JP2000200746A - Mask pattern transfer method, mask pattern transfer apparatus, device manufacturing method, and transfer mask - Google Patents

Abstract

[PROBLEMS] To improve the joining accuracy of transfer patterns by correcting a relative position between a partial transfer pattern of a transfer mask and an object to be transferred. SOLUTION: An alignment mark of a partial transfer pattern on a transfer mask is measured (step 23), and a coordinate system of an actual alignment mark is calculated based on these (step 25). A parameter representing the relative relationship between the actual coordinate system obtained in step 25 and the designed coordinate system is calculated (step 26), and based on this, the transfer position of the partial transfer pattern to the transfer target is calculated (step 26). 29) Thereafter, based on the transfer position calculated in step 29, the mask stage and the wafer stage are driven to sequentially perform exposure (steps 34 to 40).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of transferring a mask pattern when manufacturing a device by transferring a fine pattern on a mask onto a substrate such as a wafer or glass.

[0002]

2. Description of the Related Art As the degree of integration and the performance and functions of semiconductor devices increase, the resolution and the angle of view of a lithographic apparatus increase, and the transfer method is a wafer batch method and a step-and-repeat method. Has been changed to a step-and-scan method. Under such circumstances, a scanning type reduction transfer system using a transfer mask has been proposed for an exposure apparatus using a charged beam such as an electron beam, which can be expected to have a higher resolution.

However, this type of apparatus uses a divided mask in which a transfer pattern is divided into a plurality of patterns due to a remarkable decrease in image performance when the angle of view is widened, and restrictions on the structure of a mask to be used. A desired pattern is obtained by sequentially connecting and transferring the divided patterns.

[0004] The division mask is one in which transfer patterns divided on one mask substrate are arranged at predetermined positions, and is formed on a silicon wafer by a method similar to a semiconductor process such as lithography or etching.

In the transfer method using the divided mask, since the joining accuracy between adjacent transfer patterns is important, the divided transfer patterns need to be arranged at a predetermined position on the mask with very high accuracy. is there. However, in the transfer pattern of these divided masks and the arrangement thereof, there are an arrangement error, a rotation error, a magnification error, a perpendicularity error, a distortion, and the like as geometric errors generated in the process of making the mask. Further, a distortion due to a temperature change of the divided mask, a distortion due to a mechanical stress, a distortion due to a change over time, and the like also cause errors.
These errors degrade the connection accuracy described above, resulting in a chip defect.

In order to solve such a problem, for example, as disclosed in Japanese Patent No. 2,647,835, the position of the entire mask is obtained using a special mark, and a temporary mark based on the data is obtained. A method is known in which a division pattern to be exposed next is driven to an exposure position using a coordinate system, and then alignment is sequentially performed for each division transfer pattern using an alignment mark dedicated to each division pattern. This method is called a die-by-die method, and alignment is performed for each divided pattern, so that accurate positioning can be performed.

[0007]

However, in the above-described method, each divided pattern is driven once using a temporary coordinate system, and then sequentially divided and transferred using an alignment mark dedicated to the divided pattern. Since alignment is performed for each pattern, measurement and driving for alignment are required each time a divided pattern is transferred, and there is a problem that the measurement time and drive time decrease the throughput of the entire apparatus. This problem has a greater effect on throughput as the number of mask divisions increases.

Further, in the above method, the so-called 2nd
In the layer, an alignment mark is required on the object to be transferred, that is, on the wafer. However, depending on how the transfer mask is divided, a mark on the wafer side corresponding to a partial transfer pattern on the transfer mask may not exist. For example, when the array is divided into a 5 × 5 matrix array, the 3 × 3 matrix array near the center of the array is inside the actual element pattern area, and thus the alignment marks cannot be transferred in advance. Therefore, the above-mentioned method cannot practically perform alignment.

Accordingly, the present invention provides a mask pattern transfer method and apparatus for improving the joining accuracy of each partial transfer pattern when a plurality of divided partial transfer patterns are sequentially transferred to an object to obtain a desired pattern. , Device manufacturing method,
Another object is to provide a transfer mask.

[0010]

In order to solve the above-mentioned problems, a mask pattern transfer method according to the present invention uses a transfer mask having a plurality of divided and distinguishable partial transfer patterns. A mask pattern that irradiates a transfer beam onto a transfer object through the transfer unit and sequentially transfers the partial transfer patterns onto the transfer object, thereby obtaining a desired pattern in which the partial transfer patterns are connected to the transfer object In the transfer method, the transfer mask is provided with a plurality of alignment marks associated with each of the partial transfer patterns, and a relative position between the transferred object and each of the alignment marks is corrected, so that each of the partial transfer patterns is corrected. The pattern is sequentially transferred to the transfer object.

As described above, according to the present invention, even if an error due to a geometric error, a temperature change, a mechanical stress, a change over time, or the like occurring in the mask making process occurs in the transfer pattern and the arrangement of the divided masks, each partial transfer is performed. By providing a plurality of alignment marks respectively corresponding to the patterns and correcting the relative position between the transferred object and each of the alignment marks, the joining accuracy of each partial transfer pattern can be improved.

In the method of transferring a mask pattern according to the present invention, at the same time that the partial transfer pattern is transferred to the object to be transferred, each alignment mark is also transferred to the object to be transferred. After the transfer of the mark, using a transfer mask for overlay having a plurality of partial transfer patterns different from the transfer mask, which are divided and distinguishable, and a plurality of alignment marks associated with each of the partial transfer patterns. It is also possible to correct the relative position between each alignment mark transferred to the transfer object and each alignment mark of the overlay transfer mask, and sequentially transfer each partial transfer pattern of the overlay transfer mask to the transfer object. Good.

Further, the partial transfer patterns may be grouped and then transferred to an object to be transferred to improve the throughput.

Further, the mask pattern transfer apparatus of the present invention is characterized in that a beam radiating means for radiating a charged beam, a shaping means for shaping the charged beam into an arc shape, and a dividing and transferring means for transferring the charged beam to an object to be transferred. A movable first placement unit on which a transfer mask on which a plurality of distinguishable partial transfer patterns and a plurality of alignment marks associated with each of the partial transfer patterns are formed is placed; A movable second mounting unit on which an object to be transferred on which a transfer pattern including the partial transfer patterns of the mask is to be mounted, and a projection magnification and an image of the charged beam transmitted through the transfer mask A magnification rotation amount correcting means for correcting the amount of rotation to irradiate the transfer object, and measuring the positions of the respective partial transfer patterns and the respective alignment marks of the transfer mask. First measuring means, and second measuring means for measuring the positions of the transferred patterns formed by joining the partial transfer patterns transferred to the transferred object and the positions of the alignment marks, Based on output signals from the first measuring means and the second measuring means, the moving amounts of the first mounting means and the second mounting means are controlled, and the magnification rotation amount correction is performed. Control means for controlling the projection magnification and the image rotation amount of the means.

With this apparatus, the mask pattern transfer method of the present invention can be easily carried out.

Further, the device manufacturing method of the present invention may be a method of manufacturing a device by a process including the mask pattern transferring method of the present invention, or may be a method of manufacturing a device by using the mask pattern transferring apparatus of the present invention. Good.

The transfer mask of the present invention is divided and
A mask having a plurality of distinguishable partial transfer patterns, and sequentially transferring each partial transfer pattern to an object to be transferred, thereby obtaining a desired pattern in which the partial transfer patterns are joined to the object to be transferred. , A plurality of alignment marks associated with each of the partial transfer patterns are provided.

Note that "design coordinates" used in this specification.
Means coordinates formed by a grid that does not include an error. The “coordinates in production” are “coordinates in design”.
And the coordinates on the transfer mask and the transfer mask for superimposition actually mounted on the transfer mask before being placed on the mask pattern transfer device. The “actual coordinates” mean coordinates in a state where the transfer mask having the “manufactured coordinates” and the transfer mask for superimposition are placed on the mask pattern transfer device.

[0019]

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

(First Embodiment) FIG. 26 is a schematic diagram showing an example of an exposure apparatus for manufacturing a semiconductor device and a control system of the exposure apparatus for manufacturing a semiconductor device to which the transfer method of the present invention can be applied. The same device is used in the second to fifth embodiments to be described later.

The exposure apparatus for manufacturing a semiconductor device is controlled by a beam controller 2a for controlling the dose of the beam 12 emitted from the beam source 1, a mask stage controller 6a for moving the mask stage 6, a deflector. A deflector control unit 15a for controlling the magnification and a magnification correction control unit 8 for controlling a magnification correction system 8 and a rotation correction system 9 for correcting the magnification and rotation of the pattern of the transfer mask 3 irradiated onto the wafer 10;
a and a rotation correction controller 9a, and a device controller 16 having a wafer stage controller 11a for moving the wafer stage 11.

First, a charged beam 12, which is a charged particle beam, such as an electron beam or an ion beam emitted from the beam source 1, is turned on by a beam control system 2 by a beam control unit 2a.
By performing / off, the dose amount is controlled to be incident on the illumination system 4, whereby the transfer mask 3 on the mask stage 6 is
Is irradiated.

Here, a slit plate 13 in which an arc-shaped slit is formed is provided in the illumination system 4, and the charged beam 12 passing through the illumination system 4 is formed into an arc shape in a plane perpendicular to the optical axis. Molded. By forming the beam into an arc shape in this way, the imaging position can be changed to the charged beam 1 away from the optical axis.
2 is near the outer circumference. This makes it possible to enlarge the exposure area while suppressing the curvature of field to a negligible level. In the present embodiment, the charged beam 12 has an arc-shaped beam shape, but the beam shape may be any other shape.

The mask stage 6 is driven by a mask stage control section 6a based on the correction values measured and processed by the mask alignment system 5, thereby aligning the transfer mask 3 and the transfer beam 12.

The charged beam 12 applied to the transfer mask 3
After the pattern on the transfer mask 3 is deflected by the deflector 15 and projected onto the projection system 7, the projection magnification is further corrected by the magnification correction system 8, and the image rotation is corrected by the rotation correction system 9,
Irradiation is performed on the wafer 10 on the wafer stage 11, whereby the pattern of the transfer mask 3 is transferred.

The wafer stage 11 controlled by the wafer stage control unit 11a controls the wafer 1 based on the values measured and calculated by the wafer alignment system 14.
Zero alignment is performed. At the time of transfer, the mask stage 6 and the wafer stage 11 are adjusted in advance in the scanning movement direction, orthogonality, scale magnification, and the like of each other, and the coordinate system is used as a reference coordinate system and used as a reference for all subsequent processes.

Next, the transfer mask 3 shown in FIG.
The transfer method of the present invention will be described in the step of baking the wafer 10 as a t mask.

FIG. 2 shows a transfer mask 3 having partial transfer patterns divided into a 5 × 5 array, and mask shots 3 a to 3 y which are partial transfer patterns on the transfer mask 3, which are printed on a wafer 10. Each image 101 is shown.

Each of the mask shots 3a to 3y is a transfer image 10
1 is divided into a plurality. Around the mask shots 3a to 3y on the transfer mask 3, alignment marks 3a1 to 3y6, which are marks for defining the positions of the mask shots 3a to 3y on the transfer mask 3, are arranged.

For example, the position of the mask shot 3a is defined by the alignment marks 3a1-3a6. Each of the mask shots 3a to 3 on these transfer masks 3
As for y, the mask shot 3a is sequentially transferred onto the wafer 10 with no gap in such a manner that the mask shot 3b is a shot on the wafer 10 and the mask shot 3b is a wafer shot 10b, and a desired transfer image 101 is obtained. At this time, the mask shots 3a to 3e, 3f, 3j,
Of the alignment marks 3k, 3o, 3p, 3t, 3u to 3y, at the outer peripheral portion of the transferred image, for example, 3a1, 3
a2, 3a5, 3a6, etc. are also transferred onto the wafer 10 and used for alignment of the second and subsequent layers as in the embodiment described later.

FIG. 1 is a flowchart showing the details of the above steps, and is a basic flow of the present invention.

Hereinafter, description will be made in order with reference to the flow charts of FIGS. Here, it is assumed that the wafer 10 has already been loaded into the apparatus.

First, the transfer mask 3 is loaded on the mask stage 6 (step 20). Next, the position of the transfer mask 3 is measured by the mask alignment system 5, and
Is driven to perform reference alignment (step 21). The reference at this time is the above-described reference coordinate system, and the reference axis of the charged beam 12 and a separately provided device reference match the reference coordinate system. Therefore, this reference positioning is nothing but positioning of the transfer mask 3 with respect to the reference axis of the charged beam 12 or a separately provided device reference.

When the transfer mask 3 is located at the reference position,
An alignment mark used for measurement in alignment is specified (step 22). Here, mask shot 3
a, 3c, 3e, 3k, 3m, 30, 3u, 3w, 3y
Alignment marks 3a1-3a4, 3c corresponding to
1-3c4, 3e1-3e4, 3k1-3k4, 3m1
~ 3m4, 3o1 ~ 3o4, 3u1 ~ 3u4, 3w1 ~
3w4, 3y1-3y4 are designated. In this step, the alignment marks of all the mask shots may be specified, but it is necessary to keep in mind that the larger the number of marks to be measured, the longer the measurement time, and the lower the throughput of the apparatus.

Next, each of the mask shots 3a, 3c, 3e, 3k, 3
m, 3o, 3u, 3w, 3y, respectively, alignment marks 3a1-3a4, 3c1-3c4, 3e1-3
e4, 3k1-3k4, 3m1-3m4, 3o1-3o
4, 3u1-3u4, 3w1-3w4, 3y1-3y
4 are measured by the mask alignment system 5 (step 23). That is, in the mask shot 3a, the positions of the alignment marks 3a1, 3a2, 3a3, 3a4 are measured.

Here, the alignment marks 3a5, 3a6
May be specified, but the alignment mark 3a1
If the positions of .about.3a4 are known, the position of the mask shot 3a can be specified.

The mask shots 3a, 3c, 3e, 3k,
3m, 30, 3u, 3w, 3y alignment marks 3a1 to 3a4, 3c1 to 3c4, 3e1 to 3e4, 3
k1 to 3k4, 3m1 to 3m4, 3o1 to 3o4, 3u
After measuring the positions of 1 to 3u4, 3w1 to 3w4, and 3y1 to 3y4, the position of each alignment mark is stored for each mark (step 24).

Next, the alignment marks 3a1-3a4, 3c1-3c4, 3e1
3e4, 3k1-3k4, 3m1-3m4, 3o1-3
o4, 3u1-3u4, 3w1-3w4, 3y1-3y
The position of 4 is read to calculate the actual coordinate system of the mask shots 3a to 3y (step 25).

FIG. 3 shows the difference between the actual coordinate system obtained from the position data of the alignment mark and the designed coordinate system.

The actual coordinate system is derived by assuming a virtual grid and using the mask shots 3a, 3c, 3e, 3
The difference between the actual position of k, 3m, 3o, 3u, 3w, 3y and the position of each lattice point corresponding to each mask shot of the virtual lattice is evaluated by, for example, the least-squares method to determine the best fit lattice. Is adopted as the coordinate system.

Hereinafter, the derivation of the actual coordinate system will be described in detail.

The coordinate system in design is an orthogonal coordinate system at equal intervals without errors, but the coordinate system actually manufactured based on the coordinate system indicating the position of each partial transfer pattern of the transfer mask 3 is As shown in FIG. 4, an error occurs at the time of manufacture with respect to the design coordinate system. The error of the coordinate system in manufacturing may include a random component for some reason at the time of manufacturing. The manufacturing coordinate system used herein means the coordinate system of the transfer mask 3 not set in the transfer device. For this reason, an error due to an increase in temperature in the transfer device, an influence of the weight deformation due to the mask support, and the like. Refers to a coordinate system in which is not generated.

Further, when the transfer mask 3 having the above-described coordinate system in manufacturing is set in a transfer device, the transfer mask 3 is affected by the environment in which the device is placed, such as a rise in temperature inside the device and deformation of its own weight due to the support of the mask. As a result, a variation is generated in the coordinate system due to thermal expansion or stress deformation, which is a so-called linear component error. The coordinate system including this further error is the actual coordinate system.

The derivation of such an actual coordinate system is performed by fitting a lattice representing an actual coordinate system to a lattice representing a design coordinate system by using a statistical method such as an error least square method.

However, the above-described statistical method is effective for correcting fluctuations due to thermal expansion and stress deformation. However, when the above-described statistical method is executed while including a random component, if the random component is large, There may be a problem that excessive correction is performed on the random component.

In order to avoid such a problem, here, as an example of a method for deriving an actual coordinate system, which is a coordinate system including an error composed of a linear component and a random component, a two-stage coordinate system is used. It is desirable to use a method of performing grid fitting.

FIGS. 5 (a) to 5 (a) to 5 (c) are schematic diagrams relating to the fitting of the lattice of the two-stage coordinate system described above.
This will be described with reference to FIG. 6C and FIG.

FIGS. 5A to 5C are schematic diagrams showing grids in each coordinate system.

FIG. 5A is a schematic diagram showing a coordinate system in design.

FIG. 5 (b) is a schematic diagram showing a coordinate system in production manufactured based on the design coordinate system shown in FIG. 5 (a). At this stage, the error has occurred in the coordinate system because it has not been set in the transfer device, and is only a random component which is an error that occurred during manufacturing.

FIG. 5 (c) includes, in addition to the random component which is an error of the coordinate system in the production shown in FIG. 5 (b), a linear component which is an error caused by setting in the transfer device. It is a schematic diagram which shows an actual coordinate system.

First, as shown in FIG. 6, grid fitting is performed by paying attention only to the linear component which is a variation.
That is, this fitting subtracts the grid (error: random component) of the manufactured coordinate system from the grid (error: linear component and random component) of the actual coordinate system. A virtual grid, which is a grid to which only the variation is fitted, is obtained. In order to perform the fitting, alignment marks associated with each of the partial transfer patterns in production are measured in advance, and the position of each of the partial transfer patterns in production and the coordinate system in production are obtained in advance.

Next, as shown in FIG. 7, re-fitting, which is the fitting of the lattice focusing on the random component, is performed. That is, by adding the coordinate system in production to the virtual grid to which only the variation has been fitted, a final correction grid in which the variation and the random component have been fitted is obtained. In this way, the final correction grid in which the linear component and the random component have been corrected through the two-stage fitting is used as the grid of the actual coordinate system.

According to the above-described method, the position variation of the partial transfer pattern can be evaluated while eliminating the influence of the random component due to an error in the production of the partial transfer pattern at the time of the lattice fitting. Transfer becomes possible.

The derivation of the actual coordinate system is not limited to the above-described method, but may be derived by another method.

Sx and Sy are translation (shift) components in the x-axis direction and the y-axis direction, respectively. θx and θy are rotation components with respect to the x-axis and y-axis, respectively, and θx−θy may be interpreted as the orthogonality component, where θx is the rotation component of the entire coordinate system. Also, a value obtained by normalizing the length Dxr of the actual coordinate system viewed from the designed coordinate system with the designed length Dxf, in this case, Dxr / D
xf is the magnification in the x-axis direction (scale magnification), and Dyr / Dyf is the magnification in the y-axis direction for the y-axis.

These differences, that is, parameters are calculated (step 26), and these parameters are stored.
(Step 27).

In the steps up to this point, all the parameters relating to the transfer position of each mask shot have been determined and stored. In the subsequent steps, the actual exposure position of each mask shot is determined, and the mask stage 6 and the wafer stage 11 are driven based on each parameter to proceed to the actual exposure processing.

First, the stored parameters are read out (step 28), and the transfer position of each mask shot is calculated from the shift components Sx and Sy and the orthogonality components θx−θy (step 29). Is stored (step 30).

Here, the orthogonality component θx−θy is applied as a so-called staircase correction, which is treated as a stepwise shift of the x direction component with respect to the y axis. Furthermore, each scale magnification Dxr / D
xf and Dyr / Dyf are given as drive parameters from the mask stage controller 6a to the mask stage 6, and the amount of rotation of the mask stage 6 is determined based on the rotational component θx, and the mask stage on which the transfer mask 3 is placed 6 is rotated to correct the rotational component of the transfer mask 3 (step 31).

When the exposure processing is started (step 3
2) The mask shot number N is initialized to 1 (step 33), and the flow advances to the subsequent shot exposure loop. When the mask shot number N = 1, the mask shot 3a is obtained.

At the beginning of this loop, first, in step 30
The transfer position of the first mask shot 3a stored in (1) is read (step 34). Then, the mask stage 6 having the x, y, and θ axes is driven to move the mask shot 3a to the read transfer position (step 35). Simultaneously with this operation, the wafer stage 11 is driven (step 36), and the wafer 10 is moved to a position where the first mask shot 3a is to be transferred. The position of each stage driven in these steps 35 and 36 is the start position of the relative scanning. At this stage, for this shot,
The exposure is ready.

Next, the beam control system 2 is driven to irradiate the mask shot 3a with the charged beam 12 to start the transfer as the wafer shot 10a on the wafer 10.
At this time, the entire area of the mask shot 3a is transferred onto the wafer 10 while the mask stage 6 and the wafer stage 11 are relatively scanned with respect to the charged beam 12 (Step 3).
8).

When the exposure for this shot is completed,
It is determined whether or not exposure of all shots has been completed (step 39). If the exposure of all the shots has not been completed yet and the exposure of the next shot is to be performed, 1 is added to the mask shot number N (step 40), and the process returns to step.

By repeating the loop of steps 34 to 40, the same processing is performed for mask shots 3b to 3y.

When the processing for all the mask shots 3a to 3y is completed, the process exits from this loop and proceeds to the next chip processing step (step 41).

Thus, one desired chip or one pattern is transferred onto the wafer 10. Note that the wafer 1
In order to transfer a plurality of chips onto zero, steps 33 to 41 of the above flow may be repeated by the number of chips.

The difference between the transfer position of each mask shot read in step 34 and the design transfer position is obtained, and in step 35 the mask stage 6 is driven to the design transfer position and the deflector 15 is driven. Thus, the difference may be corrected.

As described above, some of the alignment marks 3a1 to 3y6 corresponding to the mask shots 3a to 3y, which are partial transfer patterns on the transfer mask 3, are designated, and the actual coordinates of each partial transfer pattern are designated from them. The transfer position of each partial transfer pattern can be corrected based on the calculated system, calculating parameters indicating the relative relationship with the coordinate system in design. As a result, the transfer image 101 with high connection accuracy can be reduced while shortening the time required for alignment.
Can be obtained.

(Second Embodiment) In the present embodiment, by focusing on shot profiles such as a magnification error and a rotation error of a mask shot itself, each item of the shot profile can be corrected. Connection accuracy has been further improved.

For example, this is effective when each mask shot itself is rotating or expanding / contracting with respect to the coordinate system of the mask shot. FIG. 11 shows a situation where each of the mask shots 3a to 3y is rotating and a situation where the mask shot 3m is rotating by θsm.

FIG. 12 is a view showing a coordinate system for manufacturing the partial transfer pattern of the transfer mask 3 '. As in the transfer mask 3 of the first embodiment, the position of each partial transfer pattern is determined at the time of manufacture. Contains an error with respect to the design coordinate system.
Also in the transfer mask 3 'used in the processes after the second layer, similarly to the first mask process, the alignment mark associated with each partial transfer pattern is measured in advance, and the position in the production of each partial transfer pattern is obtained. be able to.

Hereinafter, FIG. 2 used in the description of the first embodiment will be described.
The transfer method according to the present embodiment will be described with reference to FIG.

FIG. 9 is a flowchart illustrating the transfer method according to the present embodiment.

First, Steps 20 to 27 are the first steps.
In the same manner as in steps 20 to 27 of the embodiment, after the transfer mask 3 shown in FIG. 2 is loaded into the apparatus to perform reference alignment, measurement of the specified alignment mark is performed. The specified alignment marks are the same as in the first embodiment, but all the marks may be specified. In the present embodiment, as will be described later, another effect occurs when all the marks are specified.

The position of each mark is obtained from the measurement of the alignment mark, and the actual coordinate system of the mask shot is calculated. A parameter representing the relative relationship between this coordinate system and the designed coordinate system is obtained and stored.

Next, the position of each alignment mark stored in step 24 is read, and the shot profile of each mask shot is calculated (step 271). The profiles calculated here are the shot magnification and the shot rotation amount.

For example, let us consider a state in which the actual mask shot 3a is inclined with respect to the designed mask shot 3a ″ and is enlarged or reduced. The shot magnification is, for example, as shown in FIG. If the distance between the actual mark positions of the alignment marks 3a1 and 3a4 is Da14 and the designed distance is Daf,
Da14 / Daf is calculated as the shot magnification. Similarly, taking the mask shot 3a as an example, the amount obtained by subtracting the rotation amount θx of the coordinate system from the angle θa14 formed by the line segment passing through the two actual positions of the alignment marks 3a1 and 3a4 and the corresponding design line segment is obtained. It is calculated as the shot rotation amount θsa.

If all the mask shots have not been specified in step 22, the average values of the shot magnification and the shot rotation amount of each of the specified mask shots are obtained.
Assuming that the average shot magnification is Mm and the average shot rotation amount is θm, these Mm and θm are stored (step 2).
72). If all mask shots have been designated in step 22, all shot magnifications and shot rotation amounts of each mask shot are stored (step 27).
2).

Next, it is determined whether or not the measurement mark designation in step 22 has been made for all mask shots (step 273). Step 274 is skipped, otherwise step 274 is executed (step 27).
It branches as shown in 3). The reason is that when using the average shot magnification Mm and the average shot rotation amount θm,
This is because the same amount of correction can be performed for all mask shots by driving the magnification correction system 8 and the rotation correction system 9 in advance (step 274). On the other hand, if all the mask shots are specified in step 22, it is possible to correct the individual shot magnification and shot rotation amount that are optimal for each mask shot.

In this case, it is necessary to drive the magnification correction system 8 and the rotation correction system 9 for each shot in a shot exposure loop described later to perform correction. Therefore, if all mask shots have been designated, step 274 is skipped.

Thereafter, steps 28 to 31 are the same as steps 28 to 31 of the first embodiment, calculating and storing the transfer position of each mask shot, and further driving the θ stage of the mask stage 6 to transfer the transfer mask. After correcting the third rotation component, a series of exposure processing is started in step 32.

Further, similarly to the first embodiment, the mask shot number N is initialized to 1 (step 33), and steps 34 to 40, which are a loop of shot exposure, are performed.
Proceed to. Steps 34 to 36 are the same as steps 34 to 36 of the first actual flag form, read the transfer position of the mask shot, drive the mask stage 6, and transfer the transfer image of the mask shot onto the wafer 10. The wafer stage 11 is driven so as to be at a position to be performed, in preparation for the start of relative scanning exposure.

Next, it is determined in step 22 whether or not all the mask shots have been designated (step 371).
If the correction has been performed for all the mask shots, that is, if the correction for the shot profile is to be performed for each mask shot, the shot profile of the mask shot stored in step 272 is read out (step 372), The magnification correction system 8 is driven to correct the shot magnification, and the rotation correction system 9 is driven to correct the shot rotation amount (step 373).

If all the mask shots have not been specified in step 22, that is, the same amount of correction is given to all the mask shots using the average shot magnification Mm and the average shot rotation amount θm for the correction of the shot profile. Then, since the magnification correction system 8 and the rotation correction system 9 have already been driven in step 274 based on the average shot magnification Mm and the average shot rotation amount θm, steps 372 and 373 are skipped, and the process proceeds to the next step. . Thereafter, the relative scanning exposure is performed on the shot (step 38), and the processing for one shot is completed.

The subsequent steps are the same as those in the first embodiment.
When the exposure of all shots is completed, the process branches at step 39 to proceed to the next chip processing step.

As described above, in the present embodiment, by paying attention to the shot profile of each shot and correcting the shot magnification and the shot rotation amount of the shot itself, it is possible to obtain the transfer image 101 with higher connection accuracy.

(Third Embodiment) Next, FIG. 10 shows a shot layout, FIG. 13 and FIG. 14 show flowcharts, FIG. 15 shows a wafer shot coordinate system, and FIG. The transfer method for the second and subsequent layers will be described with reference to the drawings to be described.

The 2nd layer refers to a transfer image (not shown) transferred further superimposed on the transfer image 101 transferred onto the wafer 10 as the first layer described in the first embodiment. Point.

FIG. 10 shows the mask shots 3a to 3a of FIG.
The transfer mask 3 'having mask shots 3'a to 3'y to be superimposed on the transfer mask 3' respectively corresponding to the mask shots 3a to 3'y on the transfer mask 3 described in the first embodiment.
3y shows the transferred image 101 exposed on the wafer 10 together with the alignment marks around the pattern.

Each of the mask shots 3'a to 3'y is obtained by dividing the transferred image further superimposed on the transferred image 101 as a first layer into a plurality of layers. The transfer mask 3 'has alignment marks 3'a1-3'y6 on both sides of each of the mask shots 3'a-3'y, and further below 3'a-3'e and above 3'u-3'y. Is provided. On wafer 10, alignment marks 10-1 to 10-30 already transferred in the previous process exist.

The transfer method according to the present embodiment will be described below with reference to the flowcharts shown in FIGS.

In FIG. 13, steps 20 to 27,
Steps 271 to 274 are the same as in the second embodiment. That is, at this stage, the measurement and processing for each mask shot, the calculation of the parameters for the coordinate system of the mask shot, and the processing for the mask shot profile are completed.

In steps 122 to 127, step 2
Processing equivalent to the processing performed on each mask shot and the coordinate system of the mask shots in steps 2 to 27 is performed on each wafer shot and the coordinate system of the wafer shots.

First, the wafer shots 10a, 10c, 1
0e, 10k, 10o, 10u, 10w, and 10y are designated (step 122). Here, it should be noted that the alignment marks already transferred on the wafer are only those corresponding to the wafer shots in the peripheral portion.

Next, the wafer shots 10a, 10c, 10e, 10k, 10o, 1
The positions of the alignment marks 10-1 to 10-40 corresponding to 0u, 10w, and 10y are measured (step 12).
3) Then, it is stored (step 124).

Further, the alignment marks 10-
The actual coordinate system of the wafer shot is calculated from the positions 1 to 10-40 (step 125), and a parameter representing the relative relationship with the designed coordinate system is determined (step 126) and stored (step 127). .

At this stage, as shown in FIG. 15, a parameter equivalent to the parameter obtained in the processing for the mask shot, that is, x relating to the coordinate system of the wafer shot
Translational components Sxw, Syw, and x in the axial and y-axis directions
Rotation components θxw and θyw with respect to the x-axis and y-axis, and magnifications (scale magnifications) Dxrw / Dx in the x-axis direction and the y-axis direction
fw, Dyrw / Dyfw are obtained. Here, the rotation component of the entire coordinate system is θxw, and θxw−θy
Interpret w as an orthogonality component.

Here, the parameters representing the relative relationship were obtained by comparing the coordinate system for designing the wafer shot, which is the reference coordinate system of the apparatus, with the actual coordinate system of the wafer shot. The parameter representing the relative relationship obtained by comparing the designed coordinate system of the mask shot with the actual coordinate system of the mask shot is also a comparison with the designed coordinate system that is the reference coordinate system of the apparatus. Therefore, by comparing the actual coordinate system of the mask shot with the actual coordinate system of the wafer shot,
A parameter indicating the relative relationship may be obtained.

Steps 1271 to 1274 are processing relating to a shot profile of a wafer shot. In these processes, processes equivalent to the processes related to the profile of the mask shot (steps 271 to 274) are performed on the wafer shot.

The position of each alignment mark stored in step 124 is read out to calculate a shot profile of each wafer shot (step 1271). The profiles calculated here are the shot magnification and the shot rotation amount. The shot magnification is, for example, as shown in FIG.
Regarding wafer shot 10a, alignment mark 1
The interval between the actual mark positions of 0-1 and 10-2 is D12w
If the design interval is D12wf, then D12w /
D12wf is calculated as the shot magnification. Similarly, taking the actual wafer shot 10a as an example, a line segment passing through two points of the alignment marks 10-1 and 10-2 of the actual wafer shot 10a and the alignment mark 10-1 "of the designed wafer shot 10a" From the angle θw12 formed by the line segment passing through the two points of 10-2 ″ and the rotation amount θxw of the coordinate system
Is calculated as the shot rotation amount θsaw. If all the wafer shots are not specified in step 122, the average value of the shot magnification and the shot rotation amount of each specified wafer shot is obtained, and the average shot magnification is Mmw, and the average shot rotation amount is θmw. These Mmw and θmw are stored (step 1272). If all the wafer shots 10a to 10y are designated in step 122, the shot magnification and the shot rotation amount of each of the wafer shots 10a to 10y are all stored in step 1272.

It is determined whether or not the measurement mark designation in step 122 has been made for all the wafer shots 10a to 10y (step 1273). Step 1274 is skipped, otherwise step 1274 is executed.

In the present embodiment, since the partial transfer patterns of the transfer mask 3 of the first layer are in a 5 × 5 array, the wafer shots 10a to 10e, 10f, 10j, 10k, and 10o on the outer peripheral portion among the wafer shots 10a to 10y. ,
Only the alignment marks corresponding to 10p, 10t, 10u to 10y are transferred onto the wafer 10. Therefore, in step 122, these wafer shots 1
0a to 10e, 10f, 10j, 10k, 10o, 10
Only p, 10t, 10u to 10y can be specified, and the shot magnification and the shot rotation amount are corrected using the average shot magnification Mmw and the average shot rotation amount θmw (step 1274).

When step 1274 is skipped, that is, in step 122, all wafer shots 10a
The case where it is possible to specify ｙ10y is 1s
In a shot layout in which alignment marks corresponding to all mask shots can be transferred on the wafer 10 in the t layer, for example, a transfer mask pattern is 2 × 2
Or a case where the image is divided so as to be arranged in a line in the x direction or the y direction.

The magnification correction system 8 in step 1274
If the magnification correction system 8 and the rotation correction system 9 have already been driven in step 274, the drive of the rotation correction system 9 is additionally driven.

The subsequent steps will be described in detail with reference to the flowchart shown in FIG.

Steps 28 to 31 are the same as steps 28 to 31 of the first embodiment, and read the mask shot parameters (step 28).
The transfer position of each mask shot is calculated (step 2
9), store (step 30), give the scale magnification to the mask stage 6, and
The upper θ stage is driven to correct the rotational component of the mask (step 31).

Steps 128 to 131 correspond to steps 28 to 31 relating to the above-described mask shot.
Is performed on the wafer shot.

First, the stored parameters are read out (step 128).

Next, the transfer position of each wafer shot is calculated from the shift components Sxw and Syw and the orthogonality component θxw-θyw among those parameters (step 129).
It is stored (step 130). Where the orthogonality component θ
xw-θyw applies so-called staircase correction, which is treated as a stepwise shift of the x-direction component with respect to the y-axis. Further, each scale magnification Dxrw / Dxfw, Dyrw / Dyfw is given to the wafer stage 11 as a drive parameter, and the rotation amount of the wafer 10 is calculated based on the rotation component θxw,
Θ on wafer stage 11 on which wafer 10 is mounted
The stage is rotated to correct the rotational component of the wafer (step 131).

When the exposure processing is started (step 32), the shot number N is initialized to 1 (step 33), and the flow advances to the subsequent shot exposure loop. When the number N = 1,
The mask shot to be transferred is 3'a, and the wafer shot 10a is superimposed on the image.

Next, for the mask shots 3'a to 3'y, the transfer position of the mask shot is read out (step 34), and the mask stage 6 is driven to move the mask shot to the transfer position (step 35). ).

On the other hand, for wafer shots 10a to 10y, the position of the wafer shot on which the image of the mask shot is to be superimposed is read (step 13).
4) The wafer stage 11 is driven to move to the position of the wafer shot (step 36).

Here, these mask shots 3′a to 3′a to
The positions of 3′y and the wafer shots 10 a to 10 y are scanning start positions of the subsequent relative scanning exposure to the charged beam 12.

Next, Steps 371 to 373 correspond to Steps 371 to 3 of the second embodiment described above.
Same as 73.

Steps 1371 to 1373 correspond to mask shot profile correction steps 371 to 373, and are conditional branches depending on how to correct the shot magnification and the shot rotation amount as the wafer shot profile. .

First, all the wafer shots 10a to 10a
It is determined whether or not y is specified in step 122 (step 1371), and all wafer shots 10a to 10y are determined.
Is specified, that is, if the correction relating to the shot profile is performed for each wafer shot, the shot profile of the wafer shot stored in step 1272 is read out (step 137).
2) driving the magnification correction system 8 to correct the shot magnification,
The rotation correction system 9 is driven to correct the shot rotation amount (step 1373).

When all the wafer shots 10a to 10y are not specified, that is, correction for the shot profile is performed using the average shot magnification Mmw and the average shot rotation amount θmw, and all the mask shots 3a to 3y are used.
, The magnification correction system 8 and the rotation correction system 9 have already been driven in step 1274 based on the average shot magnification Mmw and the average shot rotation amount θmw.
Condition 1 branches at step 1371 so as to skip 3.

As described above, in the present embodiment, step 1
Since only the outer peripheral wafer shot is designated in step S22, the average shot magnification Mm has already been set in step 1274.
Since the correction has been made based on w and the average shot rotation amount θmw, steps 1372 and 1373 are performed.
Skips.

Thereafter, the beam control system 2 is driven to irradiate the mask beam 3'a with the charged beam 12, and the image of the mask shot 3'a is transferred to the wafer shot 10a on the wafer 10.
Then, transfer is started by superimposing the mask, the mask stage 6 and the wafer stage 11 are relatively moved to perform relative scanning exposure (step 38), and the processing for one shot is completed.

The subsequent steps are the same as in the first and second embodiments, and the loop of steps 34 to 40 is repeated by the number of shots. When exposure of all shots is completed, the process branches at step 39 to branch to the next step. To the chip processing step.

As described above, mask shots 3'a to 3 '
y alignment marks 3'a1-3'y6 and wafer shots 10a-10y alignment marks 10-1
By correcting the relative position with respect to 10-40 and sequentially transferring the partial transfer pattern based on the corrected position, a transfer image with high connection accuracy of each shot can be obtained, and the wafer 1
It is possible to superimpose the pattern transferred on 0 with high precision.

In the description of the present embodiment, the alignment and transfer of the partial transfer pattern have been described.
Various methods employed in a conventional light exposure apparatus or the like can be used as a method for aligning the chip on the zero.

Although the alignment of the chips on the wafer 10 will not be described in detail here, for example, in the global alignment method, the arrangement of the chips on the wafer 10 is measured, and the arrangement is compared with the theoretical arrangement. The coordinate system relating to the arrangement of is corrected. The wafer stage 11 may be driven using the corrected coordinate system, and then the present embodiment may be further applied.

(Fourth Embodiment) In the first to third embodiments, correction is performed collectively for all shots, in other words, for the entire shot array, but in this embodiment, several shot arrays are used. The feature is that the correction is performed for each group.

The details of the present embodiment applied to the second layer using the shot layout of FIG. 10 will be described below with reference to FIGS. 17 to 22.

FIGS. 17 to 19 are flowcharts for explaining a mask shot transfer method according to the present embodiment.

FIG. 20 shows mask shots 3'a to 3'y.
FIG. 21 is a diagram showing a state in which wafer shots 10a to 10y are divided into two groups. FIG. 21 is a schematic diagram illustrating a shot profile of a mask shot group, and FIG. 22 is a diagram illustrating a shot profile of a wafer shot group. It is a schematic diagram.

First, a mask shot on the transfer mask 3 'and a wafer shot 10a of the transfer image 101 on the wafer 10
The grouping of 10 to 10 y will be described.

Here, as shown in FIG. 20, two mask shot pattern groups 3′mg1 and 3′mg1
It is divided into two. Mask shot pattern group 3 '
mg1 is composed of mask shots 3'a to 3'o, and mask shot pattern group 3'mg2 is composed of mask shots 3'p to 3'y.

Similarly, wafer shots are also divided into two wafer shot pattern groups, 10 mg1 and 10 mg2. Wafer shot pattern group 10mg1
Is composed of wafer shots 10a to 10o, and the wafer shot pattern group 10wg2 is composed of wafer shots 10p to 10y.

The mask shot pattern group 3 '
Each of the grids indicated by mg1 and 3′mg2 and the wafer shot pattern groups 10wg1 and 10wg2 represents a coordinate system for each pattern group.

In this embodiment, the number of groups is 2
And the grouping is the same for mask shots and wafer shots. However, when applying the present invention, there are no restrictions on the number of groups and the method of dividing mask shots and wafer shots. Absent.

Next, after grouping the mask shots and wafer shots as described above, FIGS.
9 will be described in detail.

Step 1000 to step 1001 are
This is the same as Steps 20 to 21 in the third embodiment, the transfer mask 3 'is loaded on the mask stage 6 (Step 1000), and the mask stage 6 is driven to perform reference alignment (Step 1001).

When the transfer mask 3 ′ is arranged at the reference position, an alignment mark used for measuring the alignment of each mask shot of the transfer mask 3 ′ is specified, and at the same time, the mask shot pattern group and which group the mask shot belongs to. (Step 10
02).

Here, the mask shots 3'a, 3 '
c, 3'e, 3'k, 3'm, 3'o, 3'p, 3 '
r, 3't, 3'u, 3'w, 3'y corresponding to each alignment mark 3'a1-3'a4, 3'c1-3'c
4, 3'e1-3'e4, 3'k1-3'k4, 3'm
1-3'm4, 3'o1-3'o4, 3'p1-3'p
4, 3'r1-3'r4, 3't1-3't4, 3'u
1-3'u4, 3'w1-3'w4, 3'y1-3'y
4 is specified. Also, as described with reference to FIG.
As a mask shot pattern group, 3′mg1 composed of mask shots 3′a to 3′o and 3′mg2 composed of mask shots 3′p to 3′y are designated here.

Next, each mask shot 3'a, 3'c, 3 'on the transfer mask 3' designated in step 1002.
e, 3'k, 3'm, 3'o, 3'p, 3'r, 3 '
t, 3′u, 3′w, 3′y, respectively, alignment marks 3′a1 to 3′a4, 3′c1 to 3′c4,
3'e1-3'e4, 3'k1-3'k4, 3'm1
3'm4, 3'o1-3'o4, 3'p1-3'p4,
3'r1-3'r4, 3't1-3't4, 3'u1-
3'u4, 3'w1-3'w4, 3'y1-3'y4,
Is measured by the mask alignment system 5 (step 1003).

The mask shots 3'a, 3'c, 3'e,
3'k, 3'm, 3'o, 3'p, 3'r, 3't,
3'u, 3'w, 3'y alignment marks 3'a
1-3'a4, 3'c1-3'c4, 3'e1-3'e
4,3'k1-3'k4,3'm1-3'm4,3'o
1-3'o4, 3'p1-3'p4, 3'r1-3'r
4, 3't1-3't4, 3'u1-3'u4, 3'w
After measuring the positions of 1 to 3'w4 and 3'y1 to 3'y4, the position of each alignment mark is stored for each mark (step 1004).

Next, each of the alignment marks 3'a1-3'a4, 3'c1-3'c stored in step 1004 is stored.
4, 3'e1-3'e4, 3'k1-3'k4, 3'm
1-3'm4, 3'o1-3'o4, 3'p1-3'p
4, 3'r1-3'r4, 3't1-3't4, 3'u
1-3'u4, 3'w1-3'w4, 3'y1-3'y
4. The position of 4 is read and the mask shots 3'a to 3'y
Is calculated (step 1005).

Next steps 1006 to 1012
Is a series of processes for each mask shot pattern group.

First, the mask shot pattern group number Nmg is initialized to 1 (step 1006). That is, the mask shot pattern group 3 ′ is the first group of a series of loop processing starting from the next step.
This means that mg1 has been specified.

Next, parameters relating to the coordinate system of the mask shot pattern group indicated by 3′mg1 are calculated (step 1007) and stored (step 10).
08). The parameters calculated here and the method of deriving them are the same as in the third embodiment. As shown in FIG. 21, the parameters are translation (shift) components Sxl, Sy1, Rotation component θx of the whole coordinate system
1, orthogonality components θx1−θy1, magnification in the x-axis direction (scale magnification) Dxr1 / Dxf1, magnification Dyr in the y-axis direction
1 / Dyf1. Here, the suffix 1 indicates that the mask shot pattern group number Nmg is a parameter relating to the coordinate system of 1.

Further, the mask shot group 3 ′ mg
Average shot magnification Mm1 and average shot rotation amount θ at 1
m1 is calculated (step 1009) and stored (step 1010). Here, the reason why the average value of each profile is used is that not all the shots in the mask shot pattern group are specified in step 1002. Up to this point, a series of processing for the mask shot pattern group 3′mg1 has been completed, so 1 is added to the number Nmg of the mask shot pattern group (step 1011) to determine whether the processing has been completed for all mask shot pattern groups. A determination is made (step 1012).

If the processing has been completed for all the mask shot pattern groups, the process proceeds to the next step; otherwise, the process returns to step 1007 to return to the next mask shot pattern group, in this case, the mask shot pattern group 3′mg2. Perform processing for.
Steps 1007 to 1 until the processing is completed for all the mask shot pattern groups.
The loop of 012 is repeated.

Subsequent steps 1013 to 102
No. 4 performs processing corresponding to steps 1007 to 1012 on the mask shot pattern group on the wafer shot pattern group.

First, an alignment mark to be used for alignment measurement of each wafer shot of the transfer image 101, which is the transferred pattern, which is the first layer, is specified, and at the same time, a wafer shot pattern group and a group to which each wafer shot belongs are specified. (Step 10
13). Here, the wafer shots 10a, 10c, 1
0e, 10f, 10j, 10k, 10o, 10t, 10
u, 10w, 10y each alignment mark 1
0-1, 10-2, 10-5, 10-6, 10-9, 1
0-10, 10-37, 10-38, 10-13, 10
-14, 10-35, 10-36, 10-15, 10-
16, 10-33, 10-34, 10-17, 10-1
8, 10-29, 10-30, 10-25, 10-2
6, 10-22, and 10-21, respectively.

As described with reference to FIG. 22, wafer shot pattern group
Here, a wafer shot pattern group 10wg1 composed of wafer shots 10p to 10y and a wafer shot pattern group 10wg2 composed of wafer shots 10p to 10y are designated here.

Next, each of the wafer shots 10a, 10c, 10e, 10f,
j, 10k, 10o, 10p, 10t, 10u, 10
w, alignment marks 10- respectively corresponding to 10y
1, 10-2, 10-5, 10-6, 10-9, 10-
10, 10-37, 10-38, 10-13, 10-1
4, 10-35, 10-36, 10-15, 10-1
6, 10-33, 10-34, 10-17, 10-1
8, 10-29, 10-30, 10-25, 10-2
The positions of 6, 10, 22 and 10-21 are measured by the wafer alignment system 14 (step 1014).

The wafer shots 10a, 10c, 10e,
10f, 10j, 10k, 10o, 10p, 10t, 1
Alignment marks 10-1, 10-2, 10-5, 10-6, 10- corresponding to 0u, 10w, and 10y, respectively.
9, 10-10, 10-37, 10-38, 10-1
3, 10-14, 10-35, 10-36, 10-1
5, 10-16, 10-33, 10-34, 10-1
7, 10-18, 10-29, 10-30, 10-2
After measuring the positions of 5, 10-26, 10-22, and 10-21, the position of each alignment mark is stored for each mark (step 1015).

Each of the alignment marks 10-1, 10-2, 10-5, 10-6, 1 stored in step 1015
0-9, 10-10, 10-37, 10-38, 10-
13, 10-14, 10-35, 10-36, 1011
5, 10-16, 10-33, 10-34, 10-1
7, 10-18, 10-29, 10-30, 10-2
The positions of 5, 10-26, 10-22, and 10-21 are read to calculate the actual coordinate system of the wafer shots 10a to 10y (step 1016).

Next steps 1017 to 1024
Is a series of processes for each wafer shot pattern group.

First, the wafer shot pattern group number Nwg is initialized to 1 (step 1017). That is, as the first group of a series of loop processing starting from the next step, the wafer shot pattern group 10
This means that wg1 has been specified.

Next, parameters relating to the coordinate system of the wafer shot pattern group indicated by 10wg1 in FIG. 22 are calculated (step 1018) and stored (step 1019). The parameters calculated here and the method of deriving them are the same as in the third embodiment, and each parameter is a translation (shift) component Sx in the x-axis direction and the y-axis direction.
w1, Syw1, rotation component θxw1 of the entire coordinate system, orthogonality component θxw1-θyw1, magnification in the x-axis direction (scale magnification) Dxrw1 / Dxfw1, magnification Dyr in the y-axis direction
w1 / Dyfw1. Here, the subscript 1 indicates that the wafer shot pattern group number Nwg is a parameter relating to the coordinate system of 1.

Here, the parameters representing the relative relationship were obtained by comparing the coordinate system for design of the wafer shot, which is the reference coordinate system of the apparatus, with the actual coordinate system of the wafer shot. The parameter representing the relative relationship obtained by comparing the designed coordinate system of the mask shot with the actual coordinate system of the mask shot is also a comparison with the designed coordinate system that is the reference coordinate system of the apparatus. Therefore, by comparing the actual coordinate system of the mask shot with the actual coordinate system of the wafer shot,
A parameter indicating the relative relationship may be obtained.

Next, wafer shot group 10wg1
Shot magnification Mmw1 and average shot rotation amount θ
mw1 is calculated (step 1020) and stored (step 1021). The reason why the average value of each profile is used here is that not all the shots in the wafer shot pattern group are specified in step 1013. Up to this point, a series of processing for the wafer shot pattern group 10wg1 has been completed, so 1 is added to the number Nwg of the wafer shot pattern group (step 1023), and it is determined whether the processing has been completed for all the wafer shot pattern groups. Is performed (step 1024).

If the processing has been completed for all the wafer shot pattern groups, the process proceeds to the next step, and if not, the process returns to step 1018, and the process returns to the next wafer shot pattern group, in this case, the wafer shot pattern group 10wg2. The processing is performed for this.

Step 1018 until the processing is completed for all the mask shot pattern groups.
To repeat the loop of Step 1024.

Steps 1025 to 1031 are
In this step, the mask shot parameters are read out for each mask shot pattern group, and the transfer position of each mask shot is calculated and stored.

First, the group number Nmg of the mask shot pattern group is initialized to 1 (step 10).
25).

Next, the mask shot pattern group number Nmg, that is, each mask shot parameter of the group 3′mg1 is read (step 10).
26), the transfer position of each mask shot is calculated (step 1027) and stored (step 1028).

The parameters read out here and the procedure for calculating the transfer position of each mask shot from the parameters are the same as in the third embodiment, except that the scale magnifications Dxr1 / Dxf1, Dyr1 / D with respect to the mask stage 6.
Since it is necessary to correct the yf1 and the rotation component θx1 for each mask shot pattern group, it is not necessary to immediately set the scale magnification for the mask stage 6 or drive the θ stage on the mask stage 6 here. Can not.

Therefore, these scale magnifications Dxr1 / D
xf1, Dyr1 / Dyf1 and the rotation component θx1 are stored (1029).

Next, 1 is added to the group number Nmg of the mask shot pattern group (step 103).
0), it is determined whether the processing has been completed for all the mask shot pattern groups (step 103).
1).

If the processing has been completed for all the mask shot pattern groups, the process proceeds to the next step; otherwise, the process returns to step 1026 to return to the next mask shot pattern group, in this embodiment, the mask shot pattern. Processing is performed on pattern group 3′mg2. Step 102 is performed until the processing is completed for all the mask shot pattern groups.
The loop from 6 to 1031 is repeated.

In steps 1032 to 1038, a process similar to the process performed on the mask shot pattern group in steps 1025 to 1031 is performed on the wafer shot pattern group.

First, the group number Nwg of the wafer shot pattern group is initialized to 1 (step 10).
32).

Next, the wafer shot pattern group number Nwg, that is, each wafer shot parameter of the group 10wg1 in FIG. 22 is read (step 10).
33), calculate the position of each wafer shot (step 1)
034) and memorize it (step 1035).

The procedure for calculating the position of each wafer shot from the parameters read out here and the parameters is as follows.
Same as the third embodiment, except that the setting of the scale magnification for the mask stage 6 and the drive of the θ stage on the mask stage 6 cannot be performed immediately,
Scale magnification Dxrw1 / D for wafer stage 6
xfw1, Dyrw1 / Dyfw1, and rotation component θxw
1 needs to be performed for each wafer shot pattern group, the scale magnification is immediately set for the wafer stage 11 or the wafer stage 6
The upper θ stage cannot be driven.

Accordingly, these scale magnifications Dxrw1 /
Dxfw1, Dyrw1 / Dyfw1 and rotation component θx
w1 is stored (1036).

Next, 1 is added to the group number Nwg of the mask shot pattern group (step 103).
7) It is determined whether or not the processing has been completed for all wafer shot pattern groups (step 103).
8).

If the processing has been completed for all the wafer shot pattern groups, the process proceeds to the next step; otherwise, the process returns to step 1033 to return to the next wafer shot pattern group, in this case, wafer shot pattern group 10wg2. The processing is performed for this. Thus, the loop of Steps 1033 to 1038 is repeated until the processing is completed for all the wafer shot pattern groups.

The exposure process is started (step 103)
9) and the shot number N is initialized to 1 (step 10)
40), and proceed to the subsequent shot exposure loop. Number N =
At 1, the mask shot to be transferred is 3'a, and the wafer shot 10a is superimposed on the image.

Here, it is necessary to specify the mask shot pattern group to which the shot N belongs in order to read and correct the scale magnification and the rotation amount respectively corresponding to each mask shot pattern group. Therefore, the number of the mask shot pattern group to which the shot N belongs is substituted into the group number Nmg (step 1041).

The scale magnification Dxr of the mask shot pattern group Nmg stored in step 1029
Nmg / DxfNmg, DyrNmg / DyfNmg and rotation component θxNmg are read (step 104).
2). In this case, since the shot 1 belongs to the mask shot pattern group 3′mg1, the scale magnification Dxr
1 / Dxf1, Dyr1 / Dyf1, and rotation component θx1
Is read.

Therefore, the scale magnification Dxr1 / Dxf1,
Dyr1 / Dyf1 is set for the mask stage 6 (step 1043), the transfer position of the mask shot N is read out (step 1044), and the mask stage 6 is driven, and at the same time, based on the rotation amount θx1 read out at step 1042. Then, the θ stage on the mask stage 6 is also driven to drive the mask shot pattern group 3′mg1.
Is corrected (step 1045).

Next, the same processing as that for the mask shot pattern group is performed on the wafer shot pattern group.

To specify the wafer shot pattern group to which the shot N belongs, the number of the wafer shot pattern group to which the shot N belongs is assigned to the group number Nwg (step 1046).

Next, scale magnifications DxrwNwg / DxfwNwg, DyrwNwg / D of wafer shot pattern group Nwg stored in step 1036
The yfwNwg and the rotation component θxwNwg are read (step 1047). In this case, since shot 1 belongs to wafer shot pattern group 10wg1, scale magnifications Dxrw1 / Dxfw1, Dyrw1 / Dyfw
1 and the rotation component θxw1 are read.

Accordingly, the scale magnification Dxrw1 / Dxfw
1, Dyrw1 / Dyfw1 is set for the wafer stage 11 (step 1048), the transfer position of the mask shot N is read out (step 1049), and the wafer stage 11 is driven to move to the transfer position of the shot N. At the same time, the θ stage on the wafer stage 11 is also driven based on the rotation amount θxw1 read in step 1047 to correct the rotation component of the wafer shot pattern group 10wg1 (step 1050).

Next, the shot profile of the mask shot N stored in step 1010 is read (step 1051), the magnification correction system 8 is driven to correct the shot magnification, and the rotation correction system 9 is driven. The shot rotation amount is corrected (Step 1052).

Now, since the shot number is 1, the average shot magnification Mm1 and the average shot rotation amount θm1 for the mask shot pattern group 3′mg1 are read out, and the magnification correction system 8 and the rotation correction system 9 are read based on them. Will be driven.

Similarly, the shot profile of the wafer shot N stored in step 1021 is read out (step 1053), the magnification correction system 8 is driven to correct the shot magnification, and the rotation correction system 9 is driven. The shot rotation amount is corrected (Step 1054).

That is, the average shot magnification Mmw1 and the average shot rotation amount θmw1 with respect to the wafer shot pattern group 10wg1 are read, and the magnification correction system 8 and the rotation correction system 9 are driven based on them.

If all shots have been specified in step 1002 or step 1013, step 1051 to step 1052 or step 10
In steps 53 to 1054, the shot magnification and the shot rotation amount for each shot are read, and the magnification correction system 8 and the rotation correction system 9 are driven based on them.

Here, the beam control system 2 is driven to irradiate the mask shot 3 ′ with the charged beam 12, and the image of the mask shot 3 ′ a is changed to the wafer shot 10 a on the wafer 10.
Then, transfer is started by superimposing the mask, the mask stage 6 and the wafer stage 11 are relatively moved to perform relative scanning exposure (step 1055), and the processing for one shot is completed.

Thereafter, if exposure for all shots has been completed, the process branches (step 1056), and proceeds to the next chip processing step (step 1058). Otherwise, 1 is added to the shot number N and step 1041 is performed. Return to In this way, step 1041 is performed for the number of shots.
Step 1057 is repeated.

As described above, mask shots and wafer shots are assigned to a plurality of shot groups, for example, mask shots are assigned to mask shot pattern groups 3′mg1 and 3 ′.
mg2, the wafer shots are divided into wafer shot pattern groups 10wg1 and 10wg2, and relative correction is performed for each group using a parameter representing the relative relationship between their coordinate system and the designed coordinate system. Further precise correction in a small area becomes possible.

The greatest advantage of this embodiment is that it is possible to perform further precise correction in this small area, and it is possible to extract an area requiring higher joining accuracy from information of a transfer pattern on a transfer mask. If the partial transfer pattern in that area is applied as a pattern group, the advantages of the present embodiment can be maximized.

The information of the transfer pattern of interest here is, for example, a line width, a pattern density, a set of circuit patterns such as contact holes and memory cells, and the information of the transfer pattern of interest differs for each layer. You may. By doing so, it is possible to perform joining with higher precision.

(Fifth Embodiment) In the fourth embodiment,
The measurement of each alignment mark, the calculation of the position thereof, and the exposure of the shot are collectively performed, and the processing of the coordinate system and the processing of the shot profile are performed for each group.

In this way, a sufficient effect can be obtained by dividing the shot array into a plurality of groups.
Further, if there is a risk of time, that is, there is a possibility that the coordinate system of another group may change while exposing a shot of one group, each alignment may be changed as described below in this embodiment. From the measurement of the mark to the calculation of the coordinate parameters, the calculation of the shot profile, and the exposure of the shot, are performed for each group. And a process including the process may be started.

FIGS. 23 to 25 show flowcharts of this embodiment.

The steps in FIGS. 23 to 25 are substantially the same as the steps in the fourth embodiment. The difference from the fourth embodiment is that the order of processing and the condition of grouping, ie, the mask Since the only limitation is to make the combination of shots that are elements of the shot pattern group and the wafer shot group the same, a detailed description is omitted.

In this embodiment, the number of shots is particularly large.
For example, when the number of shots is as large as 200 shots in a 10 × 20 matrix arrangement, the environment related to the transfer mask 3 and the wafer 10 shown in FIG. However, this is suitable for a case in which these coordinate systems are also affected.

As described above, in the first and second embodiments, the first
The processing for the layer has been described, and the processing for the second layer has been described in the third to fifth embodiments. Normally, various transfer patterns can be obtained by applying any of the first to fifth embodiments.
For example, when assuming a mix-and-match with a light exposure device or the like, when the first layer is exposed at once by the light exposure device on the entire chip, or when the alignment mark in the present invention is not on the first layer , And so on.

However, even in such a case, the first or second embodiment can be treated as apparently the transfer processing of the first layer. In other words, the chip alignment on the wafer 10 side is determined by using a method adopted in the conventional exposure apparatus, for example, a global alignment method, and the position of each chip is determined.
In other words, it can be understood that the embodiment of the present invention described in the first or second embodiment should be applied to the arrangement coordinate system of the chips.

As can be seen from the first to fifth embodiments described above, the correction for each of the mask shot and the wafer shot is performed on each of the mask stage 6 and the wafer stage 11. However, the transfer mask 3
Since the wafer 10 and the wafer 10 have a certain relative relationship determined by the transfer system, the mask stage 6 and the wafer stage 11 can share the correction component. For example, of the correction items, the shift components (x, y components) can be corrected on the wafer stage 11 side, and the rotation components can be corrected on the mask stage 6 side. By selecting such sharing depending on the function, performance, driving accuracy, etc. of each stage,
Shot joining accuracy can be improved.

In the present invention, the transfer is performed by correcting the relative relationship between the mask shot and the object to be transferred. The present invention can be applied to systems such as a projection transfer system, a reduced projection transfer system, a proximity transfer system, and a contact transfer system.

In the first to fifth embodiments, the case where the present invention is applied to a so-called scanner which uses scanning exposure as a shot exposure method has been described. However, a step-and-repeat method which employs still batch transfer as a shot exposure method is described. Even if the present invention is applied to a so-called stepper, the essence of the present invention does not change at all.

Next, an embodiment of a device manufacturing method using the above-described mask pattern transfer method will be described.

FIG. 27 shows a micro device (a semiconductor chip such as an IC or an LSI, a liquid crystal panel, a CCD, a thin film magnetic head,
2 shows a manufacturing flow of a micromachine or the like. Step 10
In 1 (circuit design), a pattern of a device is designed. Step 102 (mask fabrication) forms a mask on which the designed pattern is formed. On the other hand, in step 103 (wafer manufacture), a wafer is manufactured using a material such as silicon or glass. Step 104 (wafer process) is called a pre-process, and an actual circuit is formed on the wafer by lithography using the prepared mask and wafer. The next step 105 (assembly) is called a post-process, and is a process of forming a semiconductor chip using the wafer prepared in step 104, and includes processes such as an assembly process (dicing and bonding) and a packaging process (chip encapsulation). including. In step 106 (inspection), inspections such as an operation confirmation test and a durability test of the semiconductor device manufactured in step 105 are performed. Through these steps, a semiconductor device is completed and shipped (Step 1).
07).

FIG. 28 shows a detailed flow of the wafer process. Step 111 (oxidation) oxidizes the wafer's surface. Step 112 (CVD) forms an insulating film on the wafer surface. In step 113 (electrode formation), electrodes are formed on the wafer by vapor deposition. Step 11
In step 4 (ion implantation), ions are implanted into the wafer. In step 115 (resist processing), a resist is applied to the wafer. Step 116 (exposure) uses the above-described exposure apparatus or exposure method to print the circuit pattern of the mask on the plurality of shot areas of the wafer by printing. In step 117 (developing), the exposed wafer is developed.
In step 118 (etching), portions other than the developed resist image are removed. In step 119 (resist removal), unnecessary resist after etching is removed. By repeating these steps, multiple circuit patterns are formed on the wafer. By using the production method of the present embodiment, a large-sized, high-density device, which has conventionally been difficult to produce, can be produced at low cost.

[0204]

As described above, according to the present invention,
The transfer pattern on the transfer mask is divided into a plurality of partial transfer patterns, alignment marks are arranged on those partial transfer patterns, the alignment marks and the alignment marks transferred on the substrate to be transferred, and the partial transfer pattern on the transfer mask. By using the parameters representing the relative relationship between the coordinate system and the coordinate system of the object to be transferred, by correcting the relative relationship, it is possible to transfer with high throughput while improving the connection accuracy between the partial transfer patterns. it can. As a result, a device with higher precision than before can be manufactured in a short time.

[Brief description of the drawings]

FIG. 1 is a flowchart illustrating a mask shot transfer method according to a first embodiment of the present invention.

FIG. 2 is a schematic diagram illustrating a mask shot and a wafer shot in the first embodiment.

FIG. 3 is a schematic diagram illustrating a coordinate system of a mask shot in the first embodiment.

FIG. 4 is a diagram showing a coordinate system in manufacturing according to the first embodiment of the present invention.

FIG. 5 is a schematic diagram showing each coordinate system.

FIG. 6 is a schematic diagram illustrating fitting of only a variation.

FIG. 7 is a schematic diagram illustrating fitting of a random component.

FIG. 8 is a schematic diagram illustrating a shot profile of a mask shot in the first embodiment.

FIG. 9 is a flowchart illustrating a mask shot transfer method according to a second embodiment of the present invention.

FIG. 10 is a schematic diagram illustrating a mask shot and a wafer shot in the third and fourth embodiments.

FIG. 11 is a schematic diagram illustrating a shot profile of a mask shot in the second embodiment.

FIG. 12 is a diagram showing a coordinate system in production according to a third embodiment of the present invention.

FIG. 13 is a flowchart illustrating a mask shot transfer method according to a third embodiment of the present invention.

FIG. 14 is a flowchart illustrating a method of transferring a mask shot according to a third embodiment of the present invention.

FIG. 15 is a schematic diagram illustrating a coordinate system of a wafer shot in the third embodiment.

FIG. 16 is a schematic diagram illustrating a shot profile of a wafer shot in the third embodiment.

FIG. 17 is a flowchart illustrating a mask shot transfer method according to a fourth embodiment of the present invention.

FIG. 18 is a flowchart illustrating a mask shot transfer method according to a fourth embodiment of the present invention.

FIG. 19 is a flowchart illustrating a mask shot transfer method according to a fourth embodiment of the present invention.

FIG. 20 is a schematic diagram illustrating a pattern group of a mask shot and a wafer shot in the fourth embodiment.

FIG. 21 is a schematic diagram illustrating a shot profile of a mask shot group in a fourth embodiment.

FIG. 22 is a schematic diagram illustrating a shot profile of a wafer shot group in the fourth embodiment.

FIG. 23 is a flowchart illustrating a method of transferring a mask shot according to a fifth embodiment of the present invention.

FIG. 24 is a flowchart illustrating a mask shot transfer method according to a fifth embodiment of the present invention.

FIG. 25 is a flowchart illustrating a mask shot transfer method according to a fifth embodiment of the present invention.

FIG. 26 is a schematic configuration diagram of an example of an exposure apparatus for manufacturing a semiconductor device to which the transfer method of the present invention can be applied.

FIG. 27 is a flowchart showing a device manufacturing process.

FIG. 28 is a flowchart showing detailed steps of the wafer process shown in FIG. 27.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 Beam source 2 Beam control system 2a Beam control part 3, 3 'Transfer mask 3a1-3y6, 3'a1-3'y4, 10-1-10
-40 Alignment mark 3a-3y, 3'a-3'y Mask shot 3'mg1, 3'mg2 Mask shot pattern group 4 Illumination system 5 Mask alignment system 6 Mask stage 6a Mask stage controller 7 Projection system 8 Magnification correction system 8a Magnification correction control unit 9 Rotation correction system 9a Rotation correction control unit 10 Wafer 10a to 10y Wafer shot 10wg1, 10wg2 Wafer shot pattern group 11 Wafer stage 11a Wafer stage control unit 12 Charge beam 13 Slit 14 Wafer alignment system 15 Deflector 16 Device Control unit 101 Transfer image

Claims (42)

[Claims] 1. A transfer mask having a plurality of divided and distinguishable partial transfer patterns is used to irradiate an object to be transferred with a transfer beam through the transfer mask, and each of the partial transfer patterns is transferred to the transfer target. A mask pattern transfer method for sequentially transferring the partial transfer patterns to an object to obtain a desired pattern in which the partial transfer patterns are joined to the transfer target object; a plurality of the transfer masks each being associated with each of the partial transfer patterns; A mask pattern transfer method, wherein the alignment marks are provided, the relative position between the transfer target and each of the alignment marks is corrected, and the partial transfer patterns are sequentially transferred to the transfer target. 2. The correction of the relative position includes a step of designating a transfer mask mark for designating each of the alignment marks, and a step of measuring the position of each of the alignment marks designated in the step of designating the transfer mask mark. Calculating an actual coordinate system of each of the partial transfer patterns from the first coordinate parameter, and a first coordinate parameter representing a relative relationship between a designed coordinate system of each of the partial transfer patterns and an actual coordinate system of each of the partial transfer patterns. A first relative coordinate processing step consisting of calculating the following, when sequentially transferring each of the partial transfer patterns to the transfer object, using the first coordinate parameters, the respective relative to the transfer object 2. The method according to claim 1, further comprising: a transfer mask relative movement step of relatively moving an image of the partial transfer pattern. 3. In the first relative coordinate processing step,
In the step of calculating the actual coordinate system of each of the partial transfer patterns, the measurement of the position of each of the alignment marks is performed for each of the partial transfer patterns of the transfer mask in production, and based on this. A step of defining a coordinate grid in manufacturing; and a step of measuring the position of each of the alignment marks with respect to each of the partial transfer patterns of the actual transfer mask, and defining an actual coordinate grid based on this. Subtracting the production coordinate grid from the actual coordinate grid to calculate a first variation subdivision; and a first variation subdivision based on the first variation subdivision and the production coordinate lattice. 3. The method according to claim 2, further comprising: calculating a correction grid. 4. The mask pattern transfer method according to claim 2, further comprising a step of calculating a transfer position at which each of the partial transfer patterns is transferred to the transfer target from the first coordinate parameters. 5. The transfer mask mark designating step,
When all the partial transfer patterns are specified by specifying all the alignment marks, from the positions of the alignment marks obtained in the first relative coordinate processing step,
Calculating a partial transfer pattern profile comprising a partial transfer pattern rotation amount and a partial transfer pattern magnification of each of the partial transfer patterns, and using the calculated partial transfer pattern profile, an image of the image of the partial transfer pattern on the object to be transferred 5. The mask pattern transfer method according to claim 1, further comprising a transfer mask image profile correction step of correcting an image profile including a rotation amount and an image magnification. 6. The transfer mask mark designating step,
When a part of the partial transfer pattern is specified by specifying a part of the alignment mark, from the position of each of the alignment marks obtained in the first relative coordinate processing step,
An average partial transfer pattern profile, which is an average value of the partial transfer pattern profiles including the partial transfer pattern rotation amount and the partial transfer pattern magnification of each of the partial transfer patterns, is calculated. The mask pattern transfer method according to claim 1, further comprising a transfer mask image profile correction step of correcting an average image profile including an image rotation amount and an image magnification of the image of the partial transfer pattern with respect to a transferred material. 7. Simultaneously with transferring the partial transfer pattern to the transfer target, the alignment marks are also transferred to the transfer target, and after the transfer of the partial transfer pattern and each alignment mark to the transfer target, Using a transfer mask for overlapping having a plurality of partial transfer patterns different from the transfer mask and being divided and distinguishable, and a plurality of alignment marks associated with each of the partial transfer patterns, 2. A relative position between each of the alignment marks transferred on the transfer mask and each of the alignment marks on the overlay transfer mask is corrected, and each partial transfer pattern of the overlay transfer mask is sequentially transferred to the object to be transferred. 7. The method for transferring a mask pattern according to any one of items 6 to 6. 8. Correction of a relative position between each of the alignment marks transferred to the object to be transferred and each of the alignment marks of the overlay transfer mask is performed by an overlay transfer designating each of the alignment marks of the overlay transfer mask. Mask mark designating step, measuring the position of each alignment mark designated by the overlay transfer mask mark designating step, and calculating the actual coordinate system of the partial transfer pattern of the overlay transfer mask from the measured position, Calculating a second coordinate parameter representing a relative relationship between a designed coordinate system of the partial transfer pattern of the overlay transfer mask and an actual coordinate system of the partial transfer pattern of the overlay transfer mask; A transferred object mark designating step for designating each transferred alignment mark, Measuring the position of the alignment mark, calculating the actual coordinate system of each partial transfer pattern transferred to the transfer target from the measured position, and design coordinates of each partial transfer pattern of the transfer target A second relative coordinate processing step including a step of calculating a third coordinate parameter representing a relative relationship between a system and an actual coordinate system of each partial transfer pattern transferred to the transfer target; and the overlay transfer mask. When sequentially transferring the partial transfer pattern on the object to be transferred, the second coordinate parameter and the third coordinate parameter are used for each of the partial transfer patterns transferred to the object to be transferred. The method of transferring a mask pattern according to claim 7, further comprising: a relative transfer step of the transfer mask for overlapping, which relatively moves the image of each partial transfer pattern of the transfer mask for overlay. 9. In the second relative coordinate processing step,
In the step of calculating the actual coordinate system of the partial transfer pattern of the overlay transfer mask, the position of each alignment mark is measured for the partial transfer pattern of the overlay transfer mask in production, A step of defining a coordinate grid in the production of the partial transfer pattern of the transfer mask for overlay based on the above, and measuring the position of each alignment mark is performed on the actual partial transfer pattern of the transfer mask for overlay. Defining an actual coordinate grid of the partial transfer pattern of the transfer mask for overlay based on this, and a partial transfer pattern of the transfer mask for overlay from the actual coordinate grid of the partial transfer pattern of the transfer mask for overlay. Calculating a second variation subdivision by subtracting a coordinate grid in the production of the above, and a partial transfer of the second variation subdivision and the transfer mask for superimposition. 9. The method according to claim 8, further comprising: calculating a second correction grid based on a coordinate grid used for manufacturing the pattern. 10. The mask pattern transfer according to claim 8, further comprising a step of calculating a transfer position at which each partial transfer pattern of the overlay transfer mask is transferred to the transfer target from the second coordinate parameters. Method. 11. The mask pattern transfer according to claim 8, further comprising a step of calculating a transfer position of each partial transfer pattern transferred to the transfer target from the third coordinate parameter. Method. 12. Correction of a relative position between each of the alignment marks transferred to the transfer object and each of the alignment marks of the overlay transfer mask is performed by an overlay transfer designating each of the alignment marks of the overlay transfer mask. A mask mark designating step, measuring the position of each alignment mark designated by the overlay transfer mask mark designating step, and calculating an actual coordinate system of each partial transfer pattern of the overlay transfer mask from the measured position; The transfer object mark designation step of designating each alignment mark transferred to the transfer object, measuring the position of each alignment mark designated by the transfer object mark designation step, and measuring the transfer object mark from the measured position. Calculating the actual coordinate system of each of the partial transfer patterns transferred to the transfer mask; A second relative coordinate process including a step of calculating a fourth coordinate parameter representing a relative relationship between an actual coordinate system of the pattern and an actual coordinate system of each partial transfer pattern transferred to the transfer object. A step of, when sequentially transferring the partial transfer pattern of the transfer mask for superimposition onto the transfer object, using the fourth coordinate parameter, for each partial transfer pattern transferred to the transfer object. 10. The method of transferring a mask pattern according to claim 8, further comprising: a relative movement step of the transfer mask for overlapping, wherein the image of each partial transfer pattern of the transfer mask for overlapping is relatively moved. 13. In the case where all the partial transfer patterns are designated by designating all the alignment marks in the overlay transfer mask mark designating step, the overlay transfer mask obtained in the second relative coordinate processing step is provided. From the position of each alignment mark of the mask, a partial transfer pattern profile of the transfer mask for overlay composed of a partial transfer pattern rotation amount and a partial transfer pattern magnification of each partial transfer pattern of the transfer mask for overlay is calculated and calculated. An overlay transfer mask image profile for correcting an image profile comprising an image rotation amount and an image magnification of a partial transfer pattern image of the overlay transfer mask with respect to the partial transfer pattern transferred to the transfer target using the partial transfer pattern profile 13. A method according to claim 8, further comprising a correction step.
The method for transferring a mask pattern according to the above item. 14. When the partial transfer pattern is specified by designating a part of the alignment mark in the overlay transfer mask mark designating step, the overlay obtained in the second relative coordinate processing step is designated. From the position of each alignment mark of the transfer mask for use, the average partial transfer pattern profile which is the average value of the partial transfer pattern profile consisting of the partial transfer pattern rotation amount and the partial transfer pattern magnification of each of the partial transfer patterns was calculated and calculated. A superimposed transfer mask image for correcting an average image profile consisting of an image rotation amount and an image magnification of a partial transfer pattern image of the superimposed transfer mask with respect to the partial transfer pattern transferred to the transfer object using an average partial transfer pattern profile 13. The mask according to claim 8, further comprising a profile correction step. Screen pattern transfer method. 15. The transfer object mark designation step,
When all the partial transfer patterns are specified by specifying all the alignment marks, each part of the transfer target is obtained from the position of each alignment mark of the transfer target obtained in the second relative coordinate processing step. A partial transfer pattern profile including a partial transfer pattern rotation amount of the transfer pattern and a partial transfer pattern magnification is calculated, and the image rotation of the image of the partial transfer pattern transferred to the transfer target using the calculated partial transfer pattern profile 15. The mask pattern transfer method according to claim 8, further comprising a transfer object image profile correction step of correcting an image profile including an amount and an image magnification. 16. The transfer object mark designation step,
When a part of the partial transfer pattern is specified by specifying a part of the alignment mark, the position of the alignment mark of the transfer object obtained from the second relative coordinate processing step is determined based on the position of each alignment mark of the transfer object. An average partial transfer pattern profile, which is an average value of a partial transfer pattern profile including a partial transfer pattern rotation amount and a partial transfer pattern magnification of each partial transfer pattern, is calculated, and the transferred object is calculated using the calculated average partial transfer pattern profile. 15. A transfer object image profile correcting step of correcting an average image profile comprising an image rotation amount and an image magnification of an image of a partial transfer pattern transferred to a transfer pattern.
The method for transferring a mask pattern according to the above item. 17. The correction of the relative position includes associating a plurality of partial transfer pattern groups each including a plurality of partial transfer patterns of the transfer mask with each of the plurality of partial transfer patterns and the plurality of partial transfer patterns. A transfer mask group setting step of setting by specifying each of the alignment marks specified, and measuring the position of each of the alignment marks specified by the transfer mask group setting step, and from the measured position, the partial transfer pattern group Calculating the actual coordinate system of the partial transfer pattern group, and the first representing the relative relationship between the designed coordinate system of the partial transfer pattern group and the actual coordinate system.
Calculating the group coordinate parameters of
A group relative coordinate processing step, and when sequentially transferring the partial transfer pattern group to the transfer target, a transfer mask for relatively moving the image of the partial transfer pattern group with respect to the transfer target based on the transfer position 2. The method according to claim 1, further comprising a group relative movement step. 18. The step of calculating an actual coordinate system of the partial transfer pattern group in the first group relative coordinate processing step includes: measuring a position of each of the alignment marks; A step of defining a coordinate grid in the production of a partial transfer pattern group of the transfer mask based on the transfer pattern group based on the transfer mask group; and measuring the positions of the respective alignment marks. For the partial transfer pattern group of
Defining an actual coordinate grid of the partial transfer pattern group of the transfer mask based on this; and manufacturing the partial transfer pattern group of the transfer mask from the actual coordinate grid of the partial transfer pattern group of the transfer mask. Calculating a first group variation subdivision by subtracting a coordinate lattice; and a first group correction based on the first group variation subdivision and the production coordinate grid of the partial transfer pattern group of the transfer mask. 18. The method according to claim 17, further comprising: calculating a lattice. 19. The mask pattern transfer method according to claim 17, further comprising a step of calculating a transfer position at which each of the partial transfer pattern groups is transferred to the object to be transferred from the first group coordinate parameters. 20. When all the partial transfer patterns are specified by specifying all the alignment marks in the transfer mask group setting step, each of the alignment marks obtained in the first group relative coordinate processing step is obtained. From the position, a partial transfer pattern group profile including the partial transfer pattern group rotation amount and the partial transfer pattern group magnification of each of the partial transfer pattern groups is calculated, and the calculated partial transfer pattern group profile is used to calculate the partial transfer pattern group profile. 20. A transfer mask group image profile correcting step of correcting an image profile consisting of an image rotation amount and an image magnification of the image of the partial transfer pattern group.
The method for transferring a mask pattern according to the above item. 21. In the transfer mask group setting step, when a part of the partial transfer pattern is specified by specifying a part of the alignment mark, each of the alignments obtained in the first group relative coordinate processing step. From the position of the mark, the average partial transfer pattern group profile which is the average value of the partial transfer pattern group profiles including the partial transfer pattern group rotation amount and the partial transfer pattern group magnification of each of the partial transfer pattern groups is calculated. 20. A transfer mask group image profile correcting step of correcting an average image profile including an image rotation amount and an image magnification of an image of the partial transfer pattern group with respect to the transferred object using an average partial transfer pattern group profile.
The method for transferring a mask pattern according to any one of the above items. 22. Simultaneously with transferring the partial transfer pattern group to the transfer target, the alignment marks are also transferred to the transfer target, and the partial transfer pattern group and the alignment marks are transferred to the transfer target. After transfer, different from the transfer mask, divided, and using a transfer mask for overlay having a plurality of partial transfer patterns that can be distinguished, and a plurality of alignment marks associated with each of the partial transfer patterns, A relative position between each of the alignment marks transferred to the transfer object and each of the alignment marks of the overlay transfer mask is corrected, and each partial transfer pattern of the overlay transfer mask is sequentially transferred to the transfer object. The method of transferring a mask pattern according to any one of claims 17 to 21. 23. The correction of the relative position between each of the alignment marks transferred to the transfer object and each of the alignment marks of the overlay transfer mask is performed using a plurality of partial transfer patterns of the overlay transfer mask as elements. A plurality of partial transfer pattern groups are set by specifying a plurality of partial transfer patterns of the transfer mask for overlay and each alignment mark associated with each of the plurality of partial transfer patterns of the transfer mask for overlay. The position of each alignment mark designated in the transfer mask group setting step and the overlay transfer mask group setting step is measured, and the actual coordinate system of the partial transfer pattern group of the overlay transfer mask is calculated from the measured position. A process, a design coordinate system of a partial transfer pattern group of the transfer mask for overlay and the Calculating a second group coordinate parameter representing a relative relationship between the partial transfer pattern group of the transfer mask and the actual coordinate system; and a plurality of partial transfer patterns having the plurality of partial transfer patterns transferred to the transfer object as elements. A partial transfer pattern group is set by designating a plurality of partial transfer patterns transferred to the transfer object and each alignment mark associated with each of the plurality of partial transfer patterns transferred to the transfer object. The transfer object group setting step, measuring the position of each alignment mark specified in the transfer object group setting step, and the actual coordinate system of each partial transfer pattern group transferred from the measured position to the transfer object Calculating, and the design coordinate system of each partial transfer pattern group of the transferred object and the transferred image on the transferred object. A second group relative coordinate processing step of calculating a third group coordinate parameter representing a relative relationship between each partial transfer pattern group and an actual coordinate system, and a partial transfer pattern group of the overlay transfer mask. ,
When sequentially transferring the image onto the object to be transferred sequentially, the second group coordinate parameter and the third group coordinate parameter are used for each partial transfer pattern group transferred to the object to be transferred. 23. The method of transferring a mask pattern according to claim 22, further comprising: a relative transfer step of a transfer mask for superposition for relatively moving an image of each partial transfer pattern group of the transfer mask. 24. The step of calculating an actual coordinate system of the partial transfer pattern group of the overlay transfer mask in the second group relative coordinate processing step includes measuring a position of each of the alignment marks by manufacturing. A step of defining a coordinate grid for the production of the partial transfer pattern group of the overlay transfer mask based on the partial transfer pattern group of the overlay transfer mask, and the position of each alignment mark Measuring the actual partial transfer pattern group of the overlay transfer mask, and defining the actual coordinate grid of the partial transfer pattern group of the overlay transfer mask based on this, Production of the partial transfer pattern group of the transfer mask for overlay from the actual coordinate grid of the partial transfer pattern group of the transfer mask Calculating a second group variation subdivision by subtracting the above coordinate lattice, and a second group variation subdivision based on the second group variation subdivision and the production coordinate grid of the partial transfer pattern group of the overlay transfer mask. 24. The mask pattern transfer method according to claim 23, further comprising: calculating two group correction gratings. 25. The method according to claim 22, further comprising calculating a transfer position at which each partial transfer pattern group of the overlay transfer mask is transferred to the transfer target from the second group coordinate parameters. The method for transferring a mask pattern according to the above item. 26. The mask according to claim 22, further comprising a step of calculating a transfer position of each partial transfer pattern group transferred to the transfer target from the third group coordinate parameters. Pattern transfer method. 27. The correction of a relative position between each of the alignment marks transferred to the transfer object and each of the alignment marks of the overlay transfer mask is performed using a plurality of partial transfer patterns of the overlay transfer mask as elements. A plurality of partial transfer pattern groups are set by specifying a plurality of partial transfer patterns of the transfer mask for overlay and each alignment mark associated with each of the plurality of partial transfer patterns of the transfer mask for overlay. The position of each alignment mark designated in the transfer mask group setting step and the overlay transfer mask group setting step is measured, and the actual coordinate system of the partial transfer pattern group of the overlay transfer mask is calculated from the measured position. Process, a plurality of portions having a plurality of partial transfer patterns transferred to the object to be transferred as elements A transfer pattern group is set by designating a plurality of partial transfer patterns transferred to the transfer object and each alignment mark associated with each of the plurality of partial transfer patterns transferred to the transfer object. The transfer object group setting step measures the position of each alignment mark specified in the transfer object group setting step, and calculates the actual coordinate system of each partial transfer pattern group transferred to the transfer object from the measured position. Calculating, and the actual coordinate system of the partial transfer pattern group of the overlay transfer mask,
A second group relative coordinate processing step of calculating a fourth group coordinate parameter representing a relative relationship between each partial transfer pattern group transferred to the transfer object and an actual coordinate system; The transfer mask partial transfer pattern group
When sequentially transferring the partial transfer pattern groups on the transfer object, the partial transfer pattern groups of the overlay transfer mask are used for the partial transfer pattern groups transferred to the transfer object using the fourth group coordinate parameter. 23. A relative movement step of the transfer mask for superposition for relatively moving the image of FIG.
Or the mask pattern transfer method according to 24. 28. When all the partial transfer patterns are designated by designating all the alignment marks in the overlay transfer mask group setting step, the overlay mask obtained in the second group relative coordinate processing step is provided. From the position of each alignment mark of the transfer mask, a partial transfer pattern group profile of the transfer mask group for overlay is calculated, which includes a rotation amount of a partial transfer pattern group and a magnification of the partial transfer pattern group of each partial transfer pattern group of the transfer mask for overlay. An image comprising the image rotation amount and the image magnification of the image of the partial transfer pattern group of the transfer mask for superimposition with respect to the partial transfer pattern group transferred to the object using the calculated partial transfer pattern group profile Overlay transfer mask to correct profile 28. The mask pattern transfer method according to claim 23, further comprising a loop image profile correction step. 29. When a part of the partial transfer pattern is designated by designating a part of the alignment mark in the transfer mask group setting step for superimposition, the part obtained by the second group relative coordinate processing step is obtained. From the position of each alignment mark of the overlay transfer mask, the average partial transfer pattern group profile which is the average value of the partial transfer pattern group profiles including the partial transfer pattern group rotation amount and the partial transfer pattern group magnification of each of the partial transfer pattern groups is calculated. An average comprising the image rotation amount and the image magnification of the image of the partial transfer pattern group of the overlay transfer mask with respect to the partial transfer pattern group transferred to the transfer object using the calculated average partial transfer pattern group profile. Overlay transfer mask to correct image profile 28. The mask pattern transfer method according to claim 23, further comprising a screen group image profile correction step. 30. When all partial transfer patterns are designated by designating all alignment marks in the transferred object group setting step, the transferred object obtained in the second group relative coordinate processing step is provided. Calculating the partial transfer pattern group profile including the partial transfer pattern group rotation amount and the partial transfer pattern group magnification of each of the partial transfer pattern groups of the transfer target from the positions of the respective alignment marks, and calculating the calculated partial transfer pattern group. 30. A transfer object image group profile correction step of correcting an image profile including an image rotation amount and an image magnification of an image of a partial transfer pattern group transferred to the transfer object using a profile. 2. The method for transferring a mask pattern according to claim 1. 31. When a part of the partial transfer pattern is designated by designating a part of the alignment mark in the part to be transferred group setting step, the object to be transferred obtained in the second group relative coordinate processing step is designated. The average partial transfer pattern group profile which is the average value of the partial transfer pattern group profile consisting of the partial transfer pattern group rotation amount and the partial transfer pattern group magnification of each partial transfer pattern group of the transferred object from the position of each alignment mark of the transferred object And an average image profile comprising an image rotation amount and an image magnification of an image of a partial transfer pattern group transferred to the object using the calculated average partial transfer pattern group profile. 30. The method according to claim 23, further comprising an image profile correcting step. The method for transferring a mask pattern according to claim 1. 32. The mask pattern transfer method according to claim 1, wherein after the correction of all the relative positions is completed, the desired pattern is obtained on the transfer target by performing transfer. . 33. The mask pattern transfer method according to claim 1, wherein the desired pattern is obtained on the transfer target by performing transfer every time the relative position is corrected. 34. The method according to claim 1, wherein the transfer beam is a charged beam. 35. The mask pattern transfer method according to claim 1, wherein an arc shape is formed in a plane perpendicular to an optical axis of the transfer beam. 36. A beam radiating means for radiating a charged beam; a shaping means for shaping the charged beam into an arc shape; and a plurality of divided and distinguishable partial transfer patterns for transferring to a transfer object. And a movable first transfer mask on which a plurality of alignment marks associated with each of the partial transfer patterns are formed.
Mounting means, and a movable second mounting means on which an object to be transferred on which the transfer pattern including the partial transfer patterns of the transfer mask is transferred is mounted, and which has passed through the transfer mask. Magnification rotation amount correction means for correcting the projection magnification and the image rotation amount of the charged beam to irradiate the transfer object, and measuring the positions of the partial transfer patterns and the alignment marks of the transfer mask. Measuring means; second measuring means for measuring the position of each of the alignment marks of the transferred pattern formed by joining the partial transfer patterns transferred to the transferred object; and Based on output signals from the measuring means and the second measuring means, the moving amounts of the first mounting means and the second mounting means are controlled, and the projection magnification of the magnification rotation amount correcting means is controlled. Mask pattern transfer apparatus and a control means for controlling the beauty image rotation amount. 37. The control means designates each alignment mark to be measured of the transfer mask to the first measurement means, and a position of each alignment mark which is an output from the first measurement means. An actual coordinate system for each of the partial transfer patterns of the transfer mask is calculated, and a first relationship indicating a relative relationship between a design coordinate system and the actual coordinate system for each of the partial transfer patterns of the transfer mask is calculated.
When sequentially transferring the partial transfer pattern to the transfer target, the relative movement amount of the image related to the partial transfer pattern with respect to the transfer target based on the first parameter is calculated by the first parameter. 37. The mask pattern transfer device according to claim 36, wherein the mask pattern transfer device outputs an image to the mounting means and the second mounting means. 38. The controller according to claim 1, further comprising: a controller configured to determine, based on a position of each alignment mark, which is an output from the first measuring unit, a partial transfer pattern rotation amount and a partial transfer pattern magnification for each of the partial transfer patterns. Calculating a partial transfer pattern profile, and using the calculated first partial transfer pattern profile to convert the image profile consisting of the projection magnification and the image rotation amount of the image of the partial transfer pattern onto the object to be transferred to the magnification rotation. 38. The mask pattern transfer device according to claim 36, wherein the mask pattern transfer device outputs the data to an amount correction unit. 39. The control means designates each alignment mark to be measured on the transferred object to the second measurement means, and outputs each alignment mark which is an output from the second measurement means. Calculating an actual coordinate system of the transferred pattern of the transferred object from the position, and calculating a relative relationship between an actual coordinate system of each of the partial transfer patterns of the transfer mask and the actual coordinate system of the transferred pattern. Calculating a second parameter to represent, when sequentially transferring the partial transfer pattern to the transfer target, the relative movement amount of the image related to the partial transfer pattern with respect to the transfer target based on the second parameter; The mask pattern transfer device according to any one of claims 36 to 38, wherein an output is provided to the first placement unit and the second placement unit. 40. A device manufacturing method for manufacturing a device by a process including the mask pattern transfer method according to claim 1. Description: 41. A device manufacturing method for manufacturing a device by using the mask pattern transfer device according to claim 36. 42. A plurality of divided and distinguishable partial transfer patterns, wherein each of the partial transfer patterns is sequentially transferred to the object to be transferred, whereby each of the partial transfer patterns is transferred to the object to be transferred. A transfer mask for obtaining a joined desired pattern, wherein a plurality of alignment marks associated with each of the partial transfer patterns are provided.