JPH0368127A - Exposure device - Google Patents

Abstract

PURPOSE:To enhance the overlapping alignment precision and the throughput in the case of making alignment of a mask with a wafer by a method wherein the wafer positioning is made by the feed corresponding to any exposure position. CONSTITUTION:The alignment measurement in the exposure area for several shots represented on a wafer 3 is taken on the wafer 3 loaded on a chuck. Any residual shots are interpolated and estimated by the alignment measurement data collected from the shots for the peripheral measurement. Next, the driven amounts in respective axes of a wafer stage 24 for making alignment of the shot on the wafer 3 to be exposed with a pattern drawn on a mask 2 as well as a mask stage 4 are computed. Later, the wafer whose alignment is made by blind feeding is exposed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an exposure apparatus that tightly controls a mask chuck or a wafer chuck having a drive shaft.

[Prior Art] Conventionally, in exposure apparatuses, two methods have been mainly used for alignment sequences between a mask and a wafer: a die-by-die alignment method and a global alignment method. The die-by-die alignment method is a method in which the mask and wafer are aligned for each shot and then exposed. The global alignment method measures the alignment of the mask and wafer at several representative locations on a single wafer, calculates the optimal position for other exposure shots based on this data, and
After that, all shots are exposed at once without alignment measurement.

[Problems to be Solved by the Invention] However, although the die-by-die alignment method has good overlay accuracy for each shot, it has drawbacks such as reduced throughput and cumulative chip rotation. Furthermore, although the global alignment method has better throughput than the die-by-die alignment method, it has the disadvantage that the overlay accuracy for each shot is worse than the die-by-die alignment method. Therefore, the current situation is that these methods are used depending on the situation.

Furthermore, when exposing a square using the global alignment method, the sequence is shown in FIG. 5, and in step 504 of the sequence, the alignment state between the wafer stage and the mask stage is measured However, in this case, if the coordinate axes of the mask stage and the wafer stage do not match, an overlay error occurs between the mask and the wafer. Conventionally, this error has not been considered much of a problem because the amount of drive is small, and no countermeasures have been taken against it. However, with the recent demand for finer circuit patterns, these errors can no longer be ignored.

In view of the problems of the prior art, an object of the present invention is to
An object of the present invention is to improve both the overlay accuracy and throughput during alignment between a mask and a wafer in an exposure apparatus.

[Means for Solving the Problems] In order to achieve the above object, the present invention includes means for holding and positioning a mask, and means for holding and positioning a wafer, and sequentially positions the mask and the wafer, In an exposure apparatus that sequentially projects and exposes a pattern formed on a mask onto multiple exposure positions on a wafer, it is equipped with a data table obtained by measuring the positioning error seen from the wafer side as the mask is positioned. When positioning the wafer, this data table is referred to for positioning the wafer.

Positioning the mask typically involves driving in a rotational direction,
Positioning errors seen from the wafer side are usually caused by driving in this rotational direction.

The positioning error seen from the wafer side is measured, for example, by measuring and processing the output of a light receiving element fixed to the wafer that receives light emitted from a light emitting element whose position is fixed with respect to the mask.

[Function] In this configuration, during exposure, the mask and wafer are positioned by global alignment, not by die-by-die alignment, in order to improve throughput. Therefore, positioning at other exposure positions, that is, blind feeding, is performed based on the alignment results for some exposure positions, but in this case, corrections such as rotating the mask are usually required depending on the exposure position. shall be. As a result, the X and Y positions of the mask as seen from the wafer side also change, causing errors. However, since the wafer is positioned by the feed amount according to the exposure position in consideration of the data table obtained by measuring the error in advance, the error is corrected and accurate alignment is performed.

[Example] Hereinafter, the present invention will be explained in detail using the drawings.

FIG. 1 shows the configuration of a mask θ stage portion in an exposure apparatus according to an embodiment of the present invention. stage base 106
On the other hand, a radial leaf spring 10 supports the mask 2 flexibly only in the in-plane rotation direction (θ direction) and rigidly supports it in other directions.
7 is a mask chuck for attaching and detaching the mask 2;
103 and 104 are a positioning V block and a positioning pin for mechanically positioning the mask 2, and 102 is a displacement for measuring the relative rotation angle of the mask 2 with respect to the mask θ stage base 106 approximately as a linear displacement in the circumferential direction. A sensor, 112 is a piezo element which is a driving displacement source of the mask θ stage 4, 113 is a piezo element 112
IOT is an elastic coupling that transmits the displacement expanded by the lever expansion mechanism 113 in a direction other than the circumferential direction as much as possible, and 111 is a mask θ.
This is a damper that dampens the vibrations of the stage 4.

A laser diode 108 is fixed to the mask θ stage 4, and its optical axis is projected toward the wafer stage side perpendicularly to the mask 2 surface.

FIG. 2 shows the overall configuration of the exposure apparatus in this embodiment.

In the same figure, numeral 3 indicates a wafer which is chucked into a wafer chuck (not shown) on a wafer stage 24 . 5
is a coarse θ stage, which is rotatable in the θ direction (around the Z axis) and is held on a two-tilt stage 6. The Z tilt stage 6 is movable in the Z direction and is held on the full-theta stage 7. The θ stage 7 is rotatable in the θ direction and is held on the ltx stage 8. The fine X stage 8 is movable in the X-axis direction and is held on the W&Y stage 9. The fine Y stage 9 is movable in the Y-axis direction and is held on the coarse X stage 10.

The coarse X stage 10 is movable in the X direction, and the coarse Y stage 1
It is held above 1. The rough Y stage 11 is movable in the Y direction and is held by a frame 15.

The position X, Y and attitude θ, ω×ω7 of the stage 24 are the laser interferometer beam 17 projected from the fixed position of the device.
is reflected by a mirror 16 fixed on a second tilt stage 6, and its length is measured by an interferometer.

2 is a mask, which is chucked on a mask chuck (not shown) on a mask θ stage 4. The mask θ stage 4 is rotatable within a plane normal to the exposure light 1, and is held by a frame C.

Reference numeral 12 denotes a plurality of detection optical needles (hereinafter referred to as pickups), which measure the gap between the mask and the wafer and the misalignment between the mask and the wafer.

In particular, this embodiment is equipped with four pickups. Each pickup 12 is held on a pickup stage 13 that is movable in the X and Y directions.

Reference numeral 23 indicates a light beam emitted from the pickup 12. Since the projected light beam is projected obliquely, the pickup does not block the exposure light 1 during exposure, so there is no need to retreat the pickup 12 during exposure.

A transfer pattern formed on a mask 2 by exposure light 1 is printed onto a transferred pattern on a wafer on which up to the previous layer has been transferred. A gap sensor 27 is installed on the Z tilt stage 6 and relatively measures the lower surface of the wafer chuck (not shown).

14 is a light receiving element installed on the 2-tilt stage 6;
Reference numeral 6 denotes a light projecting element fixed on the frame, and is located at a position facing the wafer stage 24.

The light receiving element 14 can detect a projection beam 23 for alignment detection projected from the pickup 12 and a projection beam projected from the light projection element 26 . Regarding the projected beam received by the light receiving element 14, the center position of the beam is calculated, and by taking into account the position data of the wafer stage 24, it is possible to accurately measure the position of the beam center.

FIG. 8 shows the structure of the light-receiving element 14. As shown in the figure, the light-receiving element 14 has an electrode 80 on a flat silicon element 802.
1a to 801d, and the electrode 801a
- 801d are enclosed in one package 805 including one end of leads 803a to 803d. The coordinate values (px, py) of the beam spot projected onto the silicon element 802 as seen from the element midpoint can be determined from the following relationship.

However, here, Xi, Yt, X2, and Y2 are the center electrode 804 of the light receiving element and each electrode 801a to 801d, respectively.
A photocurrent flows between them, and L indicates the distance from the center electrode 804 to each of the electrodes 801a to 801d.

The position of the beam spot in the stage coordinate system is determined from the values read from the light receiving element 14 (px, py), the measured value of the stage coordinates on the laser interferometer axis, and the arrangement (fixed value) of the light receiving element 14 on the wafer stage. You can ask for it.

FIG. 3 is a block diagram showing a control section of the exposure apparatus. As shown in the figure, in the exposure apparatus according to this embodiment, all functions are managed under a main processor 301 located at the top. The main processor 301 is connected to the communication path 30
2 and communication I/F 303, and in Figure 3, the hardware units including the alignment function, stage control function, and pickup stage control function are extracted and shown. be. This hardware unit is herein referred to as a main body control block.

Regarding the alignment system, the main body control block includes a pickup stage control section 30 that controls the positioning of the four pickups 12 (see FIG. 2) in two axis directions (α axis, β axis).
5. Fine AA and AF control unit 309a for aligning and aligning the wafer 3 and mask 2 on a plane
~309d.

Wafer X, Y coarse movement stage 10.11 (see Figure 2),
A stage control section 313 for controlling the positioning of the wafer movement stages 6 to 9 is connected.

The main body control unit 304 stores a program for executing a predetermined sequence, and is a control section that operates the above-mentioned control sections according to the sequence. In addition, the main body control unit 304 connects to the upper main processor unit 301 and the communication path 302,
It is connected via a communication I/F 303 and sends and receives data.

Furthermore, among the above-mentioned control units, the fine AA.

AF control units 309a to 309d and stage control unit 3
13 are respectively communication 17F306, 308°310.312
Data is exchanged with the main body control unit 304 through communication paths 307 and 311.

The main body control unit 304 uses these communication channels.
Each control unit connected to the controller has a CPU mounted therein, and all hardware signal processing is completed within this CPU. This configuration allows each control section to operate independently and to be hierarchical.

FIG. 4 is a detailed block diagram of the stage control section 313 in FIG. 3. This unit is connected to the main body control unit 304 through communication I/F 310° 312 and communication path 31.
Information is exchanged via 1. The wafer stage control unit 401 has a CPU (not shown) inside, and calculates the drive amount of each actuator mounted inside the wafer stage with respect to the target position of the stage sent from the main body control unit 304. . In addition, there is also a light receiving sensor (P) on the wafer stage.
It also has a processing function that converts data sent from the PSD length measurement system control unit 409, which receives the signal output from the SD) 14, into appropriate position data.

Reference numeral 402 denotes a laser interferometric measurement length unit, which performs position measurement on the S axis excluding the axes (two axes) normal to the surface of the sea urchin. 403 is a minute displacement sensor control unit, which drives the minute displacement sensor 27 and performs signal processing to control the second wafer stage.
Measure the amount of shaft displacement. 404 is an X, Y coarse movement controller, which controls the wafer X, Y coarse movement stage 1 in FIG.
0.11 electric cylinder control. 405 is a θ coarse motion controller. The θ coarse movement axes are the X and Y drive axes of stages 8 to 11, and the Z of stage 6, as shown in FIG.

ω8. It is mounted on the ω, drive axis and the θ fine movement axis of the stage 7, and the position change of the wafer stage 24 caused by moving the θ coarse movement axis of the stage 5 is caused by a mirror for reflecting the optical axis 17 from the interferometer. It is not reflected in the movement of 16. Therefore, in this embodiment, by using an inchworm actuator as the actuator for the θ coarse movement axis, the θ
Eliminates positional fluctuations after rough shaft alignment. 406 is operated by three piezo actuators. This is a tilt stage controller that positions ωX and ωY, and controls the position of the tilt stage 6 mounted on the X and Y drive axes and the θ fine movement axis based on commands from the wafer stage control unit 401. 407 is a fine movement stage controller, which controls the fine displacement actuators of the X and Y drive axes of stage 8.9, and 408 is a mask θ controller, which controls the mask θ fine movement stage 4 mounted on the mask stage. Let's do it.

FIG. 5 shows an exposure sequence using the global alignment method. The operation of the apparatus will be described with reference to the figure. First, in step 501, the wafer 3 loaded onto the wafer chuck undergoes alignment measurement in an exposure area corresponding to several representative shots on the wafer 3. For example, the shot whose alignment is measured is 6 in Fig. 6.
01. The remaining shots 602 are the shots shown by a to 601e and the alignment measurement was not performed.
is interpolated using alignment measurement data obtained from shots 601a to 601C for which peripheral measurements were performed,
predicted (step 502). In step 503,
A wafer stage 2 for aligning the shot on the urchin 3 to be exposed to the pattern drawn on the mask 2
4 and the drive amount of each axis of the mask stage 4 is calculated.

For example, if the coordinates of the movement target position P are given as absolute position coordinates such as P (x, y, z, θ, ωX, ωY), then the drive target position of each axis at this time (x, y, z, θ, ωX, ωY) e, z□Z2, zs) ■ Tilt stage 21. Regarding the three piezo actuators that drive toward Z2 and Zs, ■For the piezo actuator of the mask stage, eM = LM θ (4)
, ■The target A of the X and Y stages is X-x - ft (L, Zz, Zs) - fM (e M )
(5) Y-+y-gT (Z+, Zz, Zs)
−gy (e M ) (6).

However, AT is a transformation matrix determined from the positional relationship in which the three piezo actuators are arranged, and L4 is the piezo actuator 112 on the mask stage that slightly displaces.
Indicates the distance from the center of the mask to the point of application. The functions fT and gt are the amount of movement caused by the tilt stage height caused by tilting the tilt stage with respect to the X and Y axes, respectively, and fM and gM are the displacements of the mask θ stage 4 newly added as correction terms according to the present invention. This is the amount of movement of the rotation center of the mask stage in the X and Y directions caused by the drive of the mask stage. This amount of movement is not something that was originally intended, but is due to variations in the rigidity of the leaf springs 109 that support the mask stage 4, and the force generated by the piezo actuator 112 at a single point around the mask stage 4. This is caused by the expansion and contraction of a leaf spring placed in the direction of the force acting on the point of application. The functions fM and gM are created when the exposure apparatus in this embodiment is started up, and the creation procedure will be described later.

Next, in step 504, drive commands are applied to the axes of each stage to align the mask and wafer. Step 50
In step 5, the aligned wafer is exposed by blind feeding. Then, in step 506, it is determined whether or not exposure has been completed for all shots. If the exposure has been completed, the exposure for that wafer is terminated, and if not, the next shot is processed.

Figure 6 shows alignment measurement shot 60 on wafer 3 using the global alignment method! a to 601e are shown. Reference numeral 602 denotes a shot (hereinafter referred to as an unmeasured shot) that is blindly fed and exposed to a position predicted from surrounding alignment measurement shots without performing alignment measurement.

FIG. 7 shows the function f mentioned in the explanation of FIG.

It is a flowchart of the gm creation sequence.

To explain with reference to the figure, when the exposure apparatus is started, initialization of the wafer stage and mask stage (steps 701 and 702) are performed in parallel.

Next, in step 703, Ln2O on the mask stage
3 is turned on, and the light beam falls in a direction perpendicular to the mask 2. In step 704, the light receiving element 14 on the wafer stage searches for a spot of the laser beam projected from Ln2O3, and in step 705, the position of the wafer stage 24 when the same spot is captured is paired with the feeding position of the mask stage, and the wafer is stage control unit 40
Store in memory within 1.

In step 706, it is determined whether all the measurement points to be zumbling have been measured, and if all the measurement points have not been measured yet, the mask θ stage 4 is moved step by step and measurement begins at the next measurement point. (Step 707).

When all measurements are completed, the data stored in the wafer stage control unit 401 is transferred to the main body control unit 301, and a process is performed to create a table interpolated with sample points measured each time the mask stage is stepped. . In step 709 , a function (or a conversion table may be used) fw r gM is generated based on the table created in step 708 and transferred to the wafer stage control unit 401 .

[Effects of the Invention] By applying the present invention, the accuracy of exposure using the global alignment method is improved.

[Brief explanation of drawings]

1 is a configuration diagram of a mask chuck and a mask stage according to an embodiment of the present invention, FIG. 342 is a schematic configuration diagram of an exposure apparatus in which the apparatus of FIG. 1 is used, and FIG. FIG. 4 is a block diagram of the electrical hardware configuration of the stage control section in FIG. 3. FIG. Flowchart of the Uchi Yaro tip, Figure 6 is a layout diagram of global alignment measurement shots on the wafer, Figure 7 is the amount of movement fM of the rotation center of the mask stage in the X and Y directions due to the drive of the mask θ stage, A flowchart for determining gM and FIG. 8 are detailed views of the light receiving element in FIG. 2. 2: Mask, 3: Wafer, 4: Mask θ stage, 1
4: Photodetector, 24: Wafer stage, 108: Laser diode (LD), 112: Piezo actuator.

Claims (3)

[Claims] (1) Equipped with a means for holding and positioning a mask and a means for holding and positioning a wafer, the mask and wafer are sequentially positioned, and the pattern formed on the mask is sequentially projected onto multiple exposure positions on the wafer. The exposure equipment that performs exposure is equipped with a data table obtained by measuring the positioning error seen from the wafer side during mask positioning drive, and the wafer is positioned by referring to this data table when positioning between the mask and the wafer. An exposure device characterized by: (2) The exposure apparatus according to claim 1, wherein the positioning of the mask includes driving in the rotational direction, and the positioning error seen from the wafer side is caused by the driving in the rotational direction. (3) It is equipped with a light emitting element whose position is fixed with respect to the mask and a light receiving element whose position is fixed with respect to the wafer and receives the light emitted from the light emitting element, and the measurement of positioning error from the wafer side is possible. 2. The exposure apparatus according to claim 1, wherein the exposure is performed based on the output of the light receiving element.