JPWO2005085671A1 - Anti-vibration device, exposure apparatus, and anti-vibration method - Google Patents

Abstract

This is an active vibration isolator having improved vibration isolation performance. In a vibration isolator having an air damper (43) that supports a structure and a drive unit that controls its internal pressure, a position sensor (49) that measures the position of the structure, a target position and an actual position of the structure, The subtractor (51) for generating the difference information, the variable amplifier (52) for controlling the drive unit based on the difference information, and the air damper (43) and its structure using the integral characteristics of the air damper (43) And a variable amplifier (58, 60, 63) for feeding back information for controlling the natural frequency of the system to the drive unit, and an amplifier (68) for feeding back the internal pressure information of the air damper (43) to the drive unit. And have.

Description

  The present invention relates to an anti-vibration technique used to suppress vibration when supporting a structure, for example, to support an exposure apparatus used when manufacturing various devices such as semiconductor devices and liquid crystal displays. It is suitable for use in. Furthermore, the present invention relates to an exposure technique using the image stabilization technique. The present application claims priority to Japanese Patent Application No. 2004-064636 filed on March 8, 2004, the contents of which are incorporated herein by reference.

  For example, in a lithography process which is one of the manufacturing processes of a semiconductor device, a pattern formed on a reticle (or photomask) as a mask is applied to a wafer (or glass plate or the like) coated with a photoresist as a substrate. An exposure apparatus is used for transfer exposure. As the exposure apparatus, a batch exposure type (stationary exposure type) projection exposure apparatus such as a stepper or a scanning exposure type projection exposure apparatus (scanning type exposure apparatus) such as a scanning stepper is used.

  In the exposure apparatus, in order to improve exposure accuracy such as positioning accuracy and overlay accuracy of the reticle stage and wafer stage, it is necessary to eliminate the influence of vibration as much as possible. However, when each of the above stages moves, the reaction force during acceleration is transmitted to the floor and the floor may vibrate greatly. In addition, since the floor vibrates by various forces even when the related equipment in the device manufacturing factory where the exposure apparatus is installed operates, the floor constantly vibrates to some extent. Therefore, in order to prevent the vibration of the floor from being transmitted to the exposure apparatus and lowering the exposure accuracy, a vibration isolator is conventionally disposed between the exposure apparatus and the floor (installation surface).

As a conventional vibration isolator, a mechanism that supports a stage or the like with an air damper to which air is supplied so that the internal pressure is maintained substantially constant is widely used. In addition, an active vibration isolator that uses an actuator that suppresses vibration detected by an acceleration sensor arranged on a stage or the like in combination with an air damper has been used. In order to further improve the anti-vibration performance, the type of sensor that feeds back information to the actuator is increased, and the air damper also controls the pressure using the detection result of the motion sensor provided on the stage. (See, for example, Patent Document 1).
JP 2002-175122 A

In recent years, patterns of semiconductor devices and the like have been further miniaturized, and accordingly, the exposure accuracy required has been increased. Therefore, it is necessary to further suppress the influence of vibration in the exposure apparatus. For this purpose, even in the active vibration isolator, it is necessary to improve the vibration isolation performance by reducing the vibration isolation ratio, which is the ratio of the floor vibration transmitted to the exposure apparatus, in a wide frequency range.
In addition, since it is required to increase the throughput as the pattern becomes finer, in the scanning exposure apparatus, the scanning speed of the stage is gradually increased. However, when the scanning speed of the stage is increased, vibration due to reaction force during acceleration / deceleration increases, and if this vibration is transmitted to the floor via the vibration isolator, the vibration of the floor will adversely affect the exposure apparatus again. Become. Therefore, in an active vibration isolator, it is necessary to suppress vibration detected by the sensor on the stage side as efficiently as possible to improve vibration isolation performance.

In view of the above, the present invention has a first object to provide an active vibration isolation technique with improved vibration isolation performance.
A second object of the present invention is to provide an exposure technique capable of improving exposure accuracy or throughput by improving vibration isolation performance.

    A first vibration isolator according to the present invention includes a gas damper that is supplied with gas and supports a structure on an installation surface, and a drive unit that controls the pressure of the gas in the gas damper. A first sensor for measuring the position information of the structure, and a first sensor for generating difference information by subtracting the position information of the structure measured by the first sensor from the information of the target position of the structure. A feedback unit, a drive amount control unit that generates drive information of the drive unit based on the difference information of the first feedback unit, and the gas damper and the structure using the integral characteristics of the gas damper And a characteristic control unit that feeds back information for controlling the natural frequency of the system to the drive unit.

  According to the present invention, the position of the structure is actively maintained near the target position by the first feedback unit. Further, by utilizing the fact that the characteristic from the drive unit to the pressure of the gas inside the gas damper (internal pressure) is almost an integral characteristic, for example, the viscous friction coefficient (D) of the gas damper so as to lower the natural frequency, Information on at least one of the spring constant (K) of the gas damper and the mass (M) of the structure is controlled by feedback. As a result, the vibration isolation rate of the first vibration isolator is improved, and the vibration isolation performance is improved.

  In the present invention, as an example, the characteristic control unit includes at least one of information corresponding to the viscous friction coefficient of the gas damper, information corresponding to the spring constant of the gas damper, and information corresponding to the mass of the structure. Is fed back to the drive unit. In this case, if the acceleration of the structure is α and the integral characteristic of the gas damper is ignored, the viscous friction coefficient (D) of the gas damper is a coefficient of resistance force proportional to the integral (speed) of the acceleration α. The spring constant (K) of the gas damper represents a coefficient of resistance proportional to the second order integral (position) of the acceleration α, and the mass (M) of the structure is a resistance proportional to the acceleration α. Represents the coefficient of force (inertia). Therefore, when considering the integral characteristic from the drive unit to the internal pressure of the gas damper, the viscous resistance coefficient (D) can be controlled by feedback of an amount obtained by multiplying the acceleration α of the structure by a predetermined gain, and the acceleration α The spring constant (K) can be controlled by feedback of the amount obtained by multiplying the integral by a predetermined gain, and the mass (M) can be apparently controlled by feedback of the amount obtained by multiplying the derivative of the acceleration α by the predetermined gain. .

  As another example, the characteristic control unit includes a second sensor that measures acceleration information of the structure, and a viscosity of the gas damper that is obtained by multiplying the acceleration information measured by the second sensor by a predetermined coefficient. A second feedback unit that feeds back information corresponding to the friction coefficient to the drive unit, and a spring constant of the gas damper obtained by multiplying information obtained by substantially integrating acceleration information measured by the second sensor by a predetermined coefficient Corresponding to the mass of the structure obtained by multiplying the information obtained by substantially differentiating the acceleration information measured by the second sensor by a predetermined coefficient. And a fourth feedback unit that feeds back information to the drive unit.

In this case, by the feedback, for example, the viscous resistance coefficient (D) is increased to (D + ΔD), the spring constant (K) is decreased to (K−ΔK), and the mass (M) is decreased to (M + ΔM). If it is increased, its natural frequency is lowered, and the vibration isolation ratio in the high frequency region is improved.
In the present invention, it may further include a fifth feedback unit that feeds back information corresponding to a low frequency component of the pressure fluctuation of the gas in the gas damper to the drive unit. The integral characteristic of the gas damper according to the present invention is small in a low frequency range below a predetermined break frequency fc. In view of this, in the low frequency region, the pressure in the gas damper is fed back so that the integral characteristic can be obtained in the entire frequency region by the gas damper. As a result, the vibration isolation rate as a whole is further improved, and the vibration isolation performance is improved.

As an example, the first sensor includes an acceleration sensor and an integrator that integrates acceleration measured by the acceleration sensor twice. In this case, the acceleration sensor can also be used as a position sensor.
As an example, an electromagnetic damper that applies an urging force with an electromagnetic force according to the displacement of the structure may be disposed in parallel with the gas damper between the installation surface and the structure. Since the response speed of the electromagnetic damper is higher than that of the gas damper, the vibration isolation rate can be further improved particularly in a high frequency range.

  Next, the second vibration isolator of the present invention is a vibration isolator having a damper for supporting the structure on the installation surface and a drive unit for controlling the biasing force of the damper. A first sensor that measures acceleration, a second sensor that measures acceleration information of the structure, a low-frequency component of position information measured by the first sensor, and acceleration information measured by the second sensor. A combining unit that obtains the position information of the structure by adding the high frequency components of the position information obtained by the time integration, and the position information of the structure that is obtained by the combining unit from the target position information of the structure And a drive amount control unit that generates drive information of the drive unit based on the difference information of the feedback unit.

  According to the present invention, the position of the structure is actively maintained near the target position by the feedback unit. The structure can be positioned based on the position information (inertial system spatial position) obtained by integrating the measurement value of the first sensor twice, and the position information in the low frequency range of the first sensor Based on this, the relative positioning of the installation surface and the structure can be performed.

In the present invention, as an example, the cut-off frequency for obtaining the low-frequency component of the position information is substantially equal to the cut-off frequency for obtaining the high-frequency component of the position information. As a result, the position information obtained by the combining unit becomes highly accurate in a wide frequency range.
Moreover, as an example, the damper is a gas damper in which a gas is supplied. In this case, the integral characteristic of the gas damper is used to control the natural frequency of the system including the gas damper and the structure so that the viscous friction coefficient of the gas damper, the spring constant of the gas damper, and A characteristic control unit that feeds back information corresponding to the mass of the structure to the driving unit can be further provided. As a result, for example, the natural frequency can be lowered, so that the vibration isolation rate can be improved particularly in a high frequency range.

As another example, the damper is an electromagnetic damper that generates an urging force by an electromagnetic force. In this case, vibration can be suppressed at a high tracking speed.
Furthermore, a gas damper and an electromagnetic damper may be used in combination, thereby improving the vibration isolation rate in a wide frequency range.
Next, the third vibration isolator of the present invention includes a gas damper that is supplied with gas and supports the structure on the installation surface, and a drive unit that controls the pressure of the gas in the gas damper. In the vibration isolator, a first sensor that measures the position information of the structure, and difference information is generated by subtracting the position information of the structure measured by the first sensor from the information of the target position of the structure Feedback unit, a drive amount control unit that generates drive information of the drive unit based on the difference information of the feedback unit, and a compensation unit that feeds forward the difference information of the feedback unit to position information of the structure It has.

According to the present invention, since the compensation unit acts as a parallel compensator for the gas damper, a stable state can be maintained even if the gain of the drive amount control unit is increased. Accordingly, the difference between the target position of the structure and the actual position is reduced, and the vibration isolation performance is improved.
The fourth vibration isolator of the present invention includes a gas damper that is supplied with gas and supports a structure on the installation surface, and a drive unit that controls the pressure of the gas in the gas damper. In the vibration device, the first sensor that measures the position information of the structure and the position information of the structure measured by the first sensor are subtracted from the information of the target position of the structure to generate difference information. A feedback unit, a drive amount control unit that generates drive information of the drive unit based on the difference information of the feedback unit, and a compensation unit that feeds back the drive information of the drive amount control unit to the difference information It is.

  According to the present invention, the portion composed of the drive amount control unit and the compensation unit acts as a series compensator for the gas damper, so that even when the gain of the drive amount control unit is increased, the stable state Can be maintained. Accordingly, the difference between the target position of the structure and the actual position is reduced, and the vibration isolation performance is improved. In this configuration, since only the position information of the structure can be fed back to the drive amount control unit, the difference between the target position of the structure and the actual position is even smaller than when the parallel compensator is used. Become.

  In this case, as an example, in the transfer function of the series compensator including the drive amount controller and the compensator, the denominator and the numerator are expressed by quadratic functions with respect to the variable s used in Laplace transform. When the gas damper is used, the equivalent circuit of the dynamic model includes a function having a quadratic function as the denominator with respect to the variable s. Therefore, the resonance peak of the gas damper can be almost canceled by the numerator polynomial of the series compensation unit, so that the resonance peak can be moved to a high frequency region, and the stability when the gain of the drive amount control unit is increased. Improves.

As an example, the series compensation unit including the drive amount control unit and the compensation unit moves the resonance peak when there is no compensation unit to the high frequency side. This improves the stability when the gain of the drive amount control unit is increased.
In these cases, the DC gain of the series control unit may be further increased. This improves its stability.

Next, the exposure apparatus according to the present invention illuminates the first object held on the first stage with the exposure beam, and exposes the second object held on the second stage through the first object with the exposure beam. In the exposure apparatus, at least one of the first stage and the second stage is supported via any one of the vibration isolator of the present invention.
Since the vibration isolation performance is improved by using the vibration isolator of the present invention, exposure accuracy such as overlay accuracy is improved. Further, in the case of a scanning exposure apparatus, the scanning speed can be increased in a state where vibration is reduced, so that the throughput can be improved.
In this case, a projection optical system that projects an image of the pattern of the first object onto the second object, and a column structure that holds the projection optical system and the first stage are provided. You may support through any vibration isolator of this invention. Thus, by integrally supporting the projection optical system and the first stage, the influence of vibration can be further reduced.

Further, three or more of the vibration isolation devices may be used to support the second stage or the column structure. By using three or more vibration isolation devices, the second stage and the like can be stably supported.
Next, a first vibration isolating method according to the present invention is the vibration isolating method for controlling the pressure of the gas in the gas damper that supports the structure on the installation surface by supplying the gas therein. A step of measuring information, a step of generating difference information by subtracting the position information of the structure measured in the step from the information of the target position of the structure, and the difference information generated in the step Based on the integral characteristics of the gas damper, the information for controlling the natural frequency of the system including the gas damper and the structure is fed back to control the pressure of the gas in the gas damper. It has a process.

According to the present invention, at least one information of the viscous friction coefficient of the gas damper, the spring constant of the gas damper, and the mass of the structure is used to reduce the natural frequency using the integral characteristic, for example. Control by feedback. As a result, the vibration isolation rate is improved in the high frequency region, so that the vibration isolation performance is improved.
Further, a second vibration isolation method according to the present invention includes a step of measuring position information and acceleration information of the structure in the vibration isolation method for controlling the urging force of the damper that supports the structure on the installation surface; Adding the low frequency component of the position information measured in the process and the high frequency component of the position information obtained by integrating the acceleration information measured in the process twice to obtain the position information of the structure And subtracting the position information of the structure obtained by the addition from the target position information of the structure to generate difference information, and driving the damper based on the difference information generated in the process And a process of performing.

According to the present invention, the structure can be positioned based on the position information of the structure (space position of the inertial system) obtained by integrating the acceleration information of the structure twice, and the low frequency component of the position information can be obtained. Thus, the relative positioning of the installation surface and the structure can be performed.
Further, a third vibration isolating method according to the present invention is the vibration isolating method for controlling the pressure of the gas in the gas damper that is supplied with gas and supports the structure on the installation surface. A step of generating difference information by subtracting the position information of the structure measured in the step from the information of the target position of the structure, and the difference information generated in the step And feed-forwarding the position information of the structure to control the pressure of the gas in the gas damper.
According to the present invention, a stable state can be maintained even if the gain at the time of driving the gas damper is increased by feeding forward the difference information to the position information of the structure. Accordingly, the difference between the target position of the structure and the actual position is reduced, and the vibration isolation performance is improved.

According to the vibration isolation method and apparatus of the present invention, it is possible to improve the vibration isolation performance when performing active vibration isolation.
In the present invention, a characteristic control unit that feeds back information to control the natural frequency of the system including the gas damper and the structure to the drive unit using the integral characteristic of the gas damper is provided, or The position information of the structure is obtained by adding the low frequency component of the position information measured by the first sensor and the high frequency component of the position information obtained by integrating the acceleration information measured by the second sensor twice. When using the synthesis unit, the natural frequency can be lowered or the position information of the structure can be measured with high accuracy in a wide frequency range, so that the vibration isolation performance when performing active vibration isolation can be improved. .

  Further, in the present invention, when providing a compensation unit that feeds forward the difference information of the feedback unit to the position information of the structure, or when providing a compensation unit that feeds back the drive information of the drive amount control unit to the difference information, Since the parallel compensator or the series compensator is provided, it is possible to improve the vibration isolation performance when performing active vibration isolation.

It is a figure which shows schematic structure of the projection exposure apparatus of embodiment of this invention. FIG. 2 is a partially cutaway view showing a state where the projection exposure apparatus of FIG. 1 is installed on the floor. It is a figure which shows the one anti-vibration stand 35 in FIG. 2, and its control system. It is a figure which shows the dynamic model of the vibration isolator 35 of FIG. It is a figure which shows an example of the change of the vibration isolation ratio of the dynamic model of FIG. It is a block diagram which shows the 1st structural example of the vibration isolator control system 48 of the 1st Embodiment of this invention. It is a figure which shows an example of the inertance response of the vibration isolator control system 48 of FIG. It is a block diagram which shows the 2nd structural example of the vibration isolator control system 48 of the 1st Embodiment. It is a figure which shows an example of the inertance response of the vibration isolator control system 48 of FIG. It is a figure which shows the dynamic model of the vibration isolator of the 2nd Embodiment of this invention. It is a figure used for the principle explanation of the synthetic sensor used in the second embodiment. It is a figure which shows an example of the vibration isolation characteristic of the dynamic model of FIG. It is a block diagram which shows the structure of the vibration isolator control system of the 2nd Embodiment of this invention. It is a block diagram which shows the 1st structural example of the vibration isolator control system of the 3rd Embodiment of this invention. It is a block diagram which shows the 2nd structural example of the vibration isolator control system of the 3rd Embodiment. It is a figure which shows the change of the vibration of a structure at the time of using the vibration isolator control system of FIG.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 9 ... Illumination optical system, R ... Reticle, PL ... Projection optical system, W ... Wafer, RST ... Reticle stage, WST ... Wafer stage, 28 ... Pressure sensor, 32 ... Pedestal, 35 ... Anti-vibration stand, 36 ... First column , CL ... column structure, 40 ... acceleration sensor, 43 ... air damper (gas damper), 47 ... servo valve (drive unit), 48 ... vibration isolator control system, 49 ... position sensor, 50 ... voice coil motor (electromagnetic damper) ), 51 ... Subtractor, 52, 58, 60, 63 ... Variable amplifier, 59 ... Pseudo integrator, 61 ... Pseudo differentiator, 68 ... Amplifier, 76 ... Synthetic sensor, 107 ... Parallel compensator, 109 ... Series compensator

[First Embodiment]
A preferred first embodiment of the present invention will be described below with reference to FIGS. In this example, the present invention is applied to the case of performing vibration isolation of a scanning exposure type projection exposure apparatus (scanning type exposure apparatus) composed of a scanning stepper.
FIG. 1 is a block diagram showing functional units constituting the projection exposure apparatus of this example. In FIG. 1, the chamber for housing the projection exposure apparatus is omitted. In FIG. 1, a laser light source 1 composed of a KrF excimer laser (wavelength 248 nm) or an ArF excimer laser (wavelength 193 nm) is used as a light source for exposure. As the light source for exposure, a laser emitting near-infrared light from a solid-state laser light source (such as YAG or semiconductor laser) that emits laser light in the ultraviolet light at an oscillation stage, such as other F ₂ lasers (wavelength 157 nm). A lamp that emits harmonic laser light in the vacuum ultraviolet region obtained by wavelength conversion of light, or a mercury discharge lamp that is often used in this type of exposure apparatus can also be used.

  Illumination light (exposure light) IL for exposure as an exposure beam from the laser light source 1 is a homogenizing optical system 2 composed of a lens system and a fly-eye lens system, a beam splitter 3, and variable dimming for light amount adjustment. The reticle blind mechanism 7 is irradiated with a uniform illuminance distribution through the device 4, the mirror 5, and the relay lens system 6. Illumination light IL limited in a slit shape or rectangular shape by the reticle blind 7 is irradiated onto the reticle R as a mask through the imaging lens system 8, and an image of the opening of the reticle blind 7 is formed on the reticle R. Imaged. An illumination optical system 9 is configured including a homogenizing optical system 2, a beam splitter 3, a variable dimmer 4 for adjusting light quantity, a mirror 5, a relay lens system 6, a reticle blind mechanism 7, and an imaging lens system 8. Yes.

  Of the circuit pattern area formed on the reticle R, the image of the portion irradiated with the illumination light is obtained by applying a photoresist as a substrate (sensitive substrate) through the projection optical system PL with the telecentric projection on both sides and the reduction magnification β. An image is projected onto the coated wafer W. As an example, the projection magnification β of the projection optical system PL is ¼, the image-side numerical aperture NA is 0.7, and the field diameter is about 27 to 30 mm. Although the projection optical system PL is a refractive system, a catadioptric system or the like can also be used. Reticle R and wafer W can also be regarded as a first object and a second object, respectively. Hereinafter, the Z-axis is taken in parallel to the optical axis AX of the projection optical system PL, the X-axis is parallel to the plane of FIG. 1 in the plane perpendicular to the Z-axis, and the Y-axis is perpendicular to the plane of FIG. Take and explain. In this example, the direction along the Y axis (Y direction) is the scanning direction of the reticle R and the wafer W during scanning exposure, and the illumination area on the reticle R is the direction along the X axis, which is the non-scanning direction. The shape is elongated in the (X direction).

  First, a reticle R arranged on the object plane side of the projection optical system PL is a reticle stage RST (first stage) that moves at a constant speed in at least the Y direction via an air bearing on a reticle base (not shown) during scanning exposure. Is held in. The movement coordinate position of the reticle stage RST (X-direction, Y-direction position, and rotation angle around the Z-axis) is a movable mirror Mr fixed to the reticle stage RST and a laser interferometer disposed opposite thereto. Measurement is sequentially performed with the system 10, and the movement is performed by a drive system 11 including a linear motor, a fine actuator, and the like. The movable mirror Mr and the laser interferometer system 10 actually constitute a triaxial laser interferometer having at least one axis in the X direction and two axes in the Y direction. The measurement information of the laser interferometer system 10 is supplied to a stage control unit 14, and the stage control unit 14 uses the measurement information and control information (input information) from a main control system 20 comprising a computer that controls the overall operation of the apparatus. Based on this, the operation of the drive system 11 is controlled.

  On the other hand, wafer W arranged on the image plane side of projection optical system PL is held on wafer stage WST (second stage) via a wafer holder (not shown), and wafer stage WST is at least in the Y direction during scanning exposure. It is mounted on a wafer base (not shown) via an air bearing so that it can move at a constant speed and can move stepwise in the X and Y directions. Further, the movement coordinate position (X direction, Y direction position, and rotation angle around the Z axis) of wafer stage WST is fixed to reference mirror Mf fixed to the lower part of projection optical system PL and wafer stage WST. The moving mirror Mw and the laser interferometer system 12 arranged opposite to the moving mirror Mw are sequentially measured, and the movement is performed by a drive system 13 including an actuator such as a linear motor and a voice coil motor (VCM). . The movable mirror Mw and the laser interferometer system 12 actually constitute a triaxial laser interferometer having at least one axis in the X direction and two axes in the Y direction. The laser interferometer system 12 actually further includes a two-axis laser interferometer for measuring rotation angles around the X and Y axes. Measurement information of the laser interferometer system 12 is supplied to the stage control unit 14, and the stage control unit 14 controls the operation of the drive system 13 based on the measurement information and control information (input information) from the main control system 20. .

  Wafer stage WST is also provided with a Z leveling mechanism that controls the position (focus position) of wafer W in the Z direction and the tilt angles around the X and Y axes. Then, on the lower side surface of the projection optical system PL, the projection optical system 23A that projects the slit image onto a plurality of measurement points on the surface of the wafer W, and the reflected light from the surface is received to re-image those slit images. An oblique incidence type multi-point autofocus sensor (23A, 23B) constituted by a light receiving optical system 23B that detects the amount of lateral shift of the image and supplies it to the stage control unit 14 is disposed. The stage control unit 14 calculates defocus amounts from the image plane of the projection optical system PL at the plurality of measurement points using information on the lateral shift amounts of the slit images, and these defocus amounts are set to predetermined values during scanning exposure. The Z leveling mechanism in wafer stage WST is driven by the autofocus method so as to be within the control accuracy. The detailed configuration of the oblique incidence type multi-point autofocus sensor is disclosed in, for example, Japanese Patent Application Laid-Open No. 1-253603.

  Further, the stage control unit 14 optimizes the drive system 13 based on the measurement information from the reticle interferometer system 12 and the reticle side control circuit that optimally controls the drive system 11 based on the measurement information from the laser interferometer system 10. When the reticle R and the wafer W are scanned synchronously during scanning exposure, both control circuits cooperatively control the drive systems 11 and 13. Further, the main control system 20 exchanges commands and parameters with each control circuit in the stage control unit 14 and executes optimum exposure processing according to a program designated by the operator. For this purpose, an operation panel unit (not shown) (including an input device and a display device) that provides an interface between the operator and the main control system 20 is provided.

  Further, when the laser light source 1 is an excimer laser light source, a laser control unit 25 under the control of the main control system 20 is provided, and this control unit 25 is used for a pulse oscillation mode (one pulse) of the laser light source 1. Mode, burst mode, standby mode, etc.) and the discharge high voltage of the laser light source 1 is controlled in order to adjust the average amount of pulsed laser light emitted. The light quantity control unit 27 is variably dimmed so as to obtain an appropriate exposure amount based on a signal from a photoelectric detector 26 (integrator sensor) that receives a part of the illumination light divided by the beam splitter 3. The apparatus 4 is controlled, and intensity (light quantity) information of the pulse illumination light is sent to the laser control unit 25 and the main control system 20.

  In FIG. 1, irradiation of the reticle R with the illumination light IL is started, and an image of a part of the pattern of the reticle R through the projection optical system PL is projected onto one shot area on the wafer W. The pattern image of the reticle R is transferred to the shot area by the scanning exposure operation in which the reticle stage RST and the wafer stage WST are moved (synchronously scanned) in synchronization with the Y direction using the projection magnification β of the projection optical system PL as the speed ratio. Is done. Thereafter, the irradiation of the illumination light IL is stopped, and the step-and-scan method is performed by repeating the operation of moving the wafer W stepwise in the X and Y directions via the wafer stage WST and the above-described scanning exposure operation. Thus, the pattern image of the reticle R is transferred to all shot areas on the wafer W.

For this exposure, it is necessary to align the reticle R and the wafer W in advance. Therefore, the projection exposure apparatus of FIG. 1 includes a reticle alignment system (RA system) 21 for setting the reticle R at a predetermined position and an off-axis alignment system 22 for detecting marks on the wafer W. Is provided.
Next, an example of an installation state of the projection exposure apparatus of this example in a semiconductor device manufacturing factory will be described. FIG. 2 shows an example of an installation state of the projection exposure apparatus. In FIG. 2, projection exposure is performed via a plurality of (for example, four or more) columns 31 made of H-shaped steel on the floor FL of the manufacturing factory. A thick flat plate pedestal 32 is installed as a base member when the apparatus is installed, and a rectangular thin flat plate base plate 33 for installing a projection exposure apparatus is fixed on the pedestal 32.

  A first column 36 is placed on the base plate 33 via three or four support members 34 and an active vibration isolator 35 (anti-vibration device), and projection optics is projected onto the central opening of the first column 36. System PL is held. The anti-vibration table 35 includes an air damper (gas damper) and an electromagnetic damper including a voice coil motor, as will be described later, and includes a set of acceleration sensors 40 and a set of position sensors installed in the first column 36. By controlling the pressure in the air damper and the thrust of the electromagnetic damper based on the detection information (not shown), vibration isolation of the first column 36 (and members supported thereby) is actively performed. ing. In this case, vibration isolation in a relatively low frequency region is performed by the air damper, and vibration isolation in a relatively high frequency region is performed by the electromagnetic damper.

  As the acceleration sensor 40, a piezoelectric acceleration sensor that detects a voltage generated by a piezoelectric element (piezo element or the like), or a semiconductor type that utilizes a change in the logical threshold voltage of a CMOS converter according to the magnitude of strain, for example. Can be used. As the position sensor (or displacement sensor), for example, an eddy current displacement sensor can be used. In this eddy current displacement sensor, for example, an alternating current is applied to a coil wound around an insulator, and when the coil is brought close to a measuring object made of a conductor, an eddy current is generated in the conductor by an alternating magnetic field generated by the coil. Take advantage of what happens. That is, the magnetic field due to the eddy current is in the opposite direction to the magnetic field due to the current of the coil, and the intensity and phase of the current flowing through the coil changes as these two magnetic fields overlap. Since this change becomes larger as the measurement object is closer to the coil, the position or displacement of the measurement object can be detected in a non-contact manner by detecting a signal corresponding to the current flowing through the coil. As other position sensors, a capacitance-type non-contact displacement sensor that detects the distance in a non-contact manner by utilizing that the capacitance is inversely proportional to the distance between the sensor electrode and the measurement target, An optical sensor or the like that detects the position of the light beam using a PSD (semiconductor position detector) can also be used.

  Further, a reticle base 37 is fixed to the upper part of the first column 36, a second column 38 is fixed so as to cover the reticle base 37, and the illumination optical system 9 of FIG. The illumination system subchamber 39 is fixed. In this case, the laser light source 1 of FIG. 1 is installed on the floor FL outside the pedestal 32 of FIG. 2 as an example, and the illumination light IL emitted from the laser light source 1 is illuminated via a beam transmission system (not shown). Guided to the optical system 9. A reticle stage RST that holds the reticle R is placed on the reticle base 37. In FIG. 2, a column structure CL is constituted by a first column 36, a reticle base 37, and a second column 38. The column structure CL is supported on the upper surface (installation surface) of the pedestal 32 via a plurality of active vibration isolation tables 35, the projection optical system PL, the reticle stage RST (first stage), and the illumination The optical system 9 is held.

The above-described set of acceleration sensors 40 includes, for example, three Z-axis acceleration sensors that measure acceleration in the Z direction at three locations that are not substantially on the same line in the XY plane, and two X-direction acceleration sensors that are separated in the Y direction. Are comprised of two X-axis acceleration sensors that measure the acceleration in the Y direction, and two Y-axis acceleration sensors that measure the acceleration in the Y direction at two locations separated in the X direction. The acceleration sensor 40 measures the acceleration in the X direction, Y direction, and Z direction of the column structure CL and the rotational acceleration [rad / s ² ] around the X axis, Y axis, and Z axis. . Similarly, the position of the column structure CL in the X direction, the Y direction, and the Z direction and the rotation angles around the X axis, the Y axis, and the Z axis are measured by the pair of position sensors (not shown). The Based on these measured values, the air dampers and the electromagnetic dampers in the plurality of vibration isolation tables 35 are configured so that the vibration of the column structure CL is kept small, the inclination angle of the column structure CL, and the height in the Z direction. Acts to maintain a constant value.

  Further, on the region surrounded by the plurality of support members 34 on the base plate 33 on the pedestal 32 and the active vibration isolation table 35, the wafer base WB is interposed via three or four active vibration isolation tables 41. Is supported. On wafer base WB, wafer stage WST holding wafer W is movably mounted. The anti-vibration table 41 includes an air damper and an electromagnetic damper similarly to the anti-vibration table 35, and the anti-vibration table 41 supports the wafer stage WST (second stage) on the upper surface (installation surface) of the pedestal 32. The anti-vibration table 41 actively suppresses vibrations of the wafer base WB and the wafer stage WST based on measurement information from an acceleration sensor and a position sensor (not shown) on the wafer base WB.

  The anti-vibration tables 35 and 41 of this example and their control systems (described later) correspond to the anti-vibration devices, respectively. The systems including the vibration isolation tables 35 and 41 and their control systems can also be called AVIS (Active Vibration Isolation System), which is an active vibration isolation system. The anti-vibration table 35 supports the reticle stage RST and the projection optical system PL via the column structure CL, and the scanning speed of the reticle stage RST during scanning exposure is higher than the scanning speed of the wafer stage WST. The reciprocal times (for example, 4 times) the projection magnification β is faster. On the other hand, since the anti-vibration table 41 supports only the wafer stage WST via the wafer base WB, the column structure CL is more susceptible to vibration than the wafer base WB. Therefore, the vibration isolation performance of the vibration isolation table 35 can be set higher than the vibration isolation performance of the vibration isolation table 41. In this case, as an example, in the anti-vibration table 41, the air damper may only control the pressure so that the position of the wafer base WB in the Z direction becomes substantially constant, for example.

  In FIG. 2, for example, wafer base WB and wafer stage WST may be supported so as to be suspended from the bottom surface of first column 36 that holds projection optical system PL. In this case, reticle stage RST, projection optical system PL, and wafer stage WST are all supported by a plurality of vibration isolation tables 35. At this time, the column supporting the projection optical system PL and the wafer stage WST and the column supporting the reticle stage RST are separated, and these are supported via a vibration isolation table similar to the vibration isolation table 35, respectively. Also good. In addition, reticle stage RST, projection optical system PL, and wafer stage WST may be supported independently of each other via a vibration isolation table similar to vibration isolation table 35.

  As described above, the active vibration isolation tables 35 and 41 of FIG. 2 can be configured in substantially the same manner. Below, the structure of the vibration isolator 35 and its control system, and its operation will be described as a representative example. Hereinafter, a mechanism for suppressing vibration in the Z direction, which is a direction parallel to the optical axis AX of the projection optical system PL, will be described. This is a mechanism for suppressing vibration in the X direction and the Y direction, and further the X axis. It can be similarly applied to a mechanism that suppresses vibration in the rotational direction around the Y axis and the Z axis.

  3 shows one vibration isolator 35 and its control system in FIG. 2. In FIG. 3, a support member 34 is installed on a base plate 33 on a pedestal 32, and a bottom plate 42 is mounted on the support member 34. The first column 36 is placed via the air damper 43 and the upper plate 44. The air damper 43 is obtained by enclosing air in a flexible hollow bag so that the pressure can be controlled. That is, the air damper 43 (gas damper) accumulates a predetermined amount or more of air at a predetermined pressure or higher via a flexible pipe 46 equipped with a servo valve 47 (drive unit) capable of controlling the air flow rate. A connected air source 45 is connected. As the air source 45, for example, an apparatus that combines an air compressor and an air cylinder filled with air pressurized by the air compressor can be used. Further, a pressure sensor 28 for measuring information on the pressure of the air in the air damper 43 is provided on the side surface of the air damper 43, and a measured value (a signal corresponding to the pressure) of the pressure sensor 28 is a vibration isolator control system 48. Has been supplied to. As the pressure sensor 28, a sensor in which a strain gauge is fixed to a diaphragm, a sensor using deformation of a silicon substrate, or the like can be used.

  A voice coil motor 50 as an electromagnetic damper is installed between the support member 34 and the first column 36 in parallel with the air damper 43. The voice coil motor 50 includes a stator 50b fixed to the upper surface of the support member 34 and permanent magnets arranged at a predetermined pitch in the Z direction, and a mover 50a fixed to the bottom surface of the first column 36 and mounted with a coil. It consists of and. Further, the acceleration sensor 40 and the position sensor 49 are fixed to the first column 36. In the example of FIG. 3, the acceleration sensor 40 measures acceleration information in the Z direction of the first column 36, and the position sensor 49 supports the support member 34. The relative position in the Z direction of the first column 36 relative to (or the floor surface), or information on the relative displacement in the Z direction is measured. As an example, the acceleration sensor 40 is a piezoelectric acceleration sensor, and the position sensor 49 is an eddy current displacement sensor. A speed sensor may be used as a sensor for detecting acceleration information. In this case, the speed information detected by the speed sensor may be differentiated once to obtain acceleration information.

  The acceleration sensor 40 in FIG. 3 represents one sensor for measuring the acceleration of the first column 36 at a position where the air damper 43 and the voice coil motor 50 are installed. The measurement values (signals corresponding to the acceleration and position) of the acceleration sensor 40 and the position sensor 49 are supplied to the anti-vibration table control system 48. The anti-vibration table control system 48 controls the flow rate of the air passing through the servo valve 47 based on the measurement values of the pressure sensor 28, the acceleration sensor 40, and the position sensor 49, thereby The air pressure (internal pressure) in the air damper 43 is controlled so that the position of the air reaches a predetermined target position. In parallel with this, the anti-vibration table control system 48 controls the current flowing through the coil of the mover 50a of the voice coil motor 50 based on the measurement values of the acceleration sensor 40 and the position sensor 49. The thrust in the Z direction by the voice coil motor 50 is controlled so that the position in the Z direction becomes a predetermined target position.

Next, a control system for controlling the internal pressure of the air damper 43 in the vibration isolator control system 48 of FIG. 3 will be described. FIG. 4 is a dynamic model of the air damper 43 in the vibration isolator 35 of FIG. 3. In FIG. 4, the installation surface 15 corresponds to the surface of the pedestal 32 of FIG. This corresponds to the first column 36. More precisely, the structure 16 includes the first column 36, the reticle base 37, the reticle stage RST, the second column 38, the illumination system subchamber 39, the illumination optical system 9, the projection optical system PL, and the like shown in FIG. include. The mass of the portion supported by the air damper 43 in the structure 16 is M, the viscous friction coefficient of the air damper 43 is D, and the spring constant is K. At this time, the mass M is a coefficient of resistance (inertia) according to the acceleration of the structure 16, the viscous friction coefficient D is a coefficient of resistance according to the speed of the structure 16, and the spring constant K is the structure. It can be regarded as a coefficient of resistance according to the position of 16. If the displacement in the Z direction of the structure 16 when the installation surface 15 is displaced by x _{0 in the} Z direction is x, the vibration isolation rate (x / x ₀ ) of the dynamic model in FIG. 4 is as follows: Become. The variable s is a Laplace transform variable. If the frequency is f (Hz), s = i2πf in the steady state.

In this case, the natural frequency of the system including the air damper 43 and the structure 16 can be reduced by increasing the apparent mass M and the viscous friction coefficient D and reducing the spring constant K by feedback. In this example, by utilizing the fact that the characteristic from the driver for driving the servo valve 47 in FIG. 3 to the internal pressure of the air damper 43 is substantially an integral characteristic, M → M + ΔM, D → D + ΔD, K → K by feedback. Assuming −ΔK, the vibration isolation rate (x / x ₀ ) changes as follows.

The dotted curve 51A and the solid curve 51B in FIG. 5 show the vibration isolation rates of the equations (1) and (2), respectively. In FIG. 5, the horizontal axis is the frequency f (Hz), and the vertical axis is the vibration isolation rate. (DB). As can be seen from FIG. 5, the natural frequency shifts to a low range by increasing the mass M and the viscous friction coefficient D and decreasing the spring constant K, so that the high frequency range in the hatched area is shown. The vibration isolation rate is improved. Note that the natural frequency of the curve 51A in the equation (1) is more accurately {K / M−D ² / (2M ² )} ^1/2 / (2π), so that the viscous friction coefficient D increases. However, the natural frequency decreases.

However, when the spring constant K is decreased by ΔK, the vibration isolation rate in the low frequency region is deteriorated as can be seen from the curve 51B in FIG. In order to improve the vibration isolation ratio in the low frequency range, the spring constant K is substantially fixed and other characteristics are controlled, or the internal pressure of the air damper 43 is further fed back as described later.
FIG. 6 shows the vibration isolator control of FIG. 3 for reducing the natural frequency by controlling the mass M, the viscous friction coefficient D, and the spring constant K apparently by feedback using the integral characteristic of the air damper 43. A first configuration example of the system 48 is shown. In FIG. 6, the anti-vibration table 35 that is blocked represents an equivalent circuit of the dynamic model of FIG. 4. That is, when pressure (thrust) from the air damper 43 in FIG. 3 is applied to the structure 16 in the vibration isolator 35 in FIG. 6, an acceleration α proportional to 1 / M is generated in the structure 16. Then, a resistance force obtained by integrating the acceleration α (multiplied by 1 / s) by a viscous friction coefficient D and a position obtained by further integrating the speed (multiplied by 1 / s) The resistance force multiplied by the spring constant K is fed back to perform vibration isolation. In this case, acceleration information of the structure 16 is measured by the acceleration sensor 40 (second sensor), and position information of the structure 16 is measured by the position sensor 49 (first sensor). The acceleration sensor 40 and the position sensor 49 output an analog signal such as a voltage corresponding to the measurement value as an example.

In FIG. 6, a signal (usually a constant value) corresponding to the target position x ₀ in the Z direction of the structure 16 is input from a control unit (not shown) to the subtractor 51 (first feedback unit) and measured by the position sensor 49. The signal corresponding to the position x in the Z direction of the structure 16 is fed back to the subtractor 51, and the subtractor 51 converts the signal corresponding to the difference (x ₀ −x) to the variable amplifier 52 (gain k1). Is supplied to the PI compensator 53 via a drive amount control unit). The PI compensator 53 is obtained by weighted averaging a first signal obtained by multiplying an input signal by a predetermined coefficient and a second signal obtained by multiplying a signal obtained by integrating the input signal by a predetermined coefficient. Is input to the adder 54. Further, a signal corresponding to the acceleration measured by the acceleration sensor 40 is passed through the DC cut filter 56 and a signal from which a direct current component is removed is input to the secondary Butterworth filter 57, the pseudo integrator 59, and the pseudo differentiator 61. Yes. The second-order Butterworth filter 57 is a filter that can obtain a uniform gain in a predetermined band obtained by cascading two single-tuned amplifier circuits. The pseudo-integrator 59 passes a low-frequency component from a direct current component as it is and has a high frequency. The pseudo-differentiator 61 is a circuit that differentiates low-frequency components by passing high-frequency noise components as they are. Note that a normal low-pass filter may be used instead of the secondary Butterworth filter 57, or the secondary Butterworth filter 57 may be omitted. It is also possible to use an integrator instead of the pseudo-integrator 59 and use a differentiator instead of the pseudo-differentiator 61.

The output signal of the secondary Butterworth filter 57 is fed back to the adder 54 via a variable amplifier 58 (second feedback unit) having a gain k2, and the output signal of the pseudo integrator 59 is passed through a variable amplifier 60 having a gain _kk1. Is fed back to the adder 54, and the output signal of the pseudo-differentiator 61 is fed back to the adder 54 via the secondary Butterworth filter 62 and the variable amplifier 63 having a gain _km1 . The pseudo integrator 59 and the variable amplifier 60 correspond to the third feedback unit, the pseudo differentiator 61 and the variable amplifier 63 correspond to the fourth feedback unit, and the variable amplifier includes the adder 54, the acceleration sensor 40, and the DC cut filter 56. The members up to 63 correspond to the characteristic control unit. Also in this case, a normal low-pass filter may be used instead of the secondary Butterworth filter 62, and the secondary Butterworth filter 62 may be omitted. The adder 54 inputs a signal obtained by adding the output signals of the PI compensator 53 and the variable amplifiers 58, 60, 63 to the driver 55. The driver 55 controls the air flow rate in the servo valve 47 of FIG. 3 based on the input signal.

  Note that the control system for the voice coil motor 50 shown in FIG. 3 is omitted in the vibration isolator control system 48 shown in FIG. The anti-vibration table control system 48 is composed of an analog circuit. For example, by converting the output signals of the acceleration sensor 40 and the position sensor 49 into digital data, the anti-vibration table control system 48 of FIG. Alternatively, it may be realized on software of a computer. The same applies to the following embodiments. Thus, when processing as digital data, the output data of the driver 55 is converted into an analog signal and supplied to the servo valve 47 of FIG.

In FIG. 6, the air damper 43 of FIG. 3 having an integral characteristic is actually arranged between the driver 55 and the structure 16. Therefore, a signal obtained by multiplying the output signal (corresponding to the speed by the integration effect) of the acceleration sensor 40 of FIG. 6 by the gain k2 by the variable amplifier 58 is a feedback amount (ΔD) corresponding to the viscous friction coefficient D of the air damper 43. The signal obtained by multiplying the signal (corresponding to the position by the integration effect) obtained by integrating the output signal of the acceleration sensor 40 by the pseudo-integrator 59 and the gain k _k1 by the variable amplifier 60 is the spring coefficient K of the air damper 43. feedback amount according to the corresponding to ([Delta] k), (corresponding to acceleration by integration effect) differentiating signal of the output signal of the acceleration sensor 40 in the pseudo differentiator 61 is obtained by multiplying the gain k _m1 in variable amplifier 63 to the signal Corresponds to the feedback amount (ΔM) corresponding to the mass M of the structure 16.

As a result, in the anti-vibration table control system 48 of this example, the information on the viscous friction coefficient D, the spring constant K, and the mass M is fed back in quadruplicate together with the position information of the structure 16. In this case, since the adder 54 is used, in order to make D → D + ΔD, K → K−ΔK, and M → M + ΔM by feedback, the gain k2 is a negative value, the gain _kk1 is a positive value, and the gain k _m1 may be a negative value.

In vibration isolating control system 48 of FIG. 6, the gain k1 is 0.25 as an example, the gain k2 is -3 to fix the gain k _k1 to 5. At this time, the calculation results of the inertance response in FIG. 6 when the gain _km1 is sequentially set to 0, −10, −30, and −45 are respectively shown by curves 64A, 64B, 64C, and 64D in FIG. In FIG. 7, the horizontal axis represents the frequency f [Hz], and the vertical axis represents the gain [dB] of inertance (the amount of acceleration when energized with a unit force, that is, the amount corresponding to 1 / mass). From FIG. 7, it can be seen that as the gain _km1 is decreased by a negative value, the natural frequency is decreased and the vibration isolation ratio is improved as a whole.

  Next, since the integration characteristic of the air damper 43 in FIG. 3 becomes small in a low frequency region below a predetermined break frequency fc, it is not actually complete integration. Therefore, in the following second configuration example, the integral characteristic is created by feeding back the pressure in the air damper 43 in the low frequency range. Further, the feedback with the mass M of the structure 16 as (M + ΔM), the viscous friction coefficient D of the air damper 43 as (D + ΔD), the spring coefficient K of the air damper 43 as (K−ΔK), and the position feedback are applied to the preceding stage. As a whole, it is assumed that feedback is given five times as a whole.

FIG. 8 shows a second configuration example in which feedback is performed five times in the vibration isolator control system 48 for the air damper 43 in the vibration isolator 35 of FIG. 3, and the same reference numerals are given to portions corresponding to FIG. In FIG. 8 attached thereto, an air damper 43 is also shown in the block diagram of the vibration isolator 35. In FIG. 8, the subtractor 51 (first feedback unit) outputs a signal corresponding to the difference (x ₀ −x) between the target position x ₀ of the structure 16 and the position x measured by the position sensor 49 with the gain k1. The signal is input to the adder 54 via the variable amplifier 52 (drive amount control unit). Further, a signal corresponding to the acceleration measured by the acceleration sensor 40 is passed through the DC cut filter 56 and a signal from which a direct current component is removed is input to the secondary Butterworth filter 57, the pseudo integrator 59, and the pseudo differentiator 61. Yes.

The output signal of the second-order Butterworth filter 57 is fed back to the adder 54 via a variable amplifier 58 (second feedback unit) having a positive gain k2, and the output signal of the pseudo-integrator 59 is variable with a negative gain _kk2 . It is fed back to the adder 54 via the amplifier 60 (the output unit of the third feedback unit), and the output signal of the pseudo-differentiator 61 is the second-order Butterworth filter 62 and the variable amplifier 63 (the fourth feedback unit) having a positive gain _km2. Are fed back to the adder. The adder 54 inputs a signal obtained by adding the output signals of the variable amplifiers 52, 58, 60, 63 to the subtractor 65 via the PI compensator 53. Since the is different from the circuit in position 6 of the adder 54 in FIG. 8, the sign of the gain k2, k _k2, k _{m @ 2} of FIG. 8 is different from the sign of the corresponding gain in FIG. 6 .

  Also, a signal corresponding to the internal pressure of the air damper 43 measured by the pressure sensor 28 in FIG. 3 is sent through the secondary Butterworth filter 67 and the positive gain kg amplifier 68 (fifth feedback unit) in FIG. Has been supplied to. Here, instead of the second-order Butterworth filter 67, a low-pass filter having the cut-off frequency fc as a cutoff frequency may be used. The subtractor 65 inputs a signal obtained by subtracting the output of the amplifier 68 from the output of the PI compensator 53 to the driver 55 via the PI compensator 66. The PI compensator 66 is a circuit that performs proportional and integral control similar to the PI compensator 53. The driver 55 controls the flow rate of air in the servo valve 47 of FIG. 3 based on the input signal.

In vibration isolating control system 48 of FIG. 8, the gain k1 is 0.05 as an example, the gain k2 is 0.5, the gain kg 30 pressure feedback gain k _k2 is fixed to -3.5 . At this time, the calculation results of the inertance response in FIG. 8 when the gain _km1 is sequentially set to 0, 10, 20, and 25 are shown in curves 69A, 69B, 69C, and 69D in FIG. 9, respectively. In FIG. 9, the horizontal axis represents the frequency f [Hz], and the vertical axis represents the inertance gain [dB]. From FIG. 9, the natural frequency can be further reduced from about 1.4 Hz to about 1.3 Hz compared to the case of using the circuit of FIG. 6 (FIG. 7) by the fivefold feedback including the pressure feedback. It can be seen that the vibration isolation rate is further improved.

[Second Embodiment]
Next, a second embodiment of the present invention will be described with reference to FIGS. In this example, the active vibration isolator 35 of the projection exposure apparatus shown in FIGS. 2 and 3 is used. In this example, a control system that can be used not only when the air damper 43 (gas damper) of FIG. 3 but also when driving the voice coil motor 50 (electromagnetic damper) will be described.

FIG. 10 shows a mechanical model of the vibration isolator 35 of this example. In FIG. 10, a damper having a spring constant K and a viscous friction coefficient D (corresponding to the air damper 43 in FIG. 3) is provided on the installation surface 15. A structure 16 is supported. Further, thrust F is applied to the structure 16 in the Z direction by an actuator 17 (corresponding to the air damper 43 and the voice coil motor 50 in FIG. 3). In this case, the position of the installation surface 15 in the Z direction is x _f , and the position of the structure 16 in the Z direction is x _p . At this time, the acceleration dx _p ² / dt ² of the structure 16 is measured by the acceleration sensor 40, and the relative position x (= x _p −x _f ) of the structure 16 with respect to the installation surface 15 is measured by the position sensor 49. ing. It is assumed that signals such as voltages corresponding to measured values are output from the acceleration sensor 40 and the position sensor 49, respectively.

Also in this example, the thrust F is controlled so that the position x in the Z direction of the structure 16 becomes a predetermined target value x ₀ , but the position of the structure 16 is simply measured by the position sensor 49. Instead of using the value, the position x ₁ obtained by processing the measurement value of the acceleration sensor 40 and the measurement value of the position sensor 49 is used. That is, a signal corresponding to the position output from the position sensor 49 is supplied to the adder 74 via the low-pass filter 75, and a signal corresponding to the acceleration output from the acceleration sensor 40 is supplied to the integrator 71 (transfer function is 1 /. s), an integrator 72, and a high-pass filter 73. As the high-pass filter 73, it is desirable to use a secondary high-pass filter in order to increase the positional accuracy. A signal obtained by adding the two signals input in the adder 74 corresponds to the position x ₁ at which the structure 16 is measured. In order to generate the information of the position x ₁ synthesized from the acceleration and position information of the structure 16 from the acceleration sensor 40, the integrators 71 and 72, the high-pass filter 73, the position sensor 49, the low-pass filter 75, and the adder 74. The synthetic sensor 76 is configured. Further, the integrators 71 and 72, the high-pass filter 73, the low-pass filter 75, and the adder 74 correspond to the synthesis unit.

The composite sensor 76 of this example can be used in place of the position sensor 49 in the vibration isolator control system 48 that performs feedback in a fivefold manner in FIG. 8 in the first embodiment described above, for example. By using the synthetic sensor 76 of this example, the structure 16 can be maintained at the target position with high accuracy in a wide frequency range.
Next, the measurement principle of the composite sensor 76 of this example will be described with reference to the evaluation model of FIG. FIG. 11 is a block diagram showing an evaluation model of the control object and the synthetic sensor. In FIG. 11, the upper control object 35A corresponds to the vibration isolator 35 in FIG. 10, and the lower synthetic sensor 76A in FIG. It corresponds to the composite sensor 76. The input of the target position and the like of the control target 35A is u (t), and the output of the actual position and the like is y (t). X (t) represents the state of the controlled object. On the other hand, ys (t) is the output of the composite sensor 76A, and xs (t) represents the state (state estimated value) of the composite sensor 76A.

  In the control target 35A of FIG. 11, the input u (t) is input to the adder 96 via the gain B amplifier 95, and the state x (t) is also input to the adder 96 via the gain A amplifier 99. ing. The result of integrating the output of the adder 96 (Ax (t) + Bu (t)) by the integrator 97 is the state x (t), and the state x (t) is multiplied by the gain C by the amplifier 98 and output y ( t) is obtained. The output y (t) and the output ys (t) of the combined sensor 76A are supplied to the subtracter 100, and the difference 101 (= y (t) −ys (t)) obtained by the subtracter 100 is the gain 101. Is input to the adder 96A of the composite sensor 76A.

  In the combined sensor 76A, the input u (t) is supplied to the adder 96A via the gain B amplifier 95A, and the state xs (t) is also supplied to the adder 96A via the gain A amplifier 99A. Yes. A state xs (t) obtained by integrating the sum of the outputs of the amplifiers 95A, 99A, and 101 output from the adder 96A by the integrator 97A is output as an output ys (t) via the gain C amplifier 98A. 100 is entered.

In general, a linear time-invariant control object 35A is expressed in a state space as follows.
dx (t) / dt = Ax (t) + Bu (t)
y (t) = Cx (t) + Du (t) (3-1)
In the equation (3-1), consider a case where D = 0 without a direct term. On the other hand, when the synthetic sensor 76A is expressed in the state space, it is as follows.

dxs (t) / dt = (A-LC) xs (t) + Bu (t) + Ly (t)
ys (t) = Cxs (t) (3-2)
Here, when the state estimation error is defined as follows, the following holds from the equations (3-1) and (3-2).
e (t) = xs (t) -x (t)
de (t) / dt = (A-LC) e (t) (3-3)
Therefore, if the eigenvalue of the matrix (A-LC) can be arbitrarily set from the equation (3-3), the error e (t) converges to 0 when the time t is increased. This is possible when (C, A) is observable.

  When the composite sensor 76A shown in FIG. 11 is configured using the acceleration sensor 40 and the position sensor 49 shown in FIG. 10, equation (3-1) is considered as follows.

Here, v (t) is the speed of the structure 16 in FIG. 10, p (t) is the actual position of the structure 16, u (t) is the output of the acceleration sensor 40, and y (t) is the position sensor 49. Output. Further, the feedback gain of FIG. 11 is L = [l ₁ l ₂ ] ^T. Further, when the Laplace transform is performed on the expression (3-2), the following result is obtained. Note that the Laplace transforms of xs (t), u (t), and y (t) are Xs (s), U (s), and Y (s), respectively.

Xs (s) = (sI−A + LC) ⁻¹ (BU (s) + LY (s)) (3-5)
Substituting a matrix into equation (3-5) and calculating the result is as follows. Note that vs (t) and ps (t) are estimated values of the velocity v (t) and the position p (t) of the structure 16, respectively.

  The physical meaning of the first term of the expression (3-7) can be understood by modifying it as follows.

From the expression (3-8), it can be considered that the first term of the expression (3-7) is composed of two high-pass filters (HPF) and two integrators. The frequencies f ₁ and f _{2 in the} expression (3-8) represent the cut-off frequencies of the two high-pass filters. When this is determined, the feedback gains l ₁ and l ₂ are determined from the expression (3-8). . At this time, the position of the structure 16 is calculated by integrating the signal of the acceleration sensor 40 twice, and the low frequency component is further cut through the position through a secondary high-pass filter. Since the second term of the expression (3-7) is a low pass filter, the high frequency region of the signal from the position sensor 49 is cut. The sum of these two signals is the estimated position value of the structure 16 in FIG. Therefore, it can be understood that the position of the structure 16 can be estimated with high accuracy by the synthetic sensor 76 of FIG.

  In this case, it is desirable that the cut-off frequency of the high-pass filter 73 and the cut-off frequency of the low-pass filter 75 in FIG. Thereby, the position of the structure 16 can be measured with high accuracy in a wide frequency range using the synthetic sensor 76. The combined sensor 76 can detect the spatial position of the inertial system of the structure 16 with the acceleration sensor 40 and can measure the relative position between the floor FL and the structure 16 in the low frequency range with the position sensor 49. For this reason, by controlling the internal pressure of the air mount 43 based on the output value of the synthetic sensor 76, the positioning of the structure 16 in the spatial position of the inertial system and the relative relationship between the floor FL and the structure 16 in the low frequency range. Positioning.

Further, the vibration isolation rate for the structure 16 in FIG. 10 can be expressed by equation (1) as in FIG. FIG. 12 shows the result of calculating the vibration isolation rate of equation (1). In FIG. 12, the horizontal axis represents frequency f (Hz), and the vertical axis represents gain (dB) and phase (deg). In order to improve the vibration isolation performance of the equation (1), as a first method, each element of the denominator (mass M, viscous friction coefficient D, spring constant K) is changed without changing the numerator characteristics of the equation (1). What is necessary is just to add the thrust F of FIG. This can be realized by feeding back the acceleration dx ² / dt ² , the velocity dx / dT, and the position x of the structure 16 in FIG.

Similarly, as a second method, the thrust F may be applied so as to reduce each element (D, K) of the numerator without changing the characteristics of the denominator of the equation (1). In order to realize this, information on the speed and position of the installation surface 15 in FIG. 10 may be measured and fed back.
As a third method, the apparent resonance frequency of the air damper 43 can be lowered by reducing the denominator and the spring coefficient K in the numerator (the rigidity of the air damper 43 in FIG. 3). It is. Therefore, in FIG. 10, the vibration isolation performance can be improved by applying the thrust F in this way. Hereinafter, a configuration example of a control system using the synthetic sensor 76 for the vibration isolation table 35 in FIG. 10 will be described with reference to FIG. 13.

FIG. 13 shows a control system of the vibration isolation table 35 of FIG. 10. In FIG. 13, the position sensor 49 and the acceleration sensor 40 measure the position and acceleration information of the vibration isolation table 35, respectively. Similarly to FIG. 10, the adder 74 combines the signal obtained from the output of the position sensor 49 through the low-pass filter 75 and the signal obtained from the output of the acceleration sensor 40 through the two integrators 71 and 72 and the high-pass filter 73. By combining the signals, a signal corresponding to the position x ₁ of the structure 16 is generated. That is, the composite sensor 76 of FIG. 10 is incorporated in the control system of FIG.

In FIG. 13, a signal corresponding to the target position x ₀ of the structure 16 and a signal corresponding to the combined position x ₁ of the structure 16 are input to the subtractor 51 (feedback unit). Supplies a signal corresponding to the difference (x ₀ −x ₁ ) to the PI compensator 77 (drive amount control unit). The PI compensator 77 inputs an output obtained by performing proportional control and integral control to the subtractor 78. Note that the PI compensator 77 may include a low-pass filter. The subtractor 78 also receives a signal obtained by passing the output of the acceleration sensor 40 through an integrator 93 formed by connecting an integrator and a high-pass filter in cascade. The subtractor 78 inputs a signal obtained by subtracting the output of the integrator 93 from the output of the PI compensator 77 to the adder / subtractor 80 via the secondary Butterworth filter 79. Note that a PI compensator, an amplifier, or a low-pass filter similar to the PI compensator 77 can be used instead of the second-order Butterworth filter 79 having a substantially flat gain in a predetermined band.

  The adder / subtractor 80 also receives a signal obtained by passing the output of the position sensor 49 through a low-pass filter 83 and an amplifier having a gain Gp1. Further, an output from the acceleration sensor 40 is input to an adder 90 through an integrator 91 including two integrators and a high-pass filter and an amplifier 92 having a gain Gk1, and the output of the acceleration sensor 40 is converted to a notch filter. 86, a low-pass filter 87, a high-pass filter 88, and a signal obtained through the amplifier 89 having the gain Gm 1 are also input to the adder 90. The adder 90 inputs a signal obtained by adding the outputs of the amplifiers 89 and 92 to the adder / subtractor 80. In the adder / subtractor 80, a signal obtained by subtracting the output of the adder 90 from the sum of the output of the amplifier 84 and the output of the secondary Butterworth filter 79 is input to the adder 81.

  The anti-vibration table 35 of this example supports the column structure CL that supports the reticle stage RST of the scanning exposure apparatus as shown in FIG. 2, and at the start of movement of the reticle stage RST during scanning exposure and immediately before the stop. In some cases, the column structure CL may vibrate due to a reaction force (counter force) opposite to the acceleration of the reticle stage RST. Therefore, in the control system of FIG. 13, a scan counter 85 that generates a signal corresponding to the predicted reaction force of the reticle stage RST is arranged, and a signal corresponding to the reaction force generated by the scan counter 85 is added. Is supplied to the container 81. The adder 81 inputs a signal obtained by adding the output of the adder / subtractor 80 and the output of the scan counter 85 to the driver 82 having a predetermined gain. The driver 82 drives the actuator 17 (in this case, the voice coil motor 50 in FIG. 3) of the anti-vibration table 35 in FIG. 10 so that the thrust F according to the output of the adder 81 is obtained. In parallel with the driving of the voice coil motor 50, the internal pressure of the air damper 43 may be controlled.

In this example, the signal compared with the target position x ₀ in the subtractor 51 for position feedback is the position x ₁ of the structure 16 obtained by the synthetic sensor 76 of FIG. Therefore, since the actual position of the structure 16 is measured with high accuracy over a wide frequency range, the vibration isolation rate can be improved and the vibration isolation performance is improved. Further, in the control system of FIG. 13, acceleration information (output of the amplifier 89) that is an output of the acceleration sensor 40, position information obtained by integrating the acceleration twice (output of the amplifier 92), and speed obtained by integrating the acceleration. Since the information (output of the integrator 93) is fed back, each element of the denominator (mass M, viscous friction coefficient D, spring constant K) is increased without changing the numerator characteristics of the equation (1). The thrust F of FIG. 10 is controlled. Therefore, the vibration isolation rate is further improved. Further, since the scan counter 85 is provided and the thrust F is controlled by feedforward so as to cancel the influence of the reaction force of the reticle stage RST, the vibration isolation performance during scanning exposure is further improved.

[Third Embodiment]
Next, a third embodiment of the present invention will be described with reference to FIGS. In this example, the active vibration isolator 35 of the projection exposure apparatus shown in FIGS. 2 and 3 is used. Further, in this example, a control system for controlling the internal pressure of the air damper 43 will be described assuming that the air damper 43 (gas damper) in FIG. 3 is an ideal integrator.

FIG. 14 shows a block diagram of the first control system of the present example. In FIG. 14, the air damper 43 is represented as an integrator having an integral gain of A _air (transfer function is A _air / s). 14 has a transfer function obtained from a dynamic model of the air damper 43 (corresponding to the dynamic model of FIG. 4). The coefficients m, c, and k of the transfer function in the controlled object 102 represent the mass of the structure 16 (see FIG. 4), the viscous friction coefficient of the air damper 43, and the spring constant of the air damper 43, respectively. Therefore, the anti-vibration table 103 composed of the air damper 43 and the controlled object 102 corresponds to a portion obtained by removing the voice coil motor 50 from the anti-vibration table 35 of FIG. Further, the position x ₁ of the structure 16 that is the output of the control object 102 represents the position measured by the position sensor 49 of FIG. 3, for example. The position x ₁ may be a position measured by the composite sensor 76 in FIG.

In FIG. 14, a signal corresponding to the target position x ₀ of the structure and a signal corresponding to the position y obtained from the measured position x _{1 of} the structure are input to the subtractor 104 (feedback unit). The subtractor 104 supplies a signal corresponding to the difference (x ₀ −y) to the amplifier 105 (drive amount control unit) having the gain K _P. Then, by controlling the flow rate of air in a servo valve 47 (see FIG. 3) as a drive unit of the air damper 43 by the output of the amplifier 105, the air damper 43 is set so that the position difference (x ₀ −y) becomes zero. The internal pressure is controlled.

At this time, the output of the amplifier 105 is fed forward to the adder 106 via the parallel compensator 107. The parallel compensator 107 is a filter of a transfer function G (s) arranged in parallel with the anti-vibration table 103, and the transfer function G (s) is set so as to stabilize the output of the amplifier 105. . The parallel compensator 107 in FIG. 14 can also be referred to as a “parallel feedforward compensator”. The adder 106 also receives the signal at the measured position x ₁ of the structure, and the adder 106 adds the signal at the position x ₁ and the output from the parallel compensator 107 to the subtracter 104. Has been fed back.

According to the control system of FIG. 14, by increasing the gain K _{P of the} amplifier 105, the error of the position x ₁ at which the structure is measured with respect to the target position x ₀ can be reduced, and the vibration isolation performance is improved. . At this time, since the parallel compensator 107 is provided, even if the gain K _{P of the} amplifier 105 is increased, the output is prevented from becoming unstable.
Next, a control system configured to have a series compensator by equivalently converting the control system of FIG. 14 will be described with reference to FIG.

FIG. 15 is a block diagram of the second control system of this example. In FIG. 15, in which parts corresponding to FIG. 14 are assigned the same reference numerals, the structure 16 of FIG. The position x ₁ is measured by, for example, the position sensor 49 in FIG. 3 or the composite sensor 76 in FIG.
In FIG. 15, the signal corresponding to the target position x ₀ of the structure and the measured position x _{1 of} the structure are input to the subtractor 104 (feedback unit), and the subtractor 104 calculates the difference (x A signal corresponding to ₀₋ x ₁ ) is input to the subtractor 108. The output of the subtractor 108 is input to an amplifier 105 (drive amount control unit) having a gain K _P , and the output of the amplifier 15 is fed back to the subtractor 108 via a parallel compensator 107 having a transfer function G (s). The subtractor 108 inputs a signal obtained by subtracting the output of the parallel compensator 107 from the output of the subtractor 104 to the amplifier 105. Then, by controlling the flow rate of the air in the servo valve 47 (see FIG. 3) as the drive unit of the air damper 43 by the output of the amplifier 105, the air damper so that the position difference (x ₀ −x ₁ ) becomes zero. The internal pressure of 43 is controlled.

  The control system of FIG. 15 is an equivalent conversion of the control system of FIG. In FIG. 15, a circuit including the subtractor 108, the amplifier 105, and the parallel compensator 107 can be regarded as one controller, which is a series compensator 109 for the transfer function C (s). The transfer function C (s) is given by

In this equation (3), the coefficients T ₀ and ζ ₀ are determined according to the difference between the numerator and the denominator. As a result of the inventor calculating the transfer function C (s) in the expression (3) by using a numerical example, the resonance peak of the control target 102 in FIG. 15 is canceled by the numerator polynomial of the transfer function C (s). It was found that the resonance peak as moved to the high frequency band. However, the addition of the series compensator 109 also reduces the DC gain. Therefore, it is assumed that the DC gain of the series compensator 109 is recovered.

  In order to show a design example of the series compensator 109, the transfer function C (s) of the expression (3) is summarized as follows.

Here, ω _n of the numerator polynomial is a natural angular frequency, ζ _n is a damping coefficient, and these are parameters indicating the characteristics of the controlled object. The denominator polynomial ω _d needs to be designed larger than the numerator polynomial ω _n . Similarly, the coefficient of the s ¹ term including ω _d and ζ _d of the denominator polynomial needs to be set larger than that of the numerator polynomial. That is, the numerator polynomial is set from the identification of the parameter to be controlled, and each coefficient of the denominator polynomial is a parameter determined by the designer. As described above, the series compensator 109 is designed so that the transfer function of the equation (4) becomes the following equation.

The series compensator 109 may be an analog circuit, but may be realized by digital control using a DSP (digital signal processor), for example.
Further, substituting s = 0 into the equation (5) results in the following.
20log 0.27146 = -11.33 [dB] (6)
Therefore, in order to recover the DC gain to 0 [dB], the following is performed.

1 / 0.27146 = 3.68 (7)
Therefore, in FIG. 15, the transfer function C (s) of the series compensator 109 is set as shown in the equation (5), and an amplifier having a gain of 3.68 is added in series to the input side of the series compensator 109.
FIG. 16 shows experimental results showing the operation of the series compensator 109 of FIG. 15. In FIG. 16, the horizontal axis represents time t [s], and the vertical axis represents the position measured with respect to the controlled object 102 of FIG. 3 is represented by a detection signal (voltage) [V] of the position sensor 49 of FIG. In FIG. 16, a period T1 indicates a period in which oscillation is caused by removing (short-circuiting) the series compensator 109 in FIG. 15, and a period T2 is a period in which oscillation is suppressed by operating the series compensator 109 and the amplifier. Is shown. As a result, it can be seen that the series compensator 109 suppresses oscillation and can stably perform vibration isolation. Further, in the control system of FIG. 15, since the information of the position x ₁ of the controlled object 102 as compared with the control system of FIG. 14 is directly fed back to the subtracter 104, the control target 102 more accurately to the target position Can be controlled.

  In the projection exposure apparatus of the above-described embodiment, the column structure CL is installed via the vibration isolation tables 35 and 41, and then the illumination optical system and projection optical system composed of a plurality of lenses are used as the exposure apparatus body. It can be manufactured by adjusting the built-in optics, attaching a reticle stage or wafer stage consisting of many machine parts to the exposure apparatus body, connecting wiring and piping, and performing general adjustment (electrical adjustment, operation check, etc.) it can. The exposure apparatus is preferably manufactured in a clean room where the temperature, cleanliness, etc. are controlled.

  In addition, when a semiconductor device is manufactured using the projection exposure apparatus of the above-described embodiment, the semiconductor device includes a step of designing a function / performance of the device, a step of manufacturing a reticle based on the step, a wafer from a silicon material, A step of aligning with the projection exposure apparatus of the above-described embodiment to expose the reticle pattern onto the wafer, a step of forming a circuit pattern such as etching, a device assembly step (dicing process, bonding process, package process) ) And an inspection step.

  Note that the present invention can also be applied to a case where vibration isolation is actively performed with an immersion type exposure apparatus disclosed in, for example, International Publication (WO) No. 99/49504. Further, the present invention provides vibration isolation with a projection exposure apparatus that uses extreme ultraviolet light (EUV light) having a wavelength of about several nm to 100 nm as an exposure beam, and a proximity type or contact type exposure apparatus that does not use a projection optical system. It can also be applied when performing.

  In addition, the present invention is not limited to application to an exposure apparatus for manufacturing a semiconductor device, for example, an exposure apparatus for a display device such as a liquid crystal display element formed on a square glass plate or a plasma display, It can also be widely applied to an exposure apparatus for manufacturing various devices such as an image sensor (CCD or the like), a micromachine, a thin film magnetic head, and a DNA chip. Furthermore, the present invention can also be applied to an exposure process (exposure apparatus) when manufacturing a mask (photomask, reticle, etc.) on which mask patterns of various devices are formed using a photolithographic process.

Furthermore, the present invention can also be applied to the case of performing vibration isolation for equipment other than the exposure apparatus, such as a defect inspection apparatus, a photosensitive material coater / developer, and the like.
In addition, this invention is not limited to the above-mentioned embodiment, Of course, a various structure can be taken in the range which does not deviate from the summary of this invention.

  When the present invention is applied to an exposure apparatus, the vibration isolation performance of the exposure apparatus with respect to the installation surface can be improved, so that exposure accuracy such as overlay accuracy can be improved. Further, in the case of a scanning exposure apparatus, the scanning speed can be improved while suppressing vibrations, so that the throughput can be improved.

Claims (21)

In the vibration isolator having a gas damper that is supplied with gas and supports the structure on the installation surface, and a drive unit that controls the pressure of the gas in the gas damper,
A first sensor for measuring positional information of the structure;
A first feedback unit that generates difference information by subtracting position information of the structure measured by the first sensor from information on a target position of the structure;
A drive amount control unit that generates drive information of the drive unit based on the difference information of the first feedback unit;
And a characteristic control unit that feeds back information for controlling the natural frequency of a system including the gas damper and the structure to the drive unit using the integral characteristic of the gas damper. Shaker.   The characteristic control unit supplies at least one of information corresponding to a viscous friction coefficient of the gas damper, information corresponding to a spring constant of the gas damper, and information corresponding to a mass of the structure to the driving unit. The anti-vibration device according to claim 1, wherein feedback is provided. The characteristic control unit includes:
A second sensor for measuring acceleration information of the structure;
A second feedback unit that feeds back information corresponding to the viscous friction coefficient of the gas damper obtained by multiplying acceleration information measured by the second sensor to a predetermined coefficient;
A third feedback unit that feeds back information corresponding to a spring constant of the gas damper obtained by multiplying information obtained by substantially integrating acceleration information measured by the second sensor to a predetermined coefficient;
A fourth feedback unit that feeds back information corresponding to the mass of the structure obtained by multiplying information obtained by substantially differentiating acceleration information measured by the second sensor by a predetermined coefficient to the drive unit; The vibration isolator according to claim 1 or 2, characterized in that   4. The apparatus according to claim 1, further comprising a fifth feedback unit that feeds back information corresponding to a low-frequency component in the pressure fluctuation of the gas in the gas damper to the driving unit. 5. The vibration isolator as described.   5. The vibration isolator according to claim 1, wherein the first sensor includes an acceleration sensor and an integrator that integrates acceleration information measured by the acceleration sensor twice. 6. .   6. The electromagnetic damper according to claim 1, wherein an electromagnetic damper that applies an urging force by an electromagnetic force according to a displacement of the structure is disposed in parallel with the gas damper between the installation surface and the structure. An anti-vibration device according to claim 1. In a vibration isolator having a damper that supports a structure on an installation surface, and a drive unit that controls a biasing force of the damper,
A first sensor for measuring positional information of the structure;
A second sensor for measuring acceleration information of the structure;
The position of the structure is obtained by adding a low frequency component of position information measured by the first sensor and a high frequency component of position information obtained by integrating acceleration information measured by the second sensor twice. A synthesizing unit for obtaining information;
A feedback unit that generates difference information by subtracting position information of the structure obtained by the combining unit from information on a target position of the structure;
And a drive amount control unit that generates drive information of the drive unit based on the difference information of the feedback unit.   The anti-vibration device according to claim 7, wherein a cutoff frequency for obtaining a low frequency component of the position information is substantially equal to a cutoff frequency for obtaining a high frequency component of the position information. .   The vibration isolator according to claim 7 or 8, wherein the damper is a gas damper in which a gas is supplied.   The viscous friction coefficient of the gas damper, the spring constant of the gas damper, and the structure so as to control the natural frequency of the system including the gas damper and the structure using the integral characteristic of the gas damper The vibration isolator according to claim 9, further comprising a characteristic control unit that feeds back information corresponding to the mass of the drive unit to the drive unit.   The vibration isolator according to claim 7 or 8, wherein the damper is an electromagnetic damper that generates an urging force by an electromagnetic force. In the vibration isolator having a gas damper that is supplied with gas and supports the structure on the installation surface, and a drive unit that controls the pressure of the gas in the gas damper,
A first sensor for measuring positional information of the structure;
A feedback unit that generates difference information by subtracting position information of the structure measured by the first sensor from information on a target position of the structure;
A drive amount control unit that generates drive information of the drive unit based on the difference information of the feedback unit;
And a compensation unit that feeds forward the difference information of the feedback unit to position information of the structure. In the vibration isolator having a gas damper that is supplied with gas and supports the structure on the installation surface, and a drive unit that controls the pressure of the gas in the gas damper,
A first sensor for measuring positional information of the structure;
A feedback unit that generates difference information by subtracting position information of the structure measured by the first sensor from information on a target position of the structure;
A drive amount control unit that generates drive information of the drive unit based on the difference information of the feedback unit;
And a compensation unit that feeds back drive information of the drive amount control unit to the difference information.   The transfer function of a series compensation unit including the drive amount control unit and the compensation unit is characterized in that the denominator and the numerator are each expressed by a quadratic function with respect to a variable s used in Laplace transform. Anti-vibration device.   15. The series compensation unit including the drive amount control unit and the compensation unit moves a resonance peak when there is no compensation unit to the high frequency side, according to any one of claims 12 to 14. Anti-vibration device. In an exposure apparatus that illuminates a first object held on a first stage with an exposure beam and exposes the second object held on the second stage via the first object with the exposure beam,
An exposure apparatus, wherein at least one of the first stage and the second stage is supported via the vibration isolator according to any one of claims 1 to 15. A projection optical system for projecting an image of the pattern of the first object onto the second object;
A column structure that holds the projection optical system and the first stage;
The exposure apparatus according to claim 16, wherein the column structure is supported via the vibration isolator according to claim 1.   18. The exposure apparatus according to claim 17, wherein three or more of the anti-vibration devices are used to support the second stage or the column structure. In the vibration isolation method for controlling the pressure of the gas in the gas damper that supports the structure on the installation surface with the gas supplied inside,
Measuring the position information of the structure;
Subtracting the position information of the structure measured in the step from the information on the target position of the structure to generate difference information;
Based on the difference information generated in the step, using the integral characteristic of the gas damper, feedback information for controlling the natural frequency of the system including the gas damper and the structure, And a step of controlling the pressure of the gas in the gas damper. In the vibration isolating method for controlling the biasing force of the damper that supports the structure on the installation surface,
Measuring the position information and acceleration information of the structure;
Adding the low frequency component of the position information measured in the step and the high frequency component of the position information obtained by integrating the acceleration information measured in the step twice to obtain the position information of the structure When,
Subtracting the position information of the structure obtained by the addition from the information of the target position of the structure to generate difference information;
And a step of driving the damper based on the difference information generated in the step. In the vibration isolation method for controlling the pressure of the gas in the gas damper that supports the structure on the installation surface with the gas supplied inside,
Measuring the position information of the structure;
Subtracting the position information of the structure measured in the step from the information on the target position of the structure to generate difference information;
A step of feeding forward the difference information generated in the step to the position information of the structure and controlling a pressure of the gas in the gas damper.

Images