JPH088175A - Aligner and align method - Google Patents

Abstract

PURPOSE:To provide an aligner and an alignment method which can shorten the time of alignment by correcting the position of the objective coordinate in the next processed area, according to the deviation between at least one coordinate position in the processed area and the objective position. CONSTITUTION:A system controller 16 reads out the coordinate position in design in a shot region where a mark for global alignment from a data part 18, and shifts a wafer stage WST, according to this coordinate position. Then, an FIA system 11 detects the quantity of dislocation in X and Y directions of a mark for global alignment. Next, the system controller 16 reads out the value of the disposition coordinate (objective coordinate position) in design in the first shot region on a wafer W from the data part 18, and also this adds the quantity of dislocation of the mark for global alignment detected before to the position of the objective coordinate as offset so as to correct it.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an aligner for an exposure apparatus for repeatedly exposing a circuit pattern of a mask on a semiconductor wafer (hereinafter referred to as a wafer), and more particularly to a device for rough positioning and precision positioning in an alignment operation. Is.

[0002]

2. Description of the Related Art Recently, an apparatus for repeatedly exposing a circuit pattern of a mask on a semiconductor wafer (hereinafter referred to as a wafer), that is, a so-called step-and-repeat exposure apparatus has been used for manufacturing a semiconductor element. In this exposure apparatus, it is necessary to precisely align the processing target area provided on the semiconductor wafer to be exposed with the transfer position of the optical system of the exposure apparatus. This is because a circuit pattern transferred to a local region of a semiconductor wafer is composed of a plurality of sets, and each of these plurality of patterns is precisely aligned with a region to be processed on a corresponding substrate and repeatedly exposed. This is because.

The position alignment error of the step-and-repeat exposure apparatus is classified into various errors such as offset deviations in the X and Y directions, wafer rotation, pitch deviation, orthogonality, chip rotation, and trapezoid.

Therefore, some proposals have been made to eliminate the alignment error in the alignment of the exposure apparatus. For example,
Rotation error detection means for detecting this rotation error (Δθ),
In order to eliminate the detected rotation error, there has been proposed an exposure apparatus provided with a rotating means for rotating the mask and the wafer relative to each other (JP-A-60-186845).

In such a step-and-repeat type exposure apparatus, it is necessary to accurately align each processing target area with a target coordinate position. Therefore, it has also been proposed to suppress the error of these chip patterns by statistical processing for all the processed areas such as the chip patterns arranged on one wafer.

For example, when aligning a plurality of chip patterns regularly arranged on a substrate to be processed along a designed arrangement coordinate with a predetermined reference position by a step-and-repeat method, Prior to step-and-repeat alignment, the substrate to be processed is moved based on the array coordinate values in the design of the chip pattern,
After actually measuring each position when some of the plurality of chip patterns are aligned with the reference position, the design array coordinate value and the actual array coordinate value to be aligned by the step and repeat method are predetermined. Assuming that there is a unique relationship including the error parameter, the error parameter is determined so that the average deviation between the plurality of actually measured values and the actual array coordinate value is minimized, and this is determined. The actual array coordinate value is calculated based on the error parameter and the designed array coordinate value, and the substrate to be processed according to the calculated actual array coordinate value during step-and-repeat alignment. For positioning (Japanese Patent Laid-Open No. 61-444)
No. 29).

This will be specifically described. FIG. 5 is a process diagram showing each step of a positioning operation of a conventional exposure apparatus. FIG. 6 is an explanatory view showing the arrangement of measurement shots on a wafer in a conventional positioning operation. In the conventional exposure apparatus, as shown in FIG. 5, in steps 401 and 402, the wafer is placed on the wafer stage after rough positioning is completed on the wafer loader.

Then, if necessary, rough positioning is performed again on the wafer stage. After that, a coarse positioning (search alignment) operation for entering a fine positioning (fine alignment) is started. Recently, in order to reduce the mark area for search alignment, one alignment sensor is used to set two Ys at arbitrary positions on the wafer.
Orientation marks are often detected.

In steps 403, 404, 405, and 406, the wafer stage is moved in the designated distance X direction, the positions of Y ₁ and Y ₂ are detected, and the same sensor is used to determine the difference between the detected values of the two marks. The amount of deviation and the amount of rotation of the wafer in the Y direction in the controlled state are obtained. Here, the deviation in the Y direction can be solved by correcting the target position of the wafer stage at the time of step and repeat.

Regarding the rotation error, if it is less than the allowable value, the exposure is performed as it is, but if it exceeds the allowable value, several correction methods can be considered. One is to rotate the wafer itself, which is actually θ in the wafer stage.
It is realized by rotating a rotary table called a table. That is, in steps 407, 408, and 409, after the .theta. Table is rotated, the residual rotation amount is checked again with the same mark, and it is determined whether the rotation amount is within the allowable value or not and it is determined whether or not to rotate again.
It is also possible to use the first two shots of the fine alignment described below to check the residual rotation amount. As another method, it is also possible to rotate the original plate (rectile) side. In this method, it is necessary to provide an accurate measurement mechanism on the reticle stage side.

Next, in step 410, search alignment in the X direction is performed. In this case, since the measurement of the rotation direction has already been completed, one mark for detecting the X direction may be detected.

After the search alignment is completed as described above, the fine positioning mode is entered. Recent trends in fine alignment include EGA (Enhanced Global) in processes 411, 412, 413 and 414.
There is a method called Alignment) that calculates the exposure position of each shot by performing statistical calculation based on the measurement results of a plurality of shots. For example, in the case of an 8-inch wafer, if the exposure area of the exposure apparatus is set to 20 mm × 20 mm, about 70 shots can be exposed. However, by measuring about 10 shots out of all the shots, the measurement time Is being shortened.

After the EGA measurement is completed, in steps 415 and 416, an exposure operation by step and repeat is performed. still,
In addition to the EGA, after the search alignment operation, alignment mark detection is performed for each shot, and the die is moved to the exposure position based on the result and the exposure operation is performed.
There is Die by Die type alignment.

[0014]

In the case of the EGA method as described above, the search alignment performed in steps 403, 404, 405 and 406 requires two shots for search alignment on the wafer (500) as shown in FIG. It becomes (501,502). This shot can be shared with the shot for fine alignment, but in any case, two large search marks capable of search alignment are required on the wafer.

By the way, in the above-mentioned conventional technique,
After performing the two-shot alignment by the search alignment, the operation of the fine alignment was started, but considering the measurement accuracy in the rotation direction, it is better to set the interval between the two shots as wide as possible.

However, from the viewpoint of throughput, the search alignment time cannot be ignored due to the increase in the interval between two shots. For example, the time required to move two shots is about 1 second. However, one second is required for each wafer, and as the number of processed wafers increases, the alignment time becomes longer, which cannot be ignored. Also, in order to improve the positioning accuracy in the rotation direction,
It is necessary to repeat measurement and driving several times, which causes a big problem in the moving time between two shots.

The present invention has been made in view of such conventional problems, and obtains a novel alignment apparatus and alignment method having various features such as reduction of alignment time. The purpose is to

[0018]

In the alignment apparatus according to the present invention as set forth in claim 1, a substrate stage that two-dimensionally moves while holding a substrate on which N target regions are arranged;
On the stationary coordinate system when the detection center is located at a predetermined position on the stationary coordinate system that defines the moving position of the substrate stage, and the detection center and the substrate mark attached to the processing area on the substrate coincide with each other. Position detecting means for detecting the coordinate position at
Alignment with control means for controlling the movement of the substrate stage using the detected coordinate position in order to align each of the N processed regions with a predetermined point in the stationary coordinate system. In the device, of the n (2 ≦ n ≦ N) target regions whose coordinate positions should be detected, m (2 ≦ n)
In order to detect the coordinate position of the m ≦ n) -th processed region on the stationary coordinate system, when moving the substrate stage in accordance with a predetermined target coordinate position of the m-th processed region, The target coordinate position of the m-th processed area is determined according to the deviation between the coordinate position detected by the position detecting means and the target coordinate position in at least one of the (m-1) -th processed area. The control means controls the movement of the substrate stage based on the corrected target coordinate position so that the position detecting means detects the substrate mark of the m-th processed region. To do.

According to a second aspect of the alignment method of the present invention, in the alignment method using the alignment device according to the first aspect, the stationary coordinate system of each of the n processing target regions is used. The coordinate positions on the stationary coordinate system of each of the N processing target areas are calculated by statistically calculating the detected coordinate positions, and the calculated coordinate positions are calculated. It is a method of aligning each of the N processed regions with the predetermined point based on the coordinate position.

In the alignment method according to the present invention as defined in claim 3, as the m-th processed area according to claim 2, the (m-1) -th processed object in which the coordinate position is detected is detected. This is a method in which the processing region closest to the (m-1) th processing region is selected from the (n- (m-1)) processing regions other than the region.

In the alignment method according to the invention described in claim 4, in the alignment method using the alignment device according to claim 1, on the stationary coordinate system for each processing area on the substrate. This is a method of repeating the operation of detecting the coordinate position in and aligning it with a predetermined point.

In the alignment apparatus according to the invention described in claim 5, at least the first processed area out of the n processed areas described in claim 1 is larger than the substrate mark. It has a mark for global alignment of the area.

[0023]

In the present invention, the coordinate position to be detected is n.
Of (2 ≦ n ≦ N) processed regions, m (2 ≦ m ≦
When the movement of the substrate stage is controlled by the control means in accordance with the predetermined target coordinate position of the m-th processing area in order to detect the coordinate position of the n-th processing area on the stationary coordinate system. , The target of the m-th processed area according to the deviation between the coordinate position detected by the position detection means and the target coordinate position in at least one of the (m-1) th processed areas The coordinate position is corrected. Therefore, when the substrate stage is moved to detect the substrate mark of the m-th processing area, the substrate mark can be detected well.

That is, when moving to the target coordinate position of the m-th processed area, the coordinate position detected by the position detection means and the target coordinate position in at least one of the (m-1) th processed areas. By correcting the target coordinate position of the m-th processed area using the deviation from
The deviation between the target coordinate position and the coordinate position of the th th processing region can be minimized, and the substrate mark can be detected well when the substrate stage is moved to detect the substrate mark of the m th processing region. Is what you can do.

As a preferable alignment apparatus in the present invention, for example, during an exposure operation for projecting and transferring a pattern onto each of a plurality of regions to be processed on a substrate, the pattern is moved to the transfer position of the pattern. There is also an aligning device for an exposure apparatus that performs a aligning operation for aligning the position of the processed region.

Specifically, as a substrate stage, a substrate on which N regions to be processed are arrayed is held and moved two-dimensionally, and a plurality of transfer positions at which the original pattern of the mask is projected onto the substrate are transferred. A detection optical system having a stepwise step in each of the regions to be processed and having a detection center at a predetermined position on a stationary coordinate system that defines the moving position of the substrate stage as position detection means, and the detection center. A position detection system including a detection unit that detects a coordinate position on the stationary coordinate system when the substrate mark attached to the processing target region on the substrate matches, and a substrate stage is used as a control unit. X
A drive system is provided to drive the coordinate value, the Y coordinate value, and the rotation (θ) amount value, and the pattern is transferred to the plurality of target regions on the substrate during the exposure operation. There is an aligning device of an exposure apparatus that performs a aligning operation of aligning the position of the processed region stepped to the pattern transfer position.

The relationship between the number N of processed regions arranged on the substrate, the number n for detecting the coordinate position, and the number m for actually detecting the coordinate position is 2 ≦ m ≦ n ≦. It is N. That is, n pieces including the maximum N pieces are selected from N pieces of all the processing target areas, and the coordinate positions are sequentially detected up to the mth.

Specifically, specifically, n out of the N processed regions on the substrate held on the substrate stage
The target coordinate position and the detected coordinate position of each processed region are compared and examined. First, the coordinate position of one processed area is detected. In this detection, for example, the substrate mark attached to this processing area is detected. This can be done with normal search alignment shots.

Next, the coordinate position of the second processed region is detected with reference to the first processed region, and is compared with the target coordinate position obtained from the first processed region. The rotation amount of the substrate is obtained from the coordinate position of the first processed area, the coordinate position of the second processed area, and the target coordinate position.

Therefore, in order to detect the coordinate position of the m-th processed area subsequent thereto, the coordinate position detected by the position detecting means and the target in at least one of the (m-1) th processed areas. It suffices to correct the target coordinate position of this m-th processed region according to the deviation from the coordinate position and move the substrate.

According to this, conventionally, two search alignments were performed to measure the deviation of the X and Y coordinate values and the deviation of the amount of rotation of the substrate, but one search alignment and subsequent ones. Fine alignment allows good error detection, which shortens alignment time.

As a positioning method using the deviation between the specific coordinate position and the target coordinate position, as shown in the present invention of claim 2, n of the N processed regions arranged on the substrate are processed. Coordinate positions on the stationary coordinate system of each of the N processing target areas by statistically calculating the detected coordinate positions on the stationary coordinate system. Is calculated, and each of the N processed regions is aligned with a predetermined point based on the calculated coordinate position, that is, a so-called “enhanced global alignment (EGA) method” is adopted.

That is, the deviations between the coordinate positions of the n target regions selected from the N arranged on the substrate and the target coordinate positions are statistically calculated to calculate the respective N deviations. Then
After that, by aligning each of the N processed regions to a predetermined point based on the calculated coordinate position, the N processed regions are aligned to the respective predetermined points. Therefore, for example, the control means sequentially steps to each of the n processed regions selected from the N processed regions on the substrate, detects their positions, and then based on the detection result. The exposure operation can be performed by estimating the distribution of the N processed regions and sequentially stepping each of the N processed regions on the substrate to the transfer position.

In the case of the "Enhanced Global Alignment (EGA) method" of the present invention, as shown in claim 3, the coordinate position is detected as the m-th processed area (m-1). (M-1) of (n- (m-1)) processed regions other than the first to-be-processed regions
When the processing area closest to the th processing area is selected, the error at the predetermined point is the smallest.

Further, as a positioning method using the deviation between another coordinate position and the target coordinate position, as shown in claim 4, the coordinate position on the stationary coordinate system is set for each processed region on the substrate. There is a so-called “die-by-die method”, which is a method of repeating the operation of detecting and aligning the position with a predetermined point.

That is, each of the N processed regions on the substrate is sequentially stepped, the position of the stepped processed region is detected, and the transfer position of the pattern is determined based on the detection result. After matching the position with the position of the processed region, an exposure operation is performed at the position, the process proceeds to the next processed region, and this is repeated sequentially.

Since at least the first processed region among the n processed regions described in the present invention has a mark for global alignment having a larger area than the substrate mark, It becomes easy to position the processing target area.

[0038]

1 is a diagram showing a schematic structure of a projection exposure apparatus suitable for applying the present invention. In FIG. 1, illumination light (i-line, KrF or ArF excimer laser, etc.) EL from an illumination system (not shown) EL is reticle R mounted on reticle stage RST via condenser lens CL.
The pattern area PA is illuminated with a substantially uniform illuminance. The illumination light EL that has passed through the pattern area PA is incident on the projection optical system PL that is telecentric on at least the image side, and the projection optical system PL forms an image of the circuit pattern formed in the pattern area PA on the wafer W for projection.

The wafer W is mounted on a wafer stage WST which is two-dimensionally movable by a motor 13 in a plane perpendicular to the optical axis AX of the projection optical system PL. The position of wafer stage WST in the X and Y directions is constantly detected by laser interferometer 12 with a resolution of, for example, about 0.01 μm. A movable mirror MR that reflects the laser beam from interferometer 12 is fixed to the end of wafer stage WST.

Further, in FIG. 1, TTR (Through The Retic
le) system alignment system 10 and an off-axis system FIA (Field Image Alignment) system 11 are provided. The TTR alignment system 10 detects an alignment mark (reticle mark) on the reticle R and an alignment mark (wafer mark) attached to a shot area on the wafer W to detect a relative positional deviation amount. Yes, in this embodiment, the pattern area PA is sandwiched between the two areas.
It is arranged in pairs.

The TTR alignment system 10 is an objective optical system O.
With J ₁ , for example, an image of the reticle mark and an image of the wafer mark are formed on the light receiving surface of the image pickup device (CCD camera or the like). The TTR alignment system 10 has X, Y, and rotation (θ) of the reticle mark and the wafer mark.
The amount of positional deviation in the direction is detected and the information is calculated by the arithmetic unit 1
5 is output. The FIA system 11 irradiates the wafer W with illumination light (broadband light) having a predetermined wavelength width, and at the same time, an image of the wafer mark is displayed by the objective optical system OJ ₂ on an index plate arranged in a plane conjugate with the wafer. The image of the wafer mark and the image of the index mark are further formed on the light receiving surface of the image pickup device (CCD camera or the like). FIA system 11
Detects the positional deviation amount between the wafer mark and the index mark in the X and Y directions and outputs the information to the arithmetic unit 15.

It is assumed that the coordinate origin of the orthogonal coordinate system XY defined by the interferometer 12 coincides with the optical axis AX, and the detection center points of the TTR alignment system 10 and the FIA system 11 are the orthogonal coordinate XY. It is set at the predetermined position above,
The shot area on the wafer W is moved to the orthogonal coordinate system X using the upper side from the TTR alignment system 10 or the FIA system 11.
The position is aligned with a predetermined point in Y, that is, the projection device (for example, the coordinate origin) of the pattern of the reticle R.

The arithmetic unit 15 determines the offset given to the target coordinate position of the shot area to be subjected to the next coordinate measurement based on the amount of positional deviation from the TTR alignment system 10 or the FIA system 11, and further the interferometer. The position signal from 12 is also input to obtain the coordinate position of wafer stage WST when the positional deviation amount of both marks becomes zero. The amount of positional deviation from the TTR system 10 or FIA system 11, the offset, and the coordinate position are stored in the storage unit 17.

Further, the arithmetic unit 15 performs EGA calculation (statistical calculation) using the coordinate position previously obtained, and stores the calculation result (calculation parameter, array coordinate value, etc.) in the storage unit 17.
And output to the system controller 16. Ie E
When the GA method is adopted, the coordinate positions of a plurality of shot areas (sample shots, which are three or more, usually about 10 to 15) stored in the storage unit 17 and the shot position data unit 18 are input. Based on the designed array coordinate values of the sample shots, the array coordinate values of all the shot areas on the wafer W are calculated by statistical calculation.

Although not shown in FIG. 1, each piece of information from the TTR alignment system 10 and the FIA system 11 is configured to be input to the arithmetic unit 15 through the changeover switch.
And the above information is calculated by the changeover switch by the arithmetic unit 15
And the system controller 16 can be switched for input. The system controller 16 switches the switch according to the alignment mode (EGA mode or die-by-die (D / D) mode) input from the input device (keyboard, barcode reader, or the like) 19. That is, when the EGA mode is designated, the above information is input to the arithmetic unit 15, and when the D / D mode is designated, the above information is input to the system controller 16. It will be switched.

The shot position data section 18 stores the designed array coordinate values of all shot areas on the wafer W, and these coordinate values are output to the arithmetic unit 15 and the system controller 16. Furthermore, in the data section 18,
Arrangement of sample shots used for EGA calculation (number,
Position) is also entered. System controller 16
Is the wafer stage WST during alignment or during step-and-repeat exposure based on the above various data.
Determine a series of steps to control the movement of the
Overall control of the entire device. Stage controller 14
Drives the motor 13 so that the difference between the coordinate value input from the system controller 16 and the coordinate value from the interferometer 12 becomes zero, and the wafer stage WST is positioned.

When the D / D mode is set in the projection exposure apparatus of FIG. 1, the system controller 16 reads from the data section 18 the designed coordinate position of the shot area provided with the global alignment mark. FI
Wafer stage WST is moved in accordance with the coordinate position so that global alignment mark is detected by A system 11. Then, the FIA system 11 detects the amount of positional deviation of the global alignment mark in the X and Y directions.

The system controller 16 reads the designed array coordinate value (target coordinate position) of the first shot area on the wafer W from the data section 18 and starts the step-and-repeat exposure. The amount of misalignment of the global alignment mark detected at is added as an offset to the target coordinate position and corrected,
Stage controller 14 moves wafer stage WST according to the corrected target coordinate position.

Next, the TTR alignment system 10 detects the amount of positional deviation between the reticle mark and the fine alignment mark (wafer mark) in the first shot area in the X, Y and θ directions. Further system controller 16
Moves the wafer stage WST finely so that the amount of positional deviation from the TTR alignment system 10 becomes substantially zero, accurately aligns the pattern area PA of the reticle R and the first shot area, and then performs overlay exposure. Start. At this time, in order to make the positional deviation amount in the θ direction zero,
The reticle R may be slightly rotated.

Next, the system controller 16 reads out the target coordinate position of the second shot area from the data section 18, and at least the global alignment mark stored in the storage section 17 and the fine alignment mark of the first shot area. The target coordinate position is corrected using one of the positional deviation amounts, and wafer stage WST is moved in accordance with the corrected target coordinate position. Furthermore, TTR
Wafer stage WST is finely moved so that the amount of positional deviation from alignment system 10 becomes zero, and overlay exposure is started.

After that, the above-described operation is repeatedly executed to expose all (N) shot areas on the wafer W in an overlapping manner. In the above-mentioned D / D mode, it is advisable to select the shot area closest to the shot area on which overlay exposure has been performed, that is, the adjacent shot area, and perform overlay exposure. Also, D
In the / D mode, the number of shots n to be coordinate-measured is n = N. In this shot area, the m (2 ≦ m ≦ n) th shot area has 1 to (m−1) th shot areas. At least one of the detected positional deviation amounts may be used. At this time, the averaging processing of a plurality of positional deviation amounts,
One positional deviation amount may be obtained by performing weighted averaging processing or minimum situation approximation processing, and the target coordinate position may be corrected according to the obtained positional deviation amount.

Further, in the above description, the amount of positional deviation from the TTR alignment system 10 is directly applied to the target coordinate position as an offset, but as described above, the position signal from the interferometer 12 is also used to make a shot area for the reticle mark. The coordinate position of the wafer stage WST when the positional deviation amount of the fine alignment mark becomes zero and the next target coordinate position may be corrected according to the deviation between the obtained coordinate position and the target coordinate position. The method is exactly the same.

In the projection exposure apparatus of FIG. 1, when the EGA mode is set, the system controller 16 reads the designed coordinate position of the shot area provided with the global alignment mark from the data section 18, and FI
Wafer stage WST is moved in accordance with the coordinate position so that global alignment mark is detected by A system 11. Then, the FIA system 11 detects the amount of positional deviation of the global alignment mark in the X and Y directions.

Further, the system controller 16 sets a plurality of n (2 ≦ n ≦) of N shot areas on the wafer W.
Each target coordinate position of the sample shot of N)
8, the target coordinate position of the first sample shot is corrected by adding the positional deviation amount as an offset to the stage controller 14 so that the FIA system 11 detects the fine alignment mark of the first sample shot. Then, wafer stage WST is moved according to the corrected target coordinate position. Next, FI
The A system 11 detects the amount of positional deviation in the X, Y and θ directions of the fine alignment mark of the first sample shot with respect to the index mark.

Further, the arithmetic unit 15 also uses the position signal from the interferometer 12 to obtain the coordinate position of the wafer stage WST when the position deviation amount of the fine alignment mark of the first sample shot with respect to the index mark becomes zero. The calculated coordinate position and the position shift amount of the first sample shot are stored in the storage unit 17. Further, the system controller 16 corrects the target coordinate position of the second sample shot by using the positional deviation amount of at least one of the global alignment mark and the fine alignment mark of the first sample shot stored in the storage unit 17. Then, wafer stage WST is moved in accordance with the corrected target coordinate position.

Next, the FIA system 11 is used to detect the fine alignment mark of the second sample shot, and the coordinate position of the wafer stage WST when the position shift amount from the FIA system 11 becomes zero is obtained. Hereinafter, the above-described operation is repeatedly executed to obtain coordinate positions of all (N) sample shots on the wafer W, and the arithmetic unit 15 further statistically calculates a plurality of coordinate positions stored in the storage unit 17 (least squares). The calculation is performed to calculate the coordinate positions of all (N) shot areas on the wafer W.

Further, the system controller 16 sequentially positions the wafer stage WST according to the coordinate position calculated by the calculation unit 15 and exposes each shot area on the wafer W by superimposing and exposing the pattern of the reticle R.
In the above-mentioned EGA mode, when the coordinate positions of a plurality of sample shots are sequentially measured, the sample shot closest to the previous sample shot is selected and the coordinate position is measured. Further, as in the D / D mode, the m-th (2 ≦ m ≦
In the sample shot of n), at least one of the positional deviation amounts detected in each of the 1st to (m-1) th sample shots may be used. At this time, a plurality of positional deviation amounts may be averaged, weighted averaged, or least squares approximated to obtain one positional deviation amount, and the target coordinate position may be corrected according to the obtained positional deviation amount. Good.

Further, the next target coordinate position may be corrected according to the deviation between the coordinate position obtained by the arithmetic unit 15 and the target coordinate position. In addition, when the pre-alignment accuracy is sufficiently high and it is not necessary to detect the global alignment mark, the first shot area (D / D) on the wafer W is selected in both the EGA mode and the D / D mode.
In the mode, the first shot area for overlay exposure,
In the EGA mode, the wafer stage WST is moved according to the target coordinate position of the first sample shot), and at this time, the target coordinate position of the next shot is corrected using the positional shift amount obtained from the TTR alignment system 10 or the FIA system 11. You can do it like this.

The alignment method in the EGA mode of the present invention will be described in more detail. FIG. 2 is a process diagram showing an alignment sequence of an embodiment of the alignment method of the present invention. FIG. 3 is an explanatory view showing positions of search alignment shots designated in advance on the wafer shown in FIG. Figure 2
Each process will be described based on the process chart of FIG.

In step 101, pre-alignment based on the outer shape reference is first performed in the pre-alignment station on the wafer loader. That is, the linear notch (flat) of the wafer is roughly positioned so as to face a certain direction. In the next step 102, wafer W is placed on the wafer holder on wafer stage WST by the load arm. That is, the wafer W is placed on the wafer holder so that the flat of the wafer W is parallel to the X axis, and the wafer W is vacuum sucked and fixed at that position.

In the next step 103, the wafer stage is moved so that the previously designated search alignment shot is located under the alignment sensor based on the design value. At this point, a search alignment mark long in the non-measurement direction is required for the prealignment accuracy. After the stage has moved, in step 104 both Y and X measurements are made in the search alignment shot.

Next, in step 105, the shot is moved to the shot designated as the fine alignment sample shot. In the case of the EGA method as described above, generally, in the wafer
About 10 shots are instructed. In many cases, the EGA shots are arranged in concentric circles in a portion that is slightly inward of the outermost periphery and is not damaged by the process.

In this EGA shot, for example, in the case of the sample shot as shown in FIG.
A shot is selected and fine alignment of X and Y is performed in step 106. By the way, since the search shot 201 and the EGA shot 202 are sufficiently close to each other, if the X and Y measurements have been completed in the search shot 201 in step 104, even if the pre-alignment accuracy in the pre-alignment in step 101 is taken into consideration, the fine alignment can be performed. Alignment is possible even with small marks for use.

Therefore, even if the pre-alignment accuracy is poor and the fine alignment mark is not detected in step 107, after the wafer is again moved in the direction displaced by the residual rotation component in step 108, The fine alignment mark can be detected by repeating the sequence of re-measurement (steps 105 to 108).

Therefore, here, the rotation amount of the wafer can be obtained from the measurement results of the two shots, that is, the search alignment in step 104 and the fine alignment in step 106, and the θ table can be rotated by the amount. Become. Further, regarding the movement to the second fine alignment shot, more accurate measurement can be performed by using both the results of the search shot 201 and the fine alignment shot 202.

That is, in step 109, from the measurement results of the detected coordinate values of X and Y of each of the search shot 201 and the fine alignment shot 202 up to step 108, the X-axis and Y-axis directions of the coordinate system of each shot on the wafer are measured. The offset component and the rotation component are obtained. Therefore, if there is no error in the scaling of the wafer itself or the orthogonality of the array, a unique solution can be obtained.

Therefore, in step 110, the shift amount of the rotation of the wafer is calculated, and when the wafer is moved to the second fine alignment shot 203 in step 111, the deviation of the X coordinate value and the Y coordinate value and the rotation error amount of the wafer. Second, taking into account
The fine alignment shot 203 is moved to.
The success rate of alignment in this case will surely improve.

After the movement, in step 112, the deviation between the detected coordinates of the second fine alignment shot 203 and the target coordinates is measured. Further, in step 113, the process returns to step 109 until the target fine alignment shot is completed (until the EGA measurement is completed), and the rotation error amount is calculated to be included in the movement of the next fine alignment shot. That is, by using the deviation between the coordinate position detected in the measured fine alignment shot and the target coordinate position, it is moved so as to minimize the deviation between the target coordinate position and the coordinate position of the next fine alignment shot. This makes it possible to increase the accuracy in the same method up to the final shot of fine alignment.

When the fine alignment is completed up to the nth shot as described above, as a result, all the parameters of X, Y offset, X, Y scaling, orthogonality, and rotation of the shot on the wafer are obtained. From the result, in the case of the exposure apparatus having means for moving the reticle at high speed and with high accuracy, in step 114,
The reticle is rotated according to the residual rotation amount of the wafer. As a result, the effect of reducing the residual rotation for each shot in the baking result can be obtained.

The exposure preparation is completed as described above, and in step 115 and step 116, accurate overlay exposure is executed by the step-and-repeat operation performed subsequently.

In this embodiment, the fine alignment subsequent to the search alignment and the fine alignment uses the adjacent search shots to ensure the detection of the alignment mark corresponding to each shot. However, the search shots are not limited to adjacent search shots.

Further, in the present embodiment, the reticle is rotated according to the residual rotation amount of the wafer in step 114 after completion of all EGA fine alignment shots. For example, the rotation for each fine alignment shot on the wafer is measured. In that case, it may be performed after each fine alignment shot.

Further, in the present embodiment, when the subsequent fine alignment shot is moved, the deviation between the target coordinate position and the detected coordinate position of the alignment shot up to the present condition is taken into consideration. The deviation between the target coordinate position of one shot and the detected coordinate position may be taken into consideration.

FIG. 4 is an explanatory view showing the position of another search alignment shot previously designated on the wafer shown in FIG. In FIG. 3 described above, the shot 201 for search alignment and the shot 202 for fine alignment are shown.
However, as shown in FIG. 4, search alignment and fine alignment can be performed in one shot 301, and the mark area can be reduced.

Depending on the pre-alignment accuracy, it is possible to omit the search alignment sequence and directly enter the fine alignment sequence after wafer replacement. Even in that case, it is possible to repeat the EGA calculation and sequentially increase the alignment success rate in accordance with the number of measurement shots as in the above sequence.

As described above, according to the present invention, sufficient overlay accuracy can be obtained even with one search shot.
It is possible to obtain higher throughput than the conventional search alignment method using two shots.

Further, since the search alignment is performed in one shot or less, it is sufficient that the number of marks for the search alignment is one or less in the wafer, and the exposure area of the exposure apparatus can be effectively used. Since the number of shots that can be exposed on one wafer is increased, productivity is increased.

In this embodiment, the EGA alignment is shown. However, for each fine alignment shot in the embodiment, the fine adjustment of the deviations of Δx, Δy and θ is carried out to perform fine alignment and exposure D / You may perform D type exposure.

Further, the alignment apparatus and alignment method of the present invention may be any one as long as it aligns the region to be processed with the target position by the step-and-repeat method, and various lithographic apparatuses other than the reduction projection exposure apparatus. Also, the invention can be applied to a device for inspecting a photomask, a laser repair device, and the like.

[0080]

As described above, according to the present invention, of n (2 ≦ n ≦ N) target regions whose coordinate positions are to be detected,
The movement of the substrate stage is controlled in accordance with a predetermined target coordinate position of the m-th processed area in order to detect the coordinate position of the m-th processed area on the stationary coordinate system. When controlling by the means, the correction means (m-
1) The target coordinate position of the m-th processed area is corrected according to the deviation between the coordinate position detected by the position detection means and the target coordinate position in at least one of the first to-be-processed areas. . Therefore, when the substrate stage is moved to detect the substrate mark of the m-th processing area, the substrate mark can be detected well. Also,
Conventionally, two search alignments are performed and X, Y
Although the deviation of the coordinate value and the deviation of the rotation amount of the substrate have been measured, the error can be satisfactorily detected by one search alignment and the subsequent fine alignment, so that the alignment time is shortened.

Further, among the (n- (m-1)) processed areas other than the (m-1) -th processed area where the coordinate position is detected, as the m-th processed area, m-1)
When the processing area closest to the th processing area is selected, the error at the predetermined point is the smallest, so that the success rate of measurement is improved.

[Brief description of drawings]

FIG. 1 is a diagram showing a schematic configuration of a projection exposure apparatus suitable for applying the present invention.

FIG. 2 is a process drawing showing an alignment sequence of an embodiment of the alignment method of the present invention.

FIG. 3 is an explanatory diagram showing positions of search alignment shots designated in advance on the wafer shown in FIG.

FIG. 4 is an explanatory diagram showing positions of other search alignment shots designated in advance on the wafer shown in FIG.

FIG. 5 is a process diagram showing each step of a positioning operation of a conventional exposure apparatus.

FIG. 6 is an explanatory diagram showing the arrangement of measurement shots on a wafer in a conventional positioning operation.

[Explanation of symbols]

EL ... Illumination light, CL ... Condenser lens, RST ... Reticle stage, R ... Reticle, PA ... Pattern area, PL ... Projection optical system, W ... Wafer, AX ... Optical axis, WST ... Wafer stage, MR ... Moving mirror, OJ ₁ ... Objective optical system, 10 ... TTR (Through The Reticle) type alignment system, 11 ... Off-axis type FIA (Field Image Alignme)
nt) system, 12 ... Laser interferometer, 13 ... Motor, 14 ... Stage controller, 15 ... Arithmetic unit, 16 ... System controller, 17 ... Storage unit, 18 ... Shot position data unit, 101-116 ... Each process, 201, 301 ... Search shots, 202, 203 ... Fine shots

Claims (5)

[Claims] 1. A substrate stage that two-dimensionally moves while holding a substrate on which N target regions are arrayed; a detection center is provided at a predetermined position on a stationary coordinate system that defines the moving position of the substrate stage. Position detecting means for detecting a coordinate position on the stationary coordinate system when the detection center coincides with a substrate mark attached to a region to be processed on the substrate; A coordinate device for controlling the movement of the substrate stage by using the detected coordinate position in order to align each of the N processed regions, the coordinate position should be detected. In order to detect the coordinate position of the m (2 ≦ m ≦ n) -th processed area on the stationary coordinate system among the n (2 ≦ n ≦ N) -processed areas, the m-th processed area is detected. According to the predetermined target coordinate position of the processing area When the substrate stage is moved, in accordance with the deviation between the coordinate position detected by the position detecting means and the target coordinate position in at least one of the (m-1) th processed regions, the mth processed region is moved. The control means includes a correcting means for correcting the target coordinate position of the processing area, and the control means is based on the corrected target coordinate position so that the position detecting means detects the substrate mark of the m-th processed area. A positioning device for controlling the movement of the substrate stage. 2. The alignment method using the alignment device according to claim 1, wherein the coordinate position of each of the n processing regions on the stationary coordinate system is detected, and the detected coordinate position is detected. A coordinate position on the stationary coordinate system of each of the N processed regions is calculated by statistically calculating a plurality of coordinate positions, and each of the N processed regions is calculated based on the calculated coordinate position. Is aligned with the predetermined point. 3. The (n- (m-1)) processed regions other than the (m-1) -th processed region where the coordinate position is detected as the m-th processed region. , (M-
The positioning method according to claim 2, wherein the processing area closest to the 1) th processing area is selected. 4. The alignment method using the alignment apparatus according to claim 1, wherein a coordinate position on the stationary coordinate system is detected for each processed region on the substrate and aligned with a predetermined point. A positioning method characterized by repeating operations. 5. The at least first processed region among the n processed regions has a mark for global alignment having an area larger than that of the substrate mark. Alignment device.