JP2000040658A - Stage control method and scanning aligner - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress the excitation of high-frequency oscillation of stage due to acceleration, by a method wherein a stage is continuously accelerated while increasing the time variation rate from the second value to almost zero from the fifth time to the sixth time for almost zeroing the acceleration of the stage in the sixth time. SOLUTION: For the stage control in Y direction, as for the functions of the time t, following factors are to be taken into consideration, i.e., Y position Y (t), Y directional velocity VY(t), Y directional acceleration α (t) and Y directional jerk Jy (t) from the acceleration starting time ts (=t10) as the first time to the acceleration finishing time tE (=t14) as the sixth time. Firstly, during the acceleration starting time t10 to the acceleration finishing time t14, the term from the time t13 to the time t14 is set up as the increasing term of the jerk Jy to zero. Besides, jerk JY (t) in respective terms is set up as the linear functions JLY (t) of the time t.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a stage control method and a scanning type exposure apparatus, and more particularly, to scanning for transferring an image of a pattern of a mask onto a sensitive substrate while synchronously moving the mask and the sensitive substrate. The present invention relates to a mold exposure apparatus and a stage control method applied to the apparatus.

[0002]

2. Description of the Related Art When a semiconductor device or a liquid crystal display device is manufactured by a lithography process, a pattern of a photomask or a reticle (hereinafter collectively referred to as a "reticle") is exposed to light through a projection optical system under exposure light. For example, a projection exposure apparatus that projects onto a wafer is used. Conventionally, as such an apparatus, a step-and-repeat type reduction projection type exposure apparatus (so-called stepper) has been mainly used. However, with the miniaturization of patterns due to the high integration of semiconductor elements, a larger area and higher A step-and-scan type scanning exposure apparatus capable of performing accurate exposure is becoming mainstream.

In this scanning type exposure apparatus, before the exposure, it is checked whether the reticle stage for holding the reticle and the wafer stage for holding the wafer are in a correct positional relationship with each other, and then these stages are controlled by speed control. Mode, and a scan operation for exposure was performed. In this speed control mode, both stages are accelerated at the respective target positions so as to have the respective target speeds, and then the reticle stage and the wafer stage are moved in opposite directions at a speed ratio according to the projection magnification of the projection optical system. The pattern on the reticle in the area illuminated by the illumination light is sequentially transferred onto the wafer via the projection optical system.

In the acceleration, the stage is accelerated by changing the time waveform of the acceleration to a polygonal or parabolic shape in response to a request for achieving the target speed at the target position and a subsequent uniform movement during the scanning exposure. A technique for controlling the movement of the stage has been used.

[0005]

It is known that when the stage is accelerated by the stage movement control as described above, the stage vibrates at the end of the acceleration. Therefore, when the acceleration of the stage is completed, the stage is in a vibrating state,
It does not move stably at a constant speed in the scanning direction.

Although the vibration waveform of such a stage contains various frequency components, if a frequency component higher than the response frequency of the control system of the stage is present, the stage is not accelerated in order to attenuate these frequency components. It is necessary to wait for the amplitude to decrease with time. Therefore, in the scanning type exposure apparatus, prior to the start of the scanning exposure, a stage settling time is provided after acceleration, and the settling time elapses.
After the amplitude of the vibration of the stage becomes equal to or less than the allowable value, the scanning exposure is started.

On the other hand, in the scanning type exposure apparatus, the scanning speed is improved and the acceleration time is shortened in order to improve the throughput. However, when the acceleration is performed at a high acceleration, the amplitude of the high frequency vibration tends to increase. Therefore, it is necessary to provide a long settling time from the end of the stage acceleration to the start of the scanning exposure. This has been an obstacle to improving the throughput of the apparatus.

The present invention has been made under such circumstances, and a first object of the present invention is to provide a stage control method capable of suppressing high-frequency vibration excitation of a stage due to acceleration.

A second object of the present invention is to reduce the stage settling time by suppressing high-frequency vibration excitation of the stage due to acceleration, thereby improving the throughput and performing high-precision exposure. Is to provide.

[0010]

According to a first aspect of the present invention, there is provided a stage control method for controlling acceleration of a stage in a predetermined direction, wherein the stage is controlled from a first time to a second time. A first step of accelerating the stage while continuously increasing the time rate of change of acceleration in a predetermined direction from substantially 0 to a first value; and the time rate of change from the second time to a third time. A second step of accelerating the stage while continuously decreasing the first value from the first value to substantially zero; and a time period from a fourth time which is a time after the third time to a fifth time. A third step of accelerating the stage while continuously decreasing the rate of change from substantially zero to a second value.
From the fifth time to the sixth time, continuously accelerating the stage while increasing the time rate of change from the second value to substantially zero, and adjusting the stage at the sixth time. And a fourth step of making the acceleration substantially zero.

According to the findings obtained by the present inventors from the results of the research, the magnitude of the high-frequency vibration generated on the stage by acceleration is determined by the time change rate of the stage acceleration (hereinafter also referred to as “jerk”). Becomes larger as the high frequency component becomes larger.

In the stage control method according to the first aspect, in consideration of this, the acceleration of the stage is continuously changed from the first time to the sixth time, which is the acceleration period of the stage. I have. Then, at a sixth time point when the acceleration of the stage ends, the acceleration is almost zero.
And the time rate of change of acceleration, that is, jerk, also becomes zero. Therefore, the high-frequency component in the time change of the jerk is reduced, and as a result, the occurrence of the high-frequency vibration of the stage is suppressed, so that the decay time of the vibration provided from the end of the acceleration of the stage to the start of the scanning exposure, that is, the settling time of the stage is reduced. Can be set shorter.

[0013] In the stage control method according to the first aspect, there are various possible forms of the time change of the jerk (the time change rate) from the fourth time to the sixth time.
It is preferable that the time change of the time change rate from the fourth time to the sixth time be a smooth change that can be differentiated using time as a variable. In such a case, since the time change of the jerk during the acceleration period immediately before the end of the stage acceleration is a smooth change that can be differentiated, the high-frequency component related to the time change of the jerk is reduced. Therefore, the high-frequency vibration of the stage at the end of the acceleration of the stage is suppressed, so that the settling time of the stage can be shortened.

According to a second aspect of the present invention, as in the third aspect, the time change of the jerk from the first time to the third time can be differentiated using time as a variable. It is preferable that the change be made as follows. In such a case, the time change of the jerk is continuous during the entire acceleration period, and the time change of the jerk becomes a differentiable smooth change in each of the positive period and the negative period of the jerk. Therefore, the high-frequency component related to the time change of the jerk is further reduced. Therefore, the high-frequency vibration of the stage at the end of the acceleration of the stage is further suppressed, so that the settling time of the stage can be further shortened.

The stage control method according to claim 2 or 3, wherein the jerk time change from the fourth time to the sixth time or the jerk time change from the first time to the third time. Although various aspects are conceivable,
As in the invention according to claim 4, the time change of the time change rate (jerk) can be a quartic function using time as a variable.

When the period of interest (ie, the period from the fourth time to the sixth time or the period from the first time to the third time) is represented by a multi-order function using time as a variable, the function satisfies The power boundary condition generally includes a jerk value, an acceleration value, a velocity value, and a position at the start time of the period (ie, the fourth time or the first time), and the end time of the period (ie, the sixth time). Time or third time)
, Eight conditions of jerk value, acceleration value, speed value, and position.

Therefore, according to the stage control method of the present invention, the time change of jerk is a quartic function of time, that is, the time change of the position is a cubic function of time. We can always be satisfied.

In the stage control method of the first aspect, the third time and the fourth time may be the same time,
As in the invention according to claim 5, the third time and the fourth time are different from each other, and the time change rate from the third time to the fourth time is substantially equal to the third time. 0
It can be assumed that In such a case, the maximum acceleration during the stage acceleration period is set to the fourth acceleration from the third time.
Since the stage is accelerated until the time, the maximum acceleration is reduced when accelerating to the target speed in a certain period, that is, when the stage is accelerated, as compared with the case where the third time and the fourth time match. This is because the maximum value of the applied force can be reduced.

According to a sixth aspect of the present invention, in the stage control method according to the first aspect, the time from the first time to the third time is equal to the time from the fourth time to the sixth time. It is characterized by being shorter than the time.

According to this, it is possible to sufficiently secure a period during which the acceleration decreases before the end of the acceleration, that is, a time from the fourth time to the sixth time when the jerk has a negative value. Therefore, the time change of the jerk immediately before the end of the acceleration can be made gradual, so that the occurrence of the high frequency vibration generated in the stage can be suppressed, and the settling time of the stage can be shortened.

According to a seventh aspect of the present invention, there is provided a scanning exposure apparatus for transferring a pattern formed on the mask onto the sensitive substrate while synchronously moving the mask and the sensitive substrate. A substrate stage for holding the sensitive substrate; and transferring a pattern formed on the mask onto the sensitive substrate.
A control system for accelerating the mask stage and the substrate stage to target speeds in the respective synchronous movement directions by any one of the stage control methods of (1) to (6).

According to this, prior to the scanning exposure performed to transfer the pattern formed on the mask onto the sensitive substrate, the mask stage is controlled by the control system by the stage control method according to any one of claims 1 to 6. Then, the substrate stage is accelerated to the target speed in the respective synchronous movement directions.
Therefore, high-frequency vibration of the mask stage and the substrate stage at the end of the acceleration is suppressed, and the settling time after the end of the acceleration can be shortened. As a result, high-precision exposure can be performed while improving the throughput.

[0023]

DESCRIPTION OF THE PREFERRED EMBODIMENTS First Embodiment A first embodiment of the present invention will be described below with reference to FIGS.

FIG. 1 shows a schematic configuration of a scanning exposure apparatus 10 according to the first embodiment. The scanning type exposure apparatus 10 includes an illumination system 12 that illuminates a reticle R as a mask with exposure light IL, and a reticle that scans the reticle R in the Y-axis direction (the left-right direction in FIG. 1) and minutely drives the reticle R in an XY plane. Side stage system 14 and projection optical system PL arranged below reticle side stage system 14.
A wafer-side stage system 16 disposed below the projection optical system PL for two-dimensionally moving a wafer W as a sensitive substrate in an XY plane, and a microcomputer (or workstation) for controlling the entire apparatus as a whole. And a main control system 18.

The illumination system 12 includes a light source system 20, a mirror 2
2. It includes a reticle blind 24 as a field stop, a relay lens 26, a mirror 28, a condenser lens 30, and the like. The light source system 20 includes a light source such as an ultra-high pressure mercury lamp or a laser light source, an optical integrator, and the like. The reticle blind 24 is disposed at a position conjugate with the pattern forming surface on the lower surface of the reticle R.

The exposure light IL emitted from the light source system 20 is:
Mirror 22, reticle blind 24, relay lens 2
6. Illuminate the slit-shaped illumination area defined by the reticle blind 24 on the reticle R with uniform illuminance via the mirror 28 and the condenser lens 30. In this case, it is assumed that the longitudinal direction of the slit-shaped illumination area is set in the X direction (the direction orthogonal to the paper surface in FIG. 1), and the direction of relative scanning between the reticle R and the slit-shaped illumination area is the Y direction.

The reticle-side stage system 14 is arranged in parallel with the XY plane, and has a reticle-side base 32 on which a linear guide (not shown) extends in the Y-axis direction. Reticle coarse movement stage 3 as mask stage that moves along the linear guide
4 and a reticle fine movement stage 36 that is mounted on the reticle coarse movement stage 34 and holds the reticle R and performs fine movement (including rotation) in the XY plane.

A movable mirror 38 is provided on the reticle fine movement stage 36, and a reticle laser for projecting a laser beam to the movable mirror 38 and detecting the position of the reticle fine movement stage 36 by receiving the reflected light. An interferometer 40 is provided facing the movable mirror 38. Here, actually, the movable mirror is provided with a total of three movable mirrors: an X movable mirror having a reflecting surface orthogonal to the X-axis direction and two Y movable mirrors having a reflecting surface orthogonal to the Y-axis direction. Correspondingly, three reticle laser interferometers, ie, an X-axis position measurement interferometer and two Y-axis position measurement interferometers, are provided. In FIG. Mirror 3
8, shown as reticle laser interferometer 40.

The outputs of these three reticle laser interferometers are supplied to a main control system 18. The main control system 18 determines the X position of the reticle fine movement stage 36 based on the output of the X-axis direction position measurement interferometer. And the Y position of the reticle fine movement stage 36 is calculated based on the average value of the outputs of the two Y-axis direction position measuring interferometers, and based on the difference between the outputs of the two Y-axis direction position measuring interferometers. The rotation angle of the reticle fine movement stage 36 in the XY plane is calculated.

The projection optical system PL has its optical axis direction set to the Z-axis direction orthogonal to the XY plane. For example, a predetermined reduction magnification β (here, β = 1
/ 4) is used. For this reason, at the time of exposure, a reduced image of the pattern of the slit-shaped illumination area in the pattern area of the reticle R is projected onto the projection optical system P.
Through L, projection exposure is performed on an exposure area conjugate to an illumination area on the wafer W having a surface coated with a photoresist.

The wafer-side stage system 16 includes a wafer-side base 42 arranged in parallel with the XY plane, and a substrate stage that holds the wafer W on the wafer-side base 42 and freely moves in the XY two-dimensional directions. Wafer stage 4
And 4. Here, the wafer stage 44 is, in fact, levitated and supported on the wafer-side base 42 via an air pad (not shown) and moves two-dimensionally along the upper surface of the base 42, and is provided on the XY stage. A leveling stage and a Z / θ stage which is arranged on the leveling stage and holds the wafer are included in FIG. 1, but these are typically shown as a wafer stage 44. .

A movable mirror 46 is provided on the wafer stage 44. A wafer laser interferometer for projecting a laser beam onto the movable mirror 46 and detecting the position of the wafer stage 44 by receiving reflected light thereof. 48 is provided facing the movable mirror 46. Where, in practice,
The movable mirror is provided with a Y movable mirror having a reflective surface orthogonal to the Y axis direction and an X movable mirror having a reflective surface orthogonal to the X axis direction.
A total of three are provided: a Y-axis direction position measurement interferometer that receives the reflected light from the moving mirror, and an X-axis direction position measurement interferometer and a rotation measurement interferometer that receives the reflected light from the X moving mirror. However, in FIG. 1, these are typically shown as a moving mirror 46 and a wafer laser interferometer 48.

The outputs of these three wafer laser interferometers are supplied to a main control system 18, and the main control system 18 measures the X position of the wafer stage 44 based on the output of the X-axis position measurement interferometer. Then, the Y position of the wafer stage 44 is measured based on the output of the Y-axis position measurement interferometer, and the wafer stage 44 is measured based on the output of the rotation measurement interferometer with respect to the output of the X-axis position measurement interferometer. The rotation angle in the XY plane is calculated.

The main control system 18, the time of exposure, for example, in synchronism with scanning in the + Y direction of the reticle coarse motion stage 34 at a predetermined scanning speed V _R0 via the drive for the relative scanning (not shown) non The wafer stage 44 is connected via the illustrated driving device.
In the −Y direction at a scanning speed V _W0 (V _W0 = β · V _R0 ;
β = 1/4 or β = 1/5: In this embodiment, β = 1 /
4) Scanning the reticle coarse movement stage 3
The relative velocity error between the reticle R and the wafer W is accurately corrected by absorbing the relative velocity error between the reticle R and the wafer W.
The operation of the reticle fine movement stage 36 is controlled via a fine movement control drive device (not shown) so that Thereby, in synchronization with the reticle R being scanned in the + Y direction with respect to the slit-shaped illumination area illuminated by the exposure light IL, the wafer W is moved from the projection optical system PL to the exposure area conjugate with the illumination area. Scanning is performed in the −Y direction at a speed corresponding to the reduction magnification, and the image of the pattern formed on the pattern formation surface of the reticle R is sequentially transferred to the shot area on the wafer W. A specific control method of each stage during this exposure will be described later in detail.

When the exposure of one shot area is completed, the main control system 18 moves the wafer stage 44 by a predetermined distance in the non-scanning direction (in the X direction) to perform the stepping operation to the exposure start position of the next shot. Is performed, scanning exposure is performed, and exposure is performed in this manner by the step-and-scan method.

FIG. 2 is a control block diagram of a stage control system of the scanning exposure apparatus 10 according to the present embodiment. The stage control system shown in FIG. 2 is a control block diagram showing the functions (realized by software) of the main control system 18 in FIG. It may be configured by hardware.

This stage control system responds to an instruction from a main computer (not shown) in the non-scanning direction (ie,
A moving speed generator 50 for outputting a wafer stage speed command value V _{WX for} the X direction) and a wafer stage speed command value V _WY for the scanning direction (ie, the Y direction), and a speed command from the moving speed generator 50 The wafer stage speed control system 52X for controlling the speed of the wafer stage 44 in the X direction based on the value V _WX , and the speed command value V _WY from the moving speed generator 50, multiplied by 1 / β (here, 4 times). wafer stage 44 based respectively on the speed command value V _R,
The position of reticle fine movement stage 36 (and the position of reticle fine movement stage 36 based on the position information obtained by quadrupling the position of wafer stage 44, and wafer stage speed control system 52Y and reticle coarse movement stage speed control system 54 for controlling the speed of reticle coarse movement stage 34, respectively. Reticle fine movement stage control system 5 for controlling speed)
6 is provided. Here, a control system includes the wafer stage speed control system 52X, the wafer stage speed control system 52Y, and the reticle coarse movement stage speed control system 54.

More specifically, the wafer stage speed control system 52X multiplies the speed command value V _WX and the output of an integration circuit 62X, which will be described later, by a gain K _RWX times the wafer stage 44.
A subtractor 58X for calculating an X-direction speed deviation which is a difference from the speed in the X direction, and a P / P for performing a (proportional + integral) control operation using the X-direction speed deviation from the subtractor 58X as an operation signal.
I controller (indicated by transfer function G ₃ (s)) 60X
And a first integration circuit 62X for integrating the acceleration in the X direction of the wafer stage 44 driven by the thrust from the PI controller 60X.
Here, actually, the X direction position of the wafer stage 44 is measured by the wafer laser interferometer 48X, and the X direction speed of the wafer stage 44 is obtained by differentiating the measurement value of the wafer laser interferometer 48X. Then, the acceleration of the wafer stage 44 in the X direction is directly measured, and this is integrated to obtain the wafer stage 44.
The speed in the X direction is not required, but is shown in FIG. 2 for convenience of explanation and according to the convention of writing a control block diagram. Therefore, the first integration circuit 62X and the second integration circuit 64X for integrating this output do not actually exist.

The wafer stage speed control system 52Y includes:
It is configured similarly to wafer stage speed control system 52X.
That is, the wafer stage speed control system 52Y converts the speed command value V _WY and the output of an integration circuit 62Y described later into a gain K.
A subtractor 58Y for calculating a Y-direction speed deviation which is a difference from the Y-direction speed of the wafer stage 44 multiplied by _RWY, and a (proportional + integral) control operation using the Y-direction speed deviation from the subtractor 58Y as an operation signal. A PI controller 60Y (indicated by a transfer function G ₂ (s)) 60Y and a wafer stage 44 driven by thrust from the PI controller 60Y
And the integration circuit 62Y for integrating the acceleration in the Y direction. Here, actually, the position of the wafer stage 44 in the Y direction is measured by the wafer laser interferometer 48Y.
Of the wafer stage 44 by differentiating the measurement value of
Direction speed is determined, the wafer stage 4
4, the acceleration in the Y direction is not directly measured, and the acceleration of the wafer stage 44 in the Y direction is not obtained by integrating the acceleration. However, for the sake of convenience of explanation and according to the convention of writing the control block diagram, as shown in FIG. Is shown. Therefore, the integrating circuit 62Y and the integrating circuit 64Y for integrating this output do not actually exist.

The reticle coarse movement stage speed control system 54
Includes a subtractor 66 for calculating a velocity deviation which is a difference between the output of the integrating circuit 70 to be described later between the speed command value V _R of the reticle coarse movement stage 34 and the gain K _RR multiplied by the speed of the reticle coarse motion stage 34, the A PI controller (indicated by a transfer function G ₁ (s)) 68 which performs a control operation (proportional + integral) using the speed deviation from the subtractor 66 as an operation signal, and a reticle coarse movement driven by a thrust from the PI controller 68 The integration circuit 70 integrates the acceleration of the stage 34.

The reticle fine movement stage control system 5
6 is a wafer stage 44 which is an output of the integrating circuit 64Y.
Is input as a target position based on the position information in the Y direction (corresponding to the measurement value of the wafer interferometer 48Y), and the target position is input to the adder 78 described later. Output reticle fine movement stage 36
And a subtractor 74 for calculating a position deviation which is a difference from the position information (corresponding to the output of the reticle interferometer 40).
4 is a voltage obtained by multiplying the position deviation which is the output of the position loop by the gain Kp of the position loop, and converts a corresponding speed information into a response gain Kv of the speed loop and a response gain K of the speed loop.
An integrating circuit 76 that integrates the speed information output via v and converts it into position information (relative position information of the reticle fine movement stage 36 from a reference point on the reticle coarse movement stage 34).
Adder for calculating position information obtained by adding the position information output from the integration circuit 76 and the position information of the reticle coarse movement stage 34 output from the integration circuit 72 obtained by integrating the output of the integration circuit 70 described above. 78. Here, the response gain Kv of the speed loop is actually the response gain of the speed loop that also includes the reticle fine movement stage 36, and determines how many mm / sec the reticle fine movement stage 36 runs by inputting 1 V (volt). Kv, which is actually composed of a speed loop similar to that of the reticle coarse movement stage speed control system 54 described above. It is shown in such a form.

Since the reticle fine movement stage 36 actually moves on the reticle coarse movement stage 34, the position information of the reticle coarse movement stage 34, which is the output of the integration circuit 72, is input to the adder 78. The reticle coarse movement stage 34 which is the position information and the output of the integration circuit 76
The relative position information of the reticle fine movement stage 36 from the upper reference point is added. Therefore, the output of the adder 78 corresponds to the output of the reticle interferometer 40. That is, actually, the position and speed of the reticle coarse movement stage 34 are not directly measured, and only the position of the reticle fine movement stage 36 is measured by the reticle laser interferometer 38. The control block diagram is shown as in FIG. 2 in accordance with the custom of writing.
Therefore, the integrating circuit 70 and the integrating circuit 7 for integrating this output
2, and the integration circuit 76 does not actually exist.

With the above configuration, the wafer stage speed control system 50X controls the speed of the wafer stage 44 in the X direction with the speed command value V _WX output from the moving speed generator 50 as a target value, and generates the moving speed. The movement of wafer stage 44 is controlled by wafer stage speed control system 50Y controlling the speed of wafer stage 44 in the Y direction with speed command value V _WY output from unit 50 as a target value.

The reticle coarse movement stage 34 is controlled by the reticle coarse movement stage speed control system with a target value of four times the speed command value V _WY output from the moving speed generator 50 (the speed direction is opposite). Controlled. Further, the wafer stage 44 measured by the wafer laser interferometer 48
The position control of the reticle fine movement stage 36 is performed based on the position information of the reticle fine movement stage 34 and the position information of the reticle coarse movement stage 34 described above.

Next, the principle of the stage control in the present embodiment will be described with reference to FIGS. First, the stage control in the Y direction will be described.

In the stage control in the Y direction, the acceleration start time t _S (= t ₁₀ ) as the first time to the acceleration end time t _E (= t ₁₄ ) as the sixth time as a function of the time t. , The Y-direction velocity V _Y (t), the Y-direction acceleration α _Y (t), and the Y-direction jerk J _Y (t). Here, between _{these, J Y (t) = d} (α Y (t)) / dt = d 2 (V Y (t)) / d 2 t = d 3 (Y (t)) / d It is made up relationship of ³ t ... (1).

Firstly, during the period from the acceleration start time t ₁₀ to the acceleration end time t _14, the jerk J _Y as the period of (t) increases from 0, the time t ₁₁ from the acceleration start time t ₁₀ as the second time Until the jerk J _Y (t) is 0
As the period of reduced to the period from time t ₁₁ to time t ₁₂ as the third time, as the period decreases from jerk J _Y (t) is 0, from the time t ₁₂ as the fourth time fifth of the period until the time t ₁₃ as time, also as the period jerk J _Y where (t) is increased to 0, the time t
To set a period of from ₁₃ to the time t _14. Then, the jerk J _Y (t) in each period is converted to a linear function J of time t.
Set to _LY (t).

That is, as shown in FIG. 3A, J _LY (t) = a ₁₁ t + b ₁₁ (t ₁₀ ≦ t ≦ t ₁₁ ) (2) J _LY (t) = a ₁₂ t + b ₁₂ ( t ₁₁ ≦ t ≦ t ₁₂ ) (3) J _LY (t) = a ₁₃ t + b ₁₃ (t ₁₂ ≦ t ≦ t ₁₃ ) (4) J _LY (t) = a ₁₄ t + b ₁₄ (t ₁₃ ≦ t ≦ t ₁₄₎ is set to ... (5). Here, a ₁₁ , a ₁₂ , a ₁₃ , a ₁₄ , b ₁₁ ,
b ₁₂ , b ₁₃ and b ₁₄ are constants.

In this embodiment, (T ₁₂ −t ₁₀ ) <(t ₁₄ −t ₁₂ ) (6) This is the acceleration reduction period before the end of acceleration,
That jerk is enough time from the time t ₁₂ to a negative value to the time t _14, by gentle the time variation of the acceleration immediately before the end of the jerk, the occurrence of high-frequency vibration generated in the stage It is for suppressing.

Based on the above equations (2) to (5), by sequentially integrating, the acceleration α at each time is calculated.
_LY (t), velocity V _LY (t), and the position Y _L when seeking (t), as follows.

[0051] In other words, the acceleration α _LY (t) _{is, α LY (t) = (} a 11/2) t 2 _{+ B 11 t + c 11 (} t 10 ≦ t ≦ t 11) ... (7) α LY (t) = (a 12/2) t 2 _{+ B 12 t + c 12 (} t 11 ≦ t ≦ t 12) ... (8) α LY (t) = (a 13/2) t 2 A _{+ b 13 t + c 13 (} t 12 ≦ t ≦ t 13) ... (9) α LY (t) = (a 14/2) t 2 + b 14 t + c 14 (t 13 ≦ t ≦ t 14) ... (10) (See FIG. 3B). Where c ₁₁ , c ₁₂ ,
c ₁₃ and c ₁₄ are integration constants.

[0052] In addition, the speed V _LY (t) _{is, V LY (t) = (} a 11/6) t 3 _{+ (B 11/2) t} 2 + c 11 t + d 11 (t 10 ≦ t ≦ t 11) ... (11) V LY (t) = (a 12/6) t 3 + (b 12/2) t 2 + c _{12 t + d 12 (t 11} ≦ t ≦ t 12) ... (12) V LY (t) = (a 13/6) t 3 _{+ (B 13/2) t} 2 + c 13 t + d 13 (t 12 ≦ t ≦ t 13) ... (13) V LY (t) = (a 14/6) t 3 + (b 14/2) t 2 + c ₁₄ t + d ₁₄ (t ₁₃ ≦ t ≦ t ₁₄ ) (14) (see FIG. 3C). Where d ₁₁ , d ₁₂ ,
d ₁₃ and d ₁₄ are integration constants.

[0053] Also, the position Y _L (t) _{is, Y L (t) = (} a 11/24) t 4 _{+ (B 11/6) t} 3 + (c 11/2) t 2 _{+ D 11 t + e 11 (} t 10 ≦ t ≦ t 11) ... (15) Y L (t) = (a 12/24) t 4 + (b 12/6) t 3 + (c 12/2) t 2 _{+ D 12 t + e 12 (} t 11 ≦ t ≦ t 12) ... (16) Y L (t) = (a 13/24) t 4 + (b 13/6) t 3 + (c 13/2) t 2 + d _{13 t + e 13 (t 12} ≦ t ≦ t 13) ... (17) Y L (t) = (a 14/24) t 4 + (b 14/6) t 3 + (c 14/2) t 2 + d 14 t + e ₁₄ (t ₁₃ ≦ t ≦ t ₁₄ ) (18) (see FIG. 3D). Where e ₁₁ , e ₁₂ ,
e ₁₃ and e ₁₄ are integration constants.

The 20 coefficients a _1i , b _1i , b _1i ,... Of the equations (2) to (5) and (7) to (18) thus obtained are obtained.
c _1i , d _1i , and e _1i (i = 1 to 4) are determined from the following 20 boundary conditions (see FIG. 3). In equations (2) to (5), jerk J _LY (t) is
Be continuous at times t ₁₁ , t ₁₂ and t ₁₃ (boundary condition 3
Pieces). In particular, J _LY (t ₁₂ ) = 0 [m / s ³ ]. In equations (7) to (10), the acceleration α _LY (t) is
Be continuous at times t ₁₁ , t ₁₂ and t ₁₃ (boundary condition 3
Pieces). In the equations (11) to (14), the speed V _LY (t) is
Be continuous at times t ₁₁ , t ₁₂ and t ₁₃ (boundary condition 3
Pieces). In the equations (15) to (18), the position Y _L (t) is
Be continuous at times t ₁₁ , t ₁₂ and t ₁₃ (boundary condition 3
Pieces). Acceleration at time t ₁₀ is not performed, the time t ₁₀
In acceleration starts to exit the acceleration at time t ₁₄ and, since thereafter does not perform the _{acceleration, J LY (t 10) =} J LY (t 14) = 0 [m / s 3], α LY (t ₁₀ ) = α _LY (t ₁₄ ) = 0 [m / s ² ] (19) (four boundary conditions). Speed before the acceleration, i.e. the velocity at the time t ₁₀ V ₁₀
V _LY (t ₁₀ ) = V ₁₀ [m / s], V _LY (t ₁₄ ) = V _W0 [m, where V _L0 [m / s] is the target speed after acceleration, that is, the speed at time t ₁₄ . / S] (20) (2 boundary conditions). The Y _L position of the acceleration at the start and Y _10, the acceleration at the end of Y _L
Position as _{Y 14, Y L (t 10} ) = Y 10 [m], Y L (t 14) = Y 14 [m] It is ... (21) (two boundary conditions).

As described above, the wafer stage 44 at the initial Y position (= Y ₁₀ ) and the initial Y direction velocity (= V ₁₀ ) without being accelerated in the Y direction is moved to the Y direction jerk J _Y (t). ) the while temporally continuously changed, a desired Y-direction velocity (= V _W0) at a desired Y-position in the desired time _{(= t 14) (= Y} 14), Y -direction jerk J _Y (t) , Y-direction acceleration α _Y (t), Y-direction velocity V _Y (t), and Y position Y (t).

Next, as shown in FIG. 4, a jerk from the jerk and time t ₁₂ of from the time t ₁₀ to the time t ₁₂ to time t _14, smooth as differentiable when the time variable The initial Y position (= Y ₁₀ ) and the initial Y direction velocity (= V
₁₀ ), while continuously changing the Y-direction jerk J _Y (t) with time, at a desired time (= t ₁₄ ) at a desired Y position (= Y ₁₄ ), a desired Y-direction speed (= V _W0 ), Y-direction jerk J _Y (t), Y-direction acceleration α
_Y (t), Y direction velocity V _Y (t), and Y position Y (t)
Ask for. At this time, in the period from the time t ₁₀ to the time t _12, Y-direction jerk J _Y (t), _Y direction acceleration α
_Y (t), Y direction velocity V _Y (t), and Y position Y (t)
There satisfactory boundary conditions, at time _{t 10, J Y (t 10} ) = J LY (t 10) = 0 [m / s 3], α Y (t 10) = α LY (t 10) = 0 [M / s ² ], V _Y (t ₁₀ ) = V _LY (t ₁₀ ) = V ₁₀ [m / s], Y (t ₁₀ ) = Y (t ₁₀ ) = Y ₁₀ [m] (22) and, at time _{t 12, J Y (t 12} ) = J LY (t 12) = 0 [m / s 3], α Y (t 12) = α LY (t 12) = α 12 [m / s 2 ], V _Y (t ₁₂ ) = V _LY (t ₁₂ ) = V ₁₂ [m / s], Y (t ₁₂ ) = Y (t ₁₂ ) = Y ₁₂ [m] (23) is there.

Therefore, the Y position Y (t) is a seventh-order function having eight coefficients, that is, Y (t) = A ₁₇ t ⁷ + A ₁₆ t ⁶ + A ₁₅ t ⁵ + A ₁₄ t ⁴ + A ₁₃ t ³ + A ₁₂ t ² + A ₁₁ t + A ₁₀ (24), the Y-direction jerk J _Y (t) and the Y-direction acceleration α satisfying the expressions (22) and (23).
_Y (t), Y direction velocity V _Y (t), and Y position Y (t)
Can be determined.

That is, the Y position Y represented by the equation (24)
(T) is sequentially differentiated, and V _Y (t) = d (Y (t)) / dt = 7A ₁₇ t ⁶ + 6A ₁₆ t ⁵ + 5A ₁₅ t ^{4 _{+ 4A 14 t 3 + 3A 13}} t 2 + 2A ₁₂ t + A ₁₁ (25) α _Y (t) = d (V _Y (t)) / dt = 42A ₁₇ t ⁵ + 30A ₁₆ t ⁴ + 20A ₁₅ t ³ + 12A ₁₄ t ² + 6A ₁₃ t + 2A ₁₂ (26) J _Y (t) = d (α _Y (t)) / dt = 210A ₁₇ t ⁴ + 120A ₁₆ t ³ + 60A ₁₅ t ² + 24A ₁₄ t + 6A ₁₃ (27) is determined, and in the equations (24) to (27), the coefficients A _{10 to} A ₁₇ satisfying the equations (22) and (23) are determined. by, for from the time t ₁₀ to the time t ₁₂ Y directions jerk J
_Y (t), Y-direction acceleration α _Y (t), Y-direction velocity V
_Y (t) and Y position Y (t) can be determined.

In the period from time t ₁₂ to time t ₁₄ , the Y-direction jerk J _Y (t) and the Y-direction acceleration α _Y
(T), Y-direction velocity V _Y (t), and Y position Y (t) boundary conditions to be satisfied are, at time t ₁₂ (23) equation,
And J _Y (t ₁₄ ) = J _LY (t ₁₄ ) = 0 [m / s ³ ], α _Y (t ₁₄ ) = α _LY (t ₁₄ ) = 0 [m / s ² ], V _Y (t ₁₄ ) = V _LY (t ₁₄ ) = V _W0 [m / s], Y (t ₁₄ ) = Y (t ₁₄ ) = Y ₁₄ [m] (28)

Therefore, the Y position Y (t) is a seventh-order function having eight coefficients, that is, Y (t) = B ₁₇ t ⁷ + B ₁₆ t ⁶ + B ₁₅ t ⁵ + B ₁₄ t ⁴ + B ₁₃ t ³ + B ₁₂ t ² + B ₁₁ t + B ₁₀ (29) By determining the coefficients B _{10 to} B ₁₇ in the same manner as above, the jerk J _Y (t), Y in the Y direction that satisfies the equations (23) and (28) is obtained. The directional acceleration α _Y (t), the Y-direction velocity V _Y (t), and the Y position Y (t) can be determined.

By controlling the movement of the stage in the Y direction based on the physical quantities required for controlling the stage in the Y direction thus obtained, the time change of the jerk is made continuous, Since the jerk makes the time change of the jerk time differentiable in each of the negative periods, high-frequency vibration of the stage at the end of acceleration can be suppressed. As a result, the settling time to be secured from the end of acceleration can be shortened.

Further, each physical quantity required for the stage control in the X direction can be obtained in the same manner as in the stage control in the Y direction described above. And the required X
By controlling the movement of the stage in the X direction based on each physical quantity required for the stage control in the direction, the time change of the jerk is made continuous, and in each of the positive period and the negative period of the jerk, Since the time change of jerk can be differentiated with time, high-frequency vibration of the stage at the end of acceleration can be suppressed. As a result, the settling time to be secured from the end of acceleration can be shortened.

Next, the control of the reticle-side stage 14 and the wafer-side stage in the scanning exposure apparatus 10 of the present embodiment using the above-described stage control principle will be described with reference to FIGS. 5 and 6. explain.

Here, wafer stage 44 and reticle coarse movement stage 34 take a target trajectory (temporal change in target position) corresponding to time t as shown in FIG. 5, and as a result, as shown in FIG. A case will be described as an example where the shot areas S1, S2, and S3 on the wafer W are scanned and exposed on the trajectory L of the center of the exposure area Ef (the optical axis of the projection optical system PL). Although the exposure area Ef is actually stationary and scanning is performed by moving the wafer stage 44 (wafer), FIG. 6 shows that the wafer is stationary and the center of the exposure area moves for convenience. . 5A to 5C, the horizontal axis indicates time, and the vertical axis indicates a target position. 5A to 5C, the time T
₀ indicates an acceleration time in scanning exposure of the exposure target shot, T ₁ indicates a settling time, T ₂ indicates an exposure time, and T
Reference numeral ₃ denotes the movement time of each stage from the end of exposure of a certain shot to the end of acceleration to the scanning speed at the time of exposure of the next shot.

It is premised that the preparatory steps such as alignment of the reticle R using a reference mark on a reference plate (not shown) mounted on the wafer stage 44 and a reticle microscope (not shown) are completed. The region Ef is the initial position O shown in FIG. 6, that is, the first shot region S1.
At the start position (scanning start position).

In this state, the moving speed generator 50
The period of time T ₀ obtained as described above, that is, the first period
The speed command value V _WX (t) (= 0 [m / s]) and the speed command value V _WY (t) are output during the acceleration period for scanning exposure of the second shot area S1. This speed command value V _WY
(T) is determined as follows.

In the scanning exposure period (time period T ₂ ) relating to the first shot area S 1, as shown in FIG. 5B, the Y direction speed V _Y (t) is changed to V _W0 [m / s]. ], The Y-direction speed V _Y (t) of the wafer stage 44 is accelerated from 0 [m / s] to V _W0 (t).
That is, the period of time T ₀ is an acceleration period in the + Y direction. For the acceleration during the acceleration period in the + Y direction, each physical quantity to be controlled in the stage control in the Y direction is obtained based on the principle of the stage control described above. In applying the above-described stage control principle to this period, the above-described boundary conditions are the same,
Instead of Oyobi boundary _{conditions, L Y (t 10) =} L Y0, L Y (t 14 (= t 10 + T 0)) = L Y1S, V Y (t 10) = 0, V Y (t 14) = V _W0 (30), the above-described principle of the stage control can be applied.

The speed V _Y (t) in the Y direction among the physical quantities obtained in this manner is calculated by the moving speed generator 50 in FIG.
_WY (t), and accordingly, the wafer stage speed control system 52Y in FIG. 2 moves the wafer stage 44 in the + Y direction so as to take the target trajectory L _Y (t) in FIG. 5B. Drive with acceleration.

On the other hand, reticle coarse movement stage 3 at this time
The target trajectory in the Y direction of No. 4 is represented by L shown in FIG.
_RC (t), but target Y of reticle coarse movement stage 34
The direction speed V _RC (t) has a relationship with the target Y direction speed V _Y (t) of the wafer stage 44, such that V _RC (t) = − 4 V _Y (t) (31). Therefore, as shown in FIG. 2 described above, V _WY (t) generated by the moving speed generator 50 is set to 4
The target speed V _R (t) is multiplied, and the positive / negative relationship in the equation (31) is absorbed by the PI controller 60Y and the feedback coefficient K _RR . By controlling the movement in the Y direction, the reticle coarse movement stage 34 is accelerated and driven along the movement trajectory of the target position L _RC (t) shown in FIG.

A reticle fine movement stage control system 56
Are the acceleration drive of the wafer stage 44 by the wafer stage speed control system 52Y and the reticle coarse movement stage control system 5
The reticle fine-movement stage 36 is driven so as to correct the synchronization deviation with the acceleration driving of the reticle coarse-movement stage 34 by the step 4.

Thus, at the end of the period of time T ₀ , wafer stage 44 moves to its acceleration end position (= (L
_X0 , L _Y1S )) and its acceleration end speed (= (0,
V _W0 )). The reticle fine movement stage 36 is
Its acceleration end position (= L _RC1S) and become its acceleration end speed _{(= V R0 = -4V W0)} .

Next, the moving speed generator 50 sets the time T
The speed command value V during the period of ₁ , ie, the settling period
_WX (t) (= 0 [m / s]) and speed command value V
_WY (t) (= V _W0 [m / s]) is output. In response, wafer stage 44 and reticle fine movement stage 36 are synchronously moved at a constant speed. During this settling period, the vibration generated on the reticle fine movement stage 36 due to the wafer stage 44 and the reticle coarse movement stage 34 is reduced. Thus, at the end of the period of time T ₁ , wafer stage 44 _{moves to} its end position (= (L _X0 ,
L _Y1T )). Also, reticle fine movement stage 36
_Reaches its end position (= L _RC1T ). These end positions are shown as positions S1a in FIGS.

When the center of the exposure area Ef is between the point O and the position S1a immediately before the start of the exposure of the shot area S1, the illumination light IL for exposure of the illumination system 12 is applied to the illumination area on the reticle R. Is not illuminated, and the resist on the wafer W is not printed.

Subsequently, the moving speed generator 50 _{controls the} speed command value V _WX (t) (= 0 [m / s]) and the speed command value V _WY during the time T ₂ , that is, during the scanning exposure period.
(T) (= V _W0 [m / s]) is output. In response, wafer stage 44 and reticle fine movement stage 36 are synchronously moved at a constant speed. The scanning exposure period starts, the center of the exposure area Ef further advances along the trajectory L, and the illumination light E from the illumination system 12 starts from the time when the exposure area Ef reaches the shot area S1.
L illuminates the illumination area on reticle R, and exposure of the pattern on reticle R starts. During this scanning exposure, the wafer stage 44 and the reticle fine movement stage 36 are synchronously moved at a predetermined speed ratio in opposite directions in the Y direction until the center of the exposure region comes to a point S1b outside the shot region S1. that is, until the end of the period of time T ₂ (scanning exposure period), the pattern in the illumination area on the reticle R is resist to the projection exposure on sequential wafer W. By such scanning exposure, the entire pattern on the reticle R is reduced and transferred to the shot area S1 on the wafer W. Then, when the scanning exposure period ends, the illumination light IL for exposure of the illumination system 12 is set so as not to illuminate the illumination area on the reticle R.

Next, the moving speed generator 50 _{controls the} speed command value V _WX (t) during the period of time T ₃ obtained as described above, ie, during the stepping and acceleration periods for the second shot area. And the speed command value V _WY (t) is output. The speed command value V _WX (t) and the speed command value V
_WY (t) is obtained as follows.

First, how to determine the speed command value V _WX (t) will be described. When the scanning exposure for the first-th shot area is completed, in the period of time T ₃ in FIG. 5 (A), moving in the X direction (non-scanning direction) by a step distance beta _X. In the movement in the X direction, the speed V _X (t) in the X direction at the start of the movement in the X direction is 0 [m / s], and X
The speed V _X (t) in the X direction at the end of the movement in the direction is 0 [m /
s], the time period T ₃ is the same as in FIG.
As shown in (A), first, an acceleration period _Ta is set,
After that, it is necessary to set the deceleration period _Td . Incidentally, in FIG. 5 (A), although deceleration period T _d immediately after the end of the acceleration period T _a is decided to start, it is also possible to leave between the acceleration period T _a and the deceleration period T _d .

Then, for at least the deceleration period _Td , each physical quantity to be controlled in the stage control in the X direction is obtained based on the principle of the stage control described above. Here, the above-mentioned principle of the stage control is applied at least for the deceleration period _Td for the following reason. That is, the high frequency vibrations of the wafer stage 44 becomes a problem, a scanning exposure is performed for a period of time T ₂ of the after a period of time T _3, the high-frequency vibration to the wafer stage 44 in the time T ₂ previously There may occur, and the is close to the period of the most time T _2, which is why deceleration period T _d.

[0078] Incidentally, it is of course possible to also apply the principles of stage control described above for acceleration period T _a. In this case, compared to the case where only the deceleration period _Td is applied,
Furthermore it is possible to reduce high-frequency vibrations that are generated in the wafer stage 44 at the time of the end period of time T _3.

In applying the above-described principle of the stage control to the deceleration period _Td , the above-mentioned boundary conditions are the same, but instead of the boundary conditions of and, L _X (t ₁₀ ) = L _X10 , L _X (t ₁₄ (= t ₁₀ + T _d )) = L _X0 + β _X , V _X (t ₁₀ ) = V _X10 , V _X (t ₁₄ ) = 0 (32) are used as boundary conditions. Note that the constants L _X10 , V _X10
Is determined by the acceleration mode of acceleration period T _a of the above.

[0080] Further, when the application of the principles of the aforementioned stage control to acceleration period T _a, the boundary conditions of the above-
Is the same, but instead of the boundary conditions of and, L _X (t ₁₀ ) = L _X0 , L _X (t ₁₄ (= t ₁₀ + T _a )) = L _X10 , V _X (t ₁₀ ) = 0, V _X (t ₁₄ ) = V _X10 (33) is used as a boundary condition.

[0081] Among the thus obtained physical quantity, X-direction velocity V _X (t) of the moving velocity generator 50 in FIG. 2 described above occurs as the speed command value V _WX (t).

Next, how to determine the speed command value V _WY (t) will be described. When the scanning exposure for the first shot area is completed, as shown in FIG.
_In period ₃ , movement in the Y direction (scanning direction) is performed. In the movement in the Y direction, the speed V _Y (t) in the Y direction at the start of the movement in the Y direction is V _W0 [m / s] (that is, the movement of the wafer stage 44 during the scanning exposure for the first shot area). Moving speed) and Y
The velocity V _Y (t) in the Y direction at the end of the movement in the direction is −V
_W0 [m / s] (ie, the moving speed of the wafer stage 44 during the scanning exposure for the second shot area)
Since it is necessary that the period of time T ₃ is the acceleration period in the -Y direction.

For the acceleration during the acceleration period in the -Y direction, the physical quantities to be controlled in the stage control in the Y direction are obtained based on the principle of the stage control described above. In applying the above-described stage control principle to this period, the above-described boundary conditions are the same, but instead of the boundary conditions of and, L _Y (t ₁₀ ) = L _Y1E , L _Y (t ₁₄ ( = T ₁₀ + T ₃ )) = L _Y2S , V _Y (t ₁₀ ) = V _W0 , V _Y (t ₁₄ ) = − V _W0 (32) where L _{Y1E is} the scanning exposure of the first shot area. End Y position L _Y2S : Use the constant velocity movement start Y position for scanning exposure of the second shot area as a boundary condition.

Of the physical quantities thus obtained, the moving speed generator 50 determines the speed V _Y (t) in the Y direction by the speed command value V
Occurs as _WY (t).

According to the speed command value V _WX (t) and the speed command value V _WY (t) generated by the moving speed generator 50, the wafer stage speed control system 52X of FIG.
The wafer stage 44 is accelerated / decelerated in the X direction so as to take the target trajectory L _X (t) of FIG. 5A, and the wafer stage speed control system 52Y outputs the target trajectory L _Y (FIG. 5B). The wafer stage 44 is accelerated in the −Y direction so as to take t). Further, reticle coarse movement stage control system 54 accelerates reticle coarse movement stage 34 in the + Y direction so as to take target trajectory L _RC (t) in FIG. 5C, and reticle fine movement stage 56 includes a wafer stage. The reticle fine movement stage 36 is driven so as to correct the synchronization deviation between the acceleration drive of the wafer stage 44 by the speed control system 52Y and the acceleration drive of the reticle coarse movement stage 34 by the reticle coarse movement stage speed control system 54.

Thus, at the end of the period of time T ₃ , wafer stage 44 moves to its acceleration end position (= (L
_{X0 + β X, L Y2S)} ) and its acceleration end speed (= (0,
−V _W0 )). Also, reticle fine movement stage 36
Is the acceleration end position (= L _{RC 2S} ) and the acceleration end speed (= V _R0 = 4V _W0 ).

Next, the moving speed generator 50 outputs the time T
The speed command value V during the period of ₁ , ie, the settling period
_WX (t) (= 0 [m / s]) and speed command value V
_WY (t) (= −V _W0 [m / s]) is output. In response, wafer stage 44 and reticle fine movement stage 36
Are synchronously moved. During this settling period, the vibration generated on the reticle fine movement stage 36 due to the wafer stage 44 and the reticle coarse movement stage 34 is reduced. Thus, at the end of the period of time T ₁ , wafer stage 44 _{moves to} its end position (= (L _X0 +
β _X , L _Y2T (= L _Y1E ))). The reticle fine movement stage 36 has its end position (= L _RC2T (= L
_RC1E )). These end positions are shown as positions S2a in FIGS.

Subsequently, the moving speed generator 50 _{controls the} speed command value V _WX (t) (= 0 [m / s]) and the speed command value V _WY during the time T ₂ , that is, during the scanning exposure period.
(T) (= −V _W0 [m / s]) is output. As a result, in the state where the scanning directions of the exposure region Ef, the reticle fine movement stage 36, and the wafer stage 44 are opposite to the shot region S1 (see FIGS. 5B, 5C, and 6), the shot region S1 By the same scanning exposure as described above, the pattern on the reticle R is reduced and transferred to the resist in the shot area S2. Thus, the center of the exposure area reaches the point S2b outside the shot area S1.

Thereafter, in the acceleration period (period of time T ₃ ), the settling period (period of time T ₁ ), and the scanning exposure period (period of time T ₂ ) for each shot area, The reticle coarse movement stage 34, the reticle fine movement stage, and the wafer stage 44 are moved and driven, and the pattern on the reticle R is reduced and transferred to each shot area in the same manner as in the case of the above shot area.

As described above, in the exposure operation in the present embodiment, as described above, the jerk time change is continuous, and the jerk time change is continuous during the entire acceleration period. In each of the positive jerk period and the negative jerk period, the time change of the jerk is a smooth change that can be differentiated, so that the high frequency component related to the time change of the jerk is reduced. As a result, the high-frequency vibration of the stage at the end of the acceleration of the stage is suppressed, so that the settling time (T ₁ ) of the stage can be shortened. Therefore, according to the present embodiment, high-precision exposure can be performed while improving the throughput.

<< Second Embodiment >> Hereinafter, a second embodiment of the present invention will be described with reference to FIGS. This embodiment corresponds to the first embodiment, the acceleration period in (in FIGS. 3 and 4 above, a period of from the time t ₁₀ to the time t _14), the period jerk positive regarding the acceleration direction (FIGS. 3 and in 4, after a period) of from the time t ₁₀ to the time t _12, immediately jerk negative duration (FIGS. 3 and 4
In, whereas it began period) from the time t ₁₂ to time t _14, the jerk during these periods 0 [m /
s ³ ]. Hereinafter, this point will be mainly described. In the following description, the same reference numerals are given to the same elements, and the duplicate description will be omitted.

The scanning exposure apparatus of this embodiment is similar to the scanning exposure apparatus of the first embodiment shown in FIG.
0, but the acceleration of the reticle coarse movement stage 34 and the wafer stage 44 by the main control system 18 is different.

Next, the principle of stage control in this embodiment will be described. First, the stage control in the Y direction will be described.

In the description of the stage control in the Y direction, as in the first embodiment, as a function of the time t, the acceleration start time t _S (= t ₂₀ ) as the first time is changed to the sixth time.
, The Y position Y (t), the Y direction velocity V _Y (t), the Y direction acceleration α _Y (t), and the Y direction jerk J _Y (until the acceleration end time t _E (= t ₂₅ ) Consider t).

[0095] First, during the period from the acceleration start time t ₂₀ to the acceleration end time t _25, jerk J as the period _{in which Y} (t) increases from 0, the time t ₂₁ from the acceleration start time t ₂₀ as the second time Until the jerk J _Y (t) is 0
As period decreases to set the interval from time t ₂₁ to time t ₂₂ as the third time. Jerk J
_Y (t) is a period of 0, sets the period from time t ₂₂ to time t ₂₃ as the fourth time. Further, as a period during which the jerk J _Y (t) decreases from 0, a period from the time t ₂₃ to the time t ₂₄ as the fifth time, and as a period during which the jerk J _Y (t) increases to 0, Time t
To set a period of from ₂₄ up to time t _25. The jerk J _Y (t) jerk J _Y (t) the time in each period except the period from time t ₂₂ to 0 to time t ₂₃ t
_Is set to the primary function J _LY (t). Incidentally, to set the time t ₂₂ and time t ₂₃ as unknowns.

That is, as shown in FIG. 7A, J _LY (t) = a ₂₁ t + b ₂₁ (t ₂₀ ≦ t ≦ t ₂₁ ) (33) J _LY (t) = a ₂₂ t + b ₂₂ ( t ₂₁ ≦ t ≦ t ₂₂ ) (34) J _LY (t) = 0 (t ₂₂ ≦ t ≦ t ₂₃ ) (35) J _LY (t) = a ₂₄ t + b ₂₄ (t ₂₃ ≦ t ≦ t ₂₄₎ ) (36) _JLY (t) = a ₂₅ t + b ₂₅ (t ₂₄ ≦ t ≦ t ₂₅ ) (37) Here, a ₂₁ , a ₂₂ , a ₂₄ , a ₂₅ , b ₂₁ ,
b ₂₂ , b ₂₄ and b ₂₅ are constants.

In the present embodiment, (t ₂₂ −t ₂₀ ) <(t ₂₅ −t ₂₃ ) (38) This is similar to the first embodiment, the reduction period of the acceleration of the front end of acceleration, i.e. enough time from the time t ₂₃ the jerk is negative value to the time t _25, the acceleration immediately before the end This is because the time change of the jerk is moderated to suppress the occurrence of high-frequency vibration generated on the stage.

On the basis of the above equations (33) to (37),
By sequentially integrating, the acceleration α at each time
_LY (t), velocity V _LY (t), and the position Y _L when seeking (t), as follows.

[0099] That is, the acceleration alpha _LY (t) _{is, α LY (t) = (} a 21/2) t 2 _{+ B 21 t + c 21 (} t 20 ≦ t ≦ t 21) ... (39) α LY (t) = (a 22/2) t 2 _{+ B 22 t + c 22 (} t 21 ≦ t ≦ t 22) ... (40) α LY (t) = α 23 (t 22 ≦ t ≦ t 23) ... (41) α LY (t) = (a 24/2) t ² _{+ B 24 t + c 24 (} t 23 ≦ t ≦ t 24) ... (42) α LY (t) = (a 25/2) t 2 + B ₂₅ t + c ₂₅ (t ₂₄ ≦ t ≦ t ₂₅ ) (43) (see FIG. 7B). Where c ₂₁ , c ₂₂ ,
c ₂₄ and c ₂₅ are integration constants, and α ₂₃ is a constant given as the maximum acceleration.

[0100] In addition, the speed V _LY (t) _{is, V LY (t) = (} a 21/6) t 3 _{+ (B 21/2) t} 2 _{+ C 21 t + d 21 (} t 20 ≦ t ≦ t 21) ... (44) V LY (t) = (a 22/6) t 3 + (b 22/2) t 2 + c 22 t + d 22 (t 21 ≦ t ≦ _{t 22) ... (45) V} LY (t) = α 23 t + d 23 (t 22 ≦ t ≦ t 23) ... (46) V LY (t) = (a 24/6) t 3 + (B _24/2 ) t ² _{+ C 24 t + d 24 (} t 23 ≦ t ≦ t 24) ... (47) V LY (t) = (a 25/6) t 3 + (B _25/2 ) t ² + C ₂₅ t + d ₂₅ (t ₂₄ ≦ t ≦ t ₂₅ ) (48) (see FIG. 7C). Where d ₂₁ , d ₂₂ ,
d ₂₃ , d ₂₄ and d ₂₅ are integration constants.

[0102] Also, the position Y _L (t) _{is, Y L (t) = (} a 21/24) t 4 _{+ (B 21/6) t} 3 + (c 21/2) t 2 _{+ D 21 t + e 21 (} t 20 ≦ t ≦ t 21) ... (49) Y L (t) = (a 22/24) t 4 + (B _22/6 ) t ³ + (c _22/2 ) t ² _{+ D 22 t + e 22 (} t 21 ≦ t ≦ t 22) ... (50) Y L (t) = (α 23/2) t 2 _{+ D 23 t + e 23 (} t 22 ≦ t ≦ t 23) ... (51) Y L (t) = (a 24/24) t 4 + (B _24/6 ) t ³ + (C _24/2 ) t ² _{+ D 24 t + e 24 (} t 23 ≦ t ≦ t 24) ... (52) Y L (t) = (a 25/24) t 4 + (B _25/6 ) t ³ + (C _25/2 ) t ² + d ₂₅ t + e ₂₅ (t ₂₄ ≦ t ≦ t ₂₅ ) (53) (see FIG. 7D). Where e ₂₁ , e ₂₂ ,
e ₂₃ , e ₂₄ and e ₂₅ are integration constants.

(33)-(37)
And 22 unknown coefficients a _2i in the equations (39) to (53),
b _2i , c _2i , (i = 1, 2, 4, 5), d _2i , e _2i (i
= 1-5) and 24 unknowns that time t ₂₂ and time t _23, determined from the following 24 boundary conditions. In equations (33) to (37), jerk J _LY (t)
Becomes continuous at times t ₂₁ , t ₂₂ , t ₂₃ , and t ₂₄ .
J _LY (t ₂₂ ) = J _LY (t ₂₃ ) = 0 [m / s ³ ] (four boundary conditions). In equations (39) to (43), the acceleration α _LY (t)
Becomes continuous at times t ₂₁ , t ₂₂ , t ₂₃ , and t ₂₄ .
α _LY (t ₂₂ ) = α _LY (t ₂₃ ) = α ₂₃ [m / s ² ] (four boundary conditions). In equations (44) to (48), the velocity V _LY (t) is
Time _{t 21, t 22, t 23} , t 24 by a continuous (four boundary conditions). In equations (49) to (53), the position Y _L (t) is
Time _{t 21, t 22, t 23} , t 24 by a continuous (four boundary conditions). Time t ₂₀ before acceleration is not performed, and terminates the acceleration at time t ₂₅ to start the acceleration at time t _20, since thereafter not performed _{acceleration, J LY (t 20) =} J LY (T ₂₅ ) = 0 [m / s ³ ], α _LY (t ₂₀ ) = α _LY (t ₂₅ ) = 0 [m / s ² ] (54) (four boundary conditions). Speed before the acceleration, i.e. the velocity at the time t ₂₀ V ₂₀
(In FIG. 7C, V ₂₀ = 0 [m / s])
The target speed after acceleration, that is, the speed at time t ₂₅ is V
_W0 as _{[m / s], V LY} (t 20) = V 20 [m / s], it is _{V LY (t 25) = V} W0 [m / s] ... (55) (2 pieces boundary conditions) . The Y position of the accelerator at the start and Y _20, Y-position at the end of acceleration as _{Y 25, Y L (t 20} ) = Y 20 [m], Y L (t 25) = Y 25 [m] ... (56 ) (2 boundary conditions).

As described above, the stage at the initial Y position (= Y ₂₀ ) and the initial Y direction speed (= V ₂₀ ) without being accelerated in the Y direction is moved to the Y direction jerk J.
_While changing _Y (t) continuously over time, the desired time (=
At time t ₂₅ ), a Y-direction jerk J _Y (t), a Y-direction acceleration α _Y (t), and a Y-direction speed V _Y at a desired Y-direction speed (= V _W0 ) at a desired Y position (= Y ₂₄ ). (T) and an example of the Y position Y (t) are obtained.

[0104] Next, as shown in FIG. 8, when the time variation of the jerk from jerk time change and time t ₂₃ from the time t ₂₀ to the time t ₂₂ to time t ₂₅ has a variable time, differential The initial Y position (= Y ₂₀ ) and the initial Y position while not being accelerated in the Y direction while making possible smooth changes.
The wafer stage 44 at the directional velocity (= V ₂₀ ) is moved at a desired Y position (= Y ₂₅ ) at a desired time (= t ₂₅ ) while continuously changing the Y-direction jerk J _Y (t) with time. Y-direction jerk J with desired Y-direction velocity (= V _W0 )
_Y (t), Y-direction acceleration α _Y (t), Y-direction velocity V
_Y (t) and Y position Y (t) are obtained in the same manner as in the first embodiment.

[0105] That is, the time of 7 linear function of Y position Y (t) in the period from the time t ₂₀ to the time _{t 22, Y (t) =} A 27 t 7 + A 26 t 6 + A ₂₅ t ⁵ + A ₂₄ t ⁴ + A ₂₃ t ³ + A ₂₂ t ² + A ₂₁ t + A ₂₀ (57) By the following formula, the Y-direction jerk J _Y (t), the Y-direction acceleration α _Y (t), and the Y-direction velocity V _Y satisfying the boundary condition.
(T) and the Y position Y (t) are determined. Time t
7 linear function of Y position Y (t) the time in the period from ₂₃ to time _{t 25, Y (t) =} B 27 t 7 + B ₂₆ t ⁶ + B ₂₅ t ⁵ + B ₂₄ t ⁴ + B ₂₃ t ³ + B ₂₂ t ² + B ₂₁ t + B ₂₀ (58), the Y-direction jerk J _Y (t), the Y-direction acceleration α _Y (t), and the Y-direction velocity V _Y satisfying the boundary conditions.
(T) and the Y position Y (t) are determined.

In the present embodiment, as in the first embodiment, the reticle-side stage 14 and the wafer-side stage 16 are controlled using the above-described principle of the stage control of the present embodiment. That is, the moving speed generator 50
Is a speed command value V _WX (t) obtained by applying the stage control principle of the present embodiment in a period corresponding to the period of the time T _{0 and} the period of the time T ₃ in FIG. By outputting the speed command value V _WY (t), the movement operation of reticle coarse movement stage 34, reticle fine movement stage 36, and wafer stage 44 is controlled.

As a result, as in the first embodiment, scanning exposure is performed on each shot area on the wafer W while the center Ef of the exposure area follows the trajectory L similar to that in FIG. The entire pattern on the reticle R is sequentially reduced and transferred to the area.

As described above, in the exposure operation in the present embodiment, as in the first embodiment, the jerk changes over time continuously, and the jerk changes over time during the entire acceleration period. In each of the positive period and the negative period of the jerk, the time change of the jerk is a smooth change that can be differentiated, so that the high-frequency component related to the time change of the jerk is reduced. As a result, the high-frequency vibration of the stage at the end of the acceleration of the stage is further suppressed, so that the stage settling time (T ₁ ) can be shortened. Therefore, according to the present embodiment, high-precision exposure can be performed while improving the throughput.

[0109] Further, in the present embodiment, the acceleration direction in (in FIGS. 7 and 8 described above, the period from the time t ₂₀ to the time t _25), jerk about acceleration direction is in the positive period (FIGS. 7 and 8 , after a period) from the time t ₂₀ to the time t _22, immediately jerk negative duration (FIGS. 7 and 8
, The jerk is 0 [m / s ³ ] during these periods (a period from time t ₂₃ to time t ₂₅ ) (from time t ₂₂ to time t in FIGS. 7 and 8). is provided with the period) to _23, the stage is accelerated from the time t ₂₂ at the maximum acceleration during the acceleration period until time t _23.
Therefore, as compared to the first embodiment, when accelerating to the target speed (V _W0 ) in a certain period, the maximum acceleration can be reduced, that is, the maximum value of the force applied to the stage can be reduced.

In the above embodiment, the wafer stage
In both the acceleration in the X direction and the acceleration in the ± Y direction, the jerk is provided with or without the period of 0 [m / s ³ ] during the acceleration period.
s ³ ] and a jerk of 0 [m / s ³ ]
May not be provided.

In the above embodiment, the jerk in the acceleration direction can be differentiated in the positive period and the negative period during the positive period and the negative period. However, FIG. 3A and FIG. As shown, it is also possible to employ a time change indicated by a polygonal line. Also in this case, since the jerk changes over time continuously, the high-frequency vibration of the stage at the end of acceleration can be reduced as compared with the related art.

[0112]

As described in detail above, claim 1 is as follows.
According to the stage control methods according to the first to sixth aspects, the acceleration of the stage continuously changes during the acceleration period of the stage, and the acceleration and the jerk become almost zero at the end of the acceleration of the stage. Accordingly, the high-frequency component of the time change of the jerk is reduced, and the occurrence of the vibration of the stage is suppressed. Therefore, the decay time of the vibration provided from the end of the acceleration of the stage to the start of the scanning exposure, that is, the settling time of the stage is set short. be able to.

According to the scanning exposure apparatus of the present invention, the acceleration drive of at least one of the wafer stage and the reticle stage is controlled by the stage control method according to any one of the first to sixth aspects. High-precision exposure is possible while improving.

[Brief description of the drawings]

FIG. 1 is a view schematically showing a configuration of a scanning exposure apparatus according to a first embodiment.

FIG. 2 is a block diagram showing a configuration of a stage control system in the apparatus of FIG.

FIG. 3 is a diagram (part 1) for explaining the principle of stage control in the first embodiment (A to D).

FIG. 4 is a diagram (part 2) for explaining the principle of stage control in the first embodiment (A to D).

FIG. 5 is a diagram showing an example of a target position of the wafer stage in the X direction, a target position of the wafer stage in the Y direction, and an example of a target position of the reticle coarse movement stage in the Y direction.
~ C).

FIG. 6 is a plan view showing a locus of the center of an exposure area when exposure is performed using the target value of FIG. 5;

FIG. 7 is a diagram (part 1) for explaining the principle of stage control in the second embodiment (A to D).

FIG. 8 is a diagram (part 2) for explaining the principle of stage control in the second embodiment (A to D).

[Explanation of symbols]

10: Scanning exposure apparatus, 34: Reticle coarse movement stage (part of mask stage), 36: Reticle fine movement stage, 44: Wafer stage (substrate stage), 52X ...
X direction wafer stage control system (part of control system), 52Y
... Y-direction wafer stage control system (part of control system), 54
... reticle coarse movement stage control system (part of control system), 56
... Reticle fine movement stage control system, R: reticle (mask), W: wafer (sensitive substrate), PL: projection optical system.

Claims (7)

[Claims] 1. A stage control method for controlling an acceleration of a stage in a predetermined direction, wherein a time change rate of the acceleration of the stage in the predetermined direction from a first time to a second time is substantially zero to zero. A first step of accelerating the stage while continuously increasing to a value of 1; and continuously changing the time rate of change from the first value to approximately 0 from the second time to a third time. A second step of accelerating the stage while decreasing; and from a fourth time to a fifth time, which is a time after the third time.
A third step of accelerating the stage while continuously decreasing the time rate of change from substantially zero to a second value;
The time change rate is calculated from the second time to the sixth time by the second time.
A step of accelerating the stage while continuously increasing the value of the stage to approximately 0, and making the acceleration of the stage at the sixth time substantially zero. 2. The time change of the time change rate from the fourth time to the sixth time is a smooth change that can be differentiated using time as a variable. Stage control method. 3. The method according to claim 2, wherein the time change of the time change rate from the first time to the third time is a smooth change that can be differentiated using time as a variable. Stage control method. 4. The stage control method according to claim 2, wherein the time change of the time change rate is a quartic function using time as a variable. 5. The third time and the fourth time,
The times are different from each other, and the fourth time is different from the third time.
2. The stage control method according to claim 1, wherein the time rate of change until the time is approximately zero. 6. The stage according to claim 1, wherein the time from the first time to the third time is shorter than the time from the fourth time to the sixth time. Control method. 7. The method according to claim 7, wherein the mask and the sensitive substrate are moved synchronously.
A scanning exposure apparatus for transferring a pattern formed on the mask onto the sensitive substrate, comprising: a mask stage for holding the mask; a substrate stage for holding the sensitive substrate; 7. The method according to claim 1, wherein the transfer onto the sensitive substrate is performed.
A control system for accelerating the mask stage and the substrate stage to target speeds in respective synchronous movement directions by any one of the stage control methods.