawing means to each of a plurality of drawing regions provided in a drawing object, and the contents to be drawn to achieve the above object A first step of generating raster data corresponding to the above, a second step of positioning the drawing means with respect to the drawing target, and detecting the position of two or more of the plurality of drawing regions as detection target regions A third step, a fourth step of adjusting a drawing position on the drawing target by the drawing unit based on a position detection result of the detection target region, and a drawing target from the drawing unit based on the raster data. And a fifth step of drawing by irradiating the light at the drawing position, and in the fourth step, a relative first positional shift amount between the detection target areas detected in the third step is a first amount. Within 1 threshold And determining that the first displacement amount is within the first threshold value, the position detection result between the detection target area and a reference position set in advance corresponding to the detection target area A first adjustment process for adjusting the drawing position by adjusting a relative position between the drawing unit and the drawing object based on the amount of the position deviation, while the first position deviation amount is the first threshold value. Is detected as a secondary detection target region of at least one of the plurality of drawing regions different from the detection target region, and the detected position and the detection target in the third step It is determined whether or not the second positional deviation amount with respect to the position of the secondary detection target area estimated from the position detection result of the area is within a second threshold value, and the second positional deviation amount is the second threshold value. If it is determined that Performing a second adjustment process for adjusting the drawing position by performing correction on the raster data according to a positional deviation amount between a position detection result of the detection target region in three steps and the reference position; When it is determined that the second positional deviation amount exceeds the second threshold, the positions of all the drawing areas included in the drawing object are detected, and the raster data is regenerated based on the position detection result. A third adjustment process for adjusting the drawing position is executed.

  According to another aspect of the present invention, in order to achieve the above object, a holding means for holding a drawing object provided with a plurality of drawing areas, and the plurality of drawing objects held by the holding means. Based on the raster data, the position detection means for detecting the position of the detection target area, the data generation means for generating raster data corresponding to the contents to be drawn, with two or more of the drawing areas as detection target areas A drawing unit configured to irradiate the drawing target with light and a drawing position adjusting unit configured to adjust a drawing position on the drawing target by the drawing unit based on a detection result of the position detecting unit; The position adjusting means determines whether or not a relative first positional deviation amount between the detection target areas detected by the position detecting means is within a first threshold, and the first positional deviation amount is the first threshold value. One threshold And determining the relative position between the drawing unit and the holding unit based on a positional deviation amount between the position detection result of the detection target region and a reference position set in advance corresponding to the detection target region. When the first adjustment process for adjusting the drawing position is performed by adjusting the position, and when it is determined that the first misregistration amount exceeds the first threshold value, it is different from the detection target area among the plurality of drawing areas. Between the position of the secondary detection target area detected by the position detection means as at least one secondary detection target area and the position of the secondary detection target area estimated from the position detection result of the detection target area And determining whether the second positional deviation amount is within the second threshold value, and determining that the second positional deviation amount is within the second threshold value, the position detection result of the detection target region and the reference position No position between When the second adjustment process for adjusting the drawing position is performed by performing correction according to the amount on the raster data and it is determined that the second positional deviation amount exceeds the second threshold value, the drawing object is A third adjustment process for adjusting the drawing position by detecting the positions of all of the drawing areas included by the position detection unit and causing the data generation unit to regenerate the raster data based on the position detection result; A drawing device to be executed.

  In these inventions, the first to third adjustment processes are configured to be executable as the adjustment process for aligning the drawing unit and the drawing object, and the first and third adjustment processes are performed according to the position detection result of the drawing area in the drawing object. Automatically selected and executed. Of these, the first adjustment process is a physical position adjustment between the drawing means and the drawing object, and the amount of positional deviation with respect to the drawing means of a plurality of drawing areas provided on the drawing object is substantially uniform. This is an effective adjustment process. Further, the second adjustment process corresponds to the positional deviation that can be dealt with by correcting the raster data created in advance. On the other hand, the third adjustment process is an adjustment process for creating raster data again after grasping the position of each of a plurality of drawing areas provided in the drawing target, and requires more processing time than other adjustment processes. By creating raster data in accordance with the position of each drawing area, this process can handle even when the position variation of the drawing area is large.

  By appropriately using these adjustment processes according to the state of the drawing object, optimal alignment processing according to the distortion of the drawing object is executed, processing time is not wasted, and the drawing position is highly accurate. It is possible to achieve the purpose of performing an adjusted drawing.

  In order to enable this, in the present invention, the positions of several drawing areas are detected, and adjustment processing to be performed is determined using the detection results. Specifically, if the relative positional deviation amount (first positional deviation amount) between the plurality of drawing areas (detection target areas) is equal to or smaller than the first threshold, the drawing target object is displaced relative to the drawing means. Even if it occurs, the first adjustment process is selected because these drawing areas can be regarded as being displaced integrally. On the other hand, when the first positional deviation amount exceeds the first threshold value, it is considered that the respective drawing areas are displaced in different directions, that is, the drawing object is distorted.

  Therefore, the position of another drawing area (secondary detection target area) is detected, and the second position between that position and the position of the drawing area estimated from the position of each detection target area detected earlier. Evaluate the amount of deviation. If the second positional deviation amount is small, it means that the estimation accuracy is high, that is, the distortion mode of the drawing target can be estimated with a certain accuracy from the already acquired information. Therefore, the drawing position can be adjusted by correcting the raster data so that the estimated distortion is canceled. Therefore, in this case, the second adjustment process is selected.

  On the other hand, when the second positional deviation amount is large, there is a case where the raster data cannot be corrected. Further, for example, as in the case where the above-described pseudo wafer is used as a drawing target, each chip, that is, each drawing area, may be relatively shifted in a different direction. Then it is difficult to respond. Therefore, when the second positional deviation amount exceeds the second threshold value, the third adjustment process for generating raster data again is selected after detecting and grasping the position of each drawing area.

  As described above, according to the present invention, the positions of several drawing areas are detected, and the adjustment mode of the drawing position of the drawing object by the drawing means is selected and executed based on the detection result. Therefore, even if the distortion and misregistration modes differ for each drawing object, the optimum adjustment processing corresponding to the drawing can be selected, so that the drawing position can be adjusted appropriately, and excessive adjustment processing can be performed. This eliminates the problem that the processing time becomes longer.

  According to the present invention, the drawing position is adjusted in a manner selected in accordance with the distortion of the drawing object and the displacement of the drawing area, so that the processing time is not wasted and the drawing position is adjusted with high accuracy. Drawing can be performed.

It is a side view which shows the pattern drawing apparatus concerning one Embodiment of this invention. It is a figure which shows the board | substrate which is a drawing target object of this pattern drawing apparatus. It is a block diagram which shows the electrical structure of the pattern drawing apparatus of FIG. It is a flowchart which shows the pattern drawing operation | movement by a pattern drawing apparatus. It is a figure which shows the 1st thru | or 4th alignment mark. It is a figure which shows the example of the position shift of an alignment mark. It is a figure which shows the position of a 5th alignment mark. It is a figure explaining the position shift in the chip unit on a board | substrate. It is a figure which shows the modification of pattern drawing operation | movement.

  FIG. 1 is a side view showing a pattern drawing apparatus according to an embodiment of the present invention, and FIG. 2 is a view showing a substrate that is a drawing object of the pattern drawing apparatus. FIG. 3 is a block diagram showing an electrical configuration of the pattern drawing apparatus of FIG. The pattern drawing apparatus 1 is an apparatus that draws a pattern by irradiating light on the surface of a substrate W such as a semiconductor substrate or a glass substrate having a photosensitive material applied to the surface. For example, the pattern drawing device 1 is applied to a substrate on which an electronic device is built. It is used for drawing a pattern for forming a metal wiring. In order to show the directions in each figure uniformly, an XYZ orthogonal coordinate system is set as shown in FIG. Here, the XY plane is a horizontal plane, and the Z axis is a vertical axis. More specifically, the (−Z) direction indicates a vertically downward direction. Further, the rotation direction around the Z axis is defined as the θ direction.

  The pattern drawing apparatus 1 includes a drawing engine (pattern drawing unit) 100 and a data processing unit 200 that generates divided exposure data called strip data or divided drawing data to be given to the drawing engine 100. In the drawing engine 100, each part of the apparatus is arranged inside a main body formed by attaching a cover (not shown) to the main body frame 101 to constitute the main body, and the outside of the main body (in this embodiment, FIG. The substrate storage cassette 110 is disposed on the right hand side of the main body as shown in FIG. In this substrate storage cassette 110, one lot of unprocessed substrates W before pattern drawing are stored, and loaded onto the main body by the transfer robot 120 arranged inside the main body. Further, after the exposure process (pattern drawing process) is performed on the unprocessed substrate W, the substrate W is unloaded from the main body by the transfer robot 120 and returned to the substrate storage cassette 110. In addition, as for the board | substrate W for 1 lot accommodated in the cassette C, all may draw the same pattern and the board | substrate from which a drawing pattern differs may be mixed.

  In this main body, a transfer robot 120 is arranged at the right hand end inside the main body. A base 130 is disposed on the left hand side of the transfer robot 120. One end side region (right hand side region in FIG. 1) of the base 130 is a substrate delivery region for delivering the substrate W to and from the transfer robot 120, while the other end side region (FIG. 1). The left hand side area) is a pattern drawing area for drawing a pattern on the substrate W.

  A stage 160 that holds the substrate W placed on the upper surface in a substantially horizontal posture is provided on the base 130. The stage 160 is moved on the base 130 by the stage moving mechanism 161 in the X direction, the Y direction, and the θ direction. That is, the stage moving mechanism 161 has a Y-axis drive unit 161Y (FIG. 3), an X-axis drive unit 161X (FIG. 3), and a θ-axis drive unit 161T (FIG. 3) stacked on the upper surface of the base 130 in this order. The stage 160 is moved and positioned two-dimensionally in a horizontal plane. When the stage 160 holding the substrate W moves horizontally in the Y direction, the substrate W can be moved between the substrate delivery region and the pattern drawing region. Further, the stage 160 is rotated around the θ axis (vertical axis) to adjust the relative angle with respect to the optical head 170 described later for positioning. As such a stage moving mechanism 161, an XY-θ axis moving mechanism that has been widely used conventionally can be used.

  Further, a head support unit 140 is provided above the base 130 at the boundary position between the substrate delivery area and the pattern drawing area. In the head support portion 140, a pair of leg members 141 are erected apart from each other in the X direction toward the upper side from the base 130, and the beam members 143 are bridged so as to bridge the top portions of the leg members 141. Are arranged in the X direction. A camera (imaging unit) 150 is fixed to the side surface of the beam member 143 on the pattern drawing area side, and the surface of the substrate W (the drawing surface and the exposed surface) held on the stage 160 can be imaged.

  Further, a box 172 that houses the optical head 170 and its illumination optical system is fixedly attached to the pattern drawing region side of the head support portion 140 configured as described above. The optical head 170 modulates the light beam emitted from the light source based on strip data described later. Then, the optical head 170 exposes the substrate W held on the stage 160 to draw a pattern by emitting a modulated light beam downward to the substrate W moving immediately below the optical head 170. As a result, the drawing pattern is drawn on the base pattern formed on the substrate W by a process executed prior to the exposure process. In the present embodiment, the optical head 170 can simultaneously irradiate light in a plurality of channels in the X direction, and the X direction is referred to as a “sub-scanning direction”. Further, it is possible to draw a strip-like pattern extending in the Y direction on the substrate W by moving the stage 160 in the Y direction, and the Y direction is referred to as a “main scanning direction”.

  As shown in FIG. 2A, an example of a substrate W that is a drawing target of the pattern drawing apparatus 1 is one in which a plurality of chip regions CR are arranged on a substantially circular semiconductor wafer. In each chip region CR, an integrated circuit, a discrete circuit element, or the like is formed in advance, and this pattern drawing apparatus 1 is used when forming a metal wiring pattern thereon. The size of the chip region CR, the number of arrangements on the substrate W, the layout, and the like vary depending on the device to be manufactured.

  As shown in the enlarged view on the right side of FIG. 2A, each chip region CR is provided with an alignment mark AM for enabling the position of the chip region CR to be detected from the outside. Although the shape and position of the alignment mark AM are arbitrary, as shown in the figure, it is preferable that the alignment mark AM is provided at two or more locations as far as possible in the chip region CR. This is because not only the position of the chip region CR in the XY plane but also the rotation angle in the θ direction can be detected.

  On the other hand, the drawing from the optical head 170 to the substrate W is performed in units of band B1 as indicated by a broken line in FIG. That is, the optical head 170 performs scanning for one band by scanning and moving in the Y direction relative to the substrate W while simultaneously exposing the range of the length Wx in the X direction. By repeatedly performing drawing in units of band B1 while sequentially changing the relative position between the substrate W and the optical head 170 in the X direction, drawing is finally performed on the entire surface of the substrate W. The bandwidth Wx is determined by the apparatus configuration, and does not necessarily have a correlation with the size of the chip region CR on the substrate W that is a drawing target.

  The drawing data corresponding to one band is strip data. The actual drawing data is processed by being divided into divided block B2 units that are finer than the size of the band B1, as indicated by a dotted line in FIG.

  The pattern drawing apparatus according to the present embodiment includes a computer 200 as a data processing unit for supplying strip data suitable for the drawing engine 100 configured as described above. The computer 200 includes a CPU, a storage unit 201, and the like, and is disposed in an electrical rack (not shown) together with the exposure control unit 181 of the drawing engine 100. Further, functional blocks such as the raster data generation unit 202, the correction amount calculation unit 203, the data correction unit 204, the strip data generation unit 205, and the alignment mark detection unit 206 are performed by the CPU in the computer 200 according to a predetermined program. Is realized.

  For example, a drawing pattern to be drawn superimposed on a base pattern is described by design data in a vector format generated by an external CAD or the like. When the design data is input to the computer 200, the drawing pattern is written in the storage unit 201. Saved. Based on the design data 211, the raster data generation unit 202 generates raster data (bitmap data) corresponding to the entire surface of one substrate W. The created raster data 212 is written and stored in the storage unit 201.

  In addition, the computer 200 includes an alignment mark detection unit 206, a correction amount calculation unit 203, and a data correction unit 204 as functional blocks for correcting the relative positional deviation between each chip region CR of the substrate W and the optical head 170. I have. Specifically, the alignment mark detection unit 206 performs appropriate image processing on the image captured by the camera 150 and detects the XY coordinate position of the alignment mark AM included in the image.

  The correction amount calculation unit 203 includes design position information included in the design data as information indicating the XY coordinate position of each chip region CR when the substrate W is positioned at a normal position on the stage 160, and an alignment mark detection unit 206. The amount of displacement of the alignment mark AM from the normal position is calculated on the basis of the actual position detected by, and a correction amount necessary for canceling this is obtained. The target of correction is the physical position of the substrate W with respect to the optical head 170 and the drawing data. That is, when the positional deviation can be corrected by changing the position of the substrate W, the amount of movement of the stage 160 necessary for that is obtained as the correction amount. On the other hand, when correcting misalignment by correcting drawing data, a correction amount used when creating strip data from raster data is obtained.

  When the physical position correction of the substrate W is necessary, the correction amount calculated by the correction amount calculation unit 203 is given to the exposure control unit 181 of the drawing engine 100. The exposure control unit 181 instructs the X-axis drive unit 161X, the Y-axis drive unit 161Y, and the θ-axis drive unit 161T of the stage moving mechanism 161 to correct X, Y, and θ components according to the given correction amount. And the X-axis drive unit 161X, the Y-axis drive unit 161Y, and the θ-axis drive unit 161T operate to move the stage 160, whereby the position of the substrate W on the stage 160 relative to the optical head 170 is corrected.

  When the drawing data needs to be corrected, the data correction unit 204 corrects the raster data read from the storage unit 201 based on the correction amount given from the correction amount calculation unit 203. Based on the corrected raster data, the strip data generation unit 205 generates band unit strip data and sends it to the optical head 170. Thereby, drawing is performed in a state where the positional deviation of the substrate W is corrected.

  The correction of the position of the substrate W by the movement of the stage 160 and the correction of the drawing data can be used in combination. That is, the position of the substrate W can be corrected by moving the stage 160, and the drawing data can be corrected before being used for drawing. By moving the stage 160, the position of the substrate W in the X direction, the Y direction, and the θ direction with respect to the optical head 170 can be changed, but the moving direction of each chip region CR on the substrate W is uniform. For example, when the chip regions CR are displaced from each other due to distortion of the substrate W, it is impossible to eliminate this by only moving the stage. In order to cope with this, correction of drawing data is effective. When the position correction by the stage movement and the correction of the drawing data are used in combination, the correction amount of the drawing data is obtained after subtracting the positional deviation that is eliminated by the stage movement.

  Note that the process of creating the raster data from the design data and performing the data correction for correcting the positional deviation and distortion of the substrate W to create the strip data is described in Japanese Patent Application Laid-Open No. 2012-0774615. ) Is described in detail. Also in this embodiment, the data processing method described in Patent Document 1 can be preferably applied. Therefore, description of the specific contents of the data processing is omitted. In Patent Document 1, raster data is expressed as run-length data, but the expression format of raster data is arbitrary.

  Next, the pattern drawing operation by the pattern drawing apparatus 1 configured as described above will be described in detail with reference to FIG. FIG. 4 is a flowchart showing a pattern drawing operation by the pattern drawing apparatus. In the pattern drawing apparatus 1, a cassette 110 that stores one lot of unprocessed substrates W is transferred to the drawing engine 100, and the drawing pattern to be drawn on the substrate W and the arrangement of the chip regions CR on the substrate W are shown. When the design data 211 describing the design position information is given to the computer 200, the drawing engine 100 and the computer 200 operate as follows to draw a drawing pattern on each substrate W.

  When the computer 200 acquires the design data 211 (step S101), the computer 200 stores the design data 211 in the storage unit 201, and then starts RIP (raster image processing) processing for rasterizing the design data 211 to generate raster data 212 (step S102). . The created raster data 212 is stored and saved in the storage unit 201. On the other hand, the drawing engine 100 loads one unprocessed substrate W from the cassette 110 onto the main body by the transfer robot 120 and places it on the stage 160 (step S103). The execution of the RIP process and the board loading may be performed in parallel, or the RIP process may be started after the board loading.

  Subsequently, from the design position information included in the design data 211, four alignment marks selected from the alignment marks AM (FIG. 2) formed in each chip region CR on the substrate W, that is, first to fourth. The position of the alignment mark is specified, the area including the position is imaged by the camera 150, and the alignment mark detection unit 206 detects the position of the alignment mark from the obtained image (step S104). Then, a positional deviation amount from the normal position of each alignment mark is calculated (step S105).

  FIG. 5 is a diagram showing first to fourth alignment marks. As shown in FIG. 5A, four different chip regions CR1 to CR4 among the chip regions CR arranged on the substrate W are appropriately selected as detection targets and formed in these chip regions CR1 to CR4, respectively. The aligned alignment marks AM1 to AM4 are imaged. It is desirable that the chip regions CR1 to CR4 to be detected are dispersed as far as possible on the substrate W in order to grasp the amount of positional deviation as a whole of the substrate W.

  Here, as a simple example, the same positions in the Y direction between the chip areas CR1 and CR2, and between the chip areas CR3 and CR4, and between the chip areas CR1 and CR3 and between the chip areas CR2 and CR4, respectively. The chip regions CR1 to CR4 are selected so that they are at the same position in the X direction. In such an example, the virtual quadrilateral formed by connecting the alignment marks AM1 to AM4 provided in each chip region is composed of two sides parallel to the X direction and two sides parallel to the Y direction. It is a rectangle.

As shown in FIG. 5B, for the nth alignment mark AMn (n = 1, 2, 3, 4), a point on the XY coordinate plane representing the position of the alignment mark AMn detected from the image is defined as Pmn. The coordinates are represented by (xmn, ymn). On the other hand, a point corresponding to the normal position of the alignment mark AMn obtained from the design position information is Ptn, and its coordinates are represented by (xtn, ytn). In this case, the positional deviation amount Δxy of the alignment mark AMn and its X-direction component Δx and Y-direction component Δy are respectively expressed by the following equations:
Δx = xmn−xtn
Δy = ymn−ytn
Δxy = √ (Δx ² + Δy ² )
Respectively. In this way, the amount of misalignment of the alignment marks AM1 to AM4 on the substrate W can be calculated.

  In FIG. 5, in addition to the first to fourth alignment marks AM <b> 1 to AM <b> 4 described above, a fifth alignment mark AM <b> 5 is shown at the center of the substrate W. The meaning of the fifth alignment mark AM5 and how to use it will be described in detail later.

  FIG. 6 is a diagram showing an example of the positional deviation of the alignment mark. In the example shown in FIG. 6A, the shape of the rectangle Qm connecting the alignment marks AM1 to AM4 actually detected is the shape of the rectangle Qt connecting the normal positions of the alignment marks obtained from the design position information. The position is shifted in the X direction, Y direction, and θ direction while being maintained. Such a type of misalignment can be corrected by moving the stage 160. On the other hand, in the example shown in FIG. 6B, the positional deviations of the alignment marks AM1 to AM4 are not uniform, and as a result, the rectangle Qm formed by them is distorted with respect to the regular rectangle Qt. In such a case, it cannot be corrected only by moving the stage.

  These two modes can be distinguished by the relative positions between the alignment marks AM1 to AM4. That is, in the case of FIG. 6A in which each alignment mark is displaced while maintaining the regular rectangular shape, the relative position between the alignment marks AM1 to AM4 is not significantly different from the relative position at the regular position. It should be. On the other hand, in the case of FIG. 6B in which the rectangle is distorted, the relative positions between the alignment marks AM1 to AM4 are greatly changed.

  Accordingly, a relative positional relationship is obtained from the detected positions of the alignment marks AM1 to AM4, and compared with the positional relationship at the normal position, and the relative positional shift amount (first positional shift amount). Can be requested. By comparing the first positional deviation amount thus obtained with a predetermined first threshold value, it is possible to determine which case is shown in FIGS. 6A and 6B.

  Specifically, for example, an operation for obtaining a relative positional deviation amount between two alignment marks selected from four alignment marks is performed for each combination of alignment marks, and an average value, a minimum value, or a maximum value is a predetermined first value. The type of misalignment can be determined by whether or not the threshold value is exceeded. Which one of the average value, the minimum value, and the maximum value is used as the first positional deviation amount and what value the first threshold value is set can be set as appropriate according to the required alignment accuracy. is there.

  Hereinafter, as shown in FIG. 6A, a type of positional deviation in which the relative positional relationship between the alignment marks AM1 to AM4 is maintained is referred to as “linear deviation”, whereas FIG. As described above, the positional deviation of the type in which the rectangles are distorted by causing the alignment marks AM1 to AM4 to be different from each other is referred to as “non-linear deviation”. In addition, here, the concept is described using a rectangle formed by virtually connecting the alignment marks to be detected, but the number and arrangement of alignment marks to be detected are arbitrary, and the shape of the figure formed by connecting these The same idea can be applied even when is not rectangular.

  Returning to FIG. 4, the actual operation will be described. When the amount of displacement of the alignment marks AM1 to AM4 from the normal position is calculated in step S105, the amount of movement of the stage 160 necessary to correct the above-described linear displacement is calculated by the correction amount calculator 203. (Step S106). Subsequently, it is determined whether or not the positional deviation grasped from the position of the alignment mark is a linear deviation (step S107).

  The determination can be performed as follows, for example. First, an affine coefficient for mapping the point group composed of the position coordinates of the alignment marks AM1 to AM4 specified by the design position information to the point group composed of the position coordinates of the alignment marks AM1 to AM4 actually detected is calculated. . For example, a least square method can be used for the calculation. Using the affine coefficient thus obtained, a mapping of the point group specified by the design position information is obtained. The position difference between the mapped point group and the detected point group consisting of the position coordinates of the alignment marks AM1 to AM4 represents a non-linear positional shift component that cannot be corrected by moving the stage. In addition, the affine coefficient at this time represents a linear displacement component that can be corrected by moving the stage.

  Therefore, a non-linear component of misregistration can be represented by the distance between each point belonging to the point group mapped using the affine coefficient and the detected alignment marks AM1 to AM4. The maximum value of the positional deviation amounts of the alignment marks AM1 to AM4 from the mapped point or the average value of the positional deviation amounts can be set as the “first positional deviation amount”. Then, a tolerance (for example, 1 μm) allowed for the first positional deviation amount is set as a “first threshold value”, and the positional deviation is linear depending on whether or not the first positional deviation amount is within the first threshold value. Or non-linear.

  As described above, when the relative first misalignment amount between the alignment marks is equal to or smaller than the first threshold value, it is determined that the misalignment type is a linear misalignment that can be eliminated by moving the stage 160 (step S107). "YES"). In this case, the calculated stage movement amount is given to the exposure control unit 181, and the stage moving mechanism 161 is actuated in response thereto, whereby the stage 160 moves (step S 108). Of course, if the position is correct from the beginning, there is no need to move the stage. Thereby, the drawing position by the optical head 170 is adjusted to the optimum position on the substrate W. In this specification, this adjustment process is referred to as a “first adjustment process”. In this state, the substrate W is exposed by irradiating the substrate W with light from the optical head 170 (step S109).

  On the other hand, if the determination in step S107 is “NO”, that is, if it is determined that the first positional deviation amount exceeds the first threshold and is a nonlinear deviation, the position of the fifth alignment mark is estimated (step S111). Even if the type of deviation corresponds to linear deviation, the same treatment as nonlinear deviation is performed even when the deviation amount is large and the positional deviation is not eliminated by the movement of the stage 160 within the movable range. It is desirable. This requirement is satisfied by making a determination by comparing the first deviation amount with the first threshold value.

  As shown in FIG. 5A, the fifth alignment mark AM5 is an alignment mark formed in a chip region CR5 different from the chip regions CR1 to CR4 in which the first to fourth alignment marks AM1 to AM4 are formed. . In this example, the chip region CR5 located at the substantially central portion of the substrate W is selected, but the position is not limited to this and the position is arbitrary. However, it is preferable that the chip region is located as far as possible from the chip regions CR1 to CR4 where the first to fourth alignment marks AM1 to AM4 are formed, and the distances from these chip regions CR1 to CR4 are substantially the same. A certain chip area is more preferable.

FIG. 7 is a diagram showing the position of the fifth alignment mark. In the non-linear deviation, as shown in FIG. 7A, the rectangle Qm connecting the first to fourth alignment marks AM1 to AM4 has a distorted shape with respect to the regular rectangle Qt. A set Pt based on the points Ptn (n = 1, 2,...) Corresponding to the normal positions of these alignment marks and a point Pmn (n = 1, 2,...) Corresponding to the positions of the actually detected alignment marks. When considering a set Pm based on (...), the relationship between the two is expressed by the following equation using an appropriate geometric transformation f:
f: Pt → Pm
Can be represented by

  Since the point Pmn on the XY coordinate plane indicating the positions of the four alignment marks AM1 to AM4 is known, and the point Ptn corresponding to these normal positions can be calculated from the design position information, the geometry in the above equation It is possible to quantitatively determine what kind of conversion the dynamic conversion f is, at least approximately. That is, the distortion of the substrate W can be quantified. Typical geometric transformation algorithms include, for example, normalized projective transformation, affine transformation, TPS (Thin Plate Spline) interpolation, and any other transformation algorithm can be applied.

  In this way, the actual position of any point on the substrate W can be estimated from the design position information by quantitatively expressing the geometrical transformation (hereinafter simply referred to as “transformation”) f with an appropriate algorithm. Can do. Therefore, for the fifth alignment mark AM5, the position Pp5 on the substrate W is estimated using the design position information and the conversion f.

  Then, the position of the fifth alignment mark AM5 is detected on the actual substrate W (step S112). Specifically, the camera 150 captures a partial region on the substrate W centered on the estimated position Pp5, and the alignment mark detection unit 206 detects the fifth alignment mark AM5 included in the captured image. FIG. 5B shows an example of a captured image. The camera 150 captures an image at the estimated position Pp5, and the approximate center of the captured image Im corresponds to the estimated position Pp5. If the distortion of the substrate W is appropriately approximated by the conversion f, the actual fifth alignment mark AM5 should be detected in the vicinity of the estimated position Pp5.

  In other words, if the fifth alignment mark AM5 is detected in the vicinity of the estimated position Pp5, approximation by the conversion f is appropriate, and each chip on the substrate W is applied by applying the conversion f to the design position information. It is possible to estimate the position of the region CR. Therefore, in this case, for example, by applying the data correction method described in Patent Document 1 to correct the raster data, it is possible to perform drawing while canceling the distortion of the substrate W.

  Specifically, the distance from the estimated position Pp5 of the fifth alignment mark AM5 to the actually detected position is calculated as the positional deviation amount of the fifth alignment mark AM5 (step S113). As shown in FIG. 7B, if the distance from the estimated position Pp5 of the fifth alignment mark AM5 to the actually detected position Pm5 is equal to or smaller than a predetermined second threshold value V2 (“YES” in step S114). The positional deviation in this case is dealt with by correcting the raster data. That is, the correction amount calculation unit 203 calculates a necessary correction amount (step S115), and at the time of drawing (step S109), the data correction unit 204 and the strip data generation unit 205 sequentially correct the raster data in units of divided blocks B2. The strip data is created and sent to the exposure control unit 181 to cause the drawing engine 100 to perform drawing. As a specific mode of data correction, the technique described in Patent Document 1 can be used. The drawing position adjustment processing performed in this way is referred to as “second adjustment processing” in this specification.

  Even in this case, the position shift that can be eliminated by moving the stage is handled by moving the stage (step S108). Accordingly, when calculating the correction amount, the correction amount of the raster data is determined after subtracting the shift amount that is eliminated by moving the stage from the detected positional shift amount of the alignment mark. In the correction by the stage movement, all the chip regions CR can be corrected collectively, and the correction does not affect the data accuracy. Therefore, the deviation that can be eliminated by the stage movement can be dealt with by the stage movement. preferable. In particular, the shift in the θ direction of the chip region CR due to the inclination of the substrate W is difficult to cope with by data correction in the band B1 (FIG. 2) unit or the divided block B2 unit, and can be eliminated by moving the stage. desirable.

  On the other hand, the amount of displacement of the fifth alignment mark AM5 from the estimated position Pp5 may be larger than the second threshold value V2. In this case, the fifth alignment mark AM5 may be located at a position far from the estimated position Pp5 in the captured image Im, or the fifth alignment mark AM5 may not exist within the imaging range. In these cases, the position of each chip region CR on the substrate W cannot be estimated from known information. Therefore, correction based on known information cannot be performed.

  Originally, in a monolithic structure substrate in which a plurality of chip regions are formed on a single wafer, a relative positional shift occurs between each well region due to expansion / contraction or distortion of the substrate. It is relatively small and has a certain regularity in the direction and amount of displacement. Therefore, the fifth alignment mark AM5 is unlikely to greatly deviate from the estimated position Pp5 estimated from the positions of the other alignment marks AM1 to AM4. On the other hand, in the above-described pseudo wafer, that is, a substrate in which chips formed separately are collected and integrated, there is a case where the chips are displaced with no correlation to each other.

  FIG. 8 is a diagram for explaining the positional deviation in units of chips on the substrate. As illustrated in the figure, in the pseudo wafer, each chip on the wafer may be individually displaced within a predetermined range. The chip C1 in the figure is displaced in the X direction. Further, the sides of the chips C2 and C3 are inclined with respect to the XY coordinate axes, and a deviation in the θ direction occurs. Further, at the position indicated by the reference symbol P1, the interval between the chips in the X direction is larger than the other positions. The chip C4 is displaced in the X direction and the Y direction. The chips indicated by reference numerals C5 and C6 are displaced in the Y direction. As described above, in the pseudo wafer, there is a possibility that a different position shift occurs for each chip.

  In these cases, since the correlation in the tendency of deviation between the chips is low, the deviation amount obtained from the position of the alignment mark of a certain chip is not a material for estimating the positional deviation amount of another chip. Therefore, it is necessary to grasp the position for each chip and adjust the drawing position. Even in a monolithic substrate, if the amount of distortion increases, the correction of raster data may not be able to cope with it. That is, distortion can be corrected by displacing the allocation position of each divided block on the substrate W, but the amount of displacement is limited, and the direction and amount of displacement differ greatly between adjacent blocks. This is because a pattern discontinuity occurs. Even in such a case, the difference between the estimated position of the fifth alignment mark and the detected position becomes large.

  For these reasons, when the amount of misalignment of the fifth alignment mark AM5 exceeds the second threshold (“NO” in step S114), the position detection of the alignment marks in all the chip regions CR is performed again (step S121). ). At this time, as shown in FIG. 8, since two alignment marks are provided in each chip region CR, the positions of these two alignment marks are detected. Thereby, in addition to the position information of each chip region CR in the XY plane, information related to the inclination angle in the θ direction is also obtained. Of course, three or more alignment marks may be provided in each chip region.

  Then, based on the information regarding the position and inclination of each chip region CR thus obtained and the design data 211 stored in the storage unit 201, the raster data generation unit 202 executes the RIP process again (step S122). The RIP process is a process of rasterizing vector data and developing it into raster data (bitmap data). Rasterization is performed for each chip area CR in consideration of the positional deviation and inclination, so that the drawing position is assigned to each chip area. Raster data adapted to the CR position can be created. By performing the position correction using the design data in the vector data format, it is possible to avoid a reduction in drawing quality that can be caused by applying a large correction to the data after rasterization. The drawing position adjustment processing performed in this way is referred to as “third adjustment processing” in this specification.

  Then, by irradiating the substrate W with light modulated by the raster data created by the re-RIP process, appropriate drawing can be performed on each chip region CR (step S109). Note that in the re-RIP process after grasping the position of each chip, in principle, it is possible to correct the positional deviation of each chip due to the positional deviation of the substrate W. Therefore, the position deviation of the substrate W may be corrected at once by the re-RIP process, and the position deviation of the substrate W is corrected by moving the stage, and only the remaining position deviation is re-RIP processed. You may make it correspond. In this embodiment, the stage is moved prior to drawing (step S108), and the re-RIP process is performed by subtracting the misalignment that is resolved by the stage movement. When the positional deviation of the substrate W is corrected by the re-RIP process, step S108 is not necessary.

  When the drawing process for one substrate W is thus completed, the processed substrate W is returned to the substrate storage cassette 110. If an unprocessed substrate W remains in the cassette, drawing is performed by the same processing as described above, and the above processing is performed until all of the substrates W for one lot stored in the substrate storage cassette 110 have been processed. repeat.

  As described above, in this embodiment, the first to third adjustment processes are prepared in advance as the adjustment process of the drawing position on the substrate W by the optical head 170, and these are prepared according to the position detection result of the alignment mark. Selected and executed. The first adjustment process is a process of adjusting the relative position between each chip region CR on the substrate W and the optical head 170 by moving the stage 160 that holds the substrate W, and each chip region CR is uniform. This is effective when there is a positional deviation, and such a deviation can be corrected collectively. Further, since the drawing data is not processed for correction, the drawing quality is good.

  The second adjustment process is a process for correcting raster data, and is effective for misalignment caused by expansion and contraction or distortion of the substrate W. In this case, when the strip data is created from the created raster data, correction is performed as needed, so that the processing time does not increase for the correction. Although a reduction in drawing quality may be a problem by correcting the created raster data afterwards, the correction is performed in units of minute divided blocks B2 so that the device performance is affected, for example, a pattern disconnection. Large distortion can be prevented.

  On the other hand, the third adjustment process is a process for regenerating raster data after grasping the positional deviation amount of each chip region by detecting the position of the alignment mark. In this process, high-precision correction is possible regardless of the presence or absence of positional deviation regularity between chips and the magnitude of the positional deviation amount. Further, by reflecting the position of each chip area at the stage of the rasterizing process, the drawing quality does not deteriorate. Therefore, in principle, any type of displacement can be dealt with. However, since the position detection of the alignment mark and the rasterizing process after the detection are required and the processing time becomes long, the tact time in the drawing process for a plurality of substrates is increased. Therefore, it should be selected when there is a misalignment that cannot be handled by the first and second adjustment processes.

  The selection of the first to third adjustment processes is performed as follows. That is, some of the alignment marks formed in advance in each chip region CR are imaged and their positions are detected, and the relative displacement amount (first displacement amount) between them is a predetermined value (first threshold value). ) It is determined whether a small linear deviation or a non-linear deviation in which the first positional deviation amount is larger than the first threshold value occurs. If it is a linear shift, the first adjustment process is selected.

  On the other hand, in the case of non-linear deviation, the position of another alignment mark is further estimated from the position of a known alignment mark, and the actual position is detected by imaging the position. Then, if the positional deviation amount (second positional deviation amount) between the estimated position and the actual position is equal to or smaller than a predetermined value (second threshold value), the second adjustment process is performed, and the second positional deviation amount is the second threshold value. If larger, the third adjustment process is selected. That is, the second adjustment process is selected when the positional deviation amount of each other chip area can be estimated with a certain degree of accuracy from the position of the known chip area, and the third adjustment process is selected otherwise.

  By determining the adjustment process based on such a determination flow, it is possible to execute an appropriate adjustment process according to the mode of misalignment, and to prevent an increase in tact time due to the execution of unnecessary processes. . That is, the drawing position adjustment process is optimized according to the substrate W that is the drawing object. Then, drawing is performed with the drawing position adjusted in this way, and in this embodiment, good quality drawing can be performed at an appropriate drawing position on the substrate W.

  As described above, in this embodiment, the pattern drawing apparatus 1 functions as the “drawing apparatus” of the present invention, the substrate W is the “drawing object” of the present invention, and each chip region CR is the present invention. Each corresponds to a “drawing area”. Of the chip regions, the chip regions CR1 to CR4 correspond to the “detection target region” of the present invention, and the chip region CR5 corresponds to the “secondary detection target region”. The alignment marks AM1 to AM5 correspond to “identification marks” of the present invention.

  In the above embodiment, the stage 160 functions as the “holding unit” of the present invention, while the optical head 170 functions as the “drawing unit” of the present invention. Further, the camera 150 and the alignment mark detection unit 206 are integrated to function as the “position detection means” of the present invention. The raster data generation unit 202 functions as the “data generation unit” of the present invention, while the correction amount calculation unit 203, the data correction unit 204, and the strip data generation unit 205 are integrated as “drawing position adjustment unit” of the present invention. It is functioning.

  4 corresponds to the “first process” of the present invention, and step S103 corresponds to the “second process” of the present invention. Step S105 corresponds to the “third step” of the present invention, and steps S108, S111 to S115, and S121 to S122 correspond to the “fourth step” of the present invention. Further, step S109 corresponds to the “fifth step” of the present invention.

  The present invention is not limited to the above-described embodiment, and various modifications other than those described above can be made without departing from the spirit of the present invention. For example, in the above embodiment, the alignment mark AM is formed in advance in each chip region CR on the substrate W and the position is detected. However, it is sufficient to know the position of each chip region. There is no need to rely on it. That is, position detection can be performed by appropriately using a part capable of specifying the position of the chip region, such as a characteristic pattern formed in each chip region or an edge portion of the chip region. Further, the position detection method of the chip area may not depend on imaging by the camera.

  Further, for example, a part of the processing flow (FIG. 4) of the pattern drawing operation of the above embodiment may be modified and executed as follows. In FIG. 9 below, description of processing steps having the same processing content as the processing steps of FIG. 4 is omitted or the same step numbers are used and description thereof is omitted.

  FIG. 9 is a diagram showing a modification of the pattern drawing operation. When it is determined that the nonlinear deviation is based on the positional deviation amounts of the first to fourth alignment marks AM1 to AM4 ("NO" in step S107), the above embodiment uses a predetermined single geometric transformation algorithm. In this modification, the position of the fifth alignment mark AM5 is estimated using a plurality of conversion algorithms (step S201). When the position of the fifth alignment mark AM5 is detected in step S111, the amount of positional deviation from the estimated position calculated by each algorithm is obtained in step S112. Then, one of the conversion algorithms having the smallest obtained positional deviation amount is selected (step S202).

  In specifying the position of the unknown alignment mark from the position of the known alignment mark, a conversion algorithm that can appropriately approximate the distortion of the substrate W can provide high estimation accuracy, but naturally an approximation that does not match the distortion mode is estimated. Accuracy is lowered. By performing estimation using a plurality of conversion algorithms in advance and adopting a conversion algorithm that derives an estimated position closest to the actual position, correction processing according to the state of the substrate W can be performed. Specifically, in step S114, the amount of misalignment is evaluated between the actual position and the estimated position closest thereto, and when the raster data is corrected, it is selected (that is, the highest estimated accuracy is obtained). The conversion algorithm is also applied to the correction process (step S203). Thereby, the raster data is corrected according to the position of each chip region CR estimated with high accuracy, and the positional deviation between the chip region CR and the drawing pattern can be more effectively suppressed.

  Further, for example, in the pattern drawing operation of the above embodiment (FIG. 4), the stage movement amount is obtained in step S106, but the actual stage movement is executed in step S108 immediately before the drawing is performed after other processes are finished. . Instead of this, the stage movement may be performed prior to other processing as long as the calculation of the stage movement amount is completed.

  In the above embodiment, the pattern drawing apparatus integrally includes the drawing engine 100 and the computer 200. However, the existing pattern drawing apparatus having the same function as the drawing engine 100 has the same function as the computer 200. May be connected by wire or wirelessly, strip data may be generated by the data processing device, and output to an existing pattern drawing device for drawing.

  Furthermore, the application target of the present invention is not limited to an apparatus that draws a semiconductor substrate W such as a wafer by irradiating the substrate with light as a “drawing object” of the present invention. Various objects such as a glass substrate can be used as the drawing object.

  The present invention can be suitably applied to a technique for performing drawing by light irradiation on a drawing object provided with a plurality of drawing areas, and in particular, a technique for performing drawing while adjusting a drawing position on the drawing object. Suitable for the field.

1 Pattern drawing device 150 Camera (position detection means)
160 stage (holding means, stage)
170 Optical head (drawing means)
202 Raster data generation unit (data generation means)
203 Correction amount calculation unit (drawing position adjusting means)
204 Data correction unit (drawing position adjusting means)
205 Strip data generator (drawing position adjusting means)
206 Alignment mark detection unit (position detection means)
AM1-AM5 alignment mark (identification mark)
CR chip area (drawing area)
CR1 to CR4 chip area (detection target area)
CR5 chip area (secondary detection area)
W substrate (object to be drawn)

Claims (10)

In a drawing method of drawing by irradiating light from a drawing means to each of a plurality of drawing regions provided in a drawing object,
A first step of generating raster data corresponding to the content to be drawn;
A second step of positioning the drawing means with respect to the drawing object;
A third step of detecting positions of two or more of the plurality of drawing regions as detection target regions;
A fourth step of adjusting a drawing position on the drawing object by the drawing means based on a position detection result of the detection target region;
Based on the raster data, and a fifth step of drawing by irradiating the light to the drawing position of the drawing object from the drawing means,
In the fourth step, it is determined whether or not a relative first displacement amount between the detection target areas detected in the third step is within a first threshold, and the first displacement amount is When it is determined that it is within the first threshold, the drawing means and the drawing object are based on a positional deviation amount between the position detection result of the detection target area and a reference position set in advance corresponding to the detection target area. While performing the first adjustment process for adjusting the drawing position by adjusting the relative position to
When it is determined that the first positional deviation amount exceeds the first threshold, the position is detected by detecting at least one of the plurality of drawing regions as a secondary detection target region that is different from the detection target region. Determining whether a second positional deviation amount between the position and the position of the secondary detection target region estimated from the position detection result of the detection target region in the third step is within a second threshold;
When it is determined that the second positional deviation amount is within the second threshold value, correction according to the positional deviation amount between the position detection result of the detection target region and the reference position in the third step is performed on the raster data. To perform a second adjustment process for adjusting the drawing position,
If it is determined that the second positional deviation amount exceeds the second threshold value, the positions of all the drawing areas included in the drawing object are detected, and the raster data is regenerated based on the position detection result. And performing a third adjustment process for adjusting the drawing position.   The drawing method according to claim 1, wherein an identification mark for position detection is provided in advance in each of the plurality of drawing areas, and the position of the drawing area is detected by detecting the identification mark.   The drawing method according to claim 2, wherein a plurality of the identification marks are provided in each of the plurality of drawing areas.   When executing the second adjustment process in the fourth step, the second adjustment process is executed together with the first adjustment process, and between the position detection result of the detection target region in the third step and the reference position. 4. The drawing method according to claim 1, wherein the raster data is subjected to correction according to a positional deviation amount obtained by subtracting a positional deviation amount that can be corrected by performing the first adjustment process from the positional deviation amount.   When the third adjustment process is executed in the fourth step, the third adjustment process is executed together with the first adjustment process, and the position change amount by the first adjustment process is subtracted from the position detection result of each drawing area. 5. The drawing method according to claim 1, wherein the raster data is regenerated.   The drawing method according to claim 1, wherein in the first adjustment process, a relative rotation angle of the drawing object with respect to the drawing unit around an axis perpendicular to the surface of the drawing object is adjusted.   In the fourth step, when obtaining the second positional deviation amount, the position of the secondary detection target region is calculated by a plurality of different calculation methods from the position detection result of the detection target region, and the detection is performed among them. The deviation amount between the position closest to the position of the detected secondary detection target region and the detected position of the secondary detection target region is set as the second positional shift amount. The drawing method described in 1. Holding means for holding a drawing object provided with a plurality of drawing areas;
Position detection means for detecting the position of the detection target area using two or more of the plurality of drawing areas of the drawing target object held by the holding means as detection target areas;
Data generating means for generating raster data corresponding to the content to be drawn;
A drawing means for drawing the drawing object by irradiating light based on the raster data;
A drawing position adjusting means for adjusting a drawing position on the drawing object by the drawing means based on a detection result of the position detecting means;
The drawing position adjusting means includes
Determining whether a relative first positional deviation amount between the detection target areas detected by the position detection means is within a first threshold;
When determining that the first positional deviation amount is within the first threshold, based on the positional deviation amount between the position detection result of the detection target area and a reference position set in advance corresponding to the detection target area, While performing a first adjustment process for adjusting the drawing position by adjusting the relative position between the drawing means and the holding means,
When it is determined that the first positional deviation amount exceeds the first threshold, the second detection unit causes the position detection unit to detect at least one of the plurality of drawing regions that is different from the detection target region as a secondary detection target region. Determining whether a second positional deviation amount between the position of the next detection target area and the position of the secondary detection target area estimated from the position detection result of the detection target area is within a second threshold;
When it is determined that the second positional deviation amount is within the second threshold value, the rendering is performed by applying correction to the raster data according to the positional deviation amount between the position detection result of the detection target region and the reference position. Execute a second adjustment process for adjusting the position;
When it is determined that the second positional deviation amount exceeds the second threshold, the positions of all the drawing areas included in the drawing object are detected by the position detecting means, and the data generation is performed based on the position detection results. A drawing apparatus, wherein a third adjustment process for adjusting the drawing position by causing the means to regenerate the raster data is executed.   The drawing apparatus according to claim 8, wherein the drawing unit exposes the drawing object by scanning the drawing object with the light modulated based on the raster data.   The drawing apparatus according to claim 8, wherein the holding unit includes a stage that holds the drawing target, and is configured to be able to change a relative position of the stage with respect to the drawing unit.

Images